176 PDF
176 PDF
176 PDF
O
WN
Robert L. Schuster
Raymond I. Krizek
editors
Notice
The project that is the subject of this report was approved by the
Governing Board of the National Research Council, whose members
are drawn from the councils of the National Academy of Sciences,
the National Academy of Engineering, and the Institute of Medicine.
The members of the committee responsible for the report were cho-
sen for their special competence and with regard for appropriate
balance.
This report has been reviewed by a group other than the authors
according to procedures approved by a Report Review Committee
consisting of members of the National Academy of Sciences, the
National Academy of Engineering, and the Institute of Medicine.
The views expressed in this report are those of the authors and do
not necessarily reflect the view of the committee, the Transportation
Research Board, the National Academy of Sciences, or the sponsors
of the project.
Iff
Subsurface Exploration, 91 Minneapolis Freeway, 133
Geologic Reconnaissance, 91 Potrero Tunnel Movements, 134
Boring, Sampling, and Logging, 94 Seattle Freeway, 134
Test Pits and Trenches, 95 Fort Benton Slide, 135
Geophysical Studies, 95 References, 137
Correlation Representation, 98
Surface Water and Groundwater, 99
Importance of Water, 99 Chapter 6: Tien H. Wu and Dwight A. Sangrey
Surface Water, 100 STRENGTH PROPERTIES AND THEIR
Groundwater, 100 MEASUREMENT
Groundwater Observations, 101 General Principles, 139
Permeability, 102 Mohr-Coulomb Failure Criterion, 139
Springs and Seeps, 102 Effective Stress Versus Total Stress Analysis, 139
Correlation, 102 Common States of Stress and Stress Change, 140
Environmental Factors, 103 Stress-Strain Characteristics, 140
Weather, 103 Effect of Rate of Loading, 140
Human Changes Before Construction, 103 Laboratory Measurement of Shear Strength, 141
Changes Brought by Construction, 104 Simple Tests, 141
Effect of Ecosystem on Sliding, 104 Triaxial Test, 141
Effect of Sliding on Ecosystem, 104 Plane-Strain Test, 143
Field Testing, 105 Direct Shear' Test, 143
Borehole Tests, 105 Simple Shear Test, 143
Large-Scale Pit Tests, 108 Shear Strength Properties of Some Common Soils, 144
Borehole Dynamics, 109 Cohesionless Soils, 144
Geophysical Tests, 109 - Soft Saturated Clays and Clayey Silts, 144
Correlation of Data, 110 Heavily Overconsolidated Clays, 145
Areal Variations, 1.10 Sensitive Soils, 148
Cross Sections, 110 'Partially Saturated Soils, 149
Time-Based Observations and Correlations, 110 Residual Soil and Colluvium, 149
Conclusions, 110 Rocks, 150
References, 111 Soil Behavior Under Repeated Loads, 151
Repeated Load Tests, 151
Chapter 5: Stanley D. Wilson and P. Erik Mikkelsen Stress-Strain Characteristics, 15 1
FIELD INSTRUMENTATION Failure Under Repeated Loading, 152
Instrumentation Planning, 112 References, 152
Types of Measurements Required, 113
Selection of Instrument Types, 113
Surface Surveying, 113 Chapter 7: N. R. Morgenstern and Dwight A. 'Sangrey
Conventional Surveying, 113 METHODS OF STABILITY ANALYSIS
Other Types of Surface Surveying, 117 Roles of Limit Equilibrium and Deformation
Crack Gauges, 118 Analyses, 155
Tiltmeters, 118 Limit Equilibrium Analysis;i 56
Inclinometers, 118 Total and Effective Stress Analyses, 157
Inclinometer Sensors, 121 Total Stress Analysis of Soil Slopes (0 = 0), 157
Casing Installation, 122 Effective Stress Analysis of Soil Slopes, 160
In-Place Inclinometers, 124 Pore-Pressure Distribution, 165
Extensometers and Strain Meters, 124 Analysis of Rock Slopes, 165
Pore-Pressure and Groundwater Measurements, 125 Deformation Analysis, 168
Observation Well, 125 References, 169
Piezometers, 125
Piezometer Sealing, 127 •.
Systems for Monitoring Rock Noise, 128 Chapter 8: David S. .Gedney and William G. Weber, Jr.
Sensors, 128 DESIGN AND CONSTRUCTION OF SOIL
Sign al-Con ditioning Equipment, 128 SLOPES
Recording and Data-Acquisition Equipment, 129 Philosophy of Design, 172
Automatic Warning and Alarm Systems, 129 Safety Factor, 172
Data Acquisition and Evaluation, 129 Design Procedures, 173
Data-Acquisition Methods, 129 Avoid Problem, 173
Inclinometer Observations, 130 Reduce Driving Forces, 175
Inclinometer Data Reduction, 131 Increase Resisting Forces, 183
Evaluation and Interpretation, 132 Toe Erosion, 190
Examples, 133. References, 190
iv
Chapter 9: Douglas R. Piteau and F. Lionel Peckover Determination of Structural Domains and Design
ENGINEERING OF ROCK SLOPES Sectors, 198
Significant Factors in Design of Rock Slopes, 193 Development of a Rock Mass Model Depicting
Structural Discontinuities, 193 Geologic Structure, 198
Groundwater, 194 Determination of Kinematically Possible Failure
Lithology, Weathering, and Alteration, 194 Modes and Performance of Slope-Stability
Climatic Conditions, 195 Analysis, 199
Slope Geometry in Plan and Section, 196 Slope Design and Remedial Measures, 201
Time Factor and Progressive Failure, 196 Planning and RelatedProcedures, 201
Residual and Induced Stress, 196 Methods of Stabilization, 205
Existing Natural and Excavated Slopes, 197 Methods of Protection, 218
Dynamic Forces, 197 Methods of Warning, 223
Fundamental Procedures in Analysis of Rock Slopes, 197 References, 225
Determination of Structural and Other Relevant
Geologic Characteristics, 197 INDEX, 229
v
The'Ed.itors.
and Authors
David S. Gedney. Acting director, Northeast Corridor jects of landslides, rock mechanics, soil-structure interac-
Project, Federal Railroad Administration. Formerly chief, tion, and soil engineering problems in cold regions.
Construction and Maintenance Division, Federal Highway
Administration; regional engineer, Region 15, Federal F. Lionel Peckover. Geotechnical consultant, Vaudreuil,
Highway Administration; and chief, Soil and Rock Me- Quebec.. Formerly in charge of geotechnical engineering
chanics Branch, Construction and Maintenance Division, work with Canadian National Railways (1959-1976), Cana-
Federal Highway Administration. His field of specializa- than St. Lawrence Seaway Authority (1953.1959), and Di-
tion is geotechnical engineering as applied to transporta- vision of Building. Research, National Research Council of
tion problems. Canada (1947-1953). His field of specialization is geotech-
nical engineering with emphasis on railway application,
Raymond J. Krizek. Professor of civil engineering, North- particularly the treatment of unstable rock slopes.
western University, Evanston, Illinois. Formerly faculty
member at the University of Maryland and lecturer at the Douglas R. Piteau. President, Piteau and Associates, Van-
Catholic University of America. He is a specialist in soil couver, British Columbia.. He has extensive experience in
mechanics and foundation engineering and has a variety of engineering geology and rock mechanics related to railways
teaching, research, and professional engineering experience. and highways, mine development and operations, and site
evaluation ofdams and tunnels.
Ta Liang. Professor of civil and environmental engineering
and director, Remote Sensing Program, Cornell University, Harold T. Rib. Chief, Aerial Surveys Branch, Federal High-
Ithaca, New York. He has long been engaged in teaching way Administration. Formerly chief, Exploratory Tech-
and research in the fields of aerial photograph interpreta- niques Group, and highway research engineer, Federal
tion, remote sensing, and geotechnical and geological en- Highway Administration. His field of specialization is ap-
gineering and has been a consultant to industrial, national, plication of remote-sensing techniques to transportation.
and international organizations on engineering and economic engineering and terrain analysis.
development projects in many parts of the world.
David L. Royster. Chief, Soils and Geological Engineering
P. Erik Mikkelsen. Senior associate engineer, Shannon and Division, Tennessee Department of Transportation, Nash-
Wilson, Inc., Seattle. During his 15 years of geotechnical ville. He has held various positions with Tennessee Depart-
engineering experience, he has become an expert in instru- ment of Transportation since 1958. His field of specializa-
mentation used in landslides, embankments, deep excava- tion is engineering geology as applied to highway design
tions, and deep foundations. and construction.
N. R. Morgenstem. Professor of civil engineering, University Dwight A. Sangrey. Professor of civil and environmental
of Alberta, Edmonton. He has been a consultant in geo- engineering, Cornell University, Ithaca, New York. In ad-
technical engineering and applied earth sciences since 1961 dition to teaching and research at Cornell, he has been ac-
His area of professional specialization is geotechnical engi- tive in practice and as a consultant on projects involving
neering; he has made noteworthy contributions on the sub- instability of slopes. Major areas of professional activity
vi
involve dynamic loading of soils, soil sensitivity, and marine William G. Weber, Jr. Washington Department of Trans-
geotechnical engineering. portation, Yakima. Formerly research engineer, Bureau
of Materials, Pennsylvania Department of Transportation,
Robert L. Schuster. Chief, Engineering Geology Branch, and senior materials and research engineer, Materials and
U.S. Geological Survey, Denver. Formerly professor and Research Department, California Division of Highways. He
chairman, Department of Civil Engineering, University of has expertise in geotechnical design of structures involving
Idaho, and associate professor and professor, Department soft soils, groundwater flow, and stability of slopes.
of Civil Engineering, University of Colorado. Major areas
of professional specialization are engineering geology and
geotechnical engineering. Stanley D. Wilson. Consulting engineer. Seattle. Formerly
executive vice president of Shannon and Wilson, Inc. His
George F. Sowers. Senior geotechnical consultant and significant contributions in the development of field opera-
senior vice president, Law Engineering Testing Company, tional equipment, his advocacy of its use in dams and other
Marietta, Georgia, and regents professor of civil engineer- civil engineering projects, and his analyses of the results
ing, Georgia Institute of Technology, Atlanta. He is a geo- have led to better understanding of the mechanisms of land-
technical engineer specializing in the interrelation of geol- slides and of the performance of embankments and dams.
ogy, engineering design, and construction and has been a
worldwide consultant on earthfill dams, foundations, and
earth and rock construction. Tien H. Wu. Professor of civil engineering, Ohio State Uni-
versity, Columbus. Formerly professor, Department of
David J. Varnes. Geologist, Engineering Geology Branch, Civil Engineering, Michigan State University, and visiting
U.S. Geological Survey, Denver. He has had 37 years of professor, Norwegian Geotechnical Institute and National
experience as an engineering geologist with USGS, includ- University of Mexico. He has specialized in the relation of
ing being chief of the Engineering Geology Branch. Fields the engineering properties of soils to their mechanical behav-
of specialization are slope stability in rock and soil and en- ior and has been a consultant on a variety of problems in
gineering geologic mapping. soil mechanics and geotechnical engineering.
vi'
Chapter 1
Introdu, ction
Robert L. Schuster
This book is a successor to Highway Research Board Special SCOPE OF THIS VOLUME
Report 29, Landslides and EngineeringPractice (1.8). Spe-
cial Report 29, which was written by the Highway Research The scope of this volume is the same as that for Special
Board Committee on Landslide Investigations and published Report 29: to bring together in coherent form and from
in 1958, achieved an excellent reputation, both in North a wide range of experience such information as may be
America and abroad, as a text on landslides. Because of its useful to those who must recognize, avoid, control, design
popularity, the original printing was sold out within a few for, or correct landslide movement.
years after publication. Since then, there has been a con- This new version, however, introduces geologic concepts
tinuing interest in reissuing the original text or publishing and engineering principles and techniques that have been
a worthwhile successor. developed since publication of Special Report 29 so that
In 1972 the Highway Research Board organized the Task both the analysis and the control of soil and rock slopes
Force for Review of Special Report 29—Landslides. The are addressed. For example, included are new methods of
membership of this task force was selected from several stability analysis and the use of computer techniques in im-
committees within the HRB Soils and Geology Group: its plementing these methods. In addition, rock-slope engineer-
charge was ing and the selection of shear-strength parameters for slope-
stability analyses are two topics that were poorly under-
stood in 1958 and therefore were given scant attention in
To review the out-of-print Special Report 29—Landslides—
Special Report 29. Since that time,these two subjects have
and to recommend what action should be taken in response
received a significant amount of study and have become
to the high interest in revising this publication.
fairly well understood;thus, they are presented as separate
chapters in the present volume.
This task force was further instructed to act as the coordi- The book is divided into two general parts. The first part
nating unit to implement its recommendations. deals principally with the definition and assessment of the
After considerable study of the original report, the task landslide problem. It includes chapters on slope-movement
force concluded that, because of the large amount of new types and processes, recognition and identification of land-
technical information that had become available since 1958 slides, field investigations, instrumentation, and evaluation
on landslides and related engineering, the best course of ac- of strength properties. The second part of the book deals
tion would be to completely rewrite the book rather than with solutions to the landslide problem. Chapters are in-
to reprint it or to revise it in part. The task force decided cluded on methods of slope-stability analysis, design tech-
that the general format of the original volume would be re- niques, and remedial measures that can be applied to both
tained but that the contents would be expanded to include soil and rock-slope problems.
concepts and methods not available in 1958. To achieve Although considerable effort has been made to eliminate
that objective, the task force secured the aid of authors presentation of the same material in different chapters,
who have broad geotechnical expertise. They were drawn some repetition has been necessary to provide continuity
from the fields of civil engineering and geology and have of thought and to allow adequate explanation of specific
specializations in soil mechanics, engineering geology, and topics. We consider this repetition to be more acceptable
interpretation of aerial photographs. than constant referral in the text to other chapters.
DEFINITIONS AND RESTRICTIONS this volume will be found useful to any engineer whose
practice leads him to deal with landslides.
In Special Report 29 the term landslide is defined as the
downward and outward movement of slope-forming ma- ECONOMICS OF SLOPE
terials—natural rock, soils, artificial fills, or combinations MOVEMENTS
of these materials. Today the term deserves further refine-
ment because, as shown in Chapter 2, slope movements can Although individual slope failures generally are not so
now be divided into five groups: falls, topples, slides, spectacular or so costly as certain other natural catas-
spreads, and flows. As used in this text, a landslide con- trophes such as earthquakes, major floods, and tornadoes,
stitutes the group of slope movements wherein shear fail- they are more widespread and the total financial loss due
ure occurs along a specific surface or combination of sur- to slope failures probably is greater than that for any other
faces.
Although this volume deals primarily with slope failures
belonging to the group designated as slides, some attention Figure 1.1. Damage to embankment on 1-75 in Campbell County,
Tennessee, from a landslide that occurred April 1972.
is given to the other four types of slope movements. The
use of the term landslide in the title is somewhat inaccurate
in that theoretically it does not cover the five basic failure
modes described above; however, the decision was made to
use this term because it is popular and easily recognized
and because the book is mainly devoted to landslides.
In keeping with the practice followed in Special Report
29, surficial creep was excluded from consideration: how-
ever, creep of a more deep-seated nature is considered in
discussions dealing with slope movements. Also excluded
are subsidence not occurring on slopes and most types of
movement primarily due to freezing and thawing of water.
In addition, snow and ice avalanches and mass wasting due
to slope-failure phenomena in tropic and arctic climates
are not considered. Although a few examples are drawn
from other parts of the world, most of the descriptions of
slope movements and engineering techniques involve slopes
in North America.
Of the five groups of potential slope movements con-
sidered, only slides are currently susceptible to quantitative
stability analysis by use of the conventional sliding-wedge
or circular-arc techniques. These methods of slope-stability
analysis are not applicable to falls, topples, spreads, or flows.
However, enough is now known about the kinematics and
nature of development of such failures that qualitative or
statistical approaches or both can be used to make rea-
sonable assessments in problem areas or potential prob-
1cm areas. Research dealing with such problems is cur-
rently being undertaken to enable at least crude quantita-
tive stability analyses to be performed on slopes subject
to spreads and flows and possibly even to falls and top- Figure 1.2. Homes and Street damaged in October 1978 Laguna
ples. Beach, California, landslide.
Although slope-stability problems related to transpor-
tation facilities are stressed, most of the examples apply
equally well to all cases of slope failure, such as those re-
lating to coastlines, mining, housing developments, and
farmlands. As noted by Eckel in the Introduction in Spe-
cial Report 29(1.7, p. 2):
single geologic hazard to mankind. In addition, much of landslides. Indirect costs of landslides are difficult to eval-
the damage occurring in conjunction with earthquakes and uate, but they may be larger than the direct costs.
floods is due to landslides instigated by shaking or water. In 1976, Krohn and Slosson (1.16) estimated the annual
Reliable estimates of the overall costs of landslides are landslide damage to buildings and their sites in the United
difficult to obtain for geographic entities as large as the States to be S400 million (1971 dollars). This figure does
United States or Canada. In 1958, Smith (1.32) stated that not include other damages, such as those to transportation
"the average yearly cost of landslides in the United States facilities and mines, or indirect costs. In the same year,
runs to hundreds of millions of dollars," an estimate that Jones (1.13) estimated the direct landslide damage losses
was probably realistic at that time. However, in the 20 to buildings and their sites to be about S500 million an-
years since Smith assembled his data, a combination of in- nually. Based on the above estimates plus indirect costs
flation, increased construction in landslide-prone areas, and estimated damages to facilities not classed as buildings,
and use of larger cuts and fills in construction has resulted a.reasonable estimate of present-day direct and indirect
in considerably increased annual costs of landslides. For costs of slope failures in the United States exceeds $1 bil-
example, environmental and political considerations and lion/year.
right-of-way costs control the selection of highway rout- Somewhat more accurate cost estimates can be made
ing today to a much greater degree than was the case 20 for individual landslides or for landslides occurring in rela-
years ago; thus, highway planners often cannot avoid con- tively small geographic areas. For instance, the Portuguese
struction in landslide-prone areas. Landslide costs include Bend landslide in Palos Verdes Hills, California, has been
both direct and indirect losses from landslides affecting estimated to have cost more than $10 million in damage to
highways, railroads, industrial installations, mines, homes, roads, houses, and other structures between 1956 and 1959
and other public and private properties. Direct costs are (1.23). Jones, Embody, and Peterson (1.14) noted that the
those losses incurred in actual damages to installations or filling of the reservoir behind Grand Coulee Dam in the
property: examples of such damages are shown in Figures state of Washington cost taxpayers and private property
1.1, 1 .2, and 1 3. Examples of indirect costs are (a) loss owners at least $20 million to avoid and correct the damage
of tax revenues on properties devalued as a result of land- due to landslides that occurred between 1934 and 1952.
slides, (b) reduced real estate values in areas threatened by Within the United States, greater effort at detailing the
landslides, (c) loss of productivity of agricultural or forest costs of slope movements has been expended in California
lands affected by landslides, and (d) loss of industrial pro- than in any other state. In a classic study of slope-movement
ductivity due to interruption of transportation systems by costs in the San Francisco Bay area,Taylor and Brabb (1.35)
documented information on these costs for nine Bay-area Federal Highway Administration believe that $100 mil-
counties during the winter of 1968-1969. The data were lion is a conservative estimate of the total annual cost of
derived largely from interviews with planners and assessors landslide damage to highways and roads in the United
in the county government and engineers and geologists in States.
city, county, and state governments. Costs of slope move- For planning purposes, other studies have attempted
ments totaled at least $25 million, of which about $9 mil- to project costs of slope movements. In a study predict-
lion was direct loss or damage to private property (due ing the cost of geologic hazards in California from 1970
mainly to drop in market value); $10 million was direct to 2000, the California Division of Mines and Geology
loss or damage to public property (chiefly for repair or re- (1.1) estimated that the costs of slope movements
location of roads and utilities); and about $6 million con- throughout the state during that period would be nearly
sisted of miscellaneous costs that could not be easily clas- $10 billion, or an average of more than $300 million a
sified in either the public or the private sector. This is a year. This estimate is based on the assumption that loss-
tremendous expense for the relatively small area involved. reduction practices in use in California in 1970 for slope
In addition, Taylor and Brabb noted that their data are in- failures will remain unchanged. Figure 1.5 (1.1) shows
complete in that they were not able to obtain costs on a comparison of the estimated losses due to slope move-
many of the slope movements. They felt, therefore, that ments and losses due to other geologic hazards and ur-
the total cost of the 1968-1969 slope movements for the banization. Of the so-called "catastrophic" geologic
San Francisco Bay area may possibly have been several hazards included in the study, losses due to slope move-
times greater than the estimated $25 million. ments exceed those due to floods and, in turn, are ex-
A survey conducted by the Federal Highway Administra- ceeded by those due to earthquakes. California, how-
tion indicates that approximately $50 million is spent annually ever, is particularly prone to earthquake activity, and in
to repair major landslides on the federally financed portion most other parts of the United States and Canada losses
of the national highway system (1.3, 1.4). This system in- due to slope movements probably would be greater than
cludes federal and state highways but does not include most those due to earthquakes.
county and city roads or streets, private roads and streets, or Various studies have shown that most damaging land-
roads built by other governmental agencies such as the U.S. slides are human related; thus, the degree of hazard can
Forest Service. Distribution of the direct costs of major be reduced beforehand by introduction of measures such
landslides for 1973 by Federal Highway Adthinistration re- as improved grading ordinances, land-use controls, and
gions within the United States is shown in Figure 1.4 (1.3). drainage or runoff controls (1.37). For example, Nilsen
The cost for an individual region is based on both the land- and Turner (1.25) showed that in Contra Costa County,
slide risk and the amount of highway construction in the California, approximately 80 percent of the landslides have
area. In addition, the given costs represent a single year; been caused by human activity. Briggs, Pomeroy, and Davies
the average annual cost for a particular region could vary (1.2) noted that more than 90 percent of the landslides
significantly from the given cost. in Allegheny County, Pennsylvania, have been related to
Total annual costs of landslides to highways in the human activities. The study by the California Division of
United States are difficult to determine precisely because Mines and Geology (1.1) indicated that the $9.9 billion es-
of the difficulty in defining the following factors: (a) timated losses due to slope movements can be reduced 90
costs of smaller slides that are routinely handled by main- percent or more by a combination of measures involving
tenance forces; (b) costs of slides on non-federal-aid adequate geologic investigations, good engineering practice,
routes; and (c) indirect costs that are related to landslide and effective enforcement of legal restraints on land use
damage, such as traffic disruption and delays, inconve- and disturbance.
nience to motorists, engineering costs for investigation, Chassie and Goughnour (1.4) further substantiated the
and analysis and design of mitigation measures. If these concept that improved geologic and geotechnical studies can
factors are included, Chassie and Goughnour (1.3) of the significantly reduce the landslide hazard. They noted that
REGION 9 REGION 7 5 I
$1 MILLION ( $4 MILLION / $6
'\ J' MILLION
$3.5 MILLION
REGION 6 / REGION 4
$7 MILLION j' $12 MILLION
4
Figure 1.5. Predicted economic losses from geologic a debris avalanche, which started as an ice avalanche from
hazards and urbanization in California from 1970 to a glacier high on the north peak of Mt. Huascaran but soon
2000(1.1). became a mixture of ice, water, rock, and soil, roared
through valley villages, killing some 4000 to 5000 people
(1.5, 1.22). An even greater number of people were killed
in a repeat of this tragedy 8 years later,when an earthquake
LOSS OF of magnitude 7.75 occurred off the coast of Peru and
MINERAL RESOURCES
DUE TO URBANIZATION triggered another disastrous debris avalanche on the slopes
$17 BILLION of Huascarân (1.5, 1.28). This debris avalanche descended
at average speeds of roughly 320 km/h (200 mph) into the
same valley but over a much larger area and killed more
than 18 000 people. The village of Ranrahirca, which had
SLOPE MOVEMENTS
$9.9 BILLION
been rebuilt after being destroyed by the 1962 debris
avalanche in which 2700 of its people were killed, was par-
EARTHQUAKE SHAKING
$21 BILLION
tially destroyed by the 1970 avalanche. The 1962 avalanche
was prevented from flowing into the town of Yungay by a
FLOODING protective ridge, but the 1970 avalanche overtopped this
$6.5 BILLION ridge and buried the town along with an estimated 15 000
of its 17 000 inhabitants. Only the tops of a few palm
trees in the central plaza and parts of the walls of the main
cathedral were left protruding above the mud to mark the
site of this formerly prosperous and picturesque city (1.28).
EROSION $600 MILLION
In 1974, another massive landslide in the Andes Moun.
EXPANSIVE SOIL $150 MILLION tains of Peru killed approximately 450 people (1.18). This
FAULT DISPLACEMENT $76 MILLION landslide, which occurred in the valley of the Mantaro
VOLCANIC ERUPTION $49 MILLION
TSUNAMI $41 MILLION River, had a volume of 1.6 Gm3 (2.1 billion yd3), making
SUBSIDENCE $26 MILLION it one of the largest in recorded history. It temporarily
damrñed the Mantaro River, forming a lake with a depth
of about 170 m (560 ft) and a length of about 31 km (19
improved geotechnical techniques in New York State re- niiles). In overtopping this landslide dam, the river caused
duced landslide repair costs by as much as 90 percent in the extensive damage downstream, destroying approximately
7 years prior to 1976. Slosson (1.31) showed. that landslide 20 km (12 miles) of road, three bridges, and many farms.
losses sustained by the city of Los Angeles as a result of the On October 9, 1963, the most disastrous landslide in
1968-1969 winter storm were 97 percent lower for those European history—the Vaiont Reservoir slide—occurred
sites developed under modern grading codes by using mod- in northeastern Italy. A mass of rock and soil having a
ern geotechnical methods than for sites developed before volume of about 250 Mm3 (330 mfflion yd3) slid into the
1952, when no grading codes existed and engineering geol- reservoir, sending a wave 260 m (850 ft) up the opposite
ogy and geotechnical engineering studies were not required. slope and at least 100 m (330 ft) over the crest of the dam
For the state of California, Leighton (1.19) estimated that into the valley below, where it destroyed five villages and
reductions of 95 to 99 percent in landslide losses can be took 2000 to 3000 lives (1.15, 1.17).
obtained by means of preventive measures that incorporate Japan has also suffered continuing large loss of life and
thorough preconstruction investigation, analysis, and de- property from landslides and other slope movements. Al-
sign and that are followed by careful construction proce- though some slope failures in Japan have been triggered
dures. by earthquakes, most are a direct result of heavy rains
In addition to the economic losses due to slope move- during the typhoon season. Data from a Japan Ministry
ments, a significant loss of human life is directly attribut- of Construction publication (1.12) and a written commu-
able to landslides and other types of slope failures. Fa- nication in 1974 are given in Table 1.1 and show the num-
talities due to catastrophic slope failures have been re- ber of deaths and damaged houses caused by slope failures
corded since people began to congregate in areas subject in Japan for the 4-year period from 1969 through 1972.
to such failures. One such catastrophe (probably a debris
flow) was noted by Spanish conquistadors in Bolivia in
Table 1.1. Deaths and damage due to recent
the.sixteenth century (1.30). According to the priest Padre slope-failure disasters in Japan.
Calancha,who observed the event from a distance,Hanco-
Hanco, a community of about 2000 inhabitants, disap-
Deaths
peared "in a few minutes and was swallowed by the earth
Year Houses Damaged Number Percenta
without more evidence of its former existence than a cloud
of dust which arose where the village had been situated." 1969 521 82 50
In the twentieth century many individual slope failures 1970 38 27 26
1971 5205 171 54
have resulted in large numbers of fatalities. Probably the
1972 1564 239 44
best known of these catastrophic failures are the debris
avalanches of 1962 and 1970 on the slopes of Mt. Huas- a0f deaths due to slope failure in relation to deaths
caran in the Andes Mountains of Peru. In January 1962, due to all other natural disasters.
Of particular interest is the high ratio of deaths due to mentioned above can all be classified as major disasters.
slope failures to deaths from all other natural disasters, In addition to catastrophes of this magnitude, however,
including earthquakes. slope failures of lesser importance occur continually
North American slope failures have not commonly re- throughout the world. Because no systematic records of
suited in major losses of life, because most catastrophic these day-to-day slope failures have been maintained in
slope failures have occurred in nonpopulated areas. the United States and Canada (as contrasted to Japan,
However, there have been several notable exceptions in Table 1.1), ascertaining the number of deaths per year
this century. The first was in Canada in 1903, when a owing to slope failures is not possible. However, Krohn
great landslide killed approximately 70 people in the and Slosson (1.16) estimated that the total loss of life in
coal mining town of Frank, Alberta (1.21). More re- the United States from all forms of landslide activity ex-
cently, the Hebgen Lake earthquake struck southwestern ceeds approximately 25 lives per year, a greater total
Montana in 1959 and triggered the Madison Canyon land- than the average number of deaths due to earthquakes.
slide. That catastrophic landslide, shown in Figure 1.6
(1.36), had a volume of 28 Mm3 (37 million yd3) and LEGAL ASPECTS OF SLOPE
buried 26 people who were camped along the banks of MOVEMENTS
the Madison River (1.10, 1.39).
Probably the worst natural disaster in central Vir- In Special Report 29, Smith (1.32, p. 13) stated:
ginia's recorded history was the 1969 flooding and as-
sociated debris flows resulting from hurricane Camille Few legal precedents have been established to guide the
(1.38). Although no exact number of deaths due to courts in determining responsibility for landslides or in
slope movements can be ascertained, estimates are that assessing the damages caused by them. This dearth of
a substantial percentage of the 150 people who died in specific laws and legal decisions is perhaps due to two
Virginia as a result of hurricane Camille were victims of main factors—many, if not most, cases that involve pri-
debris flows resulting from the hurricane. vate companies are settled out of court; most cases
Another recent catastrophic slope failure in North against state or federal agencies are settled out of court
America was the debris flow that occurred in 1971 in or the public agency exercises its sovereign right of re-
Champlain clay in the Canadian town of Saint-Jean- fusal to consent to be sued.
Vianney, Quebec (1.34). That flow carried 40 homes
to destruction and 31 persons to their deaths. In the United States during the 20 years since this state-
The most recent major catastrophe involving slope ment was made, the number of legal cases resulting from
failure in the United States was the Buffalo Creek dam property damage due to landslides has been ever increas-
failure at Saunders, West Virginia, in 1972 (1.6). Heavy ing; the cases involve private companies and landholders
rains led to the failure of three coal-refuse impoundments, as well as public agencies. For that reason it is important
The resulting debris flow consisting of released water, coal that those who undertake activities that involve the use
wastes, and sludge traveled 24 km (15 miles) downstream, of slopes have an understanding of the legal implications
killing 125 people and leaving 4000 homeless. of that use. This section deals briefly with the legal as-
The landslides and other types of slope movements pects of landslides and provides references for those wish-
ing to explore the subject further. The discussion is based
on perusal of current literature and on substantial informa-
Figure 1.6. Madison Canyon landslide of August 21, 1959, in tion provided in a written communication in 1975 from
southwestern Montana.
C. L. Love, attorney for the Legal Division of the Califor-
nia Department of Transportation. Because of the con-
stantly changing status of litigation involving natural
hazards, some of these concepts will likely change within
the next few years.
A property owner's rights in response to invasion Liability for Injuries Sustained From
of the property by sliding material or interference with a Landslide
the lateral support of the property by construction or
maintenance of a public way and Love states that, although the courts have made it clear that
A highway traveler's rights in tort against a public a public entity is not an insurer of the safety of persons
entity for injuries sustained from a landslide that resulted using its highways, in certain circumstances travelers are
protected by law from landslides. In general, the public geologist, the grading contractor, the city, the builder,
entity will not be held liable for injuries if it can be shown the lending agency, the insurance company that insured
that the acts or omissions that created the dangerous con- the home, the former owner, and the real estate agent
dition were reasonable or that the action taken to protect handling the sale if the property was, not purchased di-
against such injuries or the failure to take such action was rectly from the developer (1.11). A typical complaint
reasonable. The reasonableness of action or inaction is de- may seek recovery on theories of strict liability (i.e.,
termined by considering the time and opportunity that the liability without fault), negligence, breach of warranty,
public employees had to take action and by weighing the negligent misrepresentation, fraud, and, if a public en-
probability and gravity of potential injury to persons fore- tity is involved, inverse condemnation (1.27). In most
seeably exposed to the risk of injury against the practical- cases, the developer is the prime target because he is sub-
ity and cost of protecting against such injury. ject to strict liability for "defects" in the construction of
In California, most actions involving personal injury to the house or grading of the lot; to establish strict liability
highway travelers as a result of landslides are based on a stat- against any of the other parties such as the soils engineer
ute that imposes liability for the dangerous condition of or the geologist is much more difficult.
public property. The injured person must prove that the
public property was in dangerous condition at the time of Liability of Engineers and Geologists
the injury, that the injury resulted from that dangerous con- for Landslide Damages
dition, and that the dangerous condition created a foresee-
able risk of the kind of injury incurred. Love points out that, There has been a certain amount of variability in legal inter-
in addition, the dangerous condition must be the result of pretations of liability of geotechnical engineers and geolo-
negligence, a wrongful act, or failure of an employee of gists in regard to landslide losses; such liability is discussed
the public entity to act within the scope of his or her em- below, but the conclusions reached are general in nature
ployment, and the public entity must have had notice of and not necessarily valid in any specific court.
the dangerous condition in sufficient time prior to the in- Liability of geotechnical engineers and engineering
jury to have taken measures to protect against it. Thus, geologists for landslide damages to home sites is based
liability depends on whether circumstances and conditions most often on the theory of negligence and occasionally
were such that the danger was reasonably foreseeable in the on negligent misrepresentation. Although allegations
exercise of ordinary care and, if so, on whether reasonable seeking to recover damages from geotechnical engineers
measures were taken by the public entity to prevent injury and geologists on the basis of strict liability, breach of war-
(39 Am. Jur. 2dHighways § 532, p. 939). ranty, or intentional misrepresentation are often included
A public entity, since it is not an insurer of the safety of in a complaint, they are not usually applicable under nor-
travelers on its highways, need only maintain highways in mal circumstances. Patton (1.27) discussed each of these
reasonably safe condition for ordinary travel under ordinary theories of liability as it applies to engineering geologists
conditions or under such conditions as should reasonably be as follows, and it is felt that Patton's line of reasoning can
expected (1.20). In the case of Boskovich v. King County be extended to include geotechnical engineers involved in
[188 Wash. 63,61 P. 2d 1299 (1936)] ,the court held that development of private property.
a motorist was not entitled to recover from the highway de-
1. Negligence
partment for injuries sustained when a landslide broke loose
from a steep hillside bordering a highway and struck his auto- The most common theory of liability alleged against engi-
mobile because there was no proof that negligence in con- neering geologists is negligence. Negligence is the omission
struction or maintenance of the highway was the cause of to do something which an ordinarily prudent person would
the landslide. have done under similar circumstances or the doing of some-
thing which an ordinarily prudent person would not have
Landslides and Property Development done under those circumstances. An engineering geologist
is required to exercise that degree of care and skill ordi-
This section discusses liability related to damages from narily exercised in like cases by reputable members of his
landslides caused by the development of private property. profession practicing in the same or similar locality at the
Detailed information on litigation related to landslides in same time under similar conditions. He has the duty to
property developments is given by Sutter and Hecht (1.33). exercise ordinary care in the course of performing his
The current trend in public policy is toward protection duties for the protection of any person who foreseeably
of the consumer, a reversal of the days when caveat emptor and with reasonable certainty may be injured by his fail-
(let the buyer beware) reflected public policy (1.27). This ure to do so.
trend has extended to home purchasers since they are pro- Although failing to comply with a state statute or with
tected by law against losses due to improper workmanship county or municipal ordinances normally is considered to
or poor planning, including certain losses due to landslides. be negligence per se, the mere compliance with the letter
The trend toward increased protection for the home- of the law in such cases does not necessarily relieve one of
owner has resulted in a drastic increase in the number of liability since it generally is recognized that statutes and
legal cases involving landslides on private property. After ordinances set forth only minimum requirements and cir-
consulting with an attorney, the owner of a home that cumstances may require more than the minimum. An en-
has suffered damage from a landslide typically files legal gineering geologist cannot rely upon the approval of a
action against the developer, the civil engineer who laid project by an inspector for a governmental agency to re-
out the development, the geotechnical engineer, the lieve him of liability.
8
Negligent misrepresentation gineers, stated in 1973 that professional liability of the
technical professional was approaching a major crossroad
Negligent misrepresentation is a species of fraud along with in its development (1.9). Fife felt that the scope of pro-
intentional misrepresentation and concealment. Negligent fessional liability in the technical disciplines was at the
misrepresentation is simply the assertion, as a fact, of that point where it would proceed either toward strict liability
which is not true by one who has no reasonable ground for under pressure from skilled plaintiffs counsel or toward a
believing it to be true. Although misrepresentations of "reasonableness standard" by which adherence to the aver-
opinions generally are not actionable, they become ac- age standards of the profession involved would constitute
tionable where the person making the alleged misrepresen- a complete defense. However, unless professional groups
tation holds himself to be specially qualified to render the become more actively involved in the process of shaping
opinion. A statement of opinion by an engineering geol- the future scope of their professional liability, eventual
ogist that no unsupported bedding occurs in a particular application of strict liability to technical professionals
slope, could be actionable as negligent misrepresentation seems inevitable.
if he has no basis for that opinion. In their book, Landslide and Subsidence Liability
(1.33), Sutter and Hecht present considerable information
Intentional misrepresentation and on strict liability in California. Since November 1913,the
concealment cutoff date for the cases included in this reference, certain
California court decisions have changed the liability of geol-
Intentional misrepresentation (the assertion, as a fact, of ogists and geotechnical engineers from strict liability to lia-
that which is not true by one who does not believe it to be bility for negligence only. The publisher, California Con-
true) and concealment (the suppression of a fact or condi- tunuing Education at the Bar, plans periodic supplements
tion by one who is bound to disclose it) are species of fraud to cover changes that have occurred since Sutter and Hecht's
which are seldom if ever applicable to engineering geologists. book was published.
Such conduct on the part of an engineering geologist is not
only legally actionable but raises serious doubts about the ACKNOWLEDGMENTS
professional integrity of the geologist involved.
Sincere appreciation is expressed to all those who contrib-
Breach of warranty uted information, both published and unpublished, with-
out which this volume could not have been written. To
Breach of warranty is generally not available as a viable theory list all contributors would be impossible; to list the most
of recovery against an engineering geologist. important would be unfair to the others. Thus, the most
reasonable and equitable approach is simply to acknowledge
Strict liability the help and cooperation of all who contributed informa-
tion in the form of data, photographs, ideas, and advice.
Although the theory of strict liability is still developing and Appreciation is also due the entire staff of the Transpor-
its limits are not as yet clearly defined, it would appear now tation Research Board. Our special thanks go to John W.
that an engineering geologist would not be liable on the Guinnee, engineer of soils, geology, and foundations, for
theory of strict liability lacking some participation as the his constant encouragement and advice, and to Mildred Clark,
developer of mass produced property. Other cases indi- senior editor, who aided immensely in the technical aspects
cate that the theory is available only against the developers of the editoral process.
of mass-produced property and would not be available REFERENCES
against the developer of a single lot or building site. There
is no clear indication as to at what point between develop- 1.1 Alfors, J. T., Burnett, J. L., and Gay, T. E., Jr. Urban
ment of a single lot and development of a tract the theory Geology: Master Plan for California. California Division
becomes applicable. It does seem clear at this time that if of Mines and Geology, Bulletin 198, 1973, 112 pp.
1.2 Briggs, R. P., Pomeroy, J. S., and Davies, W. E. Landslid-
an engineering geologist offers only professional services ing in Allegheny County, Pennsylvania. U.S. Geological
in connection with the development of even a large tract Survey, Circular 728, 1975, 18 pp.
he will not subject himself to strict liability. 1.3 Chassie, R. G., and Goughnour, R. D. National Highway
Landslide Experience. Highway Focus, Vol. 8, No. 1,
Jan. 1976, pp. 1-9.
In regard to negligence, Sowers noted:
1.4 Chassie, R. G., and Goughnour, R. D. States Intensifying
Efforts to Reduce Highway Landslides. Civil Engineer-
Unfortunately, the legal profession has been expanding the ing, Vol.46, No. 4, April 1976, pp. 65-66.
definition of negligence to any act committed by the public 1.5 duff, L. S. Peru Earthquake of May 31, 1970: Engineer-
official, engineer or contractor. Some courts have applied ing Geology Observations. Seismological Society of Amer-
ica Bulletin, Vol. 61, No. 3, June 1971, pp. 511-521.
the most extravagant standards of professional knowledge 1.6 Davies, W. E. Buffalo Creek Dam Disaster: Why it Hap-
to average run-of-the-mill design and construction. In other pened. Civil Engineering, Vol.43, No. 7, July 1973,
words, some courts would presume that every engineer must pp. 69-72.
possess the wisdom and expertise of a Terzaghi. 1.7 Eckel, E. B. Introduction. In Landslides and Engineer-
ing Practice, Highway Research Board, Special Rept. 29,
1958, pp. 1-5.
In voicing an opinion somewhat different from Patton's 1.8 Eckel, E. B., ed. Landslides and Engineering Practice.
in regard to strict liability, Fife, another California attorney Highway Research Board, Special Rept. 29, 1958,
active in litigation involving geologists and geotechnical en- 232 pp.
1.9 Fife, P. K. Professional Liability and the Public Inter- 1.26 Nordin, J. G. The Portuguese Bend Landslide, Palos
est. In Geology, Seismicity, and Environmental Impact, Verdes Hills, Los Angeles County, California.. In Land-
Association of Engineering Geologists, Special PubI., slides and Subsidence, Resources Agency of California,
1973, pp. 9-14. 1965, pp. 56-62.
1.10 Hadley, J. B: Landslides and Related Phenomena Ac- 1.27 Patton, J. H., Jr. The Engineering Geologist and Pro-
companying the Hebgen Lake Earthquake of August 17, fessional Liability. In Geology, Seismicity, and En-
1959. U.S. Geological Survey, Professional Paper 435, vironmental Impact, Association of Engineering Geol-
1964, pp. 107-138. ogists, 1973, pp. 5-8.
1.11 Hays, W. V. Panel discussion. Proc., Workshop on 1.28 Plafker, G., Ericksen, G E., and Fernandez Concha, J.
Physical Hazards and Land Use: A Search for Reason. Geological Aspects of the May 31, 1970, Peru Earth-
Department of Geology, San Jose State Univ., Calif or- quake. Seismological Society of America Bulletin, Vol.
nia, 1974, pp. 8-14. 61, No. , June 1971, pp. 543-578.
1.12 Japan Ministry of Construction. Dangerous Slope Fail- 1.29 Pollock, J. P. Discussion. In Landslides and Subsi-
ure. Division of Erosion Control, Department of River dence, Resources Agency of California, 1965, pp.
Works, 1972, 14 pp. 74-77.
1.13 Jones, D. E. Handout for Roundtable Discussions. 1.30 Sanjines, A. G. Sintesis Historica de la Vida de Ia
National Workshop on Natural Hazards, June 30-July Ciudad, 1548-1948. Primer Premio de Ia Alcaldia,
2, 1976, Univ. of Colorado, Boulder, Institute of Be- La Paz, Bolivia, 1948, 86 pp.
havioral Science, 1976. 1.31 Slosson, J. E. The Role of Engineering Geology in
1.14 Jones, F. 0., Embody, D. R., and Peterson, W. L. Land- Urban Planning. In The Governors' Conference on
slides Along the Columbia River Valley, Northeastern Environmental Geology, Colorado Geological Survey,
Washington. U.S. Geological Survey, Professional Paper Special Publ. 1,1969, pp. 8-15.
367,1961,98 pp. 1.32 Smith, R. Economic and Legal Aspects. In Land-
1.15 Kiersch, G. A. Vaiont Reservoir Disaster. Civil Engineer- slides and Engineering Practice, Highway Research
ing, Vol. 34, No. 3, 1964, pp. 32-39. Board, Special Rept. 29, 1958, pp. 6-19.
1.16 Krohn, J. P., and Slosson, J. E. Landslide Potential in 1.33 Sutter, J. H., and Hecht, M. L. Landslide and Subsi-
the United States. California Geology, Vol. 29, No. 10, dence Liability. California Continuing Education at
Oct. 1976, pp. 224-231. the Bar, Berkeley, California Practice Book 65, 1974,
1.17 Lane, K. S. Stability of Reservoir Slopes. In Failure 240 pp.
and Breakage of Rock (Fairhurst, C., ed.), Proc., 8th 1.34. Tavenas, F., Chagnon, J. Y., and La Rochelle, P. The
Symposium on Rock Mechanics, Univ. of Minnesota, Saint-Jean-Vianney Landslide: Observations and Eye-
1966, American Institute of Mining, Metallurgical and witnesses Accounts. Canadian Geotechnical Journal,
Petroleum Engineers, New York, 1967, pp. 321-336. - Vol. 8, No. 3,1971, pp. 463-478.
1.18 Lee, K. L., and Duncan, J. M. Landslide of April 25, 1.35 Taylor, F. A., and Brabb, E. E. Distribution and Cost
1974, on the Mantaro River, Peru. National Academy by Counties of Structurally Damaging Landslides in
of Sciences, Washington, D.C., 1975, 72 pp. the San Francisco Bay Region, California, Winter of
1.19 Leighton, F. B. Urban Landslides: Targets for Land-Use 1968-69. U.S. Geological Survey, Miscellaneous Field
Planning in California. In Urban Geomorphology, Geo- Studies Map MF-327, 1972.
logical Society of America, Special Paper 174, 1976, 1.36 U.S. Geological Survey. The Hebgen Lake, Montana,
pp. 37-60. Earthquake of August 17, 1959. U.S. Geological Sur-
1.20 Lewis, H., McDaniel, A. H., Peters, R. B., and Jacobs, vey, Professional Paper 435, 1964, 242 pp.
D. M. Damages Due to Drainage, Runoff, Blasting, and 1.37 Wiggins, J. H., Slosson, J. E., and Krohn, J. P. Natural
Slides. National Cooperative.Highway Research Pro- Hazards: An Expected Building Loss Assessment. J. H.
gram, Rept. 134, 1972, 23 pp. Wiggins Co., Redondo Beach, California, Draft rept. to
1.21 MCConnell, R. G., and Brock, R. W. The Great Land- National Science Foundation, 1977, 134 pp.
slide at Frank, Alberta. In Annual Report of the Canada 1.38 Williams, G. P., and Guy, H. P. Erosional and Deposi-
Department of the Interior for the Year 1902-03, Ses- tional Aspects of Hurricane Camille in Virginia, 1969.
sional Paper 25, 1904, pp. 1-17. U.S. Geological Survey, Professional Paper 804, 1973,
1.22 McDowell, B. Avalanche! In Great Adventures With 80 pp.
National Geographic, National Geographic Society, 1.39 Witkind, I. J. Events on the Night of August 17, 1959:
Washington, D.C., 1963, pp. 263-269. The Human Story. U.S. Geological Survey, Professional
1.23 Merriam, R. Portuguese Bend Landslide, Palos Verdes Paper 435, 1964, pp. 1-4.
Hills, California. Journal of Geology, Vol. 68, No. 2,
March 1960, pp. 140-153.
1.24 Morton, D. M., and Streitz, R. Landslides: Part Two.
California Division of Mines and Geology, Mineral In-
formation Service, Vol. 20, No. 11, 1967, pp. 135-140. PHOTOGRAPH CREDITS
1.25 Nilsen, T. H., and Turner, B. L. Influence of Rainfall
and Ancient Landslide Deposits on Recent Landslides Figure 1.1 Courtesy of Tennessee Department of Transportation
(1950-71) in Urban Areas of Contra Costa County, Figure 1.2 Courtesy of The Register, Santa Ana, California
California. U.S. Geological Survey, Bulletin 1388, Figure 1.3 Courtesy of Japan Society of Landslide
19759 18 pp. Figure 1.6 J. R. Stacy, U.S. Geological Survey
10
Chapter 2
Slope Movement
Types and
Processes
DavidJ. Varnes
This chapter reviews a fairly complete range of slope- having some usefulness in emphasizing features pertinent
movement processes and identifies and classifies them ac- to recognition,avoidance, control, correction,or other pur-
cording to features that are also to some degree relevant to pose for the classification. Among the attributes that have
their recognition, avoidance, control, or correction. Al- been used as criteria for identification and classification
though the classification of landslides presented in Special are type of movement, kind of material, rate of movement,
Report 29 (2.182) has been well received by the profession, geometry of the area of failure and the resulting deposit,
some deficiencies have become apparent since that report age, causes, degree of disruption of the displaced mass, re-
was published in 1958; in particular, more than two dozen lation or lack of relation of slide geometry to geologic
partial or complete classifications have appeared in various structure, degree of development, geographic location of
languages, and many new data on slope processes have been type examples, and state of activity.
published. The chief criteria used in the classification presented
One obvious change is the use of the term slope move- here are, as in 1958, type of movement primarily and type
ments, rather than landslides, in the title of this chapter of material secondarily. Types of movement (defined be-
and in the classification chart. The term landslide is widely low) are divided into five main groups: falls, topples,
used and, no doubt, will continue to be used as an all- slides, spreads, and flows. A sixth group, complex slope
inclusive term for almost all varieties of slope movements, movements, includes combinations of two or more of the
including some that involve little o-r no true sliding. Never- other five types. Materials are divided into two classes:
theless, improvements in technical communication require rock and engineering soil; soil is further divided into de-
a deliberate and sustained effort to increase the precision bris and earth. Some of the various combinations of move-
associated with the meaning of words, and therefore the ments and materials are shown by diagrams in Figure 2.1
term slide will not be used to refer to movements that do (in pocket in back of book); an abbreyiated version is shown
not include sliding. However, there seems to be no single in Figure 2.2. Of course, the type of both movement and
simple term that embraces the range of processes discussed
here. Geomorphologists will see that this discussion com-
prises what they refer to as mass wasting or mass move- Figure 2.2. Abbreviated classification of slope movements.
ments, except for subsidence or other forms of ground (Figure 2.1 in pocket in back of book gives complete classification
sinking. with drawings and explanatory text.)
The classification described in Special Report 29 is here
TYPE OF MA TERIAL
extended to include extremely slow distributed movements TYPE OF MOVEMENT ENGINEERING SOILS
BEDROCK
of both rock and soil; those movements are designated in Predominontly coOrse Predominontly fine
Debris toll
many classifications as creep. The classification also in- FALLS Rock lou Earth toll
Earth topple
TOPPLES Rock topple Debris topple
cludes the increasingly recognized overturning or toppling ROTATIONAL FEW
Rock Slump Debris slump Earth Slump
11
materials may vary from place to place or from time to fall is a fall of debris, which is composed of de'trital frag-
time, and nearly continuous gradation may exist in both; ments prior to failure. Rapp (2.131, p. 104) suggested that
therefore, a rigid classification is neither practical nor de- falls of newly detached material be called primary and those
sirable. Our debts to the earlier work of Sharpe (2.146) involving earlier transported loose debris, such as that from
remain and are augmented by borrowings from many other shelves, be called secondary. Among those termed debris
sources, including, particularly, Skempton and Hutchinson falls here, Rapp (2.131, p.97) also distinguished pebble falls
(2,154), Nemok, Paiek, and Rybâi (2.116), de Freitas and (size less than 20 mm), cobble falls (more than 20 mm, but
Watters (2.37), Záruba and Mend (2.193), and Zischinsky less than 200 mm), and boulder falls (more than 200 mm).
(2.194). Discussions with D. H. Radbruch-Hall of the U.S. Included within falls would be the raveling of a thin collu-
Geological Survey have led to significant beneficial changes vial layer, as illustrated by Deere and Patton (2.36), and of
in both content and format of the presentation. fractured, steeply dipping weathered rock, as illustrated by
The classification presented here is concerned less with Sowers (2.162).
affixing short one- or two-word names to somewhat com- The falls of less along bluffs of the lower Mississippi
plicated slope processes and their deposits than with de- River valley, described in a section on debris falls by
veloping and attempting to make more precise a useful Sharpe (2.146, p. 75), would be called earth falls (or less
vocabulary of terms by which these processes and deposits falls) in the present classification.
may be described. For example, the word creep is partic-
ularly troublesome because it has been used long and Topples
widely, but with differing meanings, in both the material
sciences, such as metallurgy, and in the earth sciences, Topples have been recognized relatively recently as a dis-
such as geomorphology. As the terminology of physics tinct type of movement. This kind of movement consists
and materials science becomes more and more applied to of the forward rotation of a unit or units about some pivot
the behavior of soil and rock, it becomes necessary to en- point, below or low in the unit, under the action of gravity
sure that the word creep conveys in each instance the con- and forces exerted by adjacent units or by fluids in cracks.
cept intended by the author. Similarly, the word flow has It is tilting without collapse. The most detailed descrip-
been used in somewhat different senses by various authors tions have been given by de Freitas and Watters (2.37), and
to describe the behavior of earth materials. To clarify the some of their drawings are reproduced in Figure 2.ldl and
meaning of the terms used here, verbal definitions and dis- d2. From their studies in the British Isles, they concluded
cussions are employed in conjunction with illustrations of that toppling failures are not unusual, can develop in a va-
both idealized and actual examples to build up descriptors riety of rock types,and can range in volume from 100 m3
of movement, material, morphology, and other attributes to more than 1 Gm3 (130 to 1.3 billion yd3). Toppling
that may be required to characterize types of slope move- may or may-not culminate in either falling or sliding, de-
ments satisfactorily. pending on the geometry of the failing mass and the orien-
tation and extent of the discontinuities. Toppling failure
TERMS RELATING TO has been pictured by Hoek (2.61), Aisenstein (2.1, p. 375),
MOVEMENT and Bukovansky, RodrIquez, and Cedri:in (2.16) and studied
in detail in laboratory experiments with blocks by Hofmann
Kinds of Movement (2.63). Forward rotation was noted in the Kimbley copper
pit by Hamel (2.56), analyzed in a high rock cut by Piteau
Since all movement between bodies is only relative, a de- and others (2.125), and described among the prefailure
scription of slope movements must necessarily give some movements at Vaiont by Hofmann (2.62).
attention to identifying the. bodies that are in relative mo-
tion. For example, the word slide specifies relative motion Slides
between stable ground and moving ground in which the
vectors of relative motion are parallel to the surface of In true slides, the movement consists of shear strain and
separation or rupture; furthermore, the bodies remain in displacement along one or several surfaces that are visible
contact. The word flow, however, refers not to the mo- or may reasonably be inferred, or within a relatively nar-
tions of the moving mass relative to stable ground, but row zone. The movement may be progressive; that is,
rather to the distribution and continuity of relative move- shear failure may not initially occur simultaneously over
ments of particles within the moving mass itself. what eventually becomes a defined surface, of rupture, but
rather it may propagate from an area of local failure. The
Falls displaced mass may slide beyond the original surface of
rupture onto what had been the original ground surface,
In falls, a mass of any size is detached from a steep slope or which then becomes a surface of separation.
cliff, along a surface on which little or no shear displacement Slides were subdivided in the classification published in
takes place, and descends mostly through the air by free 1958 (2.182) into (a) those in which the material in motion
fall, leaping, bounding, or rolling. Movements are very rapid is not greatly deformed and consists of one or a few units
to extremely rapid (see rate of movement scale, Figure 2.1u) and (b) those in which the material is greatly deformed or
and may or may not be preceded by minor movements consists ofmany semi-independent units. These subtypes
leading to progressive separation of the mass from its source. were further classed into rotational slides and planar slides.
Rock fall is a fall of newly detached mass from an area In the present classification, emphasis is put on the distinc-
of bedrock. An example is shown in Figure 2.3. Debris tion between rotational and translational slides, for that
12
Figure 2.3. Rock fall due to undercutting along shore of Las unit commonly tilts backward toward the slope (Figures
Vegas Bay, Lake Mead, Nevada (photograph taken February 2.lg, 2.1i, 2.4, 2.5, 2.6, and 2.7), but some blocks may
24, 1949) (2.182). Rock is Muddy Creek formation (Pliocene) tilt forward.
consisting here of siltstone overlain by indurated breccia.
Figure 2.6 shows some of the commoner varieties of
slump failure in various kinds of materials. Figure 2.7
shows the backward tilting of strata exposed in a longitu-
dinal section through a small slump in lake beds. Although
the rupture surface of slumps is generally concave upward,
it is seldom a spherical segment of uniform curvature. Often
the shape of the surface is greatly influenced by faults,
joints, bedding, or other preexisting discontinuities of the
material. The influence of such discontinuities must be con-
sidered carefully when the engineer makes a slope-stability
analysis that assumes a certain configuration for the surface
of rupture. Figures 2.7 and 2.8 show how the surface of
rupture may follow bedding planes for a considerable part
of its length. Upward thrusting and slickensides along the
lateral margin of the toe of a slump are shown in Figure 2.9.
Figure 2.4. Slope failure in uniform material (2.182).
The classic purely rotational slump on a surface of
1.1 DflOPd WAPFfl bI ROTATIONAL SHEAR ON smooth curvature is relatively uncommon among the
CYLINDRICAL SURFACE
many types of gravitational movement to which geologic
materials are subject. Since rotational slides occur most
frequently in fairly homogeneous materials, their incidence
among constructed embankments and fills, and hence their
interest to engineers, has perhaps been high relative to other
types of failure, and their methods of analysis have in the
THRUSTING AT THE TOE
past been more actively studied. Geologic materials are
seldom uniform, however, and natural slides tend to be
complex or at least significantly controlled in their mode
of movement by internal inhomogeneities and discontinu-
difference is of at least equal significance in the analysis of ities. Moreover, deeper and deeper artificial cuts for dam-
stability and the design of control methods. An indication sites, highways, and other engineering works have increas-
of degree of disruption is still available by use of the terms ingly produced failures not amenable to analysis by the
block or intact for slides consisting of one or a few mov- methods appropriate to circular arc slides and have made
ing units and the terms broken or disrupted for those con- necessary the development of new methods of analytical
sisting of many units; these terms avoid a possible source design for prevention or cure of failures in both bedrock
of confusion, pointed out by D. H. Radbruch-Hall, in the and engineering soils.
use of the term debris slide, which is now meant to indi- The scarp at the head of a slump may be almost vertical.
cate only a slide originating in debris material, which may If the main mass of the slide moves down very far, the
either proceed as a relatively unbroken block or lead to dis- steep scarp is left unsupported and the stage is set for a
ruption into many units, each consisting of debris. new failure (similar to the original slump) at the crown of
the slide. Occasionally, the scarps along the lateral mar-
Rotational Slides gins of the upper part of the slide may also be so high and
steep that slump blocks break off along the sides and move
The commonest examples of rotational slides are little- downward and inward toward the middle of the main slide.
deformed slumps, which are slides along a surface of rup- Figure 2.10 (2.183) shows a plan view of slump units along
ture that is curved concavely upward. Slumps, and slumps the upper margins of a slide; the longest dimensions of
combined with other types of movement, make up a high these units are parallel with, rather than perpendicular to,
proportion of landslide problems facing the engineer. The the direction of movement of the main slide. Any water
movement in slumps takes place only along internal slip that finds its way into the head of a slump may be ponded
surfaces. The exposed cracks are concentric in plan and by the backward tilt of the unit blocks or by other irregu-
concave toward the direction of movement. In many larities in topography so that the slide is kept wet con-
slumps the underlying surface of rupture, together with stantly. By the successive creation of steep scarps and trap-
the exposed scarps, is spoon-shaped (Figure 2.4). If the ping of water, slumps often become self-perpetuating areas
slide extends for a considerable distance along the slope of instability and may continue to move and enlarge inter-
perpendicular to the direction of movement, much of the inittently until a stable slope of very low gradient is at-
rupture surface may approach the shape of a sector of a tained.
cylinder whose axis is parallel to the slope (Figure 2.4). In
slumps, the movement is more or less rotational about an Translational Slides
axis that is parallel to the slope. In the head area, the
movement may be almost wholly downward and have In translational sliding the mass progresses out or down
little apparent rotation; however, the top surface of each and out along a more or less planar or gently undulatory
13
surface and has little of the rotary movement or backward tween layers of bedded deposits, or by the contact between
tilting characteristic of slump. The moving mass com- firm bedrock and overlying detritus (Figure 2.11). Several
monly slides out on the original ground surface. The dis- examples of block slides are shown in Figures 2.lj2, 2.11,
tinction between rotational and translational slides is use- 2.12, 2.13 (2.136), 2.14 (2.107), and 2.15. In many trans-
ful in planning control measures. The rotary movement of lational slides, the slide mass is greatly deformed or breaks
a slump, if the surface of rupture dips into the hill at the up into many more or less independent units. As deforma-
foot of the slide, tends to restore equilibrium in the un- tion and disintegration continue, and especially as water
stable mass; the driving moment during movement de- content or velocity or both increase, the broken or dis-
creases and the slide may stop moving. A translational rupted slide mass may change into 'a flow; however, all
slide, however, may progress indefinitely if the surface on gradations exist. Broken translational slides of rock are
which it rests is sufficiently inclined and as long as the shear shown in Figure 2.lj3 and of debris in Figures 2.1k and
resistance along this surface remains lower than the more 2.16 (2.83).
or less constant driving force. A translational slide in which
the moving mass consists of a single unit that is not greatly Lateral Spreads
deformed or a few closely related units may be called a
block slide. If the moving mass consists of many semi- In spreads, the dominant mode of movement is lateral ex-
independent units, it is termed a broken or disrupted. slide. tension accommodated by shear or tensile fractures. Two
The movement of translational slides is commonly con- types may be distinguished.
trolled structurally by surfaces of weakness, such as faults,
j9ints, bedding planes, and variations in shear strength be- 1. Distributed movements result in overall extension
r2
I' off
clay
Firm
-- -- Groben
lo erycloy/////////////
Soil (a) SLIDE BENEATH SIDEHILI. (f) FAILURE WITHIN A SIDEHILL FILL.
Shole
(g) FAILURE OF EMBANKMENT. GRAVEL COUNTER. (h) SLIDE IN FILL INVOLVING UNDERTHRUSTING OF
WEIGHT ON LEFT SIDE PREVENTS SLIDE. FIRM SURFACE MATERIAL DOWN SLOPE.
Fill
Fill
Zunterw:ht r
Fill
Soft
zone
Stable \
Firm...
14
Figure 2.6. Slump of fill, but without a recognized or well-defined controlling basal
controlled in this shear surface or zone of plastic flow. These appear to occur
instance by failure i predominantly in bedrock, especially on the crests of ridges
in underlying soil 7' (Figure 2.1m1). The mechanics of movement are not well
(2.182).
known.
2. Movements may involve fracturing and extension of
coherent material, either bedrock or soil, owing to liquefac-
tion or plastic flow of subjacent material. The coherent
upper units may subside, translate, rotate, or disintegrate,
or they may liquefy and flow. The mechanism of failure
Figure 2.7. Slump in thinly bedded lake deposits of silt and clay can involve elements not only of rotation and translation
in Columbia River valley (note backward tilting of beds above but also of flow; hence, lateral spreading failures of this
surface of rupture) (2.182).
type may be properly regarded as complex. They form,
however, such a distinctive and dominant species in certain
geologic situations that specific recognition seems worth-
while.
WA ZONE A
Movement chiefly by large-
scal, slumping along slip
surf ecu
A. A' A"
Principal slump units
i- 8,81
Narrow slump unit with
.4. .' axes perpendicular to
axes of main slump units
and parallel with length
-.4- -. of main slide
B
Island remaining after
downward movement of
unit D from area B.
lt&
'.5
I- •'- - -,
ZONE c
Toe of slide area; original
form altered by railroad
reconstruction work.
100 m
328 ft
15
Figure 2.11. Thin layer of residual debris that slid on inclined Figure 2.13. Development
strata of metasiltstone along 1-40 in Cocke County, Tennessee. of landslides in horizontal
sequence of claystone and (D -- -
coal caused by relaxation ® -
of horizontal stresses - - - -
resulting from reduction in
thickness of overlying strata
(2.136).
Ic)
(d)
CLAYSTONE Q COAL
Figure 2.12. Block slide at quarry (2.182).
Figure 2.14. Section view of translational slide
at Point Fermin, near Los Angeles, California
(see also Figure 2.15) (2107, 2.182).
Maximum average rate of movement was
3cm/week (1.2 in).
800 m
.1
2400 ft
S1!Rrure
claystone. The underlying layer became plastic and flowed taken depends on local factors. Most of the larger land-
to some extent, allowing the overlying firmer rock to break slides in glacial sediments of' northern North America and
into strips and blocks that then became separated. The Scandinavia lie somewhere within this series.
cracks between the blocks were filled with either soft ma- Lateral spreads in surficial deposits have been destructive
terial squeezed UI) from below or detritus from above. The of both life and property and have, therefore, been the sub-
lateral extent of these slides is remarkable, involving hands ject olintensive study. Examples may be cited from Sweden
several to many kilometers wide around the edges of pla- (Caldenius and Lundstrom, 2.18), Canada (Mitchell and
teaus and escarpments. The rate of movement of most lat- Markell, 2.108), Alaska (Seed and Wilson, 2J44), and
eral spreads in bedrock is apparently extremely slow. California (Youd, 2.191). Most of the spreading failures
Laterally spreading slope movements also form in fine- in the western United States generally involve less than total
grained earth material on shallow slopes, particularly in liquefact ion and seem to have been mobilized only by seis-
sensitive silt and clay that loses most or all of its shear mic shock. For example, there were damaging failures in
strength on disturbance or remolding. The failure is tisti- San Fernando Valley. California, during the 1969 earth-
ally progressive; that is, it starts in a local area and spreads. quake because of liquefaction of underlying sand and silt
Often the initial failure is a slump along a stream bank or and spreading of the surficial, firmer material. The spread-
shore, and the progressive failure extends retrogressively ing failure of Bootlegger Cove clay beneath the Turnagain
back from the initial failure farther and farther into the Heights residential district at Anchorage, Alaska, during
bank. The principal movement is translation rather than the 1964 great earthquake resulted in some loss of life and
rotation. If the underlying mobile zone is thick, the blocks extensive damage. In some areas within the city of San
at the head may sink downward as grabens, not necessarily Francisco, the principal damage due to the 1906 earth-
with backward rotation, and there may be upward and out- quake resulted from spreading failures that not only did
ward extrusion and now at the toe. Movement generally direct damage to structures but also severed principal water-
begins suddenly, without appreciable warning, and proceeds supply lines and thereby hindered firefighting.
with rapid to very rapid velocity. All investigators would agree that spreading failures in
These types appear to be members of a gradational series glacial and marine sediments of Pleistocene age present
of landslides in surficial materials ranging from block slides some common and characteristic feat tires: Movement of-
at one extreme, in which the zone of how beneath the slid- ten occurs for no apparent external reason, failure is gen-
ing mass may be absent or very thin, to earth flows or com- erally sudden, gentle slopes are often unstable, dominant
pletely liquefied mud (lows at the other extreme, in which movement is translatory, materials are sensitive, and pore-
the zone of flow includes the entire mass. The form that is water pressure is important in causing instability. All de-
16
Figure 2.15. Translational slide at Point Fermin, California. tion may be recoverable over a period of time upon release
Photograph, which was taken January 17, 1965, indicates minor of the stress, but generally most of it is not. The move-
slumping into gap at rear of main mass and imminent rock falls ment commonly is imperceptible (which is usually one of
at sea cliff. Principal motion, however, was by sliding along the essential attributes of creep as defined in geomorphol-
gently seaward dipping strata.
ogy), but increasingly sophisticated methods of measure-
ment make this requirement difficult to apply. Further-
more, the usual partition of creep into three stages—
primary (decelerating), secondary (steady or nearly so),
and tertiary (accelerating to failure)—certainly includes
perceptible deformation in the final stages. Laboratory
studies show that both soil and rock, as well as metals, can
exhibit all three stages of creep. Observations in the field,
such as those reported by Muller (2.112) at Vaiont, em-
brace within the term creep perceptible movements that
immediately preceded catastrophic failure.
There is disagreement also as to whether creep in rock
and soil should be restricted to those movements that are
distributed through a mass rather than along a defined
fracture. Authorities are about equally divided on this
point but, in keeping with the use of the term in engineer-
ing mechanics, the acceptance of this restriction is not
favored. Creep movements can occur in many kinds of
topples, slides, spreads, and flows, and the term creep
need not be restricted to slow, spatially continuous defor-
Figure 2.16. Debris slide of disintegrating soil
mation. Therefore, spatially continuous deformations are
slip variety (2.83, 2.182).
classified as various types of flow in rock, debris, and earth.
Flows in Bedrock
17
more accurately recognized as flows than those in rocks be- Figure 2.17. Debris avalanche or very rapid debris flow at
cause the relative displacements within the mass are com- Franconia Notch, New Hampshire, June 24, 1948, after several
monly larger and more closely distributed and the general days of heavy rainfall (2.182). Only soil mantle 2 to 5 m (7 to
6 ft) thick, which lay over bedrock on a slope of about 1:1,
appearance is more obviously that of a body that has be-
was involved. Scar is about 450 m (1500 ft) long; natural levees
haved like a fluid. Moreover, the fluidizing effect of water can be seen along sides of flow. US-3 is in foreground.
itself is, as a rule, a part of the process. Slip surfaces within
the moving mass are usually not visible or are short lived,
and the boundary between moving mass and material in
place may be a sharp surface of differential movement or
a zone of distributed shear.
There is complete gradation from debris slides to debris
flows, depending on water content, mobility, and character
of the movement, and from debris slide to debris avalanche
as movement becomes much more rapid because of lower
cohesion or higher water content and generally steeper
slopes. Debris slides and, less commonly, debris avalanches
may have slump blocks at their heads. In debris slides, the
moving mass breaks up into smaller and smaller parts as it
advances toward the foot, and the movement is usually
slow. In debris avalanches, progressive failure is more
rapid, and the whole mass, either because it is quite wet
or because it is on a steep slope, liquefies, at least in part,
flows, and tumbles downward, commonly along a stream
channel, and may advance well beyond the foot of the
slope. Debris avalanches are generally long and narrow
and often leave a serrate or V-shaped scar tapering uphill
at the head, as shown in Figures 2.1q3 and 2.17, in con-
trast to the horseshoe.shaped scarp of a slump.
Debris flows, called mud flows in some other classifi.
cations, are here distinguished from the latter on the basis
of particle size. That is, the term debris denotes material
that contains a relatively high percentage of coarse frag-
ments, whereas the term mud flow is reserved for an earth
flow consisting of material that is wet enough to flow rap.
idly and that contains at least 50 percent sand-, silt-, and
clay-sized particles. Debris flows commonly result from slope movements of snow or ice. Rapp (2.132) and Temple
unusually heavy precipitation or from thaw of snow or and Rapp (2.169), with considerable logic, recommend that,
frozen soil. The kind of flow shown in Figure 2.1ql often because the term debris avalanche is poorly defined, it should
occurs during torrential runoff following cloudbursts. It is be abandoned, and that the term avalanche should be used
favored by the presence of soil on steep mountain slopes only in connection with mass movements of snow, either
from which the vegetative cover has been removed by fire pure or mixed with other debris. The term debris avalanche,
or other means, but the absence of vegetation is not a pre- however, is fairly well entrenched and in common usage
requisite. Once in motion, a small stream of water heavily (Knapp, 2.86); hence, its appearance in the classification
laden with soil has transporting power that is dispropor- as a variety of very rapid to extremely rapid debris flow
tionate to its size, and, as more material is added to the seems justified.
stream by caving of its banks, its size and power increase. Recent studies have contributed much to a better un-
These flows commonly follow preexisting drainageways, derstanding of the rates and duration of rainfall that lead
and they are often of high density, perhaps 60 to 70 per. to the triggering of debris flows, the physical properties of
cent solids by weight, so that boulders as big as automo- the material in place, the effect of slope angle, the effect
biles may be rolled along. If such a flow starts on an un- of pore-water pressure, the mobilization of material and
broken hillside it will quickly cut a V-shaped channel. mechanism of movement, and the properties of the result-
Some of the coarser material will be heaped at the site to ing deposit. The reader is referred especially to the works
form a natural levee, while the more fluid part moves of Campbell (2.20), Daido (2.34), Fisher (2.46), Hutchin-
down the channel (Figure 2.17). Flows may extend many son (2.70), Hutchinson and Bhandari (2.72), Johnson and
kilometers, until they drop their loads in a valley of lower Rahn (2. 76), Jones (2.78), Prior, Stephens, and Douglas
gradient or at the base of a mountain front. Some debris (2.129), Rapp (2.131), K. M. Scott (2.141), R. C. Scott
flows and mud flows have been reported to proceed by a (2.142), and Williams and Guy (2.188). Flowing move-
series of pulses in their lower parts; these pulses presumably ments of surficial debris, including creep of the mantle of
are caused by periodic mobilization of material in the source weathered rock and soil, are shown in Figure 2.1q2, 2.1q4,
area or by periodic damming and release of debris in the and 2.1q5. Soil flow, or sohifluction, which in areas of pe-
lower channel. rennially or permanently frozen ground is better termed
The term avalanche, if unmodified, should refer only to gelifluction, takes many forms and involves a variety of
18
mechanisms that can be treated adequately only in works Figure 2.19. Earth flow near Greensboro, Florida (2.80, 2.182).
Material is flat-lying, partly indurated clayey sand of the Hawthorn
devoted to this special field, which is of great significance
formation (Miocene). Length of slide is 275 m (900 ft) from scarp
to engineering works at high latitudes and altitudes. The to edge of trees in foreground. Vertical distance is about 15 m (45
reader is referred to summaries by Dylik (2.40), Washburn ft) from top to base of scarp and about 20 m (60 ft) from top of
(2.187), and Corte (2.27); the proceedings of the Interria- scarp to toe. Slide occurred in April 1948 after year of unusually
tional Conference on Permafrost (2.111); and recent work heavy rainfall, including 40cm (16 in) during 30 d preceding
sl ide.
by McRoberts and Morgenstern (2.104, 2.105) and Emble-
ton and King (2.42).
Subaerial flows in fine-grained materials such as sand,
.-
silt, or clay are classified here as earth flows. They take a ;sf:
variety of forms and range in water content from above
saturation to essentially dry and in velocity from extremely
rapid to extremely slow. Some examples are shown in Fig-
ure 2.1rl through 2.1r5. At the wet end of the scale are
mud flows, which are soupy end members of the family of
predominantly fine-grained earth flows, and subaqueous
flows or flows originating in saturated sand or silt along
_____
shores.
441.
In a recent paper reviewing Soviet work on mud flows,
Kurdin (2.91) recommended a classification of mud flows
based on (a) the nature of the water and solid-material
supply; (b) the structural-rheological model, that is, whether
the transporting medium is largely water in the free state
or is a single viscoplastic mass of water and fine particles;
(c) the composition of the mud flow mass, that is, whether Van Wamelon, and Weinberg, 2.89). When the structure of
it consists of mud made up of water and particles less than the loose sand breaks down along a section of the bank, the
1 mm (0.04 in) in size or of mud plus gravel, rubble, sand flows rapidly along the bottom, and, by repeated
boulders, and rock fragments; and (d) the force of the mud small failures, the slide eats into the bank and enlarges the
flow as defined by volume, rate of discharge, and observed cavity. Sometimes the scarp produced is an arc, concave
erosive and destructive power. In the Soviet literature mud toward the water, and sometimes it enlarges greatly, retain-
flows include not only what are here classified as debris ing a narrow neck or nozzle through which the sand flows.
flows but also heavily laden flows of water-transported An extensive discussion and classification of subaqueous
sediment. mass-transport processes and the resulting deposits have
According to Andresen and Bjerrum (2.3), subaqueous been presented by Carter (2.21).
flows are generally of two types: (a) retrogressive flow Rapid earth flows also occur in fine-grained silt, clay,
slide or (b) spontaneous liquefaction, as shown in Figure and clayey sand, as shown in Figures 2.1 r2 and 2.19 (2.80).
2.18. The retrogressive flows, as shown in Figure 2.1r1, These flows form a complete gradation with slides involving
occur mostly along banks of noncohesive clean sand or failure by lateral spreading, but they involve not only lique-
silt. They are especially common along tidal estuaries in faction of the subjacent material but also retrogressive fail-
the coastal provinces of Holland, where banks of sand are ure and liquefaction of the entire slide mass. They usually
subject to scour and to repeated fluctuations in pore-water take place in sensitive materials, that is, in those materials
pressure because of the rise and fall of the tide (Koppejan, whose shear strength on remolding at constant water con-
tent is decreased to a small fraction of its original value.
Rapid earth flows have caused loss of life and immense
Figure 2.18. Retrogressive flow slide and spontaneous destruction of property in Scandinavia, the St. Lawrence
liquefaction (2.3). River valley in Canada, and Alaska during the 1964 earth-
quake. The properties of the material involved, which is
(a) RETROGRESSIVE FLOW SLIDE
(Mechanism after Koppejan. Van Wamelon, and Weinberg. 2.891 usually a marine or estuarine clay of late Pleistocene age,
rAFTERFLOWSLIDE WATERLEVEL BEFORE FLOW SLIDE have been thoroughly studied by many investigators during
the last 15 years. Summary papers have been written by
Bjerrum and others (2.12) on flows in Norway and by
Mitchell and Markell (2.108),and Eden and Mitchell (2.41)
—SCOURED OUT BY TIDAL CURRENT
on flows in Canada. Shoreline flows produced by the Alas-
LARGE STRAIN VELOCITIES 0.05 10 SE VERAL k,elh
k.an earthquake at Valdez and Seward have been described
IbI SPONTANEOUS LIQUEFACTION
by Coulter and Migliaccio (2.28) and Lenike (2.98). The
- SECONDARY SLIDE SECO5DARY SLIDE large failure on the Reed Terrace near Kettle Falls, Wash-
AFTER
ington, shown in Figures 2.20 and 2.21 (2.79), resembles
in some respects the earth flow at Riviere Blanche, Quebec,
shown in Figure 2.1r2 (2.146).
LSTART
EARTHQUAKEIPILE DRIVING
The somewhat drier and slower earth flows in plastic
SMALL STRAIN VELOCITIES 10,0 100 Kmjh
earth are common in many parts of the world wherever
Nole: 1 km(h 0.6 rnpl,. there is a combination of clay or weathered clay-bearing
19
rocks, moderate slopes, and adequate moisture; Figure 2.22 on falling from a cliff, have been recognized, but so far as
shows a typical example. A common elongation of the flow, is known none has been studied in detail (Figure 2.26).
channelization in depression in the slope, and spreading of Flows of bess mobilized by earthquake shock have been
the toe are illustrated in Figure 2.43 and also shown in an more destructive of life than any other type of slope failure.
actual debris flow in Figure 2.23. Those that followed the 1920 earthquake in Kansu Province,
The word flow naturally brings water to mind, and some China (Close and McCormick,2.23), shown in Figure 2.1r5,
content of water is necessary for most types of flow move- took about 100 000 lives. Apparently the normal, fairly
ment. But small dry flows of granular material are common, coherent internal structure of the porous silt was destroyed
and a surprising number of large and catastrophic flow by earthquake shock, so that, for all practical purposes, the
movements have occurred in quite dry materials. There- bess became a fluid suspension of silt in air and flowed
fore, the classification of flows indicates the complete down into the valleys, filling them and overwhelming vil-
range of water content—from liquid at the top to dry at lages. The flows were essentially dry, according to the re-
the bottom. Tongues of rocky debris on steep slopes moving port. Extensive flows of bess accompanied the Chait earth-
extremely slowly and often fed by talus cones at the head quake of July 10, 1949, in Tadzhikistan, south-central Asia,
are called block streams (Figure 2.lq5). Because of rain- and buried or destroyed 33 villages as the flows covered the
wash, a higher proportion of coarse rocks may be in the sur- bottoms of valleys to depths of several tens of meters for
face layers than in the interior. Dry flows of sand are com- many kilometers (Gubin, 2.54).
mon along shores or embankments underlain by dry gran-
ular material. In form, they may be cliannelized, as shown Complex
in Figures 2.1r4 and 2.24, or sheetlike, as shown in Figure
2.25 (2.79). Small flows of dry silt, powered by impact More often than not, slope movements involve a combina-
Figure 2.20. Reed Terrace area, right bank of Lake Roosevelt Figure 2.22. Earth flow developing from slump near Berkeley.
reservoir on Columbia River. near Kettle Falls, Washington, May California (2.182).
15, 1951 (2.182). Landslide of April 10. 1952, involving about
11 Mm3 (15 million yd3) took place by progressive slumping,
liquefaction, and flowing out of glaciofluvial sediments through
narrow orifice into bottom of reservoir.
-
- F--. / .. L'- ...............•.
..-...-.:..........................-...
......... - -
A.
Figure 2.21. Reed Terrace area, Lake Roosevelt, Washington, Figure 2.23. Old debris flow in altered volcanic rocks west of
August 1, 1952, after landslide of April 10, 1952(2.79, 2.182). Pahsimeroi River in south central Idaho.
:cQ
7T' b!::I
20
Figure 2.24. Dry sand flow in Columbia River valley (2i82). tion of one or more of the principal typesof movement de-
Material is sand over lake-bed si It; dry sand from upper terrace flowed scribed above, either within various parts of the moving
like liquid through notch in more compact sand and silt below.
mass or at different stages in development of the movements.
These are termed complex slope movements, and a few cx-
ainpies of the many possible types are illustrated in Figure
2Js1 through 2Js5.
Of particular interest regarding hazards of landslides to
life and property are large, extremely rapid rock fall-debris
flows, referred to as rock-fragment now (variety rock-fall
avalanche) in the 1958 classification (2.182). Rock slide-
and rock fall-debris flows are most common in rugged
mountainous regions. The disaster at Elm. Switzerland
(Heim, 2.58, pp. 84, 109-1 12), which took 115 lives,
4t
startetl with small rock slides at each side of a quarry on
the mountainside. A few minutes later the entire mass of
rock above the quarry crashed down and shot across the
1' ¼
valley. The movement of the rock fragments, which had
to that moment been that of a rock slide and rock fall, ap-
pears to have taken on the character of a flow. The mass
- •:i
rushed up the other side of the small valley, turned and
streamed into the main valley, and flowed for nearly I 5
. .- -s.. km (1 mile) at high velocity before stopping (Figure 2.lsl).
About 10 Mm3 (13 million yd3) of rock descended an aver-
age of 470 m (1540 ft) vertically in a total elapsed time of
x_ -.-'.-- .---.- . .-
-- about 55 s. The kinetic energy involved was enormous.
A similar and even larger rock-fall avalanche occurred at
Frank, Alberta, in 1903 and also caused great loss of life
and property (McConnell and Brock, 2103; Cruden and
Krahn. 2.33).
Figure 2.25. Shallow, dry, sand flow along shore of Lake
Roosevelt. Washington (2.79). Wave erosion or saturation of
These rock fall-debris flows are minor, however, com-
sediment by lake water caused thin skin of material to lose pared with the cataclysmic flow that occurred at the time
Jpport and ravel off terrace scarp. of the May 31, 1970, earthquake in Peru, which buried the
city of Yungay and part of Ranrahirca, causing a loss of
more than 18 000 lives. According to Plafker, Ericksen, and
Fernandez Concha (2J26), the movement started high on
Huascaran Mountain at an altitude of 5500 to 6400 ni and
- ..- involved 50 Mni3 to 100 Mm3 (65 million to 130 million
yd 3) of rock, ice, snow, and soil that traveled 14.5 km (9
A"c. miles) from the source to Yungay at a velocity between
280 and 335 km/h (175 to 210 mph). They reported strong
evidence that the extremely high velocity and low friction
of the flow were due, at least in part, to lubrication by a
cushion of air entrapped beneath the debris. Pautre,
Sabarly. and Schneider (2.122) suggested that the mass
Figure 2.26. Dry flow of silt (2.182). Material is lake-bed silt of may have ridden on a cushion of steam. A sketch of the
Pleistocene age from high bluff on right bank of Columbia River, 4
area affected is shown in Figure 2.27, taken from a paper
km (2.5 miles) downstream from Belvedere, Washington. Flow was
not observed while in motion, but is believed to result from blocks by duff (224) on engineering geology observations.
of silt falling down slope, disintegrating, forming a single high- Crandell and Fahnestock (229) cited evidence for an air
density solid-in-air suspension, and flowing out from base of cliff. cushion beneath one or more rock fall-debris flows that
occurred in December 1963 at Little Tahoma Peak and
Emmons Glacier on the east flank of Mt. Rainier volcano,
Washington.
Such flows probably cannot be produced by a few thou-
sand or a few hundred thousand cubic meters of material.
Many millions of megagrams are required: and, when that
much material is set in motion, perhaps even slowly, pre-
dictions of behavior based on past experience with small
failures become questionable. The mechanics of large,ex-
- tremely rapid debris flows, many of which appear to have
- .... been nearly dry when formed, have conic under much re-
...
- - •-- 5- ;.
21
sheet of debris of which it is composed that Shreve (2.148) or gas were available for fluidization.
believed the broken material was not fluidized but slid on Getting back to Earth, we note self-explanatory examples
an ephemeral layer of compressed air. He reported, simi- of complex movements in Figure 2-1: slump-topple in Fig-
larly, that the large landslide that was triggered by the Alaska ure 2.1s2, rock slide-rock fall in Figure 2.1s3, and the com-
earthquake of 1964 and fell onto the Sherman Glacier mon combination of a slump that breaks down into an earth
showed little large-scale mixing and did not flow like a flow in its lower part in Figure 2.1s5.
viscous fluid but instead slid like a flexible sheet (Shreve, The illustration of cambering and valley bulging in Figure
2.149). On the other. hand, Johnson and Ragle (2.77) re- 2.1s4 is adapted from the classical paper by Hollingworth
ported, and Taylor (2.65) on the Northampton Sand Ironstone in
England, their earlier paper on the Kettering district (2.64),
Many rock-snow slides that followed from the Alaska earth- and a sketch supplied by J. N. Hutchinson. The complex
quake of March 27, 1964, illustrated a variety of flow me- movements were described by Hutchinson (2.68) as follows:
chanics. The form of some slides suggests a complete turbu-
lence during flow, while the form of others gives evidence Cambering and Valley Bulging. These related features were
for steady-state flow or for controlled shearing. first clearly recognized in 1944 by Hollingworth, Taylor,
and Kellaway (see reference in Terzaghi, 1950) in the North-
From a detailed analysis of the kinematics of natural rock ampton Ironstone field of central England, where they are
fall-debris flows and from model studies, Hsü (2.66) disputed believed to have a Late Pleistocene origin. The ironstone oc-
Shreve's hypothesis that some slid as relatively undeformed curs in the near-horizontal and relatively thin Northampton
sheets on compressed air and concluded, rather, that they Sands, which are the uppermost solid rocks in the neighbor-
flowed. hood. These are underlain, conformably, by a great thick-
Obviously, there is much yet to be learned about these ness of the Lias, into which shallow valleys, typically 1200
processes, particularly as similar features indicating mass to 1500 meters wide and 45 meters deep, have been eroded.
movements of huge size have been recognized in Mariner 9 Excavations for dam trenches in the valley bottoms have
photographs of the surface of Mars (Sharp, 2.145), where shown the Lias there to be thrust strongly upward and con-
it is yet uncertain that significant amounts of either liquid torted, while open cast workings in the Northampton Sands
Figure 2.27. Area affected by May 31, 1970, Huascarn debris avalanche, which originated at point A (2.24).
Vungay was protected from January 10, 1962, debris avalanche by 180 to 240-rn (600 to 800-ft) high ridge
(point B), but a portion of May 31, 1970, debris avalanche diverted from south side of canyon wall, topped
"protective" ridge, and descended on Vungay below. Only safe place in Yungay was Cemetery Hill (point C),
where some 93 people managed to run to before debris avalanche devastated surrounding area. Moving at
average speed of 320 km (200 mph), debris arrived at point D, 14.5 km (9 miles) distant and 3660 m (12 000
ft) lower, within 3 to 4 min after starting from north peak of Huascarn. Debris flowed upstream along course
of Rio Santa (point E) approximately 2.5 km 0.5 miles). Debris continued to follow course of Rio Santa
downstream to Pacific Ocean. approximately 160 km (100 miles), devastating villages and crops occupying
floodplain.
22
occupying the inter/luves reveal a general valleyward in- Figure 2.28. Blackhawk landslide (2.147). Upslope view,
crease of dip of "camber" of this stratum, often passing southward over lobe of dark marble breccia spread beyond mouth
of Blackhawk Canyon on north flank of San Bernardino Mountains
into dip and fault structure, suggesting corresponding in southern California. Maximum width of lobe is 3.2 km 12 miles);
downward movements along the valley margins. In adjust- height of scarp at near edge is about 15 m 150 ft).
ing to these movements, the rigid cap-rock has been dislo-
cated by successive, regularly spaced fissures which paral-
lel the valley and are known as "gulls." Similar features
have been recognized in other parts of England and in Bo-
hemia. The mechanisms by which cambering and valley
bulging have been formed remain to be established.
. '-.•
Hutchinson (2. 70) also pointed out that Sharpe's defi-
nition of flow (2i46), which requires zero relative dis-
placement at the boundary of the flow (flow adheres to
the stable material), does not fit the observed distribution
of velocities at Beltinge, where a mud flow developed in a
temperate climate on a 30-m-high (98-ft) coastal cliff of ' -
stiff, fissured London clay subject to moderate marine ero- I'
sion. 1-lere the mud flow was bounded both oil the sides
and on the bottom by discrete surfaces along which shear
displacements occurred. For these kinds of movements
Hutchinson and Bhandari (2. 72) proposed the term mtid-
slides. These can be regarded as complex movements in p.
••I &'
which the internal distribution of velocities within the
moving mass may or may not resemble that of viscous
fluids, but the movement relative to stable ground is finite Figure 2.29. Main types (S) SINGLE - - -
discontinuous shear. It would seem that the material of the of rotational slide -.
a— Slope failure
(2.68).
sliding earth flow is behaving as a plastic body in plug flow, Cloy
23
moving units have a common basal surface of rupture, as Figure 2.30. Two types of multiple slide blocks (2.97).
shown in Figure 2.30.
(a) SUPERPOSED SLIDE BLOCKS
4. Successive refers to any type of multiple movements
that develop successively in time. According to Skempton
and Hutchinson (2.154), "Successive rotational slips con-
sist of an assembly of individual shallow rotational slips.
The rather sparse data available suggest that successive
- - ------ 50 m
slips generally spread up a slope from its foot." Hutchin- — IOLC.
Figure 2.31 (2.67) shows the main types of landslides in Figure 2.31. Main types of,landslides in London clay (2.67).
London clay.
Landslides that develop one on top of another are called
multistoried by Ter-Stepanian and Goldstein (2.173). Fig-
DIAGRAMMATIC CROSS SECTION
All slip eurfaces are inferred and shown by broken line TYPE -
ure 2.32 shows their illustration of a three-storied landslide Ion and bank failures
Shallow
in Sochi on the coast of the Black Sea. rotational
slips
Rate of Movement
...i............... ar:,_,
The rate-of-movement scale used in this chapter is shown at I rotational slips
the bottom of the classification chart in Figure 2.1u. Metric _aco9roY nknoken .o...._.
and
equivalents to the rate scale shown in the 1958 classifica-
Undulations or
tion have been derived by Yemel'ianova (2.190), and these
should now be regarded as the primary definitions.
24
Figure 2.32. Three-storied landslide in is adequate to describe almost any feature of a slump earth
Sochi on coast of Black Sea, USSR now. In addition. Skempton and Hutchinson (2J54) used
(2.173). Boundaries of three Stories the ratio of D/L, where D is the maximum thickness of the
of sliding are shown in section and
plan by three types of lines.
slide and L is the maxiniurn length of the slide upslope.
From Skempton's figures showing original use of this ratio
4 ELOCKS OF ARGILLITE AND SANDSTONE (2.152), it seetns probable that the intended length is that
5 CRUSHED AIDjILLITE
6 SLOW EARTH FLOW IN COLLUVIUM of a chord of the rupture surface (La ), rather than the total
length (L), as shown in Figure 2.1 t. Skempton and Hutchin-
son gave a range of D/LC values of 0.15 to 0.33 for rota-
tional slides in clay and shale, and they stated that slab slides,
which commonly occur in a mantle of weathered or colluvial
material on clayey slopes, rarely if ever have D/L, ratios
greater than 0.1 - Figure 2.33 illustrates such a shallow slab
slide. In a statistical study of the forms of landslides along
the Columbia River valley, Jones, Embody, and Peterson
(2. 79) made extensive use of the horizontal component
(I-IC) or distance from the foot Of the landslide to the
32811 crown, measured in a longitudinal section of the landslide,
and the vertical component (VC) or difference in altitude
between the foot and crown, measured in the same section.
Figure 2.33. Shallow translational slide that developed on shaly
slope in Puente Hills of southern California (2.147). Slide has low
TERMS RELATING TO GEOLOGIC,
D/L ratio (note wrinkles in surface).
GEOMORPHIC. GEOGRAPHIC, OR
CLIMATIC SETTING
25
terms may occasionally be a convenience, it is not recom- ever, landslides take place under the influence of geologic,
mended as a general practice, for the terms themselves are topographic, or climatic factors that are common to large
not informative to a reader who lacks knowledge of the areas. The causes must then be understood if other similar
locality. slides are to be avoided or controlled.
Seldom, if ever, can a landslide be.attributed to a single
TERMS RELATING TO AGE OR definite cause. As clearly shown by Zolotarev (2.196), the
STATE OF ACTIVITY process leading to the development of the slide has its be-
ginning with the formation of the rock itself, when its basic
Active slopes are those that are either currently moving or physical properties are determined, and includes all the sub-
that are suspended, the latter term implying that they are sequent events of crustal movements, erosion, and weather-
not moving at the present time but have moved within the ing. Finally, some action, perhaps trivial, sets a mass of
last cycle of seasons. Active slides are commonly fresh; material in motion downhill. The last action cannot be re-
that is, their morphological features, such as scarps and garded as the only cause, even though it was necessary in
ridges, are easily recognizable as being due to gravitational the chain of events. As Sowers and Sowers(2.161, p.506)
movement, and they have not been significantly modified point out,
by surficial processes of weathering and erosion. However,
in and regions, slides may retain a fresh appearance for In most cases a number of causes exist simultaneously and
many years. so attempting to decide which one finally produced failure
Inactive slopes are those for which there is no evidence is not only difficult but also incorrect. Often the final fac-
that movement has taken place within the last cycle of sea- tor is nothing more than a trigger that set in motion an
sons. They may be dormant, in which the causes of failure earth mass that was already on the verge of failure. Calling
remain and movement may be renewed, or they may be sta the final factor the cause is like calling the match that lit
bilized, in which factors essential to movement have been the fuse that detonated the dynamite that destroyed the
removed naturally or by human activity. Slopes that have building the cause of the disaster.
long-inactive movement are generally modified by erosion
and weathering or may be covered with vegetation so that In this connection, however, the determination of all the
the evidence of the last movement is obscure. They are of- geologic causes of a landslide should not be confused with
ten referred to as fossil (Záruba and Mend, 2.193; Klengel determination of legal responsibility. The interrelations of
and Paek, 2.85; Nossin, 2.118) or ancient (Popov, 2.127) landslide causes are lucidly and graphically presented by
landslides in that they commonly have developed under dif Terzaghi (2.175). His work, that of Sharpe (2i46), Ladd
ferent geomorphological and climatic conditions thousands (2.92), and Bendel (2.6), and that of more recent researchers,
or more years ago and cannot repeat themselves at present. such as Ziruba and Mend (2.193), Skempton and Hutchin-
son (2.154), Krinitzsky and Kolb (2.90), Rapp (2.131),and
FORMING NAMES Legget (2.96) were used in the preparation of this section.
All slides involve the failure of earth materials under
The names applied to slope movements can be made pro- shear stress. The initiation of the process can therefore be
gressively more informative, as more data are obtained,by reviewed according to (a) the factors that contribute to in-
building up a designation from several descriptor words, creased shear stress and (h) the factors that contribute to
each of which has a defined meaning. For example,a slow, low or reduced shear strength. Although a single action,
moist, translational debris slab slide means material moving such as addition of water to a slope, may contribute to
along a planar surface of a little-disturbed mass of frag- both an increase in stress and a decrease in strength, it is
mented material having a D/LC ratio of 0.1 or less, contain- helpful to separate the various physical results of such an
ing some water but none free, and moving at a rate between action. The principal factors contributing to the sliding
1.5 rn/month and 1.5 rn/year (5 ft/month or year). Once of slope-forming materials are outlined in the following
all these particulars are established in the description, the discussion. The operation of many factors is self-evident
movement could be referred to thereafter simply as a de- and needs no lengthy description; some factors are only
bris slide. discussed briefly, or reference is made to literature that
gives examples or treats the subject in detail.
CAUSES OF SLIDING SLOPE
MOVEMENTS Factors That Contribute to
Increased Shear Stress
The processes involved in slides, as well as in other slope
movements, comprise a continuous series of events from Removal of Lateral Support
cause to effect. An engineer faced with a landslide is pri-
marily interested in preventing the harmful effects of the The removal of lateral support is the commonest of all fac-
slide. In many instances the principal cause of the slide tors leading to instability, and it includes the following ac-
cannot be removed, so it may be more economical to alle- tions:
viate the effects continually or intermittently without at-
tempting to remove the cause. Some slides occur in a unique 1. Erosion by (a) streams and rivers, which produce
environment and may last only a few seconds. The damage most natural slopes that are subject to sliding (Hutchinson,
can be repaired, and the cause may be of only academic in- 2.67; Jones, Embody, and Peterson, 2.79; Eyles, 2.43;
terest unless legal actions are to be taken. More often, how- Fleming, Spencer, and Banks, 2.48; California Division of
26
Highways, 2.19), (b) glaciers, which have deeply cut Removal of Underlying Support
and oversteepened many valleys in mountainous regions
that have been the sites of large slides and debris flows Examples of removal of underlying support include
(Plafker, Ericksen, and Fernandez Concha, 2.126), (c)
waves and longshore or tidal currents (Wood ,2.189; Ward, - Undercutting of banks by rivers (California Division
2.186; Hutchinson, 2.71; Koppejan, Van Wamelon, and of Highways, 2.19) and by waves;
Weinberg, 2.89), and (d) subaerial weathering, wetting and Subaerial weathering, wetting and drying, and frost
drying, and frost action; action;
Previous rock fall, slide (Kenney and Drury, 2.81), Subterranean erosion in which soluble material, such
subsidence, or large-scale faulting that create new slopes; as carbonates, salt, or gypsum is removed and granular ma-
and terial beneath firmer material is worked out (Ward, 2.186;
Work of human agencies in which (a) cuts, quarries, Terzaghi, 2.1 74);
pits, and canals (Van Rensburg, 2.181; Piteau, 2.124; Mining and similar actions by human agencies;
Patton, 2J21; Cording, 2.26) are established, (b) retaining Loss of strength or failure in underlying material; and
walls and sheet piling are removed, and (c) lakes and reser- Squ.eezing out of underlying plastic material (Záruba
voirs are created and their levels altered (Muller, 2.112; and Mend, 2.193, pp. 68-78).
Jones, Embody, and Peterson, 2.79; Lane, 2.93; Dupree
and Taucher, 2.39). Lateral Pressure
Surcharge also results from both natural and human agen- Water in cracks and caverns,
cies. The surcharge from natural agencies may be Freezing of water in cracks,
Swelling as a result of hydration of clay or anhydrite,
Weight of rain, hail, snow, and water from springs; and
Accumulation of talus overriding landslide materials; Mobilization of residual stress (Bjerrum, 2.9; Krinitzsky
Collapse of accumulated volcanic material, producing and Kolb, 2.90).
avalanches and debris flows (Francis and others, 2.50);
Vegetation (Gray, 2.53; Pain, 2.120); and Volcanic Processes
Seepage pressures of percolating water.
Stress patterns in volcanic edifices.and crater walls are mod-
The surcharge from human agencies may be ified by general dilation due to inflation or deflation of
magma chambers, fluctuation in lava-lake levels, and in-
Construction of fill; crease in harmonic tremors (Tilling, Koyanagi, and Holcomb,
Stockpiles of ore or rock; 2.178; Moore and Krivoy, 2.109; Fiske and Jackson, 2.47).
Waste piles (Bishop, 2.8; Davies, 2.35; Smalley, 2.156); Factors That Contribute to Low or
Weight of buildings and other structures and trains; Reduced Shear Strength
and
Weight of water from leaking pipelines, sewers, The factors that contribute to low or reduced shear strength
canals, and reservoirs. of rock or soil may be divided into two groups. The first
group includes factors stemming from the initial state or in-
Transitory Earth Stresses herent characteristics of the material. They are part of the
geologic setting that may be favorable to landslides, exhibit
Earthquakes have triggered a great many landslides, both little or no change during the useful life of a structure, and
small and extremely large and disastrous. Their action is may exist for a long period of time without failure. The
complex, involving both an increase in shear stress (hori- second group includes the changing or variable factors that
zontal accelerations may greatly modify the state of stress tend to lower the shear strength of the material.
within slope-forming materials) and, in some instances, a
decrease in shear strength (Seed, 2.143; Morton, 2.110; Initial State
Solonenko, 2.160; Lawson, 2.95; Hansen, 2.57; Newmark,
2.117; Simonett, 2.151; Hadley, 2.55; Gubin, 2.54). Vi- Factors in the initial state of the material that cause low
brations from blasting, machinery, traffic, thunder, and ad- shear strength are composition, texture, and gross structure
jacent slope failures also produce transitory earth stresses. and slope geometry.
Composition
Regional Tilting
Materials are inherently weak or may become weak upon
A progressive increase in the slope angle through regional change in water content or other changes. 'Included espe-
tilting is suspected as contributing to some landslides cially are organic materials, sedimentary clays and shales,
(Terzaghi,2.175). The slope must obviously be on the decomposed rocks, rocks of volcanic tuff that may weather
point of failure for such a small and slow-acting change to to clayey material, and materials composed dominantly of
be effective. soft platy minerals, such as mica, schist, talc, or serpentine.
Texture pressure and friction. Intergranular pressure due to capillary
tension in moist soil is destroyed upon saturation. Simple
The texture is a loose structure of individual particles in softening due to water and suffusion and slaking are dis-
sensitive materials, such as clays, marl, bess, sands of low cussed by Mamulea (2.101).
density, and porous organic matter (Aitchison, 2.2; Bjerrum Changes can occur because of natural actions, such as
and Kenney, 2.11 ; Cabrera and Smalley, 2.17). Roundness rainfall and snowmelt, and because of a host of human ac-
of grain influences strength as compressibility and internal tivities, such as diversion of streams, blockage of drainage,
friction increase with angularity. irrigation and ponding, and clearing of vegetation and de-
forestation. -
Gross Structure and Slope Geometry Crozier (2.30, 2.31), Shroder (2.150), and Spurek (2.163)
discuss the general effect of climate; Temple and Rapp
Included in gross structure and slope geometry are (2.169) Williams and Guy (2.188), Jones (2.78), and So
(2.158), catastrophic rainfall; Conway (2.25), Denness
Discontinuities, such as faults, bedding planes, folia- (2.38), and Piteau (2.124), effect of groundwater; Gray
tion in schist, cleavage, joints, slickensides, and brecciated (2.53), Bailey (2.4), Cleveland (2.22), Rice, Corbett, and
zones (Skempton and Petley, 2.155; Fookes and Wilson, Bailey (2.134), and Swanston (2.165), deforestation; Peck
2.49; Komarnitskii, 2.88; St. John, Sowers, and Weaver, (2.123) and Hirao and Okubo (2.60), correlation of rainfall
2.138; Van Rensburg, 2.181; Jennings and Robertson, and movement;and Shreve (2.148), Voight (2.185), Kent
2.74; Bjerrum and J$rstad, 2.10); (2.82), and Goguel and Pachoud (2.52), gaseous entrain-
Massive beds over weak or plastic materials (Zãruba ment or cushion.
and Mend, 2.193; Nemok, 2.113);
Strata inclined toward free face; Changes in Structure
Alternation of permeable beds, such as sand or sand-
stone, and weak impermeable beds, such as clay or shale Changes in structure may be caused by fissuring of shales
(Henkel, 2.59); and and preconsolidated clays and fracturing and loosening of
Slope orientation (Rice, Corbett, and Bailey, 2.134; rock slopes due to release of vertical or lateral restraints in
Shroder, 2.150). valley walls or cuts (Bjerrum, 2.9; Aisenstein, 2.1; Ferguson,
2.45; Matheson and Thomson, 2.102; Mend, 2.106). Dis-
Changes Due to Weathering and Other turbance or remolding can affect the shear strength of ma-
Physicochemical Reactions terials composed of fine particles, such as bess, dry or sat-
urated loose sand, and sensitive clays (Gubin, 2.54; Youd,
The following changes can occur because of weathering and 2.191; Smalley, 2.156; Mitchell and Markell, 2.108).
other physicochemical reactions:
Miscellaneous Causes
Softening of fissured clays (Skempton,2.153; Sangrey
and Paul, 2.139; Eden and Mitchell, 2.41); Other causes of low shear strength are (a) weakening due
Physical disintegration of granular rocks, such as to progressive creep (Suklje, 2.164; Ter-Stepanian, 2.172;
granite or sandstone, under action of frost or by thermal Trollope,2.180; Piteau,2.124) and actions of tree roots
expansion (Rapp, 2.131); (Feld, 2.44) and burrowing animals.
Hydration of clay minerals in which (a) water is ab.
sorbed by clay minerals and high water contents decrease REFERENCES
cohesion of all clayey soils, (b) montmorillonitic clays swell
and lose cohesion, and (c) bess markedly consolidates upon 2.1 Aisenstcin, B. Some Additional Information About De-
saturation because of destruction of the clay bond between consolidation Fissures of Rock on Steep Slopes. Proc.,
silt particles; 2nd Congress, International Society of Rock Mechanics,
Belgrade, Vol. 3, 1970, pp. 37 1-376.
Base exchange in clays, i.e., influence of exchangeable 2.2 Aitchison, G. D. General Report on Structurally Un-
ions on physical properties of clays (Sangrey and Paul, stable Soils. Proc., 8th International Conference on
2.139; Liebling and Kerr, 2.99; Torrance, 2.179); Soil Mechanics and Foundation Engineering, Moscow,
Migration of water to weathering front under elec- Vol. 3, 1973, pp. 161-1 90.
2.3 Andresen, A., and Bjerrum, L. Slides in Subaqueous
trical potential (Veder, 2.184);
Slopes in Loose Sand and Silt. In Marine Geotechnique
Drying of clays that results in cracks and loss of co- (Richards, A. F., ed.), Univ. of Illinois Press, Urbana,
hesion and allows water to seep in; 1967, pp. 221-239.
Drying of shales that creates cracks on bedding and 2.4 Bailey, R. G. Landslide Hazards Related to Land Use
shear planes and reduces shale to chips, granules, or smaller Planning in Teton National Forest, Northwest Wyoming.
Intermountain Region, U.S. Forest Service, 1971, 131 pp.
particles; and 2.5 Beck, A. C. Gravity Faulting as a Mechanism of Topo-
Removal of cement by solution. graphic Adjustment. New Zealand Journal of Geology
and Geophysics, Vol. 11, No. 1, 1967, pp. 191-1 99.
Changes in Intergranular Forces Due to 2.6 Bendel, L. lngenieurgeologie: Ein Handbuch fuer
Water Content and Pressure in Pores Studium und Praxis. Springer-Verlag, Vienna, Vol. 2,
1948, 832 pp.
and Fractures 2.7 Bishop, A. W. Progressive Failure, With Special Refer-
ence to the Mechanism Causing It. Proc., Geotechnical
Buoyancy in saturated state decreases effective intergranular Conference on Shear Strength Properties of Natural
28
Soils and Rocks, Norwegian Geotechnical Institute, 2.29 Crandell, D. R., and Fahnestock, R. K. Rockfalls and
Oslo, Vol. 2, 1967, pp. 142-150. Avalanches From Little Tahoma Peak on Mount Rainier
2.8 Bishop, A. W. The Stability of Tips and Spoil Heaps. Volcano, Washington. U.S. Geological Survey, Bulletin
Quarterly Journal of Engineering Geology, Vol. 6, No. 1221-A, 1965,30 pp.
4, 1973, pp. 335-376. 2.30 Crozier, M. J. Earthfiow Occurrence During High Inten-
2.9 Bjerrum, L. Progressive Failure in Slopes of Over- sity Rainfall in Eastern Otago (New Zealand). Engineer-
Consolidated Plastic Clay and Clay Shales. Journal of ing Geology, Vol. 3, No. 4, 1969, pp. 325-334.
Soil Mechanics and Foundations Division, American 2.31 Crozier, M. J. Some Problems in the Correlation of
Society of Civil Engineers, Vol. 93, No. SM5, 1967, Landslide Movement and Climate. In International
pp. 3-49. Geography, Proc., 22nd International Geographical
2.10 Bjerrum, L., and Jjrstad, F. A. Stability of Rock Slopes Congress, Montreal, Univ. of Toronto Press, Ontario,
in Norway. Norwegian Geotechnical Institute, PubI. 79, Vol. 1, 1972, pp. 90-93.
1968, pp. 1-11. 2.32 Crozier, M. J. Techniques for the Morphometric Analy-
2.11 Bjerrum, L., and Kenney, T. C. Effect of Structure on sis of Landslips. Zeitschrift fuer Geomorphologie, Vol.
the Shear Behavior of Normally Consolidated Quick Clays. 17, No. 1,1973, pp. 78-101.
Proc., Geotechnical Conference on Shear Strength Proper- 2.33 Cruden, D. M., and Krahn, J. A Reexamination of the
ties of Natural Soils and Rocks, Oslo, Norwegian Geotech- Geology of the Frank Slide. Canadian Geotechnical
nical Institute, Vol. 2, 1967, pp. 19-27. Journal, Vol. 10, No. 4, 1973, pp. 581-591.
2.12 Bjerrum, L., Lken, T., Heiberg, S., and Foster, R. A 2.34 Daido, A. On the Occurrence of Mud-Debris Flow.
Field Study of Factors Responsible for Quick Clay Slides. Bulletin of the Disaster Prevention Research Institute,
Norwegian Geotechnical Institute, PubI. 85, 1971, pp. 17- Kyoto Univ., Japan, Vol. 21,Part 2, No. 187, 1971,
26. pp. 109-1 35.
2.13 Blong, R. J. The Underthrust Slide: An Unusual Type of 2.35 Davies, W. E. Coal Waste Bank Stability. Mining Con-
Mass Movement. Geografiska Annaler, Vol. 53A, No. 1, gress Journal, Vol. 54, No. 7, 1968, pp. 19-24.
1971, pp. 52-58. 2.36 Deere, D. U., and Patton, F. D. Slope Stability in Resid-
2.14 Blong, R. J. A Numerical Classification of Selected Land- ual Soils. Proc., 4th Pan-American Conference on Soil
slides of the Debris Slide-Avalanche-Flow Type. Engineer- Mechanics and Foundation Engineering, San Juan,
ing Geology, Vol. 7, No. 2, 1973, pp. 99-114. American Society of Civil Engineers, New York, Vol. 1,
2.15 Blong, R. J. Relationships Between Morphometric Attri- 1971, pp. 87-170.
butes of Landslides. Zeitschrift fuer Geomorphologie, 2.37 de Freitas, M. H., and Watters, R. J. Some Field Ex-
Supp. Band 18, 1973, pp. 66-77. amples of Toppling Failure. Geotechnique, Vol. 23,
2.16 Bukovansky, M., RodrIquez, M. A., and Cedrôn, G. No. 4, 1973, pp. 495-514.
Three Rock Slides in Stratified and Jointed Rocks. Proc., 2.38 Denness, B. The Reservoir Principle of Mass Movement.
3rd Congress, International Society of Rock Mechanics, Institute of Geological Sciences, United Kingdom, Rept.
Denver, Vol. 2, Part B, 1974, pp. 854-85 8. 72/7, 1972, 13 pp.
2.17 Cabrera, J. G., and Smalley, I. J. Quickclays as Products 2.39 Dupree, H. K., and Taucher, G. J. Bighorn Reservoir
of Glacial Action: A New Approach to Their Nature, Landslides, South-Central Montana. In Rock Mechanics,
Geology, Distribution, and Geotechnical Properties. The American Northwest (Voight, B., ed.), Expedition
Engineering Geology, Vol. 7, No. 2, 1973, pp. 115-133. Guide for 3rd Congress, International Society of Rock
2.18 Caldenius, C., and Lundstrom, R. The Landslide at Surte Mechanics, Denver, Pennsylvania State Univ., 1974, pp.
on the River Gta ilv. Sweden Geologiska undersdkningen, 59-63.
Series Ca, No. 27, 1955, 64 pp. 2.40 Dylik, J. Solifluxion, Congelifluxion, and Related Slope
2.19 California Division of Highways. Bank and Shore Protec- Processes. Geografiska Annaler, Vol. 49A, No. 24, 1967,
tion in California Highway Practice. California Division pp. 167-177.
of Highways, Sacramento, 1960, 423 pp. 2.41 Eden, W. J., and Mitchell, R. J. The Mechanics of Land-
2.20 Campbell, R. H. Soil Slips, Debris Flows, and Rainstorms slides in Leda Clay. Canadian Geotechnical Journal, Vol.
in the Santa Monica Mountains and Vicinity, Southern 7, No. 3, 1970, pp. 285-296.
California. U.S. Geological Survey, Professional Paper 2.42 Embleton, C., and King, C. A. M. Periglacial Geomor-
851,1975,51 pp. phology. Wiley, New York, 1975, 203 pp.
2.21 Carter, R. M. A Discussion and Classification of Sub- 2.43 Eyles, R. J. Mass Movement in Tangoio Conservation
aqueous Mass-Transport With Particular Application to Reserve, Northern Hawke's Bay. Earth Science Journal,
Grain-Flow, Slurry-Flow, and Fluxoturbidites. Earth- Vol.5, No. 2, 1971, pp. 79-91.
Science Reviews, Vol. 11,No. 2, 1975, pp. 145-177. 2.44 Feld, J. Discussion of Slope Stability in Residual Soils
2.22 Cleveland, G. B. Fire + Rain = Mudflows: Big Sur 1972. (Session 2). Proc., 4th Pan-American Conference on Soil
California Geology, Vol. 26, No. 6, 1973, pp. 127-135. Mechanics and Foundation Engineering, San Juan,
2.23 Close, U., and McCormick, E. Where the Mountains American Society of Civil Engineers, New York, 1971,
Walked. National Geographic Magazine, Vol. 41, 1922, Vol. 3,p. 125.
pp. 445-464. 2.45 Ferguson, H. F. Valley Stress Release in the Allegheny
2.24 Cluff, L. S. Peru Earthquake of May 31, 1970: Engi- Plateau. Bulletin, Association of Engineering Geologists,
neering Geology Observations. Seismological Society Vol. 4, No. 1,1967, pp. 63-71.
of America Bulletin, Vol. 61, No. 3,1971, pp. 511-521. 2.46 Fisher, R. V. Features of Coarse-Grained, High-
2.25 Conway, B. W. The Black Ven Landslip, Charmouth, Concentration Fluids and Their Deposits. Journal of
Dorset. Institute of Geological Sciences, United King- Sedimentary Petrology, Vol.41, No. 4, 1971, pp. 916-
dom, Rept. 74/3, 1974,4 pp. 927.
2.26 Cording, E. J., ed. Stability of Rock Slopes. Proc., 13th 2.47 Fiske, R. S., and Jackson, E. D. Orientation and Growth
Symposium on Rock Mechanics, Univ. of Illinois at of Hawaiian Volcanic Rifts: The Effect of Regional
Urbana-Champaign, American Society of Civil Engineers, Structure and Gravitational Stresses. Proc., Royal
New York, 1972, 912 pp. Society of London, Series A, Vol. 329, 1972, pp. 299-
2.27 Corte, A. E. Geocryology and Engineering. In Reviews 326.
in Engineering Geology (Varnes, D. J. and Kiersch, G. A., 2.48 Fleming, R. W., Spencer, G. S., and Banks, D. C. Empir-
eds.), Geological Society of America, Vol. 2, 1969, pp. ical Study of Behavior of Clay Shale Slopes. U.S. Army
119-185. Engineer Nuclear Cratering Group, Technical Rept. 15,
2.28 Coulter, H. W., and Migliaccio, R. R. Effects of the Vols. 1 and 2, 1970, 397 pp.
Earthquake of March 27, 1964, at Valdez, Alaska. U.S. 2.49 Fookes, P. G., and Wilson, D. D. The Geometry of Dis-
Geological Survey, Professional Paper 542-C, 1966, continuities and Slope Failures in Siwalik Clay. Geotech-
36 pp. nique, Vol. 16, No.4, 1966, pp. 305-320.
29
2.50 Francis, P. W., Roobol, M. J., Walker, G. P. L., Cobbold, Vol. 20, No. 4, 1970, pp.412-438.
P. R., and Coward, M. The San Pedro and San Pablo Volca- 2.71 Hutchinson, J. N. The Response of London Clay Cliffs
noes of Northern Chile and Their Hot Avalanche Deposits. to Differing Rates of Toe Erosion. Geologia Applicata
Geologische Rundschau, Vol.63, No. 1,1974, pp.357-388. e ldrogeologia, Vol. 8, Part 1, 1973, pp. 221-239.
2.51 Frühlich, 0. K. General Theory of Stability of Slopes. 2.72 Hutchinson, J. N., and Bhandari, R. K. Undrained Load-
Geotechnique, Vol. 5, No. 1, 1955, pp. 37-47. ing: A Fundamental Mechanism of Mudflows and Other
2.52 Goguel, J., and Pachoud, A. Gêologie et Dynamique de Mass Movements. Geotechnique, Vol. 21, No.4, 1971,
l'écroulement du Mont Granier, dans le Massif de pp. 353-358.
Chartreuse, en Novembre 1248. France Bureau de 2.73 Jahn, A. Slopes Morphological Features Resulting
Recherches Géologiques et Minieres Bulletin, Section From Gravitation. Zeitschrift fuer Geomorphologie,
3—Hydrog6ologie-C6ologie de l'lngenieur, No. 1, 1972, Supp. Band 5, 1964, pp. 59-72.
pp. 29-38. 2.74 Jennings, J. E., and Robertson, A. M. The Stability of
2.53 Gray, D. H. Effects of Forest Clear-Cutting on the Sta- Slopes Cut Into Natural Rock. Proc., 7th International
bility of Natural Slopes. Bulletin, Association of Engi- Conference on Soil Mechanics and Foundations Engi-
neering Geologists, Vol. 7, No. 1-2, 1970, pp. 45-66. neering, Mexico City, Vol. 2, 1969, pp. 585-590.
2.54 Gubin, I. Y. Regularity of Seismic Manifestations of 2.75 Johnson, A. M. Physical Processes in Geology. Freeman,
the Tadzhikistana Territory. USSR Academy of Sciences, Cooper, and Co., San Francisco, 1970, 577 pp.
Moscow, 1960, 463 pp. (in Russian). 2.76 Johnson, A. M., and Rahn, P. H. Mobilization of Debris
2.55 Hadley, J. B. Landslides and Related Phenomena Accom- Flows. Zeitschrift fuer Geomorphologie, Supp. Band 9,
panying the Hebgen Lake Earthquake of August 17, 1959. 1970, pp. 168-186.
U.S. Geological Survey, Professional Paper 435, 1964, pp. 2.77 Johnson, N. M., and Ragel, R. H. Analysis of Flow Char-
107-138. acteristics of Allen II Slide From Aerial Photographs. In
2.56 Hamel, J. V. Kimbley Pit Slope Failure. Proc., 4th Pan- The Great Alaska Earthquake of 1964: Hydrology, Na-
American Conference on Soil Mechanics and Foundation tional Academy of Sciences, Washington, D.C., 1968, pp.
Engineering, San Juan, American Society of Civil Engi- 369-373.
neers, New York, Vol. 2, 1971, pp. 117-1 27. 2.78 Jones, F. 0. Landslides of Rio de Janeiro and the Serra
2.57 Hansen, W. R. Effects of the Earthquake of March 27, das Araras Escarpment, Brazil. U.S. Geological Survey,
1964, at Anchorage, Alaska. U.S. Geological Survey, Professional Paper 697, 1973,42 pp.
Professional Paper 542-A, 1965,68 pp. 2.79 Jones, F. 0., Embody, D. R., and Peterson, W. L. Land-
2.58 Heim, A. Bergsturz und Menschenleben. Fretz and slides Along the Columbia River Valley, Northeastern
Wasmuth Verlag, Zurich, 1932, 218 pp. Washington. U.S. Geological Survey, Professional Paper
2.59 Henkel, D. J. Local Geology and the Stability of Natu- 367, 1961,98 pp.
ral Slopes. Journal of Soil Mechanics and Foundations 2.80 Jordan, R. H. A Florida Landslide. Journal of Geology,
Division, American Society of Civil Engineers, Vol. 93, Vol. 57, No. 4, 1949, pp. 418419.
No. SM4, 1967, pp. 437446. 2.81 Kenney, T. C., and Drury, P. Case Record of the Slope
2.60 Hirao, K., and Okubo, S. Studies on the Occurrence and Failure That Initiated the Retrogressive Quick-Clay Land-
Expansion of Landslides Caused by Ebino-Yoshimatsu slide at Ullensaker, Norway. Geotechnique, Vol. 23, No.
Earthquake. Report of Cooperative Research for Disas- 1, 1973, pp. 3347.
ter Prevention, Japan Science and Technology Agency, 2.82 Kent, P. E. The Transport Mechanism in Catastrophic
No. 26, 1971, pp. 157-189 (in Japanese with Engiish Rock Falls. Journal of Geology, Vol. 74, No. 1, 1966,
summary). pp. 79-83.
2.61 Hoek, E. Recent Rock Slope Stability Research of the 2.83 Kesseli, J. E. Disintegrating Soil Slips of the Coast Ranges
Royal School of Mines, London. Proc., 2nd Interna- of Central California. Journal of Geology, Vol. 51, No. 5,
tional Conference on Stability in Open Pit Mining, Van- 1943, pp. 342-352.
couver, 1971, Society of Mining Engineers, American 2.84 Kjellman, W. Mechanics of Large Swedish Landslips. Geo-
Institute of Mining, Metallurgical and Petroleum Engi- technique, Vol. 5, No. 1, 1955, pp. 74.78.
neers, New York, 1972, pp. 23-46. 2.85 Klengel, K. J., and Paek, J. Zur Terminologie von
2.62 Hofmann, H. The Deformation Process of a Regularly Hangbewegungen. Zeitschrift fuer Angewandte Geologie,
Jointed Discontinuum During Excavation of a Cut. Vol. 20, No. 3,1974, pp. 128.132.
Proc., 2nd Congress, International Society of Rock 2.86 Knapp, G. L., ed. Avalanches, Including Debris Avalanches:
Mechanics, Belgrade, Vol. 3, 1970, pp. 267-273 (in A Bibliography. Water Resources Scientific Information
German). Center, U.S. Department of the Interior, WRSIC 72-216,
2.63 Hofmann, H. Modellversuche zur Hangtektonik. 1972, 87 pp.
Geologische Rundschau, Vol. 62, No. 1, 1973, pp. 16-29. 2.87 Kojan, E., Foggin, G. T.,III, and Rice, R. M. Prediction
2.64 Hollingworth, S. E., and Taylor, J. H. An Outline of the and Analysis of Debris Slide Incidence by Photogram-
Geology of the Kettering District. Proc., Geologists' As- metry: Santa-Ynez-San Rafael Mountains, California.
sociation, Vol. 57, 1946, pp. 204-233. Proc., 24th International Geological Congress, Montreal,
2.65 Hollingworth, S. E., and Taylor, J. H. The Northampton Section 13—Engineering Geology, 1972, pp. 124-131.
Sand Ironstone: Stratigraphy, Structure, and Reserves. 2.88 Komarnitskli, N. I. Zones and Planes of Weakness in
U.K. Geological Survey, Memoirs, 1951, 211 pp. Rocks and Slope Stability. Consultants Bureau, New
2.66 Hsu, K. J. Catastrophic Debris Streams (Sturzstroms) York, 1968, 108 pp. (translated from Russian).
Generated by Rockfalls. Geological Society of America 2.89 Koppejan, A. W., Van Wamelon, B. M., and Weinberg,
Bulletin, Vol. 86, No. 1, 1975, pp. 129-140. L. J. H. Coastal Flow Slides in the Dutch Province of
2.67 Hutchinson, J. N. The Free Degradation of London Zeeland. Proc., 2nd International Conference on Soil
Clay Cliffs. Proc., Geotechnical Conference on Shear Mechanics and Foundations Engineering, Rotterdam,
Strength Properties of Natural Soilsand Rocks, Nor- Vol. 5, 1948, pp. 89-96.
wegian Geotechnical Institute, Olso, Vol. 1, 1967, pp. 2.90 Krinitzsky, E. L., and Kolb,C. R. Geological Influences
113-118. on the Stability of Clay Shale Slopes. Proc., 7th Sympo-
2.68 Hutchinson, J. N. Mass Movement. In The Encyclo- sium on Engineering Geology and Soils Engineering,
pedia of Geomorphology (Fairbridge, R. W., ed.), Moscow, Idaho, Idaho Department of Highways, Univ.
Reinhold Book Corp., New York, 1968, pp. 688-696. of Idaho, and Idaho State Univ., 1969, pp. 160-175.
2.69 Hutchinson, J. N. Field Meeting on the Coastal Land- 2.91 Kurdin, R. D. Classification of Mudflows: Soviet Hy-
slides of Kent, 1-3 July, 1966. Proc., Geologists' Asso- drology, Vol. 12, No. 4, 1973, pp. 310-316.
ciation, Vol. 79, Part 2, 1968, pp. 227-237. 2.92 Ladd, G. E. Landslides, Subsidences, and Rockfalls.
2.70 Hutchinson, J. N. A Coastal Mudflow on the London Proc., American Railway Engineering Association, Vol.
Clay Cliffs at Beltinge, North Kent. Geotechnique, 36, 1935, pp. 1091-1162.
30
2,93 Lane, K. S. Stability of Reservoir Slopes. In Failure Mountains. Proc., 24th International Geological Con-
and Breakage of Rock (Fairhurst, C., ed.), Proc., 8th gress, Montreal, Section 13-Engineering Geology, 1972,
Symposium on Rock Mechanics, American Institute of pp. 132-141.
Mining, Metallurgy and Petroleum Engineers, New York, 2.115 Nemok, A. Gravitational Slope Deformations in the
1967, pp. 321-336. High Mountains of the Slovak Carpathians. Sbornik
2.94 Laverdiêre, C. Quick Clay Flow-Slides in Southern Geologickch véd, Rada HIG, No. 10, 1972, pp. 31-
Quebec. Revue de Geographie de Montreal, Vol. 26, No. 37.
2, 1972, pp. 193-198 (French-English vocabulary). 2.116 Nemëok, A., Pasek, J., and Rybâ, J. Classification of
2.95 Lawson, A. C. The California Earthquake of April 18, Landslides and Other Mass Movements. Rock Mechanics,
1906: Landslides. Carnegie Institute of Washington, Vol.4, No. 2, 1972, pp. 71-78.
Washington, D.C., Vol. 1, Part 1, 1908, reprinted 1969, 2.117 Newmark, N. M. Effects of Earthquakes on Dams and
pp. 384-401. Embankments. Geotechnique, Vol. 15, No. 2, 1965,
2.96 Legget, R. F. Geology and Engineering. McGraw-Hill, pp. 139-159.
New York, 2nd Ed., 1962, 884 pp. 2.118 Nossin, J. J. Landsliding in the Crati Basin, Calabria,
2.97 Leighton, F. B. Landslides and Hillside Development. In Italy, Geologie en Mijnbouw, Vol.51, No. 6, 1972,
Engineering Geology in Southern California, Association pp. 591-607.
of Engineering Geologists, Special PubI., 1966, pp. 149- 2.119 Ostaficzuk, S. Large-Scale Landslides in Northwestern
207. Libya. Acta Geologica Polonica, Vol. 23, No. 2, 1973,
2.98 Lemke, R. W. Effects of the Earthquake of March 27, pp. 231-244.
1964, at Seward, Alaska. U.S. Geological Survey, Pro- 2.120 Pain, C. F. Characteristics and Geomorphic Effects of
fessional Paper 542-E, 1966,43 pp. Earthquake-Initiated Landslides in the Adelbert Range,
2.99 Liebling, R. S., and Kerr, P. F. Observations on Quick Papua New Guinea. Engineering Geology, Vol. 6, No.
Clay. Geological Society of America Bulletin, Vol. 76, 4,1972, pp. 261-274.
No. 8, 1965, pp. 853-878. 2.121 Patton, F. D. Significant Geologic Factors in Rock Slope
2.100 La, K. Y. An Approach to the Problem of Progressive Stability. In Planning Open Pit Mines (Van Rensburg,
Failure. Canadian Geotechnical Journal, Vol. 9, No. P. W. J., ed.), Proc., Open Pit Mining Symposium, Johan-
4, 1972, pp. 407-429. nesburg, South African Institute of Mining and Metallurgy,
2.101 Maniulea, M. A. Suffusion and Slaking: Physical Pro- 1970, pp. 143-151.
cesses Prompting the Mass Movements. Bulletin, Inter- 2.122 Pautre, A., Sabarly, F., and Schneider, B. L'effet
national Association of Engineering Geologists, No. 9, d'echelle dans les 6croulements de falaise. Proc., 3rd
1974, pp. 63-68 (in French with English summary). Congress, International Society of Rock Mechanics,
2.102 Matheson, D. S., and Thomson, S. Geological Implica- Denver, Vol. 2, Part B, 1974, pp. 859-864.
tions of Valley Rebound. Canadian Journal of Earth 2.123 Peck, R. B. Stability of Natural Slopes. Journal of Soil
Sciences, Vol. 10, No. 6, 1973, pp. 961-978. Mechanics and Foundations Division, American Society
2.103 McConnell, R. G., and Brock, R. W. The Great Land- of Civil Engineers, New York, Vol. 93, SM4, 1967, pp.
slide at Frank, Alberta. In Annual Report of the Canada 403-417.
Department of the Interior for the year 1902-03, Sessional 2.124 Piteau, D R. Geological Factors Significant to the Sta-
Paper 25, 1904, pp. 1-17. bility of Slopes Cut in Rock. In Planning Open Pit
2.104 McRoberts, E. C., and Morgenstern, N. R. Stability of Mines (Van Rensburg, P. W. J., ed.), Proc., Open Pit
Thawing Slopes. Canadian Geotechnical Journal, Vol. Mining Symposium, Johannesburg, South African Insti-
11, No.4, 1974, pp.447-469. tute of Mining and Metallurgy, 1970, pp. 33-5 3.
2.105 McRoberts, E. C., and Morgenstern, N. R. The Stability 2.125 Piteau, D. R., Parkes, D. R., McLeod, B. C., and Lou,
of Slopes in Frozen Soil: Mackenzie Valley, N.W.T. J. K. Overturning Rock Slope Failure at Hell's Gate,
Canadian Geotechnical Journal, Vol. 11, No.4, 1974, British Columbia. In Geology and Mechanics of Rock-
pp. 554-573. slides and Avalanches (Voight, B., ed.), Elsevier, New
2.106 Mend, V. EngineeringGeological Importance and Pos- York (in press).
sible Origin of the Stress Relief of the Rocks of the 2.126 Plafker, G., Ericksen, G. E., and Fernandez Concha, J.
Cordillera Blanca, Peru. Bulletin, International Associa- Geological Aspects of the May 31, 1970, Peru Earth-
tion of Engineering Geologists, No. 9, 1974, pp. 69-74. quake. Seismological Society of America Bulletin, Vol..
2.107 Miller, W. J. The Landslide at Point Fermin, California. 61, No. 3, 1971, pp. 543-578.
Scientific Monthly, Vol. 32, No. 5, 1931, pp. 464-469. 2.127 Popov, I. V. A Scheme for the Natural Classification
2.108 Mitchell, R. J., and Markell, A. R. Flowsliding in Sensi- of Landslides. Doklady of the USSR Academy of
tive Soils. Canadian Geotechnical Journal, Vol. 11, No. Sciences, Vol. 54, No. 2, 1946, pp. 157-159.
1, 1974, pp. 11-31. 2.128 Prior, D. B., Stephens, N., and Archer, D. R. Composite
2.109 Moore, J. G., and Krivoy, H. L. The 1962 Flank Erup- Mudflows on the Antrim Coast of Northeast Ireland.
tion of Kilauea Volcano and Structure of the East Rift Geografiska Annaler, Vol. 50A, No. 2, 1968, pp. 65-78.
Zone. Journal of Geophysical Research, Vol. 69, No. 2.129 Prior, D. B., Stephens, N., and Douglas, G. R. Some
10, 1964, pp. 2033-2045. Examples of Modern Debris Flows in Northeast Ireland.
2.110 Morton, D. M. Seismically Triggered Landslides in the Zeitschrift fuer Geomorphologie, Vol. 14, No. 3, 1970,
Area Above the San Fernando Valley. In The San Fer- pp. 276-288.
nando, California, Earthquake of February 9, 1971, 2.130 Radbruch-Hall, D. H. Large-Scale Gravitational Creep
U.S. Geological Survey, Professional Paper 733, 1971, of Rock Masses on Slopes. In Geology and Mechanics
pp. 99-104. of Rockslides and Avalanches (Voight, B., ed.), Elsevier,
2.111 National Research Council, Canada, and National Re- New York (in press).
search Council, U.S. North American Contribution to 2.131 Rapp, A. Recent Development of Mountain Slopes in
2nd International Conference on Permafrost, Yakutsk, Kärkevagge and Surroundings, Northern Scandinavia.
USSR, July 1973. National Academy of Sciences, Wash- Geografiska Annaler, Vol. 42, No. 2-3, 1960, pp. 71-
ington, D.C. 1973, 783 pp. 199.
2.112 Muller, L. New Considerations on the Vaiont Slide. 2.132 Rapp, A. The Debris Slides at Ulvda1, Western Norway:
Felsmechanik und Ingenieur Geologie, Vol. 6, No. 1-2, An Example of Catastrophic Slope Processes in Scandi-
1968, pp. 1-91. navia. Nachrichten der Akademie der Wissenschaften
2.113 Nemöok, A. The Development of Landslides on the Gottingen, Mathematisch-Physikalische Klasse, Vol. 13,
Boundaries of Geological Formations. Sbornik 1963, pp. 195-210.
Geologick'ch véd, Rada HIG, No. 5, 1966, pp. 87- 2.133 Reiche, P. The Toreva-Block: A Distinctive Landslide
105 (in Slovak with English summary). Type. Journal of Geology, Vol. 45, No. 5, 1937, pp.
2.114 Nemêok, A. Gravitational Slope Deformation' in 'High 538-548.
31
2.134 Rice, R. M., Corbett, E. S., and Bailey, R. G. Soil Slips Soils and Rocks, Norwegian Geotechnical Institute, Oslo,
Related to Vegetation, Topography, and Soil in Southern Vol. 2, 1967, pp. 2946.
California. Water Resources Research, Vol. 5, No. 3, 2.156 Smalley, I. J. Boundary Conditions for Flowslides in Fine-
1969, pp. 647-659. Particle Mine Waste Tips. Trans., Institution of Mining
2.135 Romani, F., Lovell, C. W., Jr., and Hair, M. E. Influence and Metallurgy, London, Vol. 81, Sec. A, 1972, pp. A31-
of Progressive Failure on Slope Stability. Journal of Soil A37.
Mechanics and Foundations Division, American Society 2.157 Snopko, L. Study of Deformation Elements Developed
of Civil Engineers, New York, Vol. 98, No. SM1 1, 1972, in the Handlova Landslide. Geologickê Pracê, Vol. 28,
pp. 1209-1223. No. 28, 1963, pp. 169-183.
2.136 Rybâ, J. Slope Deformations in Brown Coal Basins Un- 2.158 So, C. L. Mass Movements Associated With the Rain-
der the Influence of Tectonics. Rock Mechanics, Vol. 3, storm of June 1966 in Hong Kong. Trans., Institute of
No. 3, 1971, pp. 139-158 (in German with English sum- British Geographers, No. 53, 1971, pp. 55-65.
mary). 2.159 Sokolov, N. I. Types of Displacement in Hard Frac-
2.137 Rybái, J., and Dobr, J. Fold Deformations in the North- tured Rocks on Slopes. In The Stability of Slopes
Bohemian Coal Basins. Sbornik geologickch véd, Rada (Popov, I. V., and Kotlov, F. V., eds.), Trans., F. P.
HIG, No. 5, 1966, pp. 133-140. Savarenskii Hydrogeology Laboratory, Vol. 35, 1963,
2.138 St. John, B. J., Sowers, G. F., and Weaver, C. E. Slicken- pp. 69-83; Consultants Bureau, New York (translated
sides in Residual Soils and Their Engineering Significance. from Russian).
Proc., 7th International Conference on Soil Mechanics 2.160 Solonenko, V. P. Seismogenic Destruction of Mountain
and Foundation Engineering, Mexico City, Vol. 2, 1969, Slopes. Proc., 24th International Geological Congress,
pp.591-597. Montreal, Section 13-Engineering Geology, 1972, pp.
2.139 Sangrey, D. A., and Paul, M. J. A Regional Study of Land- 284-290.
sliding Near Ottawa. Canadian Geotechnical Journal, Vol. 2.161 Sowers, G. B., and Sowers, G. F. Introductory Soil
8, No. 2, 1971, pp. 315-335. Mechanics and Foundations. Macmillan, New York,
2.140 Savarensky, F. P. Experimental Construction of a Land- 3rd Ed., 1970,556 pp.
slide Classification. Geologo-Razvedochnyi Institut 2.162 Sowers, G. F. Landslides in Weathered Volcanics in
(TSNIGRI), 1935, pp. 29-37 (in Russian). Puerto Rico. Proc., 4th Pan-American Conference on
2.141 Scott, K. M. Origin and Sedimentology of 1969 Debris Soil Mechanics and Foundation Engineering, San Juan,
Flows Near Glendora, California. U.S. Geological Sur- American Society of Civil Engineers, New York, 1971,
vey, Professional Paper 750-C, 1971, pp. C242-C247. pp. 105-115.
2.142 Scott, R. C., Jr. The Geomorphic Significance of De- 2.163 Spirek, M. Retrospective Analysis of the Climatic Slid-
bris Avalanching in the Appalachian Blue Ridge Moun- ing Agent. Sbornik Geologickch véd, Rada HIG, No.
tains. Univ. of Georgia, PhD dissertation, 1972; Uni- 7, 1970, pp. 61-79 (in Czech with English summary).
versity Microfilms, Ann Arbor, Mich. 2.164 Sukljc, L. A Landslide Due to Long-Term Creep. Proc.,
2.143 Seed, H. B. Landslides During Earthquakes Due to Soil 5th International Conference on Soil Mechanics and
Liquefaction. Journal of Soil Mechanics and Founda- Foundation Engineering, Paris, Vol. 2, 1961, pp. 727-
tions Division, American Society of Civil Engineers, New 735.
York, Vol. 94, No. SMS, 1968, pp. 1053-1122. 2.165 Swanston, D. N. Slope Stability Problems Associated
2.144 Seed, H. B., and Wilson, S. D. The Turnagain Heights With Timber Harvesting in Mountainous Regions of the
Landslide, Anchorage, Alaska. Journal of Soil Mechan- Western United States. U.S. Forest Service, General
ics and Foundations Division, American Society of Civil Technical Rept. PNW-21, 1974, 14 pp.
Engineers, Vol. 93, No. SM4, 1967, pp. 325-353. 2.166 Tabor, R. W. Origin of Ridge-Top Depressions by Large-
2.145 Sharp, R. P. Mass Movements on Mars. In Geology, Scale Creep in the Olympic Mountains, Washington.
Seismicity, and Environmental Impact, Association of Geological Society of America Bulletin, Vol. 82, 1971,
Engineering Geologists, Special PubI., 1973, pp. 115- pp. 1811-1822.
122. 2.167 Takada, Y. On the Landslide Mechanism of the Ter-
2.146 Sharpe, C. F. S. Landslides and Related Phenomena: tiary Type Landslide in the Thaw Time. Bulletin of the
A Study of Mass Movements of Soil and Rock. Colum- Disaster Prevention Research Institute, Kyoto Univ.,
bia Univ. Press, New York, 1938, 137 pp. Japan, Vol. 14, Part 1, 1964, pp. 11-21.
2.147 Shelton, J. S. Geology Illustrated. W. H. Freeman and 2.168 Tavenas, F., Chagnon, J. Y., and La Rochelle, P. The
Co., San Francisco, 1966, 434 pp. Saint-Jean-Vianney Landslide: Observations and Eye-
2.148 Shreve, R. L. The Blackhawk Landslide. Geological witnesses Accounts. Canadian Geotechnical Journal,
Society of America, Memoir 108, 1968,47 pp. Vol. 8, No. 3, 1971, pp. 463478.
2.149 Shreve, R. L. Sherman Landslide. In The Great Alaska 2.169 Temple, P. H., and Rapp, A. Landslides in the Mgeta
Earthquake of 1964: Hydrology. National Academy of Area, Western Uluguru Mountains, Tanzania. Geografiska
Sciences, 1968, pp. 395-401. Annaler, Vol. 54A, No. 34, 1972, pp. 157-193.
2.150 Shroder, J. F. Landslides of Utah. Utah Geological and 2.170 Ter-Stepanian, G. On the Long-Term Stability of Slopes.
Mineralogical Survey Bulletin, No. 90, 1971,51 pp. Norwegian Geotechnical Institute, PubI. 52, 1963, 14 pp.
2.151 Simonett, D. S. Landslide Distribution and Earthquakes 2.171 Ter-Stepanian, G. The Use of Observations of Slope
in the Bewani and Toricelli Mountains, New Guinea: Sta- Deformation for Analysis of Mechanism of Landslides:
tistical Analysis. In Landform Studies From Australia Problems of Geomechanics. Trans., Department of
and New Guinea (Jennings, J. N., and Mabbutt, J. A., eds.), Geomechanics, Armenian SSR Academy of Sciences,
Cambridge Univ. Press, 1967, pp. 64-84. No. 1, 1967, pp. 32-51.
2.152 Skempton, A. W. Soil Mechanics in Relation to Geology. 2.172 Ter-Stepanian, G. Depth Creep of Slopes. Bulletin,
Proc., Yorkshire Geological Society, Vol. 29, Part 1,1953, International Society of Engineering Geologists, No.
pp. 33-62. 9, 1974, pp. 97-102.
2.153 Skempton, A. W. Long-Term Stability of Clay Slopes. 2.173 Ter-Stcpanian,G.,and Goldstein, M. N. Multi-Storied
Geotechnique, Vol. 14, No. 2, 1964, pp. 77-101. Landslides and Strength of Soft Clays. Proc., 7th In-
2.154 Skempton, A. W., and Hutchinson,J. N. Stability of Nat- ternational Conference on Soil Mechanics and Founda-
ural Slopes and Embankment Foundations. Proc., 7th In- tion Engineering, Mexico City, Vol. 2, 1969, pp. 693-
ternational Conference on Soil Mechanics and Founda- 700.
don Engineering, Mexico City, State-of-the Art Vol., 2.174 Terzaghi, K. Earth Slips and Subsidences From Under-
1969, pp. 291-340. ground Erosion. Engineering News-Record, Vol. 107,
2.155 Skempton, A. W., and Petley, D. J. The Strength Along July 16, 1931, pp. 90-92.
Structural Discontinuities in Stiff Clays. Proc., Geotech- 2.175 Terzaghi, K. Mechanism of Landslides. In Application
nical Conference on Shear Strength Properties of Natural of Geology to Engineering Practice (Paige, S., ed.),
32
Geological Society of America, Berkey Vol., 1950, pp. ment of Landslides and Rockfalls as the Basis for the
83-1 23. Theory of Their Study and Prognosis. Géologie de
2.176 Terzaghi, K., andpeck, R. B. Soil Mechanics in Engi- l'Ingénieur, Societe Géologique de Beligique, Liege,
neering Practice. Wiley, New York, 2nd Ed., 1967, 729 pp. 1974, pp. 211-235.
2.177 Thomson, S., and Hayley, D. W. The Little Smoky
Landslide. Canadian Geotechnical Journal, Vol. 12,
No. 3, 1975, pp. 379-392. BIBLIOGRAPHIES
2.178 Tilling, R. I., Koyanagi, R. Y., and Holcomb, R. T.
Rockfall-Seismicity: Correlation With Field Observa- 2.1 Collins, T. Bibliography of Recent Publications on Slope
tions, Makaopuhi Crater, Kilauea Volcano, Hawaii. Stability. Landslide, The Slope Stability Review, Vol. 1,
U.S. Geological Survey, Journal of Research, Vol. 3, No. 1, 1973, pp. 28-37.
No. 3, 1975, pp. 345-361. 2.2 Fisher, C. P., Leith, C. J., and Deal, C. S. An Annotated
2.179 Torrance, J. K. On the Role of Chemistry in the De- Bibliography on Slope Stability and Related Phenomena.
velopment and Behavior of the Sensitive Marine Clays North Carolina State Highway Commission and U.S.
of Canada and Scandinavia. Canadian Geotechnical Bureau of Public Roads, 1965, 89 pp; NTIS, Springfield,
Journal, Vol. 12, No. 3,1975, pp. 326-335. Va., PB 173 029.
2.180 Trollope, D. H. Sequential Failure in Strain-Softening 2.3 Hoek, E. Bibliography on Slope Stability. In Planning
Soils. Proc., 8th International Conference on Soil Me- Open Pit Mines (Van Rensburg, P. W. J., ed.), Proc., Open
chanics and Foundation Engineering, Moscow, Vol. 2, Pit Mining Symposium, Johannesburg, South African In-
Part 2, 1973, pp. 227-232. stitute of Mining and Metallurgy, 1971, pp. 365-388.
2.181 Van Rensburg, P. W. J., ed. Planning Open Pit Mines. 2.4 Holtz, W. G. Bibliography on Landslides and Mudslides.
Proc., Open Pit Mining Symposium, Johannesburg, Building Research Advisory Board, National Academy
South African Institute of Mining and Metallurgy, 1971, of Sciences, Washington, D.C., 1973.
388 pp. 2.5 Knapp, G. L., ed. Avalanches, Including Debris Ava-
2.182 Varnes, D. J. Landslide Types and Processes. In Land- lanches: A Bibliography. Water Resources Scientific In-
slides and Engineering Practice (Eckel, E. B., ed.), HRB, formation Center, U.S. Department of the Interior,
Special Rept. 29, 1958, pp. 20-47. WRSIC 72-216, 1972, 87 pp.
2.183 Varnes, H. D. Landslide Problems of Southwestern 2.6 Larew, H. G., and others. Bibliography on Earth Move-
Colorado. U.S. Geological Survey, Circular 31, 1949, ment. Research Laboratory for Engineering Science,
13 pp. Univ. of Virginia, Charlottesville, 1964, 239 pp.; NTIS,
2.184 Veder, C. Phenomena of the Contact of Soil Mechan- Springfield, Va., AD 641 716.
ics. International Symposium on Landslide Control, 2.7 Spurek, M. Historical Catalogue of Slide Phenomena.
Kyoto and Tokyo, Japan Society of Landslide, 1972, Institute of Geography, Czechoslovak Academy of
pp. 143-1 62 (in English and Japanese). Sciences, Brno, StudiaGeographica 19, 1972, 178 pp.
2.185 Voight, B. Architecture and Mechanics of the Heart 2.8 Tompkin, J. M., and Britt, S. H. Landslides: A Selected
Mountain and South Fork Rockslides. In Rock Mechan- Annotated Bibliography. HRB, Bibliography 10, 1951,
ics, The American Northwest (Voight, B., ed.), Expedi- 47 pp.
tion Guide, 3rd Congress of the International Society of 2.9 Záruba, Q., and Mend, V. Landslides and Their Control:
Rock Mechanics, Pennsylvania State Univ., 1974, pp. Bibliography. Elsevicr, New York, and Academia,
26-36. Prague, 1969, pp. 194-202.
2.186 Ward, W. H. The Stability of Natural Slopes. Geograph-
ical Journal, Vol. 105, No. 5-6, 1945, pp. 170-1 91.
2.187 Washburn, A. L. Periglacial Processes and Environments. PHOTOGRAPH AND
St. Martin's Press, New York, 1973, 320 pp. DRAWING CREDITS
2.188 Williams, G. P. and Guy, H. P. Erosional and Deposi-
tional Aspects of Hurricane Camille in Virginia, 1969.
U.S. Geological Survey, Professional Paper 804, 1973, Figure 2.3 Courtesy of U.S. Bureau of Reclamation
Figure 2.7 F. 0. Jones, U.S. Geological Survey
80 pp.
2.189 Wood, A. M. Engineering Aspects of Coastal Landslides. Figure 2.8 F. 0. Jones, U.S. Geological Survey
Proc., Institution of Civil Engineers, London, Vol. 50, Figure 2.9 Courtesy of Tennessee Department of Transportation
1971, pp. 257-276. Figure 2.11 Courtesy of Tennessee Department of Transportation
2.190 Yemel'ianova, Ye. P. Fundamental Regularities of Land- Figure 2.15 John S. Shelton
slide Processes. Nedra, Moscow, 1972, 308 pp. (excerpts Figure 2.17 Courtesy of New Hampshire Department of Public
translated by D. B. Vitaliano for U.S. Geological Survey). Works and Highways
2.191 Youd, T. L. Liquefaction, Flow, and Associated Ground Figure 2.19 Courtesy of Journal of Geology
Failure. U.S. Geological Survey, Circular 688, 1973, Figure 2.20 F. 0. Jones, U.S. Geological Survey
Figure 2.21 0. Jones, U.S. Geological Survey
12 pp.
2.192 Zâruba, Q. Periglacial Phenomena in the Turnov Region. Figure 2.22 K. Gilbert, U.S. Geological Survey
Sbornik UstednIho Ustavu Geologického, Vol. 19, 1952, Figure 2.23 John S. Shelton
pp. 157-168 (in Czech with English and Russian sum- Figure 2.24 F. 0. Jones, U.S. Geological Survey
maries). Figure 2.25 F. 0. Jones, U.S. Geological Survey
2.193 Zâruba, Q., and Mend, V. Landslides and Their Control. Figure 2.26 F. 0. Jones, U.S. Geological Survey
Elsevier, New York, and Academia, Prague, 1969, 205 pp. Figure 2.28 From Geology Illustrated by John S. Shelton (2.147),
2.194 Zischinsky, U. On the Deformation of High Slopes. W. H. Freeman and Company, copyright 1966
Proc., 1st Congress, International Society of Rock Me- Figure 2.29 From the Encyclopedia of Geomorphology, edited by
chanics, Lisbon, Vol. 2, 1966, pp. 179-1 85. R. W. Fairbridge, copyright 1968 by Litton Educa-
2.195 Zischinsky, U. Uber Bergzerreissung und Talzuschub. tional Publishing, Inc., reprinted by permission of
Geologische Rundschau, Vol. 58, No. 3, 1969, pp. 974- Van Nostrand Reinhold Company
983 Figure 2.33 From Geology Illustrated by John S. Shelton (2.147),
2.196 Zolotarev, G. S. Geological Regularities of the Develop- W. H. Freeman and Company, copyright 1966
33
Chapter 3
Recognition
and
Identif ication
Harold T. Rib and Ta Liang
The adage that half the solution to a problem is the recogni- The use of aerial techniques for evaluating landslides is
tion that a problem exists is especially appropriate to land- emphasized in this chapter because of their proven value
slides. The recognition of the presence or the potential de- and the unique advantages they offer. No other technique
velopment of slope movement and the identification of the can provide a three-dimensional overview of the terrain
type and causes of the movement are important in the de- from which the interrelations existing among slope, drain-
velopment of procedures for-the prevention or correction age, surface cover, rock type and sequence, and human ac-
of a slide. tivities on the landscape can be viewed and evaluated. In
Several basic guidelines developed through years of ex- addition, the availability of new types of aerial imagery,
perience in investigating landslides form the basis for the including satellite, infrared, radar, and microwave radiom-
present-day approach to landslide investigations. etry, extend the advantages of this technique. (In this
chapter, the term imagery is used to describe .those data
Most landslides or potential failures can be predicted collected from sensor systems other than cameras.)
if proper investigations are performed in time.
The cost of preventing landslides is less than the cost TERRAIN EVALUATION FOR
of correcting them except for small slides that can be LANDSLIDE INVESTIGATIONS
handled by normal maintenance procedures.
Massive slides that may cost many times the cost of Basic Factors
the original facility should be avoided in the first place.
The occurrence of the initial slope movement can Recognition and identification of landslides are as complex
lead to additional unstable conditions and movements. as are materials and processes that cause them. The basic
causes of sliding movements are given in Chapter 2. Be-
This chapter discusses techniques for recognizing the cause, as Chapter 2 states,a landslide can seldom be attrib-
presence or potential development of landslides and the ob- uted to a single definite cause, the overall terrain must be
servable features that aid in identifying the types of slope analyzed and the individual factors and the interrelations
movements and their probable causes. The techniques dis- among those factors distinguished before a potential slope
cussed include (a) a review of topographic maps and geo- movement can be recognized and identified. For this pur-
logic, pedologic, and engineering reports and maps; (b) the pose, the factors contributing to slides are more conveniently
analysis of aerial photography and other forms of aerial grouped into categories relating to features that can be de-
images; and (c) preliminary field reconnaissance surveys. lineated on maps and photographsor that can be measured
These three techniques complement one another and to- and quantified, rather than into the mechanisms of failure
getherforni the, basis of the preliminary analysis in a land- listed in Chapter 2. The basic factors considered in evaluat-
slide investigation. The follow-up detailed field investiga- ing the terrain and the major elements included within each
tions required to accurately delineate the landslides or are given in Table 1 and discussed below.
landslide-prone areas, to identify the causative factors, and
to determine the physical and chemical properties required Geologic Factors
for the design of corrective measures are discussed in sub-
sequent chapters. The present-day landscape—including its topography, geo-
34
logic structure, and composition—is the result of millions of susceptible to slope movement because of the nature of
years of development and modification. The topographic their development and the stage of their evolution.
features exposed on the surface are relatively young in age,
but the geologic structures and composition from which Environmental Factors
the features were carved can be quite old. For example,
the structural features that characterize the Rocky Moun-
tains culminated at the close of the Cretaceous period (ap- The development of landforms and the occurrence of slope
proximately 60 million to 70 million years ago), but little movements are greatly affected by environmental factors.
of the topography in that area dates beyond the Pliocene Significantly different landscapes develop from the same
(several million years), and the present canyons and details geologic materials in different climatic zones. For example,
of relief are of Pleistocene or Recent age (less than 1 mil- a limestone bedrock will usually be found as a cliff or ridge
lion years old) (137). Thus, to properly evaluate the in an and area, but will commonly form a low undulating
present-day landscape requires an appreciation of the geo- plain in a humid area. The former situation would be much
logic and climatic changes that occurred during recent geo- more susceptible to landslides than the latter. Similarly,
logic periods (a determinant of topography) as well as an landslides occurring in arid regions are usually distinct and
understanding of the development of the geologic structure easily recognized because of lack of cover, while those oc-
and composition. The science that deals with landscape de- curing in wet tropical climates are weathered, more subdued,
velopment is geomorphology. covered by vegetation, and difficult to discern.
The basic unit of the landscape distinguished in geomor- Variations in microclimate, such as differences in altitude,
phology is the landform. The distinctive development of exposure to moisture-bearing winds, and exposure to sun-
landforms depends on three factors: (a) initial composition light, can cause significant differences in the geomorphic
and structure; (b) processes that act to modify the initial processes. Various observers have reported that some
composition and structure; and (c) stage of development. south-facing slopes of east-west valleys in the northern
A change in any one of these factors will produce a uniquely hemisphere are less steep than adjacent north-facing slopes.
different landform. North-facing slopes have snow cover longer, experience
To evaluate landforms, one must understand their geo- fewer days of freeze and thaw, retain their soil moisture
logic composition and structure including mineralogy and longer, and probably have a better vegetative cover, all of
lithology; physical hardness of the constituent materials which result in less active erosion and steeper slopes (3.37).
and their susceptibility to weathering; mode of deposition Numerous investigators have demonstrated that there is
and subsequent stress history; structural attitude and pres- a relation between slope movement and precipitation.
ence of structural discontinuities and weaknesses such as Záruba and Mend (3.40) evaluated rainfall records for sev-
joints, bedding planes, faults, and folds; and permeability eral different sites in Czechoslovakia covering periods of
of constituent materials or layers. 50 to more than 75 years. They were able to correlate the
The many physical and chemical ways by which the orig- years of heaviest precipitation with the most active periods
inal composition and structure are modified are called geo- of slope movements.
morphic processes. The most important of these processes Sudden changes in the landscape, often represented by
are associated with changes due to (a) actions of water, slope movement, have resulted from catastrophic occur-
wind, or ice; (b) weathering, mass wasting, and erosion; and rences, such as earthquakes, hurricanes, and floods. Figure
(c) diastrophism and vulcanism. 3.1 shows numerous slides that occurred in western Virginia
The modification and eventual destruction of the land- along the path of hurricane Camille in 1969. These slides
forms are considered to occur in stages that geomorpholo- were rapidly documented from the analysis of aerial photo-
gists generally designate as youth, maturity, and old age. graphs taken after the storm (3.39). Other case histories
Qualifying adjectives such as early and late are often used documenting the effect of catastrophic occurrences are
to designate substages. Chronological age is not inferred, referenced in Chapters 1 and 2. Undercutting by stream
but rather a relative stage of development. and wave action and the erosion and mass movement of
The concept of classifying distinct landforms is of prime slopes by seepage, wetting and drying, and frost action are
importance in recognizing and identifying landslides. Ex- also well-known environmental phenomena resulting in
perience has demonstrated that certain landforms are more slope movement.
Environmental Climate and hydrology Rainfall; stream, current, and wave actions; groundwater flow; slope exposure; wetting and
drying; frost action
Catastrophes Earthquakes; volcanic eruptions; hurricanes, typhoons, and tsunamis; flooding; subsidence
Human Human activity Construction; quarrying and mining; stripping of surface cover; over loading, vibrations
Tempora a
aCommon to all categories and factors
35
Human Factors way to initiate preliminary landslide investigations. Geo-
morphologists have divided various regional areas into phys-
Human activities, such as construction, quarrying, and min- iographic regions, that is, regional areas within which the
ing, have caused drastic landscape changes. Much has been method of deposition of earth materials and soils is approx-
said about the aesthetic damage (wastelands and scarred imately the same, the landforms are similar, and the climate
landscapes resulting from strip mining and open-pit quarry- is approximately identical.
ing) that human activities have caused, but far more damag- Early efforts to rate the landslide severity of the various
ingto life and property have been the instabilities occasioned physiographic regions of the United States were reported
by those activities. Oversteepening of slopes, removal of by Baker and Chieruzzi (3.3). Their ratings of landslide se-
support, removal of protective cover, overloading, blockage verity were based on information gathered for the various
of drainage, increasing moisture levels in the ground, and physiographic regions with regard to frequency of occurrence,
vibrations are some of the human activities that have caused size of moving mass, and dollars expended per year. This
slope movements. information was obtained from responses to a questionnaire
by state highway organizations and from various case his-
Temporal Factor tories and published records. Baker and Chieruzzi further
noted in their development of the correlation of landslide
Time is common to all other factors. It is a basic condition severity to physiographic regions that specific geologic for-
of nature to wear away the high areas by weathering, ero- mations were usually associated with the landslides in par-
sion, and mass wasting and to fill in the low areas in a pro- ticular regions. A listing of some of the more common
cess of leveling. Even the occurrence of landslides by cata- landslide -susceptible formations is included in their report.
strophic events normally can be related to conditions that A map by Radbruch-Hall and others (3.24) is a recent
have been building during a period of time. The time pe- effort at rating the severity of landslides in the United States.
riod of importance is the stage of development rather than This map, a portion of which is illustrated in Figure 3.2,
an absolute time period. For example, for a given geologic shows areas of relative incidence of landslides and areas sus-
material, evolution from youth to old age proceeds more ceptible to landslides. The accompanying text discusses
rapidly under tropical wet conditions than under arid con- the slope stability characteristics of the physiographic re-
ditions. Thus, in the same elapsed time, the landform unit gions of the United States and the geologic formations and
in a tropical wet area will be much further advanced than a geologic conditions that favor landsliding in the various
similar unit in an and area. provinces.
The stage of development achieved in a given area is ex- Figure 3.2, together with the information provided by
pressed by its topographic characteristics. Thornbury Radbruch-Hall and others and by Baker and Chieruzzi, are
(3.37) lists typical landscape characteristics for each stage offered only as a guide for use in preliminary evaluation of
of development for various idealized geomorphic cycles. landslide potential from a regional concept. They are not
The following major topographic features are typical of intended to depict precise boundaries and conditions. The
the respective stages of development in the common fluvial delineation of areas of low incidence means not that exten-
geomorphic cycle. sive landslides do not occur in those areas but that they are
negligible in comparison with occurrences and magnitudes
I. In youth, topography is relatively undissected, and of those in the other areas. It should also be recognized
only a few streams exist. Interstream tracts are extensive that, even though as noted in these reports certain geologic
and poorly drained. Valleys have V-shaped cross sections, formations are normally associated with landslides, no for-
and their depths depend on the altitude of the region. mation has developed slides throughout the entire extent
Waterfalls or rapids exist where streams cross resistant rock, of its outcrop area. A more detailed investigation is re-
In maturity, a well-integrated drainage system de- quired to pinpoint the actual slides or vulnerable locations.
velops. Topography consists mostly of hillsides and valley-
sides. Drainage divides are sharp, and the maximum possi- Landforms Susceptible to
ble relief exists. Vertical cutting ceases, and lateral destruc- Landslides
tion becomes important.
In old age, valleys are extremely broad and gently The designation of a specific landform connotes both a ge-
sloping both laterally and longitudinally. Development of netic classification and a type of landscape. For example,
floodplains is considerable, and stream meandering prevails. a sand dune landform denotes deposits formed by wind
Interstream areas have been reduced in height, and stream movement and sorting, which form unconsolidated, smooth,
divides are not so sharp as in maturity. flowing hills and ridges. An appreciation of the genetic as-
pects of landforms enables one to estimate their potential
According to Sharpe (3.29) and others, landslides are most susceptibility for movement. The type of landscape of each
prevalent during youth and the transition into maturity landform provides a basis for separating the various land-
when valley walls are steepest and down-cutting the most forms and thus recognizing those most prone to sliding.
active. Describing the genetic characteristics of landforms is be-
yond the scope of this presentation; however, excellent de-
Regional Approach to Landslide scriptions can be found in textbooks on geomorphology
Investigations (3.14, 3.37). In this chapter, landscape characteristics of
landforms are used as the basis for recognizing landslides
Analyzing the regional geology and terrain is the appropriate and landslide-prone areas.
I1
Figure 3.1. Catastrophic debris avalanches in Nelson County. Virginia, after torrential rains associated with hurricane Camille dumped
69cm (27 in) of water August 19-20, 1969. Light-toned scars on tree-covered hillsides indicate locations of debris avalanches.
Light-toned bands along stream floors are veneers of rubble and debris slide material.
Figure 3.2 Portion of USGS preliminary landslide overview map of Landslides can occur in almost any land iorm If the con-
the conterminous United States (3.24). ditions are right (e.g., steep slopes, high moisture level, no
vegetative cover). Conversely, landslides may not occur On
the most landslide-susceptible terrain if certain conditions
are not present (e.g., clay shales on flat slopes with low
moisture levels). Experience in observing and working with
various landforms, however, has demonstrated that land-
slides are common in some landforms and rare in others.
Table 3.2 provides a key to landforms and their susceptibil-
ity to landslides. The subdivisions are based on topographic
expression and, in the case of hilly terrains, also on drainage
patterns. This table gives only those landforms in which
landslides are most common and is not meant to be all in-
clusive. Illustrations of some of these landforms and a
brief description of their landscape characteristics are in-
chided later in this chapter in the section on landforms
susceptible to landslides. Almost all landforms rated as
highly susceptible to landslides are composed of alternate
layers of pervious and impervious materials (rock or soil),
a fact that needs to be specifically recognized.
Vulnerable Locations
The discussion of landslide-susceptible terrain to this point
has progressed (a) from a regional concept in which the Se-
verities of landslide occurrence for various physiographic
regions in the United States were indicated, (b) through
the listing of geologic formations in the United States
where landslides are common, and (c) to the rating of in-
AREA OF HIGH
dividual landforms as to their susceptibility to landslides.
4Z AREA OF HIGH
LANDSLIDE INCIDENCE
LANDSLIDE
SUSCEPTIBILITY
AREA OF MODERATE
This progression from the general, regional overview to
specific landforms can be carried one step further. Within
AREA OF MODERATE LANDSLIDE the susceptible landforms are certain natural, vulnerable
I tj LANDSLIDE INCIDENCE SUSCEPTIBILITY
locations that are conducive to sliding. Typical vulnerable
[ IAREA OFLOW
locations include areas of steep slopes, cliffs or banks being
_]
LANDSLIDE INCIDENCE undercut by stream or wave action, areas of drainage con-
37
centration and seepage zones, areas of hummocky ground, several factors, such as a sudden heavy rainfall (Figure 3.1)
and areas of fracture and fault concentrations. Special at- or an excavation at the toe of the slope (Figure 3.3) may
tention should be directed to those locations when maps or result in sliding of the overlying soil mass. A study of cut-
aerial photographs are examined and field studies are per- slope failures in North Carolina (3.12) reported that about
formed. In addition, areas that have recently slid require two-thirds of the slides occurred in weathered soil mate-
immediate and close scrutiny because additional movement rials and one-third occurred in rock slopes.
may occur.
Cliffs and Banks Undercut by Streams
Steep Slopes or Waves
If slopes are steep enough,movement can occur on any land- Landslides are common in cliffs or banks that are subject
form. However, on landforms highly susceptible to land- to attack by streams or waves. If the banks are made up of
slides, other factors being equal, the steepest slopes are the soil or other unconsolidated materials, the weakest (and
most vulnerable locations. Only slopes of similar materials hence the most favorable) slide position is often located at
should be compared. For example, a slope cut in earth or the point of maximum curvature of the stream. At this
talus should not be compared with a rock cliff in an adja- point, the bank receives the greatest impact from the water.
cent landform, and slopes in bedrock generally are more In areas of rock outcrops, on the other hand, the exposure
stable, even though steeper, than slopes in adjacent soil at and near the point of maximum curvature is often hard
areas. rock, and the weak spots are to be found upstream and
The most common cause of the large number of slides downstream of this point. These conditions are shown in
that occur on steep slopes is residual or colluvial soils slid- Plate 3.1 and Figure 3.23.
ing on a bedrock surface. The loose, unconsolidated soils Many landslides occur along the edges of oceans and
cannot maintainas steep a slope as the underlying rock sur- lakes because of undercutting by waves. Locating the point
face and are, consequently, in a delicate balance. Any of of maximum water impact is more complex and difficult
38
Figure 3.3. Slides (1) caused by
undercutting of slopes by coal
stripping operations in Washington
County. Ohio (3.22).
- ;.c
., ,—
1;, '
.-_.I ' - 11•i '-
along lake and ocean shores than along stream banks. Fac- sometimes aided by the identification of near-surface chan-
tors to be considered include shape and slope of the shore- nels, wet areas, tall vegetation on the slope, and displaced
line, direction of wave action, and frequency and magni- or broken roads adjacent to the slope. Delineating the drain-
tude of storms producing large waves. Data obtained at age network, especially the presence of seeps or springs, is
different periods of time are often of value in the analysis extremely important for planning new construction. Many
of these factors.
Figure 3.4. Oblique aerial view of extensive seepage zones (1) on
Areas of Drainage Concentration scarp face of massive landslide in Kittitas County, Washington.
and Seepage Seepage was major cause of slide in these unconsolidated sediments.
Remnant of dry sand flow (2) is seen above scarp face and is re-
sult of seepage at higher level. Landslide disrupted highway at base
A survey conducted by the Federal Highway Administration
of slope and water canal carried through slope.
of major landslides on the federal-aid highway system in the
United States revealed that "water is the controlling or a
major contributing factor in about 95 percent of all land-
slides" (3.5). Thus, careful study of the drainage network
and areas of concentration or outfall of water is extremely
important. Close scrutiny of existing slide scars often indi-
cates that a line connecting the scars points to drainage
channels on higher ground. Such drainage may appear as
seepage water, which is responsible for the damage. An ex-
ample of this condition is shown in Figure 3.4
Seepage with subsequent sliding is likely to occur in
areas below ponded depressions, reservoirs, irrigation canals,
and diverted surface channels (Figure 3.4). Such circum-
stances are sometimes overlooked on the ground because
the water sources may be far above the landslide itself,but
they become obvious in aerial photographs. The importance
of recognizing the potential danger in areas below diverted
surface drainage, especially in fractured and porous rocks,
needs particular emphasis. Extensive field experience has
proved repeatedly that within an unstable area one of the
most dangerous sections is the lower part of an interstream
divide through which surface water seeps from the higher
to the lower stream bed. The recognition of seepage is
39
Figure 3.5. Landslide in highway
fill section of Ohio-22 in Jefferson
County. Fill slid en masse as
block slide with fill above block
slumping into resultant hole as dNL
aben. Ohio Department of
Transportation staff report
attributed failure to saturation
of colluvial materials on side
slopes beneath fill by springs
outcropping on hillsides. Few
small drainageways commencing
part way down slope (1) and
small slide (2) on natural hillside
indicate presence of seepage
zones and instability of natural
slopes.
highway fills have failed because the natural drainage was ing should not discourage construction unconditionally, be-
blocked by the fill and allowance Was not made for drain- cause the unstable condition of the past may not necessarily
age. Figure 3.5 shows such a situation. exist today. In some parts of the western United States,
for example, railroads built in extensive old landslide areas
Areas of Hummocky Ground
have been stable for a long time. Nevertheless, special care
The presence of huminocky ground whose characteristics should be taken in construction on old slides. Figure 3.6
are inconsistent with those of the general regional slopes shows aerial photographs of the construction of a road on
an existing slide and the consequences.
and the presence of a scan) surface (sometimes not very dis-
tinct) at a higher elevation are often indications of an exist- Areas of Concentration of Fractures
ing landslide. The older the landslide is, the more estab- and Bedding Planes
lished the drainage and vegetation become on the slide mass.
The drainage and vegetation thus help in determining the Movements of slopes may be structurally controlled by sur-
relative age and stability of the slide. faces or planes of weaknesses, such as faults, joints, bedding
Once an old landslide is found, it serves as a warning that planes, and foliation. These structural features can divide a
the general area has been unstable in the past and that new rock mass into a number of individual units, which may act
disturbances may start new slides. However, such a warn- independently of one another. The result can be an incor-
40
Figure 3.6. Before (top) and '
after (bottom) construction -
of Ohio78 in Monroe
County on unstable ground
(3.22). Before construction,
presence of hummocky
I AN
.V
,. -.'. -' . q• ,
.--2
Vk
— ?'
--• T /1 it
\
'
4-4
(
rect slope design because the designer considered the rock very steep slope. The scarp face rapidly ret rogrades uphill
to be one continuous mass rather than a series of individual by continued slumping until a more stable condition occurs.
blocks. These planes of weakness also provide egress for Thus, a new landslide should be investigated as soon as pos-
water and vegetation, which further weaken the individual sible not only to determine corrective measures but also to
units by wedging action, frost heave, and reduction of slid- look for evidence of possible continued movement. The
ing friction. A careful search should be made to locate most significant sign of possible further instability is the
areas with close spacing of faults and joints, especially presence of cracks on the crown of the slide. Figure 3.7
where they cross and divide the rock mass into smaller (top) shows some telltale signs at a recent slide. Figure 3.7
blocks. (bottom) shows the slide area 5 months later. Additional
movement occurred in the area where the telltale signs were
Recent Landslides evident in the earlier photograph.
The occurrence of a landslide does not mean that final ad- Procedure for Preliminary
justments to the unstable conditions have occurred and no Investigations of Landslides
further movement will occur. In many cases, especially in
unconsolidated deposits, the materials present in the scarp Not only must the presence or potential development of
face remain in an unstable condition because they are on a landslides be recognized, but the types and causes of
41
Figure 3.7. Sequential
photography for use in I
evaluation of recent slide
development along US-95 in
Idaho County, Idaho, and
4
investigation of possible further
movements. Aerial photography
(top) taken in February 1974
thortly after occurrence of
slide (1) shows features present
at crown of slide that forewarn
of further movement fissures
- IF .
-
in photograph above
40
lie
-.. p
I,
' IF
.:.4
movement must be identified so that preventive or correc- A typical procedure, which has been developed over the
tive action can be taken. Chapter 2, and specifically Figure years, for performing terrain investigations is used for a va-
2.1 ,gives the classification of the various types of landslides riety of programs, such as engineering soil surveys, location
based on the dominant types of movement and the types of of construction materials, and landslide investigations. The
materials involved. In this chapter, the emphasis is on how following steps are generally performed. Some excellent
those various types of slides appear on the various data case studies demonstrating the application of this step-by-
sources used in terrain investigations, step procedure are reported by Mintzer and Struble (3.20).
42
Obtain aerial photography and other special coverage. the terrain as to landslide susceptibility is included.
Small-scale photography for regional overview and large- Maps have certain inherent advantages for landslide
scale coverage for detailed study are obtained from available studies. They are prepared at a uniform scale and, there-
sources. If landslides are already present, photographs taken fore, are subject to direct quantitative measurements. Fur-
both before and after the occurrence should be obtained to thermore, planimetric information is often included, thereby
aid in locating causative factors that may have been obliter- minimizing uncertainties in superimposing other informa-
ated by the slide. Special types of photography should be tion. Their disadvantage is that they are outdated as soon
acquired as needed (the section on use of aerial photog- as they are published. Unless the maps are updated period-
raphy gives details on types and scales of photography ically or no changes occur, the maps will not show the most
needed). recent terrain and cultural features.
Review literature and maps. A review is made of ex- The various maps discussed in this section are published
isting topographic maps, geologic maps and reports, water mainly by governmental and public agencies and may be ob-
well logs, agricultural soil survey reports, and other litera- tained free or at nominal costs. The published maps can
ture to develop an area] concept of the area under investiga- also be viewed at many governmental and university li-
tion. braries and in some large public libraries. Included in the
Analyze the photography and other special coverage. following discussions of the various types of maps is infor-
The patterns on the photographs are analyzed, and land- mation on map sources.
forms are identified and related to the areal concept de-
veloped from the literature review. A careful examination Topographic Maps
is made of all vulnerable locations. Existing and potential
landslide sites are delineated, and a three-dimensional con- Topographic maps show the size, shape, and distribution of
cept of the terrain is developed. In this step, a mosaic is features of the surface of the earth. These maps depict the
usually prepared and landform and drainage maps are de- features of relief, drainage, vegetation, and culture, usually
veloped. Sites that offer the best opportunities for confirm- in a color format. Some of the more recent topographic
ing or extending the information developed are selected for maps prepared in the United States use orthophotographs
field verification. as the base for depicting drainage and culture as they actu-
Perform field reconnaissance. A field reconnaissance ally appear and superimpose contours on the orthophoto-
of the area is performed to verify the three-dimensional con- graphs to depict the relief; these are published as ortho-
cept developed in the earlier steps, to fill in information in photomaps.
questionable areas, and to observe the surface features and Major landslide areas that are clearly evident are some-
details that could not be determined from other data times labeled on topographic maps. On some maps, the
sources. The type of landslide movement is classified from boundaries of the slide and arrows pointing toward the di-
the available data and from a study of the surface features rection of movement are also shown. Small slides, the types
and crack patterns observed in the field. more commonly encountered in highway and other engi-
Conduct final analysis and plan field investigations. neering works, are not usually labeled on such maps. Iden-
A final analysis of the photography is performed based on tification of these smaller slides or unlabeled larger slides
the results obtained from the field reconnaissance. A deter- can be accomplished by noting the following features on
mination is made as to what additional information is topographic maps (Figure 3.8):
needed to fully define the three-dimensional model of the Topographic expression observable, for example,
site and what samples and test data are required to design steep slope (closely spaced contours) at scarp head of slide,
the corrective procedures. Based on those needs, the field hummocky topography in slide mass (irregular nonsymmet-
investigation program is planned (this process is described rical contour patterns with shallow depressions), and pres-
fully in Chapter 4). ence of detached mass and flow characteristics at the lower
end;
MAP TECHNIQUES FOR Wavy contour lines, uneven or broken local roads,
LANDSLIDE DETECTION and other artificial lineaments such as transmission lines;
and
The acquisition and the analysis of various types of maps Minor movements or irregularities at "vulnerable lo-
constitute one of the first steps in landslide investigations. cations," as discussed in the section in this chapter on vul-
Maps depicting topography, geology, agricultural soils, and
nerable locations.
other special terrain and cultural features are available for
many areas of the world. The maps represent, in a two- The potential for identifying landslides on topographic
dimensional format, the authors' interpretations of terrain maps is essentially limited by the scale and contour interval
and cultural features determined from field investigations, of the maps. On small-scale maps, even moderate-sized
a study of available literature and photography, and map- landslides significant to engineering works may be such
making procedures. The nature and quality of the informa- microfeatures that they are not identifiable. The U.S. Geo-
tion relating to the presence of landslides or potential for logical Survey (USGS) publishes topographic maps at various
landslides that can be derived from existing maps depend scales from 1:20 000 to 1:1 000 000. Indexes to the status
on the purpose, type, scale, and detail used in preparing the of topographic mapping for the United States are available
map. From most maps, only a general indication of land- free from USGS. Information on status of mapping in
slide susceptibility can be derived. In recent large-scale other countries can be obtained from similar mapping or-
maps, more detail on locating existing landslides and rating ganizationsin the respective countries.
43
Figure 3.8. Portion of USGS topographic map of Laguna Quadrangle, New Mexico, on which typical landslide characteristics described in
text are evident (1). Figure 3.31, which shows landscape within same region but not same area, illustrates appearance of this landslide
topography on aerial photograph.
C U BE R.ô—
GJANT.
Geologic Maps and Reports particularly when surface and near-surface fo rnial ions are
of significance. Bedrock geology maps show only the bed-
The principal sources of geologic information in the United rock and do not indicate surficial deposits. These maps
States are the maps and publications of USGS and the state must be used with care to avoid erroneous assumptions
geologic surveys. Government geologic maps range from about surface conditions. For landslide investigations, bed-
general, small-scale maps covering en tire continents to rock geology maps are useful when lit tie or no soil overlics
large-scale maps covering counties and quadrangles. Not the bedrock or when slides are deep-seated.
all the geologic literature published annually. however, is Other maps of special importance for landslide investi-
found in government publications. Numerous articles are gations are the USGS Miscellaneous Investigation Series
published by geologic organizations. whose publications and Miscellaneous Field Studies Series. Maps are now being
are referenced in two periodicals: Bibliography ojNorth produced that directly indicate landslides and rate the va-
American Geology, published by USGS, for areas in North riotis terrain features as to their susceptibility to sliding.
America, the 1-lawa iia n I sian (Is. and Guam: and Bibliography Figure 3.10 shows a portion of this type of map. Other
and Index of Geology Exclusive of Non/i A inerica, pub- maps in these series indicate items or features closely re-
lished by the Geological Society of America. for other areas. lated to landslides, such as engineering geology, \vater re-
In addition, USGS publishes iiidexcs of- ologic maps and sources, and groundwater. These maps are also of some
listings of maps and reports for each of the states. value in landslide investigations. Unfortunately, most of
The most extensive fo mi of geologic maps published in the maps in these miscellaneous series are not well catalogued
the United States is the quadrangle map series, which in- and are difficult to locate.
cludes surficial geology maps, bedrock geology maps, and Even though geologic maps and reports may not specifi-
standard geologic maps depicting both surface and bedrock cally indicate the presence of landslides or landslide-
units present in the map area. Although the recent quad- susceptible terrain, this informal ion can be extracted from
rangle maps show landslide deposits, only relatively large the geologic literature by noting the physiographic regions,
slides are shown because of the small scale and large con- the geologic format ions mapped, the topogra hic a it d strati-
tour intervals. An example of this type of map is shown in graphic setting, and the landforms present. As discussed in
Figure 3.9. The surficial geology maps show only the sur- the earlier section on regional approach to landslide invest i-
face materials present and are usually limited to areas ex- gations. certain physiographic regions and geologic forma-
tensively covered by complex surficial deposits, such as in tions are noted for landslides. The presence of these suscep-
glaciated areas. These maps are useful for landslide st udies. tible format ions can be determined from the geologic lite ra-
44
Figure 3.9. Black-andwhite repoduction of a portion of USGS geologic quadrangle map GQ-970, Tyler Peak Quadrangle, Washington.
Slide areas are identified by map symbol 01 and are outlined by dashed line. On original color map, surficial deposits including slide areas
are shown in yellow and are more easily distinguishable.
-
I-
\ •;:\
Qsu Ql 1 f1
Q.l
L QOrn
Surficial deposits
Qsu, undil'ided .uo:ficial deposits
QI, landslde deposits
Qym, young alpine moraine deposit; pattern
denotes morai -ne crest
Qal, alluvium
Qt. talus
Qc, continental glacial deposits
Qom, old alpine moraine deposit; pattern
deiwtes moraine crest
7_F -- -4D'3? 30
45
Figure 3.11. Portion of soil survey report for Chilton County, Alabama (3.18). Information on susceptibility to sliding in situ is indicated in column, Soil Features Affecting Highway
Location. Presence of possible seepage zonesis also designated in this column. Some indication of slide susceptibility of soils placed in fill or embankment section is indicated in columns,
Road Fill Material or Farm Ponds, Embankments.
Masada: MnA --------------------- - Good to lair -------- - Poor: no sand or Fair to good------- - Oecnsii,sal flooding--------elan ,reiage.......' Fair stability------- - No dr:iinage needed; Sloi.,'s 010102 pert-en?; No terraces or No wn(erw&y*.
gravel availobk-. l,edroek at a o,o,l.'r;tle intake; diversions oeodsd,
• depth of 30 In 50 I eeadrratc asnilable seeded,
inches. water capacity.
Myatt: Ml. MpA------------------ Foir---------------- Pane: no sand Sr Fair: high water high seater t5lsle----------.'lOO'St'5lt5ge - - - - Fair to good 111gb n'atvr tattle; : Sit.15,'. of 0 to 2 percent; No ierraeeo ne No o'aterwayn
For Bibb part s.f MyA, refer to gravel available, tattle; th,oding. viability. . e,otivrate to toad. : locate to elan' diversions seeded.
Blbbneeies. ' I i croivlyoioo'
prr.neabrlrty;
.ske; tone soil for
crops.
needed.
outlets difficult
to eatablish.
Ora: OeA.OrB. 0e62. 0cC. OrC2... Good ---- ---------- Poor: no sand oe Good- ------------- - Seepngn in eats aloof Moderate seepage.) Fair nhearseresgth; No drainage nrcdcd.. Slopes of 010 10 inches; Moderate erodi.
gravel available. Modrato erodi.
the lrogspan layer. fate stability. . ssodvraieintcke; bility. bitity; fe*glpso.
t,.enteotsilty; medium
os'ailablo water
capacity.
Rock land: Ru.
No Interpretations; properties too
senolable.
nflo,ton. R.A. RoB. RaC. RsC2. Good ............... Poor: no sand or Good ------------- - No limitation, except Moderate In rapid : Moderate shear No drainage needed,,' Slops's of 010 IS inches; Moderate erodi-
R,D2, RuE. RIP, gravel available, elopes. seepage. , strength; pone to Moderate erodt.
,,ndeeatn to ropid bility. bility.
For Bhubuta nnd Troop to parts of . lose stability, irtalse; i,,cdiam
RIE and RIP, refe, Shabuta : available a'oter
and Troop melee, rapacity. -
Baffell: SuB. SaC. S.D .............. Fair to poor: Fair to poor: us- Good ............... No listilations except .Rapid seepage. .. Poor to fair otability. No drainage needed,,' Slopes of 210 IS inehes;
grave ly. dcrinto b- sand Moderate eeodt. Gravelly; lIttle
. elopes. notlrrate to ropid
and gravel at a I i bility. motstare.
, lislatin; medium to
depth of 4 toO '
feet; grovel gand
for road eon.
I I
-.
lose available .tcr
capacity,
otroctina tnatcriol.
8hubuIn: ShD2 ..................... Fair to good......... Poor: no sand or . Fair: high clay Sosceptible to sliding; Slow seepage.-..... Lose shear strength; No drainage aeedrd. Slopes of 2 to 15 inches' I High crodibility..
gravel available---- contest; very seepage in cttts; . lair to poor Highcoodiblltty.
moderate to slow -
plastic, moderate sheiol.. , ntohility. 'itstake; medium
well poteotiat. ' available macor -
capacity,
ture. In addition, as noted in the sections oil landforms sus- provide lists for the respective states. For copies of the pub-
ceptible to landslides and on vulnerable locations, certain lished reports, the unpublished field sheets, and the most
landlorms and topographic settings are especially suscep- up-to-date information about surveys in progress, one should
tible to landslides. For example, one of the most favorable check with the local soil conservat ion service office in the
settings for landslides is the presence of permeable or solu- county of interest. The Canada Department of Agriculture
ble beds overlying or interbedded with relatively impervious provides lists of Canadian soil survey maps and reports.
beds in an elevated topographic setting. The beds could be Sources of information on soil surveys in foreign countries
either rock or soil. Conditions such as this can be deter- include World Soil Map Office, SCS: libraries in major ag-
mined from the geologic literat ure. ricultural colleges and universities: and departments of agri-
culture in other countries.
Agricultural Soil Survey Reports
Special Maps and Reports
The Soil Conservation Service (SCS). U.S. Department
of Agriculture, in cooperation with state agricultural experi- The susceptibility of the terrain to landslides is also dis-
ment stations and other federal and state agencies, has per- cussed in special maps and reports prepared by various or-
formed and published soil surveys since 1899. The agricul- ganizations. Included in this group are engineering soil maps,
tural areas in the United States are, therefore, covered by engineering soil surveys, geologic reconnaissance surveys, and
fairly detailed soil surveys, and much of the remaining area engineering guides to agricultural soil series.
is mapped in less detail. Similarly, agricultural areas in Engineering soil maps depict the relation of the landforni
other countries are covered by soil surveys with varying and engineering soils. The lanclform - which const it utes the
degrees of detail and are available in published form. foundation of this mapping, is subdivided on the basis of
Although SCS soil survey reports are compiled primar- its engineering characteristics, such as soil textuic, drainage
ily for agricultural purposes, those published since 1957 are conditions, and slope category. Although these reports gen-
of special value for landslide investigations. Most of these erally do not iidicate landslide conditions, estimates of
soil surveys contain information on the engineering uses of landslide susceptibility can be determined by evaluating the
the soils. In addition, most of the soil maps published in landfornl types, soil characteristics, and slope conditions.
these reports are printed on a photomosaic base at a scale In the United States, engineering soil mapping is performed
of I : 15 840 or 1:20 000, and the soil information is super- at a variety of regional levels. For example, New Jersey and
imposed on the mosaic. This provides for easier identifica- Rhode Island are completely covered: Indiana and Tennessee
tion of natural and cultural features. are preparing maps on a county basis: and Kansas, Louisiana.
Soil surveys provide a three-dimensional concept of the and New Mexico prepare maps on a project-by-project basis.
land and contain information about the areal extent and For most engineering projects, preliminary engineering
vertical profile for each soil unit. The insight given of each soil surveys and geologic reconnaissance surveys are per-
soil unit and its relation to adjacent soil units enables one formed. In many of these surveys, landslides or landslide-
to estimate the environmental setting and potential stabil- susceptible areas are depicted prior to construction. Figure
ity. 3.12 shows ,- in example of' such an investigation performed
The engineering sections included in the recent reports for a route location in Kansas. The results of the survey
contain engineering test data for the major soils mapped in influenced the final location of the highway (3.31).
the counties: a table of estimated engineering properties of Since large parts of the United States are covered by ag-
the soils, including depth to water, depth to bedrock, typi- ricultural soil surveys and extensive data are available for
cal profile, and grain-size distribution: a table of engineer- the basic soil units mapped (the soil series), several state
ing interpretation of the soils; and accompanying text dis- highway organizations have (lCvCIoped information on t he
cussing the special engineering features of the soils. Es-
tinlates of susceptibility to landslides for the Various soils
mapped can be made from the data furnished. In some Figure 3.12. Delineation of landslide terrain during preliminary
cases, these estimates are indicated in the table on engineer- route location in Lincoln and Ellsworth counties, Kansas, aided in
ing interpretations. Figure 3.11 shows a portion of such a minimizing damage due to landslides (3.31).
table from the Chil ton Cowi ty, Alabama. soil survey re-
'N
port (3.18). 1-laying identified from this table the soils sus- .
,—i'? J''"-.'
ceptible to sliding, one can then locate those soil units on
the photomosaic and thus delineate the most susceptible
areas.
4L/L TM c
Other county soil surveys and generalized reconnaissance
surveys can also provide useful information for landslide in-
vestigations, even though engineering sections are not in-
cluded. Those surveys provide information on parent ma-
terials and their geologic origin, climate, physiographic set-
ting, and soil profiles.
The status of availability of agricultural soil survey re-
ports in the United States is published annually by SCS in
List of Published Soil Surveys. In addition, those state soil
conservation services that publish their own soil surveys
47
engineering characteristics of the soil series mapped in their special applications, the additional data are not cost-
states. The information generally provided for each soil effective. That being the case, the ensuing discussion on
series includes origin,topographic setting, soil profile, drain- the use of remote-sensing techniques to recognize and iden-
age characteristics, and engineering significance for items tify landslides deals mostly with the use of aerial photog-
such as location, slope stability, foundation and pavement raphy. The special applications in which other forms of
support, and source of construction materials. Several remote-sensing data are useful are also discussed. Some
states, such as Indiana, Michigan, New York, and Ohio, comments are included on systems that have potential
have summarized this information in report form. In many value but have yet to be demonstrated for this application.
areas, and particularly in densely developed areas, existing
reports for geologic and engineering investigations may be Use of Aerial Photography
available and should be acquired for reference. Detailed
boring logs, well logs, and similar valuable soil and geologic The material in this section expands and updates Chapter 5
information can be obtained at little cost. in the book on landslides published by the Highway Re-
search Board in 1958 (3.10).
REMOTE-SENSING TECHNIQUES The interpretation of aerial photography has proved to
FOR LANDSLIDE DETECTION be an effective technique for recognizing and delineating
landslides. No other technique can provide a three-
Remote sensing is the collecting of information about an dimensional overview of the terrain from which the inter-
object or phenomenon by the use of sensing devices not in relations of topography, drainage, surface cover, geologic
physical or intimate contact with the subject under investi- materials, and human activities on the landscape can be
gation. The distance of separation might be as close as a viewed and evaluated. Aerial photographs at suitable scales
few millimeters, as in the case of nuclear moisture-density are available for almost all of the United States and a large
devices, or as far as 800 km (500 miles) or more, as in the part of the world. New photography of various types and
case of satellites. The devices range from cameras to vari- scales can readily be obtained. In addition, new techniques
ous scanners and radiometers. in production and interpretation processes have continued
Most remote-sensing devices collect information from to extend the advantages of aerial photography.
the electromagnetic spectrum (Figure 3.13), which extends The value of aerial photography in landslide investigations
through many decades of wavelength from very long radio has been reported by many investigators (3.13, 3.16, 3.19,
waves to extremely short gamma and cosmic radiation waves. 3.28, 3.31). The effectiveness of this technique was well
As shown in Figure 3.13, various types of sensors are avail- demonstrated by Maruyasu and others (3.16) in analyzing
able for obtaining information in the different regions of a mountainous region where landslides were common in
the spectrum. The Manual of Remote Sensing (3.25) gives Japan. Using photo-interpretation techniques, they iden-
greater detail on the electromagnetic spectrum and sensor tified 365 landslides. After the region was field checked,
systems. an additional 68 landslides were identified, resulting in an
The attractiveness of these systems to various investiga- 84 percent accuracy in overall identification. An accuracy
tors is that they greatly expand the scope of possibilities of 96 percent was reported in identification of landslides in
for uniquely and more accurately identifying subjects of in- areas such as paddy fields not covered by trees and shrubs.
terest. In addition to the sensors that provide information The major advantages of using aerial photography in
on the spectral reflectance phenomena of objects are those landslide investigations include the following:
that provide information on the thermal-emission proper-
ties of objects in the infrared region, the backscatter of Photographs present an overall perspective of a large
radar energy in the microwave region, and the reflectance area (when examined with a pocket or mirror stereoscope,
and fluorescence phenomena in the ultraviolet region. Ana- overlapping aerial photographs provide a three-dimensional
lyzmg spectral regions individually provides special data view);
about the properties of objects in that region (e.g., thermal Boundaries of existing slides can readily be delin-
characteristics or reflectance properties). eated on aerial photographs;
Collecting data simultaneously in several spectral regions Surface and near-surface drainage channels can be
and comparing the various responses in those regions, how- traced;
ever, provide more data than can be obtained in each region Important relations in drainage, topography, and
individually and increase the accuracy of interpretation. other natural and man-made elements that seldom are cor-
This approach is referred to as multisensor analysis and is related properly on the ground become obvious on the
based on the following principle: Although two or more photographs;
objects may have similar responses in one region, they will A moderate vegetative cover seldom blankets de-
not respond similarly in all regions. Thus, evaluating the tails to the photo interpreter as it does to the ground ob-
response of various objects in different regions of the server;
spectrum increases the ability to uniquely identify each Soil and rock formations can be. seen and evaluated
of the objects. in their undisturbed state;
In almost all engineering evaluations of various sensor Continuity or repetition of features is emphasized;
systems, the conclusion has been that photography is the Routes for field investigations and programs for sur-
best sensor system and provides the most information. Al- face and subsurface exploration can be effectively planned;
though other sensor systems provide some useful data that Recent photographs can be compared with old
cannot be obtained from photography alone, except for ones to examine the progressive development of slides;
48
Figure 3.13. Electromagnetic Frequency in cycles /second
spectrum and types of
remote-sensing systems used for 020 iO° 06 io0 02 to'° 08 02
I I I
collecting data in each major I I I I
INFRARED I I
spectral region. I I -j
I UV I I I MICROWAVE VHF I IF
RAYS X RAYS
I MEDIUM ond FAR I
I
I Z
E S SEE E P
P 8
0
Wovelenglh -
o I
-
2 2 0
00—
I I I I
en I I
I
I I
'N
Aerial photographs can be studied at any time, in for delineating the presence of water on the surface and for
any place, and by any person; and giving clues to subsurface water conditions by showing the
Through the use of aerial photographs, information vigor of the surface vegetative cover. This has made color
about slides can be transmitted to others with a minimum infrared film especially valuable for locating the presence
of ambiguous description. of seepage zones at or near the surface.
The format commonly used for landslide investigations
Factors Influencing Aerial-Photograph is 23 x 23-cm (9 x 9-in) vertical photographs taken with
Interpretation an aerial mapping camera. Consecutive photographs having
an average overlap of 60 percent are obtained to provide
Success in the application of aerial-photograph interpreta- stereoscopic coverage. In addition to their use for interpre-
tion techniques for landslide investigations depends on four tation purposes, these photographs can be used for prepar-
major items: (a) qualifications of the interpreter; (b) photo- ing maps of the slide areas by photogrammetric methods.
graphic parameters; (c) natural factors; and (d) equipment Details on flight planning and mapping are given in the
and analysis techniques used. Analysis by competent inter- - Manual of Photogrammetry (3.35). Oblique photographs,
preters with suitable equipment will certainly be an impor- such as those shown in Figures 3.1 and 3.4, are also ob-
tant factor in the successful application of this technique. tained for landslide investigations. The view they provide
However, a proper understanding of the photographic param- of the unstable slopes and valley walls is more unobstructed
eters and natural factors that can influence the data is im- than that provided in vertical photographs. In recent years,
portant to the proper use of aerial photography; these two oblique views of landslide areas are being obtained with
items are discussed below. small format, nonmapping cameras in low-flying planes and
helicopters; 35 and 70-mm cameras are commonly used.
Photographic Parameters Scale of photography can be critical to the interpreter's
ability to identify landslides in that it influences his or her
Photographic parameters include types of photography, for- ability to denote topographic features indicative of land-
mat, scale, coverage, quality, and availability.. General infor- slides. In some cases, the landslides are so massive [some-
mation concerning these parameters is given in many stan- times involving more than 4km3 (1 mile3)] of material
dard references, such as those published by the American that they are virtually impossible to detect on the ground
Society of Photogrammetry (3.7, 3.25, 3.30).The following or on large-scale photographs (3.9). Small-scale photographs
discussion covers-only those items pertinent to landslide in- (1:48 000 or smaller) are required in order to view stereo-
vestigations. scopically the full extent of massive slides. However, most
Aerial photographs typically used for landslide studies landslides encountered in engineering construction are small.
include panch.romatic and infrared black-and-white films Cleaves (3.6) stated: "Within the scope of the author's ex-
and natural color and color infrared films. Panchromatic perience, more than 90 percent of all landslides requiring,
black-and-white film is the most common type used because control or correction have been small, averaging less than
it is low cost, convenient to handle, and readily available, 500-feet in width from flank to flank, and 100 to 200 feet
but natural color and color infrared fihns are now being more in length from crown to toe." Hence, many small landslides
extensively used. Color photography is especially valuable would not be readily discernible on normal-scale photography
for outlining differences in moisture, drainage, vegetation (1:15 000 to 1:30 000) and would require large-scale photog-
conditions, and soil and rock contacts. Plates 3.1 through raphy (1:4800 or larger) for identification. At these large
3.6 show the uses and advantages of colorfilms. The strati- scales, however, the area covered in each photograph is
fication in exposed soils and rocks is most easily recognized limited, and the overall perspective is more difficult to grasp.
on natural color film. Color infrared films are most helpful Whenever practicable, it is desirable to examine both small-
49
Plate 3.1. Slides in Douglas County,
Oregon, caused by undercutting of
slopes by stream. Slides largely :1
...
composed of weathered colluvial
materials are evident at almost every
bend of stream where it undercuts
naturally steep slopes (1). Advantages
provided by use of color film include
distinction of potentially more unstable
colluv jal material from more stable
bedrock surfaces; ease of distinguishing
seepage areas; and indication of
vegetation under stress due to
insufficient moisture, which could
occur because of slope movement.
50
Plate 3.4. Black-and-white
. .. :
photographs taken October 25,
1954, of delta deposits overlying
fr -• - i
-
,
- ,,•
41,
-
.•
; s4
'A
Plate 3.5. Natural color
photographs taken April 9.
1976, of area shown in
Plate 3.4. Presence of
seepage at contact between
delta and lake-bed deposits
(2) and change in color tones
of soils of these two
materials make it easier to
distinguish between these
two deposits: Darker brown
tones are seepage zones;
bluish gray tones are lake
bed; and lighter tan tones
are delta deposits.
51
and large-scale photographs: the smaller scale (1:20 000 or Agricultural Stabilization and Conservation Service (ASCS),
smaller) to provide the regional setting and landform boun- U.S. Department of Agriculture, periodically publishes the
daries and the larger scale (1:8000 or larger) to provide the ASCS Aerial Photography Status Maps and the Comprehen-
details. sive Listing of Aerial Photography, which indicate the cov-
Sufficient coverage is required in order to determine the erage available from ASCS for each county in each of the
full extent of the slide, the potential areas affected, and the 50 states. The National Archives in Washington, D.C., is
changes occurring with time, In some instances the active the depository of all nitrate-base photography collected
slide is present only at the lower or toe end of a slope while during the years 1936 to 1941. The EROS Data Center
unnoticed far up the mountainside are tension cracks that has USGS and National Aeronautics and Space Administra-
indicate the instability of the entire slope. With time, the tion photography.
slides may retrogress up the slope until the whole mountain-
side is affected. Clearly, the plans for stabilization will dif- National Cartographic Information Center
fer depending on whether one is dealing with a small con- U.S. Geological Survey
fined slide area or with a whole mountainside. A situation 507 National Center
of this type can be dealt with satisfactorily if two scales of Reston, Virginia 22092
photography are obtained, as previously recommended.
Aerial Photography Field Office
Coverage at the larger scale should include the full height
and depth (mountain top to valley bottom) of the area Agricultural Stabilization and Conservation Service
U.S. Department of Agriculture
under investigation and at least three to five times the width
2222 West, 2300 South
of the present unstable area.
Post Office Box 30010
Many areas, particularly in the United States, have been
Salt Lake City, Utah 84125
photographed at various times since the 1930s, and this se-
quential photography is valuable for noting the changes in Forest Service
terrain conditions, such as drainage, slope stability, land use, U.S. Department of Agriculture
and vegetation, that have occurred during a period of time. Washington, D.C. 20250
These changes provide a clue to the origin and irate of move- Headquarters, Defense Mapping Agency
ment of a slide. Besides denoting changes, such photographs U.S. Naval Observatory
were probably taken at different seasons, times of day,,and Building 56
moisture conditions and will often show detailed features Washington, D.C. 20305
that might otherwise have been obscured. These older photo-
graphs can also be of value in planning new coverages. Coastal Mapping Division, C-34 15
The quality of the photography obtained has a direct, in- National Ocean Survey
fluence on the amount of detail that an interpreter can de- National Oceanic and Atmospheric Administration
rive from the photographs. The quality depends on the 6001 Executive Boulevard
products used and the various processes the images pass Rockville, Maryland 20852
through. Proper selection and control of the film-filter EROS Data Center
combinations require close cooperation between a knowl- U.S. Geological Survey
edgeable interpreter who can specify the requirements of Sioux Falls, South Dakota 57198
the end product needed and an experienced contractor who
Audiovisual Department
knows the capabilities and limitations of the various systems
National Archives and Records Service
and can aid in selecting the optimum combination for the
specific job. Eighth Street and Pennsylvania Avenue, NW
Washington, D.C. 20408
Availability and access to existing photography are ex-
cellent in the United States and Canada and in much of the Maps and Surveys Branch
rest of the world. In the United States, most aerial photog- Tennessee Valley Authority
raphy has been taken for federal and state agencies. Many 210 Haney Building
areas in the United States have been covered several times Chattanooga, Tennessee 37401
byone or more agencies. The National Cartographic Infor- National Airphoto Library
mation Center (NCIC) of USGS publishes a catalog, Aerial Canada Department of Energy, Mines and Resources
Photography Summary Record System Catalog, of existing, 601 Booth Street
in-progress, and planned aerial photography of the conter- Ottawa, Canada K1AOE8
minous United States held by several federal agencies. The
catalog includes a series of outline map indexes, indicating Several state highway and transportation departments
coverage within a 75 x 7.5 -min geographic cell, and acorn- similarly have available for public review index maps that in-
puter listing on microfiche film that shows amount of cloud dicate the type and extent of coverage held by state organi-
cover, camera used, specific scale, date of coverage, film zations. Some state agencies publish lists of all the aerial
type and format, and agency holding the photography. The coverage available for their states from all sources (3.2).
NCIC address and other U.S. and Canadian agencies that Using existing photography in a landslide investigation
have large holdings of aerial photography are given below. is of course less expensive than acquiring new coverage.
Several of the government agencies whose holdings are Existing photographs, however, may be several years old,
listed in the NCIC catalog also publish status maps of aerial may not show the present terrain conditions, or may be at
photography held by their organizations. For example, the a scale too small for performing detailed analyses, and new
52
coverage may thus be necessary. The existing photography These optimum conditions are not always present at one
can then be useful in the initial analysis and planning for the time, and a compromise may be required to obtain as many
new coverage. of the ideal conditions as possible. Some detailed investiga-
When the organization performing the investigation does tions may require photographs to be taken several times
not have the facilities for obtaining new photographic cov- during the year.
erage, the best approach is to engage a reputable firm with a
base of operation close to the study site. In some cases, a Principles of Aerial-Photograph
small plane can be hired locally, and a hand-held camera can Interpretation
be used to make oblique photographs that are economical
and suitable for preliminary analysis and planning. The interpretation of aerial photographs involves three major
steps:
Natural Factors Examine the photographs to get a three-dimensional
perception,
Consideration of natural factors is important in planning Identify ground conditions by observing certain ele-
aerial photography for landslide studies. Flight planning is ments appearing in the photographs, and
certainly controlled by meteorological conditions, but con- Using photo-interpretation techniques, analyze spe-
sideration must also be given to time of year and time of cific problems by the association of ground conditions with
day in which the photography is obtained. one's background experiences.
Meteorological conditions include cloud cover, haze, tur-
bulence, and solar angle, all of which affect the quality of The quality and the reliability of any interpretation are, of
definition of images on the film. Suitable conditions for course, enhanced in direct ratio to the interpreter's knowl-
color photography are when the sun angle exceeds 300 from edge of the soils and geology of the area under study. The
the horizontal, cloud cover is less than 10 percent, visibility acquirement of such knowledge, either by field examination
at aircraft altitude when one looks toward the sun exceeds or by study of available maps and reports, should, therefore,
24km (15 miles), and the atmosphere is free of turbulence be considered an essential part of any photo-interpretation
(3.30, p. 43). job.
Time of year is also critical to the quality of aerial photog- The technique used in interpreting aerial photographs is
raphy. Although no clear-cut recommendations can be made referred to as pattern analysis. This approach is based on
to fit all situations or localities, some of the desirable con- the premise that the landforms, the basic units of the land-
ditions are discussed below. scape, have distinct patterns in aerial photographs. Those
landforms developed by the same geologic processes and in
the same environmental setting will have similar patterns,
1. Optimum soil-moisture contrasts. The optimum con- while those developed by different processes will have dif-
trast of soil moisture occurs under two different conditons: ferent patterns. These patterns are composed of several
(a) when the soils are wet but not totally saturated or (b) major elements that are evaluated in the process of interpre-
when the moisture levels are low. The former occurs in the tation: topographic expression,drainage, erosion, soil tones,
spring or early summer because of high water conditions. vegetation, and culture. These elements are discussed briefly
Wet areas and seepage zones present in unstable or poten-
in the following paragraphs; more thorough treatments are
tially unstable areas are directly visible outcropping on hill-
given by Lueder (3.15), Way (3.38), and Colwell (3.7).
sides or are evident by the presence of luxuriant vegetative
growth over those wet areas. The latter occurs during dry
periods in late summer or fall or over disrupted areas. Dur- Topographic Expression
ing dry periods, vegetation in wet areas or seepage zones
has more growth and color than that in areas that have nor- The features of importance in describing topographic expres-
mal moisture levels. Areas disrupted by cracks and fissures sion include topography, shape, and relative relief. Topog-
are better drained and thus drier, and vegetation located raphy represents the three-dimensional aspect of the land,
over those areas shows evidence of stress earlier. Contrast- scape, such as hill, basin, ridge, or mountain. Shape refers
ing vegetation and moisture conditions are best noted on to a unique character of the landscape, such as conical hill,
color infrared photography, as illustrated in Plates 3.2 and sinuous ridge, or A-shaped mountain, that aids in describing
3.3. or separating different forms of topography. Relative relief
Optimum ground cover. A minimal cover of snow refers to the comparative position of the landform in rela-
and tree foliage is desirable. In humid, temperate climates, tion to other landforms in the vicinity, for example, the po-
this usually occurs in early spring after the snow melts and sition (or relative relief) of the sloping fan-shaped plain lying
before the deciduous trees leaf or in late fall after the leaves at the foot of A-shaped hills and sloping downward to a level
fall and before snow falls. basin area. In this example, three different landforms are in-
Optimum shadow conditions. Photographs taken when dicated based on their relative relief: the mountainous area,
the sun angle is high and shadows on the hillsides and slopes the sloping plain area, and the basin area. Topographic ex-
are minimal are the best for interpretation. However, in pression of landforms is easily determined by a stereoscopic
some special cases, such as in areas of low topography, small examination of the photographs. In such an examination,
slope failures are enhanced by the presence of long shadows various landforms can be separated and clues obtained as to
produced when the photography is taken with a low sun the nature and stability of the materials constituting the
angle (either early in the morning or late in the afternoon). landform.
53
Drainage and Erosion photographs by dark spots or "tails," are a clue to seepage
on slopes. Cultivated fields, as well as natural growths, are
The density and pattern of drainage channels in a given area good indicators of local soil conditions. Thus, an orchard
directly reflect the nature of the underlying soil and rock. is often found on well-drained soils; black ash, tamarack,
The drainage pattern is obvious in some cases, but more of- and white elm are normally found on wet, fine-grained
ten the channels must be traced on a transparent overlay to soils; and vegetation is sparse in nonproductive serpentine
allow the pattern to be successfully studied. Under other- soils, where landslides are common.
wise comparable conditions, a closely spaced drainage sys- The use of color photography, especially color infrared,
tem denotes relatively impervious underlying materials; greatly facilitates vegetation analysis and thus the overall in-
widely spaced drainage, on the other hand, indicates that terpretation. The vigor of vegetative growth and the pres-
the underlying materials are pervious. In general, a treelike ence of vegetation stress due to detrimental soil and mois-
drainage pattern develops in flat-lying beds and relatively ture conditions are best noted on color infrared photography.
uniform material; a parallel stream pattern indicates the pres- Point 2 in Plates 3.2 and 3.3 shows areas of seepage, as in-
ence of a regional slope; rectangular and vinelike patterns dicated by lush vegetation, and point 3 indicates vegetation
composed of many angular drainageways are evidence of con- under stress on the unstable slide debris.
trol by underlying bedrock; and a disordered pattern, inter-
rupted by haphazard deposits, is characteristic of most gla- Culture
ciated areas. Other patterns have develope4 in response to
special circumstances. A radial pattern, for instance, is Cultural patterns reflect how humans adjust to the natural
found in areas where there is a domal structure, and a terrain on which they live and work. An understanding of
featherlike pattern is common in areas where there is severe these patterns and the reasons for their presence can be valu-
erosion in rather uniform silty material, such as bess. able indicators of soil, moisture, and other terrain conditions.
The shapes of, gullies appearing in aerial photographs pro- For example, drainage ditches and tile fields indicate the
vide valuable information regarding the characteristics of presence of a high water table; steep, vertical cuts along
surface and near-surface materials. Typically, long, smoothly highways denote the presence of bedrock; sloping cuts usu-
rounded gullies indicate clays; U-shaped gullies, silts; and ally typify unconsolidated soils; and contour farming or an
short, V-shaped gullies, sands and gravels. irregular wavy plowing pattern characterizes sloping terrain
or erodible soils.
Soil Tones
Interpretation of Landslides on Aerial
Soil tones are recognizable in photographs unless there is a Photographs
heavy vegetative cover. On black-and-white photographs,
the tones are merely shades of gray, ranging from black to As previously described in the section of this chapter on
white. Because gray tones are highly respondent to soil- landforms susceptible to landslides and the section on vul-
moisture conditions on the ground, they are an important nerable locations, landslides are more prevalent in certain
aerial-photographic element in landslide investigations. landforms and occur most frequently at certain vulnerable
A soil normally registers a dark tone if it has a high mois- locations. The interpretation of aerial photographs by the
ture content and a light tone if it has a low moisture con- pattern-analysis technique is a valuable tool for delineating
tent. The moisture condition is the result of the physical landslide-susceptible landforms and pinpointing the vulner-
properties of the soil, its topographic position, and the po- able locations. The identifying pattern elements on the
sition of the groundwater table. The degree of sharpness photographs and significant features about these landforms
of the tonal boundary between dark and light soils aids in are summarized in the following discussion and illustrated
the determination of soil properties. Well-drained, coarse- in the accompanying figures.
textured soils show distinct tonal boundaries whereas poorly
drained, fine-textured soils show irregular, fuzzy tonal Diagnostic Patterns of Landslides
boundaries. on Aerial Photographs
Color photography greatly increases the number of tones
that can be distinguished by the interpreter and improves An investigator already familiar with the appearance of land-
the amount of detail and accuracy of the information de- slides on the ground should become oriented to the aerial
rived. In addition, the use of special visual, photographic, view of landslides by an examination of photographs of
or electronic enhancement techniques increases the effective- some known examples. The difference between an aerial
ness of analysis of the tonal pattern element. view and a ground view results chiefly from the fact that the
former gives a three-dimensional perspective of the entire
Vegetation slide area, but at a rather small scale. Ground photographs,
on the other hand, show only two dimensions, but on a
Vegetative patterns reflect both regional and local climatic larger scale.
conditions. The patterns in different temperature and rain- The following features discernible on aerial photography
fall regions can be recognized in aerial photographs. Locally, are typical of landslides or landslide-susceptible terrain, but
a small difference in soil moisture condition is often detected not all features are evident for each landslide. Most of these
by a corresponding change of vegetation. A detailed study features are shown in Figure 3.14; those features not evident
of such local changes is helpful in landslide investigations. are shown in other figures in this chapter. The numbers in
For instance, areas of wet vegetation, represented on the Figure 3.14 correspond to those in the list below.
54
Figure 3.14. Typical
characteristic features of
a landslide as shown in
aerial photographs (3.6).
Slide occurred in Yakima
County, Washington.
Numbered items are
discussed in text.
- 5'
••-
It.
:Y
Land masses undercut by streams (Plate 3.1): highway is built on unstable soil, the irregular outlines and
Steep slopes having large masses of loose soil and nonuniform tonal patterns of broken or patched pavements
rock: are often visible (Figure 3.15).
Sharp line of break at the scarp (head end) or pres-
ence of tension cracks or both; Identification of Vulnerable Locations
Hummocky surface of the sliding mass below the
scarp; Many slides are too small to be detected readily in photog-
Unnatural topography, such as spoon-shaped trough raphy at the scales normally available (i.e., I : 15 000 to
in the terrain: 1:40 000). Consequently, the photographs should be
Seepage zones; closely examined for signs that indirectly indicate the pres-
Elongated undrained depressions in the area; ence of slides or, if signs are not visible, for the vulnerable
Closely spaced drainage channels; locations where slides usually occur. Typical vulnerable lo-
Accumulation of debris in drainage channels or cations include areas of steep slopes, cliffs or banks being
valleys; undercut by stream or wave action, areas of drainage con-
Appearance of light tones where vegetation and centration, seepage zones, areas of hummocky ground, and
drainage have not been reestablished (Figure 3.1 and Plates areas of fracture and fault concentrations. The characteris-
3.2 and 3.3); tics of these locations have been discussed in an earlier see-
II . Distinctive change in photograph tones from lighter tion of this chapter in which vulnerable locations are de-
to darker, the darker tones indicating higher moisture con- scribed.
tent; Aerial photographs are valuable aids in identifying the
Distinctive changes in vegetation indicative of changes vulnerable locations. The shape and slope of the terrain are
in moisture (Plates 3.2 and 3.3); and readily discernible from the stereoscopic examination of
Inclined trees and displaced fences or walls due to the photographs. In fact, the vertical appearance of the ter-
creep (Plate 3.3). rain is exaggerated when viewed with a lens stereoscope.
Moderate slopes appear steep, and steep slopes appear al-
In sonic cases the slide itself is not discernible, but indirect most vertical, making them easier to delineate. In addition,
evidence of its existence is noted. For example, where a the slopes can be measured on the photographs by using
55
Figure 3.15. Presence of
limited sections of patched
pavement (1) in generally
unpatched roadway can be
indicative of possible
slope movement in area.
Presence of fissures on
hillside above this patched
section (2) in Washington
County, Ohio, confirms
instability of slope.
"'
V
-4 r : ;.
simple measuring devices, such as an engineer's scale or a Most of the forms susceptible to landslides are readily recog-
parallax bar, and by applying photogrammetric principles. nizable on aerial photographs. The identifying elements on
Details of this technique are described by Ray (3.23). black-and-white photography and significant facts about the
The presence of drainageways, seepage zones, fractures, landforms are summarized and illustrated in the following
and fault zones is also readily evident on aerial photographs. sections. Table 3.2 gives the landslide potential of various
By means of stereoscopic examination, the complete drain- landforms/geologic materials grouped on the basis of their
age network can be mapped, including the intermittent topographic form. The order of presentation in this section
streams and small gullies. The presence of wet zones or follows a sequence based on origin and character of the ma-
seepage areas is evidenced by darker tones caused by the terials.
higher moisture content in the soils or by a more luxuriant
vegetative growth over the wet areas. Areas of drainage or Sedimentary Rocks and Their
water concentration above a slope should be closely ex- Residual Soils
amined because they are vulnerable locations. Subsurface
seepage from these areas can lead to slope failures: Figure The discussion of rocks and their residual soils is combined
3.4 shows such a case. Fracture and fault zones are indi- in this and the following two sections because the recogni-
cated on the photographs by dark linear or curvilinear lines. tion of types of residual soils depends primarily on the rec-
The darker tones are usually due to the better growl h of ognition of the landform developed on the parent rocks.
vegetation along the fracture zones where it is easier for The determination of depth of residual soil requires consid-
the roots of plants to grow and where moisture levels are erable judgment - However, investigators working constantly
usually higher. In delineating fracture zones, care must be in their own regions should have no difficulty in estimating
taken not to interpret man-made features, such as fence the depth once they are familiar with local conditions.. In
lines or field boundaries, as fracture zones. Generally, fea- general, rounded topography, intricate, smoothly curving
tures having straight lines, right-angle intersections, and drainage channels, and heavy vegetation are indicators of
standard geometric patterns are man-made. Several of the probable deep soils: in contrast, sharp, steep, resistant ridges
above features are illustraled in succeeding figures that and rock -cont rolled angular channels are commonly found
show landslides in various landforms. in areas of shallow soils. The local climatic and erosion pat-
terns should be considered in the interpretation.
L.andforms/Geologic Materials Many landslides occur in residual soils and weathered
Susceptible to Landslides rocks; they are usually in the form of slides or flows. For
example, in a study in North Carolina (3i2), two-thirds of
Landslides are rare in some landforms and common in others the slides involved in cut-slope failures occurred in the
56
Figure 3.16. Pierre shale,
Lyman County, South
Dakota, a formation that
is most highly susceptible
to landslides (3.4).
Numerous slides are
evident throughout area,
some due to undercutting
by streams (1) and others
to steep slopes, seepage,
or stress release along joints
(2).
weathered soil material. Of the sedimentary rocks, shales and medium dark tones. They usually are heavily cultivated
are the most susceptible to landslides. A typical example is in humid areas.
shown in Figure 3.16, Shales are especially susceptible when Interbedded sedimentary rocks show a combination
interbedded with pervious sandstones or limestones or other of the characteristics of their component beds. When hori-
pervious rock types. This geologic setting of alternate layers zontally bedded, they are recognized by their uniformly dis-
of pervious and impervious materials on a slope is conducive sected topography, contourlike stratification lines, and tree-
to sliding. like drainage: when tilted, parallel ridge-and-valley topog-
Landslides in massive horizontal sandstones or limestones raphy, inclined but parallel stratification lines, and trellis
are uncommon unless they are interbedded with shales or drainage are evident.
other soft rocks (Figure 3.17). In steeply tilted positions,
any sedimentary rock may fail by sliding (Figure 3.18). De- The identification of land form as a means of detect ing
pending on the (lip angle, joint syste iii, and climate, slides associated landslides is important in the flat-lying sedimen-
may take one or a combination of forms, such as rock falls, tary group because the slides there are often small and,
rock slides, topples, block slides, slumps. debris falls, debris therefore, not very obvious on aerial photographs. This is
slides, and earth flows. River undercutting and artificial ex- particularly true for slides in colluvial deposits formed from
cavation are important factors in initiating landslides in both sedimentary rocks. These conditions are common since sedi-
horizontal and tilted rocks (Figures 3.3 and 3.16). mentary rocks are the most widespread of all surface rocks.
Methods of identification of sedimentary rocks in aerial
photographs are well established. Igneous Rocks and Their Residual Soils
I . I-lard sandstones are evidenced by their high relief, Basaltic lava flows are among the most common extrusive
massive hills, angular and sparse drainage, light tones, and igneous rocks, and they are readily identifiable on aerial
general lack of land use. photographs. Basalts and other volcanic flows are highly
Clay shales are noted for their low rounded hills, well- susceptible to different types of landslides (Figures 3.19 and
integrated treelike drainage system, medium tones, and 3.31). They often form the caprock on plateaus or mesas:
gullies of the gentle swale type. They are generally farmed their sharp,jagged (saw-toothed) cliff lines are clearly visible
in regiuns with sufficient rainfall. on photographs. Surface irregularities or flow marks, sparse-
Soluble limestones are characteriied by low rolling ness of surface drainage and vegetation, and (lark tones are
topography with sinkhole development in temperate, humid con firming characteristics.
areas, by rugged karst topography in some tropical regions, If it basalt flow is underlain by or interbedded with soft
by a general lack of surface drainage, and by mottled light layers, particularly if it occupies the position of a bold es-
57
carpment, a favorable condition for large slumps is present the foot of cliffs. Disturbance of talus slopes during road
The joints and cracks in basalt give rise to springs and seep construction has caused some large slides of talus materials.
age zones and greatly facilitate movement. Rock falls and Old slides and fissures indicating incipient slides often can
rock slides along rim rock are usually favored by vertical be seen on photographs. In areas of relatively deep weather-
jointing of basalt and by undercutting of basaltic cliffs. big, the landscape is somewhat modified. A more rounded
Talus accumulations of various magnitudes are found at topography and heavier vegetation develop, although dark
slialematerial .. -. E
'.•J, .'
underlying sandstone
cap rock in Morgan ),
County, Ohio Almost
all hillsidt.s show
evidence of instability
(1). Evidence of slides . .• . . ..
in capping sandstone .. . . ' s .•.
due to undermining .. . . '.: . .
..
by sliding out of the . ... ... ...., ... . .•4 &., .... -. / ••.i .,
shask is also apparent 2
- 41
.
. :''••. ...S.....
,.i
k... ." . .
:.)
. '. • ..
S.
. ... .
. .;... ....
.
.•
wasting is indicated by
I
oversteepened slopes
I
having light tones (1); . __. .. .
bulging of lower slopes
(2); and accumulation
3
of loose scree materials r• \ '. ' •. .
at base of slopes (3) .
- 3
58
tones still predominate. Slumps of both large and small can be distinguished on aerial photography by an indication
size are common in basaltic soils. of banding or lineation of the topography and drainage net-
Granite and related rocks constitute the most widely oc- work, and landslides, when present, are clearly evident. Fig-
curring intrusive igneous rock types. The landslide poten- ure 3.21 shows an example of landslides in metamorphic
tial of granitic rocks varies widely, depending on the com- terrain (mica schists). The attitude of the foliation seen at
position of the rock and its fracture pattern, the topography, point 3 indicates that the dip of these layers may have been
and the moisture conditions. In granites that are highly re- a significant factor in the presence of numerous slides (point
sistant to weathering or are of low relief, there is generally I) found just across the river where the layers are steeply
no slide problem. In hilly country where the granite is dipping toward the river.
deeply weathered, slumps in cut slopes, as well as in natural Within the metamorphic group, many slides are associated
steep slopes, are common. Fractures in the rock and high with serpentinite areas, which are identified on aerial photo-
moisture conditions undoubtedly are favorable factors in graphs by sinuous ridges, smoothly rounded surfaces, short
producing landslides. Figure 3.20 shows an example of an steep gullies, poor vegetative cover, and dull gray tones.
extensive slide in granitic materials. Granitic masses are Within a general area, local conditions, such as vegetation,
identified on aerial photographs by the rounded (old) to moisture, and slope, may create special, favorable circum-
A-shaped (young), massive hills and by the integrated tree- stances for landslides. In genera!, low relief and low rain-
like drainage pattern with characteristic curved branches. fall are among the factors responsible for the stability of
The presence of criss-cross fracture patterns and light tones some of the serpentinite slopes.
and the absence of stratification and foliation aid in confirm-
ing the material. Glacial Deposits
Metamorphic Rocks and Their Landslides are common in some glacial and glaciofluvial de-
Residual Soils posits. Although most of the distinct glacial forms are
easily identified on aerial photographs, there are complex
As reported by Leith and others (112) and other investiga- areas that require a high degree of skill for their identifica-
tors, the frequency of landslides per unit area is greater in tion. Moraines are found in nearly all glaciated areas. They
metamorphic rocks than in most other rock types. The pres- are identified on aerial photographs by jumbled, strongly
ence and attitude of the foliation and joints in these rock rolling to hilly terrain. Moraines, particularly in semiarid
types greatly influence the stability of the slopes.. Although areas, contain a large proportion of untilled ]and. Deranged
the characteristics of the major types of metamorphic rocks drainage patterns, irregular fields, and winding roads are con-
gneiss, schist, slate, and serpentinite—have been identified on firming clues. Minor slumps, debris slides, and earth flows
aerial photographs for specific areas, they are not sufficiently are common in cut slopes iii moraines as a result of un-
consistent to develop typical pattern elements. Neverthe- drained depressions and seepage zones in the mass. Because
less, the presence and attitude of the foliation and joints morainic hills are usually small, these slides are not exten-
_ .,v
,.
ft.. .'.. .
.et!
59
sive. They are, nevertheless, large enough to cause contin- have been oversteepened by glaciation (Figure 3.23). The
ual trouble to many maintenance engineers (Figure 3.22). topography of such areas is basically that of the underlying
In areas subject in part to mountain glaciation, transpor- bedrock with slight local modifications, depending on the
tation routes must follow valleys formerly occupied by thickness of the mantle.
glaciers and be built on their deposits. Under these condi- Slides seldom occur in other glacial or glaciofluvial de-
tions slides in moraines and colluvium are common. Slides posits, such as kames, deltas, eskers, outwash plains, and
also occur in the shallow mantle overlying bedrock and take till plains. They have occurred, however, in places where
the form of slumps, debris slides, and debris falls: they often these materials are underlain by fine-grained impervious
contribute to failures in artificial fills placed on them. The materials, such as lake-bed deposits, when these deposits
landslides most commonly occur along valley walls that are present in elevated positions (Plates 3.4, 3.5, and 3.6).
Ta
Water-Laid Deposits flats, marshes, and swamps; and the presence of broad, shal.
low, tidal stream channels. The upper coastal plain is iden-
Water-deposited landforms most susceptible to landslides tified by its rolling to rugged topography: an integrated
include coastal plain deposits, river terraces, and lake beds. drainage system with wide, vegetation-filled main drainage-
Coastal plain deposits can be subdivided into three dis- ways: and irregular land-use patterns. It is also associated
tinct subforms for analysis by aerial photographic methods: with coastal features and appears on aerial photographs to
the dissected or upper coastal plain, the undissected or be somewhat similar to areas underlain by consolidated sedi-
lower coastal plain, and the beach zone. Landslides are un- mentary rocks. In the lower coastal plains, landslides offer
common in the beach zone. The lower coastal plain is iden- a problem only in the construction of canals or similar struc-
tified by its low, flat topography; its association with tidal tures that require deep excavation in flat lands. In the upper
61
coastal plains, however, slumps in natural hill slopes, as well ral equilibrium is disturbed by artificial installations, slides
as in road cuts, are common (Figure 3.24). The stratified will occur. Slides in terraces naturally start on unsupported
and unconsolidated nature of the sands, silts, and clays that slopes facing the low land (Figure 3.25).
characterize most coastal plains provides a favorable situa- Lake-bed deposits generally display flat topography un-
tion for landslides. less they are dissected. Although generally composed of
Terraces are easily recognized on aerial photographs as clays, lake beds have little chance to slide except when ex-
elevated flat land along major or minor valleys. Terraces of posed in valley walls or cuts (Figure 3.26). Slides of con-
gravel and sand are usually stable, maintaining clean slopes siderable magnitude have occurred in lake clays under each
on their faces. However, where terraces are composed of of the following circumstances: (a) where lake clays are in-
fine sands or interbedded silts and clays or where the natu- terbedded with or, especially, are overlain by granular de-
Figure 3.23. Thin till deposit overlying bedrock in glaciated valley in Broome County, New York. Highway cuts in hillsides oversteepened
by glaciation create slides (A and B). Slides generally extend to full depth of soil material until competent rock bed (in this case sandstone)
is encountered. Between A and B, section of highway entirely in rock has not experienced any slides. In this respect, landslides serve as
indicators of depth of soil to rock. On other side of river, site of future slide is indicated at C. an area of Concentration of drainage, loose
fill, and lack of bedrock control.
0 ..
2/-:. '
: .
I.
- t- ...
62
posits and (b) where lake clays overlie bedrock at shallow Eolian or Wind Deposits
depth and the base level of erosion of the area is generally
lowered. Examples of flow slides in these types of deposits Loess, or wind-deposited silt, can be identified unmistakably
are shown in Figures 3.27, 3.33, and 3.34. Undissected lake in aerial photographs by its vertical-sided gullies, which are
clays are easily identified in aerial photographs by their char- evenly spaced along wide, flat-bottomed tributaries to show
acteristic broad level tracts, dark gray tones, and artificial a featherlike drainage pattern. Such a landform is confirmed
drainage practices. Dissected and complex lake-bed areas are by equal slopes on hills and valleys (an indication of uniform
relatively difficult to identify, particularly by those not fa- material), heavy vegetative cover in dissected areas, extensive
miliar with the local geologic conditions. Again, the presence farming in undissected areas, and soft gray tones (Figure 3.28).
of existing slides is the most reliable warning signal. Earth flows and minor slumps, generally referred to as cat
1
:
1 ) ci "' ?•
( •t
I -
t 1• J t
63
steps, are commonly found in bess. The cat steps appear Gravity Deposits
as fine, roughly parallel, light tone contours on the aerial
photographs. Because of their small size, they are not al- Loose, unconsolidated talus and colluvial materials formed
ways evident. The individual steps of these small slumps by weathering of parent soil and rock materials and moved
are commonly about a meter wide and several centimeters downslope by gravity are found on steep slopes. These de-
to a meter high. These subtle features are shown in Figure posits are easily identified on aerial photographs as bare
3.29. slopes in mountainous areas, but they are not so obvious on
I
' \t
64
Figure 3.28. Highly dissected
nature of the bess terrain in
Lincoln County, Nebraska,
illustrates great erodibility of t kn~
these materials (3.10). -
is indicated by broad
amphitheater scarp faces
at crown, steep side slopes
of gullies, and light-toned
deposits in gulley bottoms. '
&q ....
Numerous little erosional
scars and slumps can be
seen on hillsides (2).
'it
got II 9.
Section (3) is shown 4t. OXN
enlarged in Figure 3.29. 5: L;
Ov '1
- ' -"-
41 -,
<
AO
.14 AN
ij
i
-[;;
-
:::
j :
" •' - -
- I
y '1% -
65
vegetated lower slopes. Excavation into these materials usu- investigations have been performed by several investigators,
ally creates stability problems. Old landslide deposits are including Alfäldi (3.1), Gagnon (3.11), Rib (3.26), Stallard
themselves part of this group (Figure 3.30). and Myers (3.32), and Tanguay and Chagnon (3.33). The
general consensus of those studies is that large-scale aerial
Complex Forms photography (most preferred color) provides the most in-
formation on terrain conditions, specifically for landslide
Most of the landforms previously described can be called detection. Several of the reports did indicate, however,
simple forms because they consist predominantly of one that satellite and infrared imagery offered some unique in-
type of material in each unit. In nature, however, complex formation that would prove useful for landslide investiga-
or superimposed forms are numerous and of common oc- tions.
currence, especially in glaciated areas, as mentioned pre-
viously. They are further emphasized here because of their Satellite Imagery
significance in landslide studies. A change of material ver-
tically or horizontally in complex areas often affects the in- Since July 1972, multispectral satellite coverage has been
ternal drainage characteristics and creates slope stability obtained for the United States, most of Canada, and many
problems. The common situation most favorable to slides other areas of the world. Data are collected in four bands
is one in which pervious formations are underlain by rela- onan 18-d cycle:
tively impervious beds.
Band Color Wavelength (nm)
Procedure for Landslide Investigation
Using Aerial Photographs Green 0.5 to 0.6
Red 0.6 to 0.7
Near infrared 0.7 to 0.8
The following procedure is recommended for photographic Near infrared 0.8 to 1.1
studies of landslides.
A second satellite launched in 1975, as well as others sched-
Lay out sites of planned facility on photographs. uled for launching, provides data on a shorter time cycle.
Delineate areas that show consistent characteristics Each satellite scene covers an area of approximately 33 000
of topography, drainage, and other natural elements and 2 (10 000 miles2) and has a resolution of about 80 m
classify into landform types. Large and obvious slides are (250 ft). The satellite scene can be provided as a 70-mm
identified at this stage. film at a scale of 1:3 369 000; as a 185 x 185-mm (7.3 x
Evaluate the general landslide potential of the land- 7.3-in) black-and-white print of each band; as a color com-
form types. Use Table 3.2 and the section on landforms posite at a scale of I : 1 000 000; and on computer-compatible
susceptible to landslides as a guide. tapes for computer processing. The scenes can also be pro-
Make a detailed study of cliffs or banks adjacent to vided on 35-mm slides.
river bends and all steep slopes. Compare slopes within the In addition to the LANDSAT images provided by these
same landform type: For instance, slopes in bedrock land- two satellites, the National Aeronautics and Space Admin-
forms are more stable, even though steeper, than slopes in istration (NASA) has obtained photographic and image coy-
adjacent soil areas. Because slides are usually small in size erages of considerable areas of the United States and
on photographs, look carefully and inspect slopes in minute scattered areas of many foreign countries during the manned
detail. Give particular attention to the following features: satellite SKYLAB missions (May 1973 through February
1974). SKYLAB data of special interest to landslide inves-
Slide Feature tigators are those collected by (a) the multispectral camera
Existing Hillside scarps and hummocky topography
(S-190A) at a scale of 1:2 850 000 with 30 to 79-m (100 to
Parallel spoon-shaped dark patches on hillsides 260-ft) resolution and (b) the earth terrain camera (S-190B)
Irregular outlines of highways and random cracks or at a scale of 1:950 000 with 17 to 30-m (55 to 100-ft) reso-
patches on existing pavements. lution. Black and white, black-and-white infrared, color,
Potential Ponded depressions and diverted drainageways
and color infrared products are included. Detailed informa-
Seepage areas suggested by faint dark lines, which may
mean near-surface channels, and fan-shaped dark tion about film types, spectral characteristics, and areal coy-
patches, probably reflecting wet vegetation erages has been published by NASA (3.21). Compared with
LANDSAT products, SKYLAB products provide better reso-
Relatively new slides appear in light tones because vegeta- lution but lack universal coverage and repeated sequence.
tion and drainage are not well established. The parallel At the small scale of satellite imagery, only extremely
spoon-shaped dark patches on hillsides are likely to reflect large landslides can be identified directly. Figure 3.31
vegetation in minor depressions. Lines drawn through the shows a typical stereographic aerial photograph coverage
axes of the scarps in the slides often point to drainageways of a large landslide zone. Figure 3.32 shows the satellite
on higher ground that contribute to landslide movement. scene covering this landslide zone. The scalloped edges of
Ground check all suspected slides. the scarp slopes and the hummocky topography of this
large slide are evident on the satellite scene.
Use of Other Remote-Sensing Since most landslides are much smaller than the land-
Systems slide shown in Figure 3.31,they are not directly identifiable
on satellite imagery. However, the value of satellite imagery,
Multisensor investigations for terrain analysis and landslide as noted by Alföldi (3.1) and Gagnon (3.11), is that the
66
Figure 3.30. Unstable talus
slopes in Yakima County,
Washington, are indicated
by steep slopes either bare of
vegetation or with streaks of
vegetation remnants (1). Some
debris avalanches (2) are also
evident on steep valley walls.
Instability of slopes was
recognized at time of location
of road. Alternate location on
opposite side of valley was
considered worse because of
numerous deep drainage
channels in addition to talus
slides.
landslide susceptibility of an area can be determined indi- most beneficial for landslide investigations.
rectly from some of the features that are identifiable at
those scales. Regional physiography, geologic structure, Infrared Imagery
and most landfornis as well as land-use practices and distri-
bution of vegetation are evident on the satellite imagery. The infrared region extends from approximately 0.7 pm to
These features in conjunction with the tonal patterns pres- I mm (Figure 3.13). Atmospheric interference, however,
ent on the imagery provide clues to the types of surface ma- limits the areas within the infrared region available for inves-
terials present, the surface moisture conditions, and the pos- tigation to only certain clear zones or "windows." Some of
sible presence of buried valleys. Correlating these factors the common windows used in infrared surveys are in the
to geology and topography and using local experience in a following bands: 1.0 to 1.4 pm, 1.5 to 1.8 pm, 2.0 to 2.6
region make it possible to rate the susceptibility of various pm, 3.0 to 5.0 pm, and 8 to 14 pm. Daytime infrared sur-
areas to sliding. For example, Alföldi noted in his study of veys collecting data in the region below 3.0 to 3.5 pm re-
landsliding in eastern Ontario that on the satellite image the cord infrared reflectance phenomena from various objects
clay plains were easy to spot because they are almost 100 in the scene. Daytime and nighttime surveys collecting data
percent cultivated; the till plains were recognizable because in the region above 3.0 to 3.5 pm record infrared heat-
they form a poorer agricultural area and field and forest sec- emission phenomena.
tions are intermixed; and the elevated sand plains of the old Since two basically different phenomena are being re-
Ottawa River delta (which overlay the clay plains) are kept corded in the infrared region (i.e., reflectance and heat emis-
mostly in forest. sion), a distinction is made in the terminology used to indi-
An additional advantage noted by Alföldi for satellite cate these phenomena. Images recorded in the bands below
imagery was the frequent coverage available. Seasonal 3.0 to 3.5 pm are referred to as infrared reflectance and
changes in vegetative cover and moisture levels—as indicated those recorded in the bands above as infrared imagery. The
by tonal changes—can be evaluated to increase the accuracy most useful window for terrain analysis is the 8 to 14-pm
of interpretation of terrain conditions. Also, any changes band. Further discussions on the characteristics of infrared
noted during the year in the landslide-susceptible zones, surveys are given by Reeves (3.25) and by Rib (3.26).
such as urban expansion, clear-cutting of forest, forest fires, Infrared imagery offers some unique information that
and draining of swamps, might presage renewed or new land- cannot be obtained directly from the analysis of aerial
slide activities. This could alert the interpreter to the neces- photography. The combination of aerial photography and
sity for a more detailed investigation in these areas. infrared imagery provides a more accurate and complete
Satellite photography and imagery for large parts of the portrayal of terrain conditions than can be obtained from
world are available from the EROS Data Center of USGS. either system alone. Infrared imagery provides the follow-
They can be ordered by providing the geographic coordi- ing types of supplemental information that is valuable for
nates (latitude and longitude) of the area of interest or by evaluatingexisting landslide and landslide-susceptible terrain:
indicating particular frame numbers from the Single Land-
sat Coverage Map available from the above-noted organiza- Surface and near-surface moisture and drainage con-
tion. Canadian imagery can be obtained from the Canada ditions;
Centre for Remote Sensing, Department of Energy, Mines Indication of the presence of massive bedrock or bed-
and Resources. Addresses of these organizations are given rock at shallow depths;
earlier in this chapter. Experience has indicated that band Distinction between loose colluvial materials that are
5 (0.6 to 0.7 pm) or the infrared color composite is the present on steep slopes and are susceptible to landslides
67
and the massive bedrock that is more stable on steep slopes; seepage areas, runoff, and standing water are dark on both
and images. On the predawn infrared imagery in Figure 3.34,
4. Diurnal temperature changes that occur in soil masses standing water (points a and f) is warmer and has light
(these provide clues to the soil-water mass conditions). tones while seepage zones and runoff (points c and d) are
cooler and have darker tones. On the photography, all of
Tanguay and Chagnon (3.33) demonstrated the value of these areas have dark tones. Vegetated areas occurring be-
infrared imagery and aerial photography for evaluating the tween the areas of standing water in this figure have a me-
moisture and drainage regime associated with a landslide. dium tone on the imagery and appear as white specks on a
A flow slide had occurred within the crater of a former dark background on the photography.
slide in clay lake beds in the vicinity of Saint-Jean-Vianney, Figure 3.35 shows a nighttime (predawn) infrared image
Quebec. To plan a drilling program to evaluate the poten- of an area along a railroad line being investigated for locat-
tial of further movements required that the areas of seepage, ing potential areas for landslides. The railroad had been
water runoff, and wet soils be identified. By means of plagued for years with landslide problems. The circled
photography alone, these features could not be uniquely darker areas were interpreted as zones of seepage and high
separated from areas of standing water, topographic shad- moisture levels -potential for landslides. Based on this anal-
ows, and dense vegetation (brush and forested zones) be- ysis, the circled areas were drained, and no further slides
cause they all produced similar dark tonal patterns. 1-low- have occurred in those areas.
ever, the combination of photography and daytime and Another example of the value of infrared imagery is
nighttime imagery made it possible to separate these va- shown in Figure 3.36, which illustrates nighttime infrared
rious features and identify the critical items for planning imagery of an area of tilted sedimentary rocks. The mas-
the drilling program. sive bedrock areas (point 2) are indicated by light tones.
Figures 3.33 and 334 show some of the results reported The fractured rock zones and colluvial slopes, which are
by Tanguay and Chagnon. Figure 3.33 shows daytime ther- more susceptible to landslides, are indicated by medium
mal infrared imagery obtained in the 8 to 14-jim band and dark tones (point 3). These types of data would be useful
black-and-white infrared photographs of a portion of the in conjunction with aerial photography for rating the land-
area shown in the imagery. Several of the seepage areas slide susceptibility of the terrain.
(points a and b) interpreted from this figure were drilled Other remote-sensing techniques, such as multispectral
and indicated very deep and soft clay starting from the sur- imagery and microwave radiometry, offer potential for in-
face. The c points indicate both seepage and runoff in a dicating the presence of landslides. Multispectral systems
farm field: the d points show standing water coming from obtain simultaneous coverage of data in several spectral re-
snowmelt and surface drainage: and the e points depict the gions- usually a portion of the ultraviolet, the visible, and
boundary of the recent slide. Point f indicates the upper- both infrared reflectance and infrared emission regions.
most boundary of an older slide surface. In this figure, The data are collected on magnetic tape, which makes them
6
amenable to the use of computer analysis techniques. Micro- time either further development is needed or the systems
wave radiometers collect radiomet nc (temperature varying) are too costly in comparison to the level of information
data in the microwave or radar regions. The particular ad- furnished. However, they do offer some future potential
vantage of this system is that at the longer wavelengths in- for landslide investigations.
formation from subsurface layers is included in the data.
Rib and others (3.27) have demonstrated that, under cer- FIELD RECONNAISSANCE
tain condit ions, informal ion on subsurface moist tire condi- TECHNIQUES
tions (i.e., presence of zones of high moisture level) can be
determined by comparative evaluation of photography, The discussion in this section is based to a large extent on
nighttime infrared imagery, and microwave radiometry. informal ion in Chapter 4 in the book on landslides published
These techniques are not described in detail because at this by the Highway Research Board in 1958 (3.10).
Figure 3.32. LANDSAT-1 satellite scene 1657-17031, band 5, May 11, 1974, north central New Mexico. Arrow pointsto edge of small
basalt remnant shown in Figure 3.31. Massive landslides are evident on satellite scene by irregular, scalloped edges. Hummocky
appearance of slide material can be discerned by close examination of scarp edges.
-
'r
.
- • .
-- .
'•.i
69
Figure 3.33. Daytime thermal infrared imagery in 8 to 14-11m band (above) and mosaic prepared from black-and-white
infrared photography (below) taken in Ontario, Canada (3.33). Lines AB and AB' represent equivalent coverages on
the two image formats. Lettered areas are discussed in text.
B,
IA'
C ,,
b-
Figure 3.34. Predawn infrared imagery in 410 5-,m band (above) and mosiic prepared from black•and•white infrared
photography (below) taken in Ontario, Canada (3.33). Line c-c represents common line on both Images. Lettered
areas are discussed in text.
dbb
: Ak Al-
P. 1kIJ e
70
Figure 3.35. Nighttime infrared imagery.
0645 h, station 560 to 580, Muskingum
Mine Railroad in Ohio. Dark-toned regions
within circled areas are interpreted as zones
of seepage and high moisture levels and
have been associated with slides along
railroad right-of-way. Light-toned areas
(1) represent standing water on surface.
An important phase in the preliminary investigation is mnent. Evidence may also be found of landslide movement
field reconnaissance to verify the three-dimensional concept that has not yet affected the highway but that may do SO in
of the terrain developed from the literature review and the time: Minor failure in an embankment, material that falls
analysis of remote-sensing data. Many of the fine details and on the roadway from an upper slope, or even the progres-
more subtle evidences of slope movement cannot be identi- sive failure of the region below a fill may presage a larger
fied at the scales of photographs and maps analyzed and can landslide that will endanger the road itself. Other evidences
be detected only in a field survey. Until that survey is made, of movement are to be found in broken pipe or power lines,
part, or even most, of the concepts developed remains con- spalling or other signs of distress in concrete structures,
jecture. Performing a field survey at the time corrective closure of expansion joints in bridge plates or rigid pave-
measures are being planned is especially important. Many merits (Figures 3.37 and 3.38), or loss of alignment of build-
landslides are complex and as time passes frequently change ing foundations. In many eases, arcuate cracks and minor
their physical characteristics and marks of identification. scarps in the soil give advance notice of serious failure.
For instance, a landslide examined a year after its occurrence Table 3.3 gives the chief evidence of movement in the va-
may have changed remarkably from the conditions immedi- rious parts of each type of landslide. The following discus-
ately following the original movement. If a landslide de- sion elaborates on the significance of cracks in recognizing
veloped as a slump slide and eventually turned into a flow, and classifying slides and on features that aid in identifying
the orinal report on the slide would be invalid as a basis the various types of landslides.
for planning a correction of the slide. The identification of The ability to recognize small cracks and displacements
type should be made at the same time the slide is to be cor- in the surface soils and to understand their meaning de-
re c ted. serves cultivation because it can produce accurate knowl-
edge of the cause and character of movement that is pre-
Field Evidence of Movement requisite to correction. Surface cracks are not necessarily
normal to the direction of ground movement, as is com-
The term landslide, by definition, implies that movement nionly assumed by some. For example, cracks near the
has taken place: hence, an analysis of the kind and amount head of a slump are indeed normal to the direction of hori-
of movement becomes a key to the nature of an active slide. zontal movement, but the cracks along its flank are nearly
It is imperative to learn to recognize the features that first parallel to it.
indicate the onset of ground movement and to learn to rec- Small en echelon cracks commonly develop in the stir-
ognize the kind of slide that already exists. Quite com- face soil before other signs of rupture take place: thus, they
monly the first visible sign of ground movement is recorded are particularly important in the recognition of potential
by settlement of the roadway or, depending on the location or incipient slides. They result from a force couple in which
of the roadway within the moving mass, a bulge of the pave- the angle between the direction of motion and that of the
71
cracks is a function of the location within the landslide area ample, in a slump the walls of cracks are slightly curved in
Thus, in many cases a map of the en echelon cracks will de- the vertical plane and concave toward the direction of move-
lineate the slide accurately, even though no other visible ment; if the rotating slump block has an appreciable vertical
movement has taken place (Figure 3.39). offset, the curved cracks wedge shut at depth. In block
In addition to indicating incipient or actual movement, slides, on the other hand, the cracks are nearly equal in
cracks in surface soils are locally useful in helping to deter- width from top to bottom and do not wedge out at depth
mine the type of slide with which one is dealing. For cx- because failure in a block slide begins with tension at the
base of the block and progresses upward toward the surface.
Figure 3.37. Early signs of impending debris slide along Clear Creek,
The presence of a few major breaks in the upper parts of a
Colorado (3. 10). Displacement of fence, bulging of pavement, and block slide (Figure 3.40) distinguish it from lateral spreading,
distress in bridge abutment (Figure 3.38) all give early indications which is characterized by a maze of intersecting cracks. The
of movement at toe of incipient slide. In several places, now inclination of cracks is commonly almost vertical in block
covered by patching, centerline stripe was offset along cracks.
slides in cohesive soils, regardless of the dip of the slip plane,
but depends on the joint systems in the rock in block slides
q
.k
t;-
of rock.
Distinguishing between incipient block slides and slumps
is one of the most helpful applications of a study of cracks.
If the outline of the crack pattern is horseshoe-shaped in
plan, with or without concentric cracks within it, a slump
is almost certainly indicated. If, on the other hand, most
of the surface cracks are essentially parallel to the slope or
ft' cliff face, a block slide is probably in the making. In either
..iI..
; ,. case, additional cracks may develop as major movement gets
under way, but these will generally conform to the earlier
crack pattern (Figures 2.14 and 2.15).
Evidence of soil creep and of "stretching" of the ground
surface should also be sought. Stretching indicates compar-
atively deep-seated movement whereas soil creep is of sur-
ticial origin. The phenomenon of stretching is most corn-
.;
nionly observed in noncohesive materials that do not form
or retain minor cracks readily. The best evidence of stretch-
ing consists of small cracks that surround or touch some
Figure 3.38. Right wing wall shown in lower center of Figure 3.37 rigid body, such as a root or boulder, in otherwise homno-
(3. 10). Distress in bridge abutment indicates incipient slide. In
addition to offset of wing wall shown here, rockers beneath bridge
gencous materials: these cracks form because the tension
girders were tipped. forces tend to concentrate at or near the rigid bodies.
/
EN ECHELON
/ CRACES
ZONE
OF UPLIFT
-
FOOT
'\
ME
72
place or is in progress, the next essential step is to identify rather than concave, cracks. Lateral spreading with few, if
the type of movement. Table 3.3 gives the surface features any, cracks, on the other hand, is characteristic of earth
that aid in the field identification of slope movements. flows.
This table gives only the most common types and not all Translational rock slides are generally easy to recognize
of the types shown in Figure 2.1 . Further generalizations because they are composed wholly of rock, boulders, or
are given in the following paragraphs. rock fragments. Individual fragments may be very large and
may move great distances from their source. Rock slides
Falls and Topples
Figure 3.40. Block slide in cohesive materials, near Portage.
Although the initial mecha ii isms of falls and topples are dif-
Montana (3. 10). The slide, in alluvium, colluvium, and some wind-
ferent (Chapter 2), their final appearances after movement deposited silt, is moving out over surface of alluvium-filled stream
are similar. Falls and topples are best recognized by the ac- channel with little or no rotation of block and without developing
cumulation of material that is not derived trorn the under- zone of uplift at foot. Parallel step scarps along main scarp and
drag effects along flanks are both characteristic of block slides.
lying slope and that is foreign to normal processes of erosion.
Arrows indicate overbreak cracks that develop after main scarp is
In most cases, this material consists of blocks of rock, de- formed; possibly because of abrupt break in slope, these are more
bris. or earth scattered over the surface or forming a talus sharply curved than are those above most slump slides.
slope. If undercutting by lake or stream waters has caused
the fall, the rate of failure is proportional to the ability of
the water to remove the fallen material. A fast-moving
stream may remove material almost as fast as it falls, thus
removing the evidence but encouraging continuing falls. :
On the other hand, a lake or some parts of an ocean shore
must depend only on wave action to disintegrate and re- .--- - - ' '
move the fallen material: hence, the evidence tends to re-
main in sight but continued falls tend to be inhibited. Most
of the material yielded by rock topples and rock falls is
necessarily close to the steep slopes from which it came,
but some may bound down the slope and come to rest far
from its present source.
In active or very recent slides, the parent cliff is com-
mon!)' marked by a fresh irregular scar that lacks the cres-
cent shape that is characteristic of slumps. Instead, the ir- Figure 3.41. Rock slide and rock fall along Penn-105
regularity of its surface is controlled by the joints and bed- (3. 10). Beds are nearly flat; slide is controlled by joints
ding planes of the parent material (Figures 2.3 and 3.41). dipping at steep angle toward road.
Some idea of the intensity and state of activity of a rock
fall or rock topple can be inferred from the presence or ab-
sence of vegetation on the scarp and by the damage done to
trees by the falling rocks. In active areas, the trees are
scarred and debarked or show evidence of healed wounds.
If the rock fall or rock topple is severe or of long duration -
conifers and other long-lived trees are absent: their places
may or may not be taken by aspen or other fast-growing
species. Many rock falls follow chutes or dry canyons that
can usually be differentiated from normal watercourses or
paths cut by snow avalanches (Figure 3.42).
Some debris and earth falls and topples exhibit most of
the characteristics of rock falls and rock topples. Others
proceed by mere spallirig of the surface, but even this activ-
ity, it' long continued, can lead to removal of considerable
quantities of material. Any fall or topple may • of course.
presage major landslide movement in the near future.
Slides
73
Table 3.3. Features that aid in recognition of common types of slope movements.
Falls, topples Rock Consists of loose rock; probably Is usually almost vertical, irregu- Are mostly bare edges of rock
- has cracks behind scarp; has ir- lar, bare, and fresh; usually con-
regular shape controlled by local sists of joint or fault surfaces
joint system
Soil Has cracks behind scarp Is nearly vertical, fresh, active, Are often nearly vertical
and spalling on surface
Slides
Rotational
Slump Soil Has numerous cracks that are Is steep, bare, concave toward Have striae with strong vertical
mostly curved concave toward slide, and commonly high; may component near head and
slide show striae and furrows on sur- strong horizontal component
face running from crown to near foot; have scarp height
head; may be vertical in upper that decreases toward foot;
part may be higher than original
ground surface between foot
and toe; have en echelon
cracks that outline slide in
early stages
Rock Has cracks that tend to follow Is steep, bare, concave toward Have striae with strong vertical
fracture pattern in original rock slide, and commonly high; may component near head and
show striae and furrows on sur- strong horizontal component
face running from crown to near foot; have scarp height
head; may be vertical in upper that decreases toward foot;
part may be higher than original
ground surface between foot
• and toe; have en echelon
cracks that outline slide in
early stages
Translational
Block Rock Has cracks most of which are Is nearly vertical in upper part Have low scarps with vertical
or nearly vertical and tend to fol- and nearly plane and gently to cracks that usually diverge
Soil low contour of slope steeply inclined in lower part downhill
Rock Rock Contains ioose rock; has cracks Is usually stepped according to Are irregular
between blocks spacing of joints or bedding
planes; has irregular surface in
upper part and is gently to
steeply inclined in lower part;
may be nearly planar.or com-
posed of rock chutes
Flows
Dry Rock Consists of loose rock: probably Is usually almost vertical, irregu- Are mostly bare edges of rock
has cracks behind scarp; has ir- lar, bare, and fresh: usually con-
regular shape controlled by local sists of joint or fault surfaces
joint system
Soil Has no cracks Is funnel shaped at angle of repose Have Continuous curve into main
scarp
Wet
Debrisavalanche Soil Has few cracks Typically has serrated or V-shaped Are §1eep and irregular in upper
Debris flow upper part: is long and narrow, part; may have Ieveesbuilt up
bare, and commonly striated in lower parts
Earth flow Soil May have a few cracks Is concave toward slide; in some Are curved; have steep sides
types is nearly circular and slide
issues through narrow orifice
Sand flow • Soil Has few cracks Is steep and concave toward slide; Commonly diverge in direction
Silt flow may have variety of shapes in of movement
outline: nearly straight, gentle
arc, circular, or bottle shaped
Note: This table updates Table 1 in book on landslides published in 1958 by Highway Research Board (3.10, chap. 4 by A. M. Ritchie, pp. 56.57)
74
Parts That Have Moved
Is usually not well defined; con- Falls:. Has irregular surface of Is commonly buried; if visible, Is irregular pile of debris or talus
sists of fallen material that jumbled rock that slopes away generally shows evidence of rea- if slide is small; may have
forms heap of rock next to from scarp and that, if large son for failure, such as promi- rounded outline and Consist of
scarp and if trees or material of Con- nent joint or bedding surface, broad, curved transverse ridge
trasting colors are included, underlying weak rock, or banks if slide is large
may show direction of move- undercut by water
ment radial from scarp; may
contain depressions
Topples: Consists of unit or
units tilted away from crown
Is usually not well defined; con Is irregular Is commonly buried; if visible, Is irregular
sists of fallen material that generally shows evidence of rea-
forms heap of rock next to son for failure, such as promi-
scarp nent joint or bedding surface,
underlying weak rock, or banks
undercut by water
Has remnants of land surface flatter Consists of original slump blocks Commonly has transverse cracks Is often a zone of earth flow of
than original slope or even tilted generally broken into smaller developing over foot line and lobate form in which material
into hill, creating at base of masses; has longitudinal cracks, transverse pressure ridges de- is rolled over and buried: has
main scarp depressions in which pressure ridges, and occasional veloping below foot line; has trees that lie flat or at various
perimeter ponds form; has trans- overthrusting; commonly de- zone of uplift, no large individ- angles and are mixed into toe
verse cracks, minor scarps, velops small pond just above ual blocks, and trees that lean material
grabens, fault blocks, bedding foot downhill
attitude different from surround-
ing area, and trees that lean up-
hill
Has remnants of land surface flatter Consists of original slump blocks Commonly has transverse cracks Has little or no earth flow; is often
than original slope or even tilted somewhat broken up; has little developing over foot line and nearly straight and close to foot:
into hill, creating at base of plastic deformation: has longi- transverse pressure ridges de- may have steep front
main scarp depressions in which tudinal cracks, pressure ridges, veloping below foot line: has
perimeter ponds form; has trans- and occasional overthrusting: zone of uplift, no large individ-
verse cracks, minor scarps, commonly develops small pond ual blocks, and trees that lean
grabens, fault blocks, bedding just above foot downhill
attitude different from surround-
ing area, and trees that lean up-
hill
Is relatively undisturbed and has Is usually composed of single or Has none and no zone of uplift Plows or overrides ground surface
no rotation few units; is undisturbed ex-
cept for common tension cracks
that show little or no vertical
displacement
Has many blocks of rock Has rough surface of many lisually has none Consists of accumulation of rock
blocks, some of which may be fragments
in approximately their original
attitude but lower if movement
was slow translation
Has none Has irregular surface of jumbled Has none Composed of tongues; may over-
rock fragments sloping down ride low ridges in valley
from source region and gener-
ally extending far out on valley
floor: shows lobate transverse
ridges and valleys
Usually has none Is conical heap of soil, equal in Has none
volume to head region
May have none Consists of large blocks pushed Is absent or buried in debris Spreads laterally in lobes; if dry,
along in a matrix of finer mate- may have a steep front about a
rial: has flow lines: follows drain- meter high
ageways and can make sharp
turns; is very long compared to
breadth
Commonly Consists of a slump Is broken into many small pieces: Has none Is spreading and lobate; consists
block shows flow structure of material rolled over and
buried; has trees that lie flat
or at various angles and are
mixed into toe material
Is generally under water Spreads out on underwater floor Has none lsspreading and lobate
75
commonly occur only on steep slopes. The numerous small Figure 3.42. Rock fall and rock slide near Skihist, British
block units with random rotation are mixed in a matrix of Columbia, on Canadian National Railroad (3.10). Note bare
active slopes, closely spaced jointing of rocks, rock chutes, and
finer grained material; the large rock fragments tend to float
absence of water. Wooden and concrete sheds bypass debris
on or in the matrix. Rock slides have translational failure falling over tracks from above.
surfaces rather than the concave surfaces of slumps, and
they do not move as unrotated multiple units, like block
slides. Most rock slides are controlled by the spacing of
joints and bedding planes in the original rock. There is par-
ticular danger of forming a rock slide of serious proportions
if construction is undertaken in an area marked by a system
of strongly developed joints or bedding planes that dip
steeply outward toward the natural slope. This is especially
true if the natural slope angle is steeper than the angle of re-
pose of the broken rock. Water is not so important a factor
in causing rock slides as in causing slides in soil. In sonic in-
stances, however, water helps to weaken bedding or joint
planes that otherwise would offer high frictional resistance.
Any seepage that is apparent after a rock slide has taken
place is most likely to be visible in the scarp region or, per-
haps, in the slide material itself.
Many rock slides are thinner than either slumps or block
slides because they commonly are restricted to the weathered
zone in bedrock or to surficial talus. The shape of the shear
zone, therefore, conforms to the unweathered bedrock sur-
face and is not controlled to any large extent by joints or even so they may be indicated by evidence of surface stretch-
bedding planes in the bedrock. Many talus slopes produce ing, by lines of rock fragments that have been displaced, or
rock slides by failure within the body of talus. even by blades of grass that have been pulled down into the
Rotational slumps rarely form from solid. hard rocks, cracks by sand or other loose surface material as it sifted
although special combinations of factors have been known into them. The head region of a slump or an earth block
to produce them. Slumps are widespread, however, in slide with surficial slump characteristics may also be recog-
sands, silts, and clays and in the weaker bedded rocks. nized by the presence of slump grabens that have experi-
There they can be identified readily from surface indica- enced some rotation (Figures 3.19 and 3.43). These are de-
tions, though only after considerable movement has taken pressed fault blocks of soil or rock, caused in part by de-
place. The head region of a slump is characterized by steep crease in curvature of the shear plane. This produces ten-
escarpments and by visible offsets between separate blocks sion and ultimate failure in the main slump block because
of material (Figure 3.19). The highest escarpment is com- of lack of support on its uphill side.
monly just below the crown: because the crests of the flanks The amount and direction of rotation that any slump
are lower than the crown, they can be recognized as flanks, block has undergone can usually be ascertained by deter-
even if the escarpment on one of them happens to be larger mining the slope of its surface as compared with that of the
than the one below the crown. If the landslide is active or original slope. Comparison of the (lip and strike of bedded
has been active recently, the scarp is bare of vegetation and material within the slump block with the original attitude
may be marked by striations or grooves that indicate the di- of the unslumped material is an even more exact and fool-
rection of movement that has taken place. Striations at the proof method of determining the amount of rotation, be-
head tend to reflect downward movement whereas striations cause erosional alternation of the surfaces may give an er-
along the flanks may be nearly horizontal. If the slump is roneous impression. In a general way, the amount of ro-
compound, its several crescent-shaped scarps will appear as tation is a measure of the amount of displacement.
a scalloped edge in plain view (Figure 2.21). The part just above the foot of a slump is a zone of com-
Undrained depressions and perimeter lakes, bounded on pression. The slumped material is con fined by the foot and
the uphill side by the main scarp, characterize the head re by the flanks of the main scarp SO that it is compressed by
gions of many slumps: even if internal drainage prevents the load above into the bot torn part of the howl-shaped sur-
such ponds from holding water for long periods, their de- face of rupture. In this region, therefore, there are no open
pressions may be evident. In humid regions, the head area cracks. The foot region is marked by a zone of tension and
may remain greener than surrounding areas because of the uplifl because the slumped material is required to stretch
swampy conditions. In the San Juan region of south- over the foot before it passes farther downhill. This stretch-
western Colorado. for instance, groves of aspen trees are ing destroys any remaining slump blocks because of the
commonly good indications of wet ground conditions and change in direction of the forces. The small blocks and frag-
hence of slides and unstable ground. In northern West Vir- ments that result tend to weather to rounded forms, produc-
ginia, the swampy areas in the head remain green during ing hummocky ground. Pavement uplifts and cracks, so
the winter whereas the well-drained low areas are brown. disconcerting to the motorist and highway engineer. are
The tension cracks near the head of a slump are generally also most likely to take place in the foot zone. Seeps,
concentric and parallel to the main scarp. Many such cracks springs, and marshy conditions commonly mark the foot
are obscured by rubble or other noncohesive materials, but and toe of a slump. Moreover, trees tend to be tilted down.
76
Figure 3.43. Graben on slump slide (3. 10). Rotation of slump without rotation, into the space left by removal of the
block is uphill, resulting in flattening of original ground surface, plastic layer.
but graben block, which breaks off from slump block, rotates
downhill. Grabens do not form if slump block has sheared on surface
Dry flows are not difficult to recognize after they have
that approaches arc of circle; instead, they form on slump blocks that occurred, but it is virtually impossible to predict them in ad-
slide over principal surface of rupture having marked decrease in its vance. They are commonly very rapid and short-lived. Dry
curvature, causing greater horizontal movement for each unit of flows are rarely composed of rock fragments; they usually
vertical offset in center portion of slide than in head region.
consist of uniformly sized silt or sand. They exhibit no
cracks above the main scarp and have poorly developed or
nonexistent flow lines. Except for sand runs, they have no
HEAVY LINE INOICATES
GRV8EN / ORIGINAL SURFACE well-defined foot. Dry bess may become fluid and flow
down a slope because of an earthquake or other external
vibrations. Sand runs also behave somewhat as fluids, but
MP J// in the latter part of their course the sand particles are more
likely to slide than to flow.
Wet flows occur when fine-grained soils, with or without
coarser debris, become mobilized by an excess amount of
water. Most of them behave like wet concrete in a chute
(differences are due to water content), but the flow of
some wet silts and fine sands is triggered by shocks. "Quick"
Figure 3.44. General orientation of trees on slump landslide (3. 10). or sensitive clays may be liquefied by the leaching of salts
As a result of rotation, trees on blocks are bowed uphill because or by other causes that are not completely understood.
tree tops tend to grow vertically while stump portion changes with Wet flows are generally characterized by their great
rotating land surface. Contrast head and toe regions and compare
lengths, generally even gradients and surfaces, absence of
with Plates 3.2 and 3.3.
tension cracks, and lack of blocky units and minor scarps.
If tension cracks are present, they are bowed in the direc-
tion of movement (Figure 3.45), showing the effects of
movement of wetter materials at depth beneath the drier
crust. An older flow that has had time to dry may show
large shrinkage cracks or flow lines. In many cases, the
main scarp area is emptied by the removal of all flow ma-
terial and resembles a glacial cirque. In other cases, the gra-
dation may be imperceptible downward from soil creep to
mud flow.
The rate of flow is dependent on the total amount of
material that feeds the slide; the accumulated debris im-
hill rather than uphill as they near the head (Figure 3.44 poses a load on the entire mass below it and tends to main-
and Plate 3.3). tain a constant rate of movement. The flow is under pres-
The approximate age of some slumps can be inferred by sure everywhere from the material above it; consequently,
study of the bent trees. If older trees are bent but younger the mass shows few, if any, cracks over the foot. Flows can
ones are straight, for instance, the slide probably has not and do make sharp turns and move around firmly fixed ob-
moved during the life of the younger trees. On the other stacles that appear in their paths. A very wet flow moving
hand, the sizes of the tree trunks at the points where the at high velocity may leave debris marks high on trees or
bends occur give a running history of the rotation. other objects and ridges of debris called torrent levees along
Just below the foot of a slump the ground is commonly its sides.
marked by long transverse ridges that are separated from
one another by open tension cracks. These cracks seldom Hidden Landslides
remain open for long periods of time, and they do not form
scarps or other evidence of displacement because the mate- Among the most difficult kinds of slides to recognize and
rial is no longer confined but spreads out laterally and de- guard against are old landslides that have been covered by
velops radial cracks at the toe. glacial till or other more recent sediments. To predict, in
detail, the existence of old buried slides or the effect that
Lateral Spreads and Flows they may have on new construction work that exposes
them is probably impossible. One who knows the recent
In lateral spreads, a coherent upper block is broken into geologic history of the region intimately, however, may be
strips or blocks and extended or stretched by plastic flow able to make some educated guesses as to the probable ex-
or liquefaction of the lower layer. The mechanism of fail- istence of such slides and even as to where they are most
ure can involve not only rotation and translation, but also likely to be found. One example is shown in Figure 3.46;
flow, and thus surface characteristics will contain elements the indicated unstable body of soft shale and chalk is
of both slides and flows. One form of lateral spread, in clearly related to a fault in the bedrock, but it also has the
which a plastic layer is squeezed out by the weight of an characteristics of a surface landslide. Since this slide took
overlying rigid layer, is here termed a piston slide. In place, the area was covered by two or more layers of bess,
this type, a part of the upper layer may drop vertically, which completely obscured the landslide until the rising
VIA
Figure 3.45. Crack pattern in slump on south
side of Reservoir Hill, Dunbar, West Virginia,
indicates flowage at depth beneath harder
material at surface (3. 10). Broken pipes from
reservoir at top of hill dumped large amount EDGE OF
of water into old slide and reactivated it. Elm 49m
Horseshoe-shaped scarp is imperfect,
differentiating it from that of true slump.
Greatest movement is near center of slide,
as indicated by arrangement of cracks and
of standing water. Presence of convex
cracks is indicative of flow movement at
depth.
Etev. 28 m
Elev. 23 m
Elev, 23 m
EI~
lm
,Etev. 17 m
oo
EIev. 12 m
30 m
EIev. 9 m
,
/ bOlt
/
Figure 3.46. Hidden landslide exposed when overburden of bess one is most likely to occur. The evidence for distressed con-
was removed by bank-cutting along shore of Fort Randall Reservoir
ditions that may be present or induced lies chiefly in evi-
in South Dakota (3. 10). Buried soil profiles indicate two periods
of bess deposition after faulting and landsliding took place in
dence of movements, minor or major, that have already
underlying shale and chalk of Cretaceous age. All surface evidence taken place or of geologic, soil, and hydrologic conditions
of landsliding was obliterated by bess until lake waters cut new that are likely to cause movement in the future. Once the
face. fact of land movement, actual or potential, has been estab-
lished, the next essential step is to identify the type of land-
slide. One would not apply the same corrective procedure
to a rock fall and a block slide any more than one would at-
tempt to prevent a slide without knowing the kind of slide
expected. If maximum benefit is to be accrued from the
preventive or corrective measures finally employed, it is
imperative to learn to recognize the kind of slide that ex-
-
- --- -
ists or is expected.
One final point needs emphasizing. Even if the prelim-
*---
inary examination of the general environment has indicated
that no landslide movements have yet taken place or are
imminent, the investigator must still determine whether
the ground to be disturbed by the proposed construction
will prove reasonably stable. No one is capable, nor is
money available, of studying in detail and of guaranteeing
waters of the lake undercut the bank and exposed the old the stability of all slopes for most large construction proj-
slide. ects. As a general rule, the amount of investigation that is
warranted is a function of the landslide susceptibility of the
CONCLUSIONS surrounding country and also of the degree of damage that
might be expected to occur to persons or installations if a
All landslide investigations must start with either the recog- slide should occur. In other words, the more serious the
nition of a distressed condition on the natural or artificial consequences of a landslide are, the more detailed the
slope or the determination of the vulnerable locations where search for potential slides should be.
78
REFERENCES eastern Ohio. Ohio Department of Highways, Columbus,
Vols. 1 and 2, 1965.
3.23 Ray, R. G. Aerial Photographs in Geologic Interpretation
and Mapping. U.S. Geological Survey, Professional Paper
3.1 Alföldi, T. T. Regional Study of Landsliding in Eastern 373, 1960, 230 pp.
Ontario by Remote Sensing. Department of Civil Engi- 3.24 Radbruch-Hall, D. H., Colton, R. B., Davies, W. E., Skipp,
neering, Univ. of Toronto, MS thesis, 1974,79 pp. B. A., Lucchitta, I., and Vanes, D. J. Preliminary Land-
3.2 Anderson, R. R., Hoyer, B. E., and Taranik, J. V. Guide slide Overview Map of the Conterminous United States.
to Aerial Imagery of Iowa. Iowa Geological Survey, U.S. Geological Survey, Miscellaneous Field Studies Map
Public Information Circular 8, Sept. 1974, 142 pp. MF-771, 1976.
3.3 Baker, R. F., and Chieruzzi, R. Regional Concept of 3.25 Reeves, R. G., ed. Manual of Remote Sensing. American
Landslide Occurrence. Highway Research Board, Bulletin
Society of Photogrammetry, Falls Church, Va., Vols. 1
216, 1959, pp. 1-16. and 2, 1975, 2144 pp.
3.4 Bruce, R. L., and Scully, J. Manual of Landslide Recog- 3.26 Rib, H. T. An Optimum Multisensor Approach for De-
nition in Pierre Shale, South Dakota. South Dakota De-
tailed Engineering Soils Mapping. Purdue Univ., Lafayette,
partment of Highways, Pierre, Final Rept. of Research
Ind., Joint Highway Research Project 22, Vols. 1 and 2,
Project 615(64), Dec. 1966, 70 pp. 1966, 406 pp.
3.5 Chassie, R. G., and Goughnour, R. D. National Highway 3.27 Rib, H. T., Spencer, J. M., Jr., Falls, C. P., and Koca, J. F.
Landslide Experience. Highway Focus, Vol. 8, No. 1,
Evaluation of Aerial Remote Sensing Systems for Detect-
Jan. 1976, pp. 1-9.
ing Subsurface Cavities in Kansas. Federal Highway Ad-
3.6 Cleaves, A. B. Landslide Investigations: A Field Hand-
ministration, Rept. FHWA RD-75-119, 1977.
book for Use in Highway Locations and Design. U.S. 3.28 Scully, J. Landslides in Pierre Shale in Central South
Bureau of Public Roads, 1961,67 pp.
Dakota. South Dakota Department of Transportation,
3.7 Colwell, R. N., ed. Manual of Photographic Interpreta-
Pierre, Final Rept. of Project 635(67), Dec. 1973, 707
tion. American Society of Photogrammetry, Falls
pp.; Federal Highway Administration.
Church, Va., 1960, 868 pp. 3.29 Sharpe, C. F. S. Landslides and Related Phenomena: A
3.8 Denny, C. S., Warren, C. R., Dow, D. H., and Dale, W. J. Study of Mass Movements of Soil and Rock. Columbia
A Descriptive Catalog of Selected Aerial Photographs of
Univ. Press, New York, 1938, 137 pp.
Geologic Features in the United States. U.S. Geological 3.30 Smith, J. T., Jr., ed. Manual of Color Aerial Photography.
Survey, Professional Paper 590, 1968, 79 pp.
American Society of Photogrammetry, Falls Church, Va.,
3.9 Dishaw, H. E. Massive Landslides. Photogrammetric En-
1968, 550 pp.
gineering, Vol. 32, No. 6, June 1967, pp. 603-609. 3.31 Stallard, A. H., and Anschutz, G. Use of the Kelsh Plotter
3.10 Eckel, E. B., ed. Landslides and Engineering Practice.
in Geoengineering and Allied Investigations in Kansas.
Highway Research Board, Special Rept. 29, 1958, 232 pp.
Highway Research Board, Highway Research Record 19,
3.11 Gagnon, H. Remote Sensing of Landslide Hazards on 1963, pp. 53-107.
Quick Clays of Eastern Canada. Proc., 10th International 3.32 Stallard, A. H., and Myers, L. D. Soil Identification by
Symposium of Remote Sensing of Environment, Environ- Remote Sensing Techniques in Kansas: Part I. State
mental Research Institute of Michigan, Ann Arbor, 1975, Highway Commission of Kansas, Topeka, 1972, 107 pp.;
pp. 803-810. Federal Highway Administration.
3.12 Leith, C. J., Fisher, C. P., Deal, C. S., Gupton, C. P., and 3.33 Tanguay, M. G., and Chagnon, J. Y. Thermal Infrared
Yorke, C. A. An Investigation of the Stability of High- Imagery at the St. Jean-Vianney Landslide. Proc., 1st
way Cut Slopes in North Carolina. North Carolina State Canadian Symposium on Remote Sensing, Canada Center
Univ., Raleigh, Project ERO-110-U, June 1964, 129 pp. for Remote Sensing and Canadian Institute of Surveying,
3.13 Liang, T. Landslides: An Aerial Photographic Study. Ottawa, 1972, pp. 387-402.
Cornell Univ., Ithaca, N.Y., PhD thesis, 1952, 274 pp. 3.34 Terzaghi, K., and Peck, R. B. Soil Mechanics in Engineer-
3.14 Lobeck, A. K. Geomorphology. McGraw-Hill, New York, ing Practice. Wiley, New York, 1948, 566 pp.
1939, 731 pp. 3.35 Thomas, M. M., ed. Manual of Photogrammetry. Ameri-
3.15 Lueder, D. R. Aerial Photographic Interpretation: Prin- can Society of Photogrammetry, Falls Church, Va., Vols.
ciples and Applications. McGraw-Hill, New York, 1959, 1 and 2,3rd Ed., 1966, 1199 pp.
462 pp. 3.36 Thompson, G. A., and White, D. E. Regional Geology
3.16 Maruyasu, T., Nakano, T., Nishio, M., Takeda, Y., of the Steamboat Springs Area, Washoe, Nevada. U.S.
Kawasaki, T., Kimata, K., Ozarki, Y., and Fuchimoto, M. Geological Survey, Professional Paper 458 A, 1964.
Statistical Analysis of Landslides and Related Phenomena 3.37 Thornbury, W. D. Principles of Geomorphology. Wiley,
on Aerial Photographs. Journal of Japan Society of New York, 1954, 618 pp.
Photogrammetry, Special Vol. 1, June 1964, pp. 93-100. 3.38 Way, D. S. Terrain Analysis: A Guide to Site Selection
3.17 Mathur, B. S., and Gartner, J. F. Principles of Photo In- Using Aerial Photographic Interpretation. Dowden,
terpretation in Highway Engineering Practice. Materials Hutchinson and Ross, Stroudsburg, Penn., 1973, 329 pp.
and Testing Division, Ontario Department of Highways, 3.39 Williams, G. P., and Guy, H. P. Erosional and Depositional
1968, 236 pp. Aspects of Hurricane Camile in Virginia, 1969. U.S. Geo-
3.18 McNutt, R. B., and others. Soil Survey of Chilton County, logical Survey, Professional Paper 804, 1973, 80 pp.
Alabama. U.S. Soil Conservation Service, U.S. Forest Ser- 3.40 Zâruba, Q., and Mend, V. Landslides and Their Control.
vice, Alabama Agricultural Experiment Station, and Ala- Elsevier, New York, and Academia, Prague, 1969, 205 pp.
bama Department of Agriculture and Industries, Oct.
1972.
3.19 Mintzer, 0. W. Application of Photo Interpretation to PHOTOGRAPH CREDITS
Highway Engineering Design. Engineering Experiment
Station, Ohio State Univ., Columbus, Final Rept. EES Plate 3.1 Courtesy of Region 10, Federal Highway Administra-
196, Oct. 1966, 213 pp. tion
3.20 Mintzer, 0. W., and Struble, R. A. Manual of Terrain In- Plate 3.2 Courtesy of Washington State Highway Commission
vestigation Techniques for Highway Engineers. Engineer- Plate 3.3 Courtesy of Washington State Highway Commission
ing Experiment Station, Ohio State Univ., Columbus, Rept. Plate 3.4 Courtesy of U.S. Department of Agriculture
196-2 and Appendix I, Oct. 1965, 384 pp. Plate 3.5 Courtesy of National Aeronautics and Space Adminis-
3.21 National Aeronautics and Space Administration. Skylab tration
Earth Resources Data Catalog. National Aeronautics and Plate 3.6 Courtesy of National Aeronautics and Space Adminis-
Space Administration, Houston, 1974, 359 pp. tration
3.22 Norell, W. F. Air Photo Patterns of Landslides in South- Figure 3.1 Courtesy of Virginia Division of Mineral Resources
79
Figure 3.3 Courtesy of Ohio Department of Transportation Figure 3.27 Courtesy of Ontario Ministry of Transportation and
Figure 3.4 Courtesy of Washington State Highway Commission Communications
Figure 3.5 Courtesy of Ohio Department of Transportation Figure 3.28 Courtesy of U.S. Department of Agriculture
Figure 3.6 Courtesy of Ohio Department of Transportation Figure 3.30 Courtesy of U.S. Department of Agriculture
Figure 3.7 Courtesy of Idaho Transportation Department Figure 3.31 Courtesy of U.S. Geological Survey
Figure 3.12 Courtesy of Kansas Department of Transportation Figure 3.32 Courtesy of National Aeronautics and Space Adminis-
Figure 3.14 Courtesy of U.S. Department of Agriculture tration
Figure 3.15 Courtesy of Ohio Department of Transportation Figure3.33 Courtesy of M. G. Tanguay
Figure 3.16 Courtesy of South Dakota Department of Transporta- Figure 3.34 Courtesy of M. G. Tanguay
tion Figure 3.35 Courtesy of D. Rubin, American Power Service Corpo-
Figure 3.17 Courtesy of Ohio Department of Transportation ration, and F. J. Buckmeier, Texas Instruments, Inc.
Figure 3.18 Courtesy of U.S. Geological Survey Figure 3.36 Courtesy of Federal Highway Administration
Figure 3.19 Courtesy of U.S. Department of Agriculture Figure 3.37 D. J. Varnes, U.S. Geological Survey
Figure 3.20 Courtesy of U.S. Geological Survey Figure 3.38 D. J. Varnes, U.S. Geological Survey
Figure 3.21 Courtesy of U.S. Department of Agriculture Figure 3.40 E.K. Maughan, U.S. Geological Survey
Figure 3.22 Courtesy of U.S. Department of Agriculture Figure 3.41 Courtesy of Pennsylvania Department of Transporta-
Figure 3.23 Courtesy of U.S. Department of Agriculture tion
Figure 3.24 Courtesy of U.S. Department of Agriculture Figure 3.42 F. 0. Jones, U.S. Geological Survey
Figure 3.25 Courtesy of U.S. Department of Agriculture Figure 3.45 Robert C. Lafferty, consulting geologist
Figure 3.26 Courtesy of North Dakota State Highway Department Figure 3.46 C. F. Erskine, U.S. Geological Survey
Me
Chapter 4
Field,
Investigation
George F. Sowers and David L. Royster
The field investigation is the central and decisive part of a Land form
study of landslides and landslide-prone areas (4.9). The in- Anomalous patterns (jumbled, scarps, bulges)
vestigation serves two essential purposes: (a) to identify B. Surface drainage
areas subject to sliding when future construction is being Continuous
planned and (b) to define features of and environmental Intermittent
factors involved in an existing slide. Unstable areas prone C. Profiles of-slope
to sliding usually exhibit symptoms of past movement and Correlate with geology (II)
incipient failure; most of these can be identified in a de- Correlate with contour map (IA)
tailed field investigation before design. Such investigations D. Topographic changes
can show how to prevent or at least minimize future move- 1. Rate of change by. time
ments, and they can suggest alternate routes that are less 2. Correlate with groundwater (III), weather (IV),
likely to slide. Once a landslide has developed (either be- and vibration (V)
fore the construction of a facility or after work is under II. Geology
way), the investigation is made to diagnose the factors re- A. Formations at site
sponsible for the movement and to determine what correc- 1. Sequence of formations
tive measures are appropriate to prevent or minimize con- 2. Colluvium
tinuing movements. Bedrock contact
Because landslides are continually changing phenomena, Residual soil
the field investigation cannot be considered an isolated or 3. Formations with bad experience
easily defined activity; instead, it is iterative. New data gen 4. Rock minerals susceptible to alteration
erate new questions that require more data for answers. Al. B. Structure: three-dimensional geometry
though the investigation must continue for a period of time I. Stratification
consistent with the shifting topography and changing en- 2. Folding
vironment, it is constrained by the timely needs for preven- 3. Strike and dip of bedding or foliation
tive or corrective design. Thus, field investigations should Changes in strike and dip
commence long before construction is anticipated and Relation to slope and slide
sometimes continue long after the area has been changed 4. Strike and dip of joints with relation to slope
by the anticipated construction. 5. Faults, breccia, and shear zones with relation
to slope and slide
SCOPE OF FIELD INVESTIGATIONS C. Weathering
Character (chemical, mechanical, and solution)
A number of features require study in a field investigation; Depth (uniform or variable)
these are enumerated below as a checklist for planning a III. Groundwater
study. A. Piezometric levels within slope
I. Normal
I. Topography 2. Perched levels, relation to formations and struc-
A. Contour map ture
3. Artesian pressures, relation to formations and 3. Evidence in topography
structure 4. Photographs (see Chapter 3)
B. Variations in piezometric levels-correlate with Oblique
weather (IV), vibration (V), and history of slope Aerial
changes (VI) Stereoptical data (photographic)
Response to rainfall Spectral changes
Seasonal fluctuations 5. Instrumental data (see Chapter 5)
Year-to-year changes Vertical changes, time history
Effect of snowmelt Horizontal changes, time history
C. Ground surface indications of subsurface water Internal strains and tilt, including time his-
Springs tory
Seeps and damp areas D. Correlations of movements
Vegetation differences Groundwater-correlate with groundwater (III)
D. Effect of human activity on groundwater Weather-correlate with weather (IV)
Groundwater utilization Vibration-correlate with vibration (V)
Groundwater flow restriction Human activity-correlate with human-induced
Impoundment and additions to groundwater vibration (VB)
Changes in ground cover and infiltration oppor-
tunity The techniques for obtaining the data will be discussed at
Surface water changes greater length later in this chapter. The extent to which any
E. Groundwater chemistry one feature requires evaluation is difficult to determine in
Dissolved salts and gases advance. Whether any more data are needed will depend
Changes in radioactive gases on the amount of information already available. However,
IV. Weather the process is seldom completed; instead, the work stops
A. Precipitation when time, money, and patience commensurate with risks
Form (rain or snow) and potential costs have been exhausted.
Hourly rates
Daily rates Topography
Monthly rates
Annual rates The topography or geometry of the ground surface is an
B. Temperature overt clue to past landslide activity and potential instability.
Hourly and daily means More detail than that shown on existing topographic or
Hourly and daily extremes project design maps is usually required for landslide studies.
Cumulative degree-day deficit (freezing index) Moreover, interpretations made by topographerswho.are
Sudden thaws not specifically looking for landslide features can obscure
C. Barometric changes the special geometric or topographic forms that are diag-
V Vibration nostic of landslides. Therefore, special mapping is usually
A. Seismicity necessary. Because the topography of a landslide is con-
Seismic events tinually changing, the area must be mapped at several dif-
Microseismic intensity ferent times: (a) several years before construction (if data
Microseismic changes are available), (b) at the time a specific landslide investiga-
B. Human induced tion is initiated, (c) at appropriate intervals during the prog-
Transport ress of the investigation, and (d) after remedial measures
Blasting are undertaken. Ultimately, the effectiveness of corrective
Heavy machinery measures is expressed by whether the topography changes.
History of slope changes Because of the long time span involved, the location of
A. Natural process the area should be referenced with some standard location
Long-term geologic changes system. Latitude and longitude may be sufficient for
Erosion studies in remote areas or studies encompassing large sites.
Evidence of past movement State survey coordinate systems and road stationing out-
Submergence and emergence side the area of study can provide an effective frame for lo-
B. Human activity cation. In addition, nearby prominent topographic fea-
Cutting tures should be referenced because man-made features often
Filling change and those who must use the data referenced to them
Changes in surface water may not be aware of those changes.
Changes in groundwater
Changes in vegetation cover, clearing, excava- Geology
tion, cultivation, and paving
Flooding and sudden drawdown of reservoirs The geometry of the subsurface is the most important sin-
C. R.ate of movement gle factor in the analysis of a landslide. First, the various
1. Visual accounts soil and rock units must be identified. Although it is de-
2. Evidence in vegetation sirable to associate the geologic units with the named geo-
82
logic formations that have been defined or previously or piezometric level. The groundwater is sometimes perched,
studied in the area, this activity frequently dissipates more leaving the overlying and underlying strata only partially
time and effort than are relevant to the evaluation of the saturated. The formation or fissure can be carrying water
landslide. Because the name of a formation seldom re- under sufficient confined pressure to cause significant re-
flects its engineering behavior, determining the structural, duction in effective stresses within the soil or rock between
lithologic, and engineering properties of the formation is blocks of impervious materials separated by fissures. To
far more important than determining its exact age and the geologist interested in rock formations and to the geo-
identification. However, names do have value in compar- technical engineer interested in soil samples, groundwater
ing landslide activity in one area with that in another that aquifer identification and piezometric pressure studies are
includes the same formations, and relative age can help in often considered to be of secondary interest, but, in many
describing structural features. cases, they constitute the most valuable data gained in the
Geologic structure is frequently a major factor in land- investigation.
slides. Although this includes the major large-scale struc-
tural features such as folds and faults to which most impor- Physical Properties
tance is attached in regional geologic studies, the minor
structural details, including joints, small faults, and local Evaluating the stability of a zone of questionable move-
shear zones, may be even more important. Changes in the ment or determining the effectiveness of various corrective
geologic conditions with location give important clues to measures requires a knowledge of the physical properties
areas of distress and to potential landslide activity. Off- of the soil and rock strata. The required data include the
sets in strata, changes in joint orientation, and abrupt properties of both the intact formation and the formation
changes in dips and strikes are indicators of nonuniform after it has been subject to water pressure changes and to
geologic conditions. These must be identified both within large strains.
the area of potential or observed movement and far enough The physical properties can be determined in three ways.
beyond that area to predict the effect of any planned con- The conventional approach, which is to test undisturbed
struction surrounding the landslide zone. samples in the laboratory, has the advantage that stresses
and water pressures can be controlled to simulate future
Water environmental changes produced by the anticipated con-
struction and climate variations. However, the process of
Water is a major factor in most landslides. Concentrations sampling, transportation to the laboratory, and preparation
of surface water from rainfall runoff, seeps, and springs are for testing causes definite changes in the character of the
overt indications of topographic changes in landslide-prone soil and rock before their laboratory testing. A second ap-
areas. For example, unusually deep erosion gullies some- proach, which is to test the soil and rock in place in the
times suggest that soils and rocks have been weakened by ground, minimizes the disturbing effects of sampling. How-
landslide activity. Moreover, surface water entering cracks ever, the range in environmental changes that can be induced
and fissures caused by earth movement aggravates instability. in the field is severely limited. A third approach is to use
Seeps and springs that serve as groundwater exits are indirect tests that measure some physical property related
another dimension of the water problem. Sometimes, how- to the engineering parameters involved in the landslide.
ever, too much attention is paid to the exit and not enough For example, the compression-wave velocity in a formation,
to the source. If there is a groundwater exit, there also as determined by a seismic refraction geophysical investiga-
must be a path of seepage leading to it; thus, the exit of tion, is related to the joint spacing, density, and rigidity of
groundwater is a clue to internal water pressures elsewhere. the soils and rocks through which the wave passes. A for-
Unfortunately, the layperson sometimes considers seeps mation exhibiting a low compression-wave velocity is gen-
and springs to be the- causeof earth movements. Seeps up- erally weaker and less rigid than one exhibiting a high ve-
hill from the slide serve as a source of surface water that locity. Unfortunately, the indirect tests require subjective
can infiltrate back into the soil within the slide, thus con- interpretation; therefore, their use in measuring the physi-
tributing to instability. However, within the slide zone and cal properties of the materials involved is limited by the
downhill from it, the opposite is usually true. Less pressure experience and imagination of the interpreter.
builds up when water is seeping out of the ground than
when the exits for groundwater are blocked. For example, Ecological Factors
in one major slide area, landslide activity was always pre-
ceded by a stoppage of spring discharge near the slide toe; A landslide, even one that occurred in the geologic past,
the cessation of movement was marked by an increase in has a distinct influence on life systems in the surrounding
spring discharge. area. For example, a landslide in a steep, fractured rock
The various joints, fissures, and more pervious strata mass can produce a loose, jumbled mass of colluvium that
that conduct water underground must be identified. These is more easily saturated by rain and therefore supports a
features constitute a part of the subsurface geometry that more luxuriant growth of vegetation than does the undis-
is not always reflected by the strike and dip of the forma- turbed slope. Thus, the ecosystem of an area can provide
tions. Even the most insignificant water-bearing stratum a subtle hint of landslide activity. Conversely, the ecosys-
can sometimes be the source of trouble and is too often tem effects changes in landslides. For example, deep
overlooked in favor of the more obvious, but perhaps less masses of roots can provide sufficient reinforcement to
dangerous, aquifers. Within each water-bearing formation distort the geometry of the failing soil mass; trees with
or fissure is a definite (but changing) groundwater pressure deep tap roots may even curtail severe movement. A thick
83
cover of vegetation minimizes rainwash and provides more Time Span
uniform infiltration of moisture. In contrast, concentrated
gullying occurs when an area has been exposed by excava- Since most landslides are influenced by climatic changes, a
tion or when vegetation has been destroyed by fire. There. minimum period for investigation should include one sea-
fore, the ecosystem, and particularly the vegetation within sonal cycle of weather—i year in most parts of the world.
the landslide area, should be compared to that within the However, because long-term climatic cycles that occur every
surrounding area. Since life systems are changing, a single ii or 22 years are superimposed on the yearly changes, a
study at a particular time may not be sufficient; continuing landslide investigation could be necessary for more than 2
studies during an extended period usually are necessary in decades. Such a long investigation is almost impossible,
order to define the significance of the interrelation between however, because of the need to draw conclusions and take
the earth movements and the ecology of an area. corrective action. Investigations made during a period in
which the climatic conditions are less severe than the maxi-
mum will prove too optimistic, and those made during a pe-
PLANNING INVESTIGATIONS riod of bad climate may appear too pessimistic. The worst
climatic conditions that develop during the life of the proj-
Area of Investigation ect control the risk to engineering construction. Experience
has indicated that many false conclusions have been reached
The area of an investigation is controlled by the size of the regarding the causes of landslides and the effectiveness of
project and by the extent of the topographic and geologic corrective measures because worsened climatic changes were
features that are to be involved in the landslide activity. At not considered by the engineers and geologists concerned.
sites where there is potential for movement that has not yet
developed, the area that must be investigated cannot be Stages of Investigation
easily defined in advance. The extent of the investigation
can be better defined once a landslide has occurred. How- An investigation of landslides is a continuing process but,
ever, in either case, the area studied must be considerably from the practical point of view, may be divided into four
larger than that comprising the suspected activity or known stages.
movement for two reasons: (a) The landslide or potential
landslide must be referenced to the stable area surrounding The first stage is a preliminary investigation or recon-
it, and (b) most landslides enlarge with passage of time naissance in which a general overall view of the problem
(moreover, many landslides are much larger than first sus- area is gained. The work begins with a review of the pub-
pected from the obvious overt indications of activity). As lished geology in the area and accounts of past land insta-
a crude rule of thumb, the area studied should be two to bility. The field study is largely visual and the interpreta-
three times wider and longer than the area suspected. In tion is highly subjective. The results of this preliminary
some mountainous areas, it is necessary to investigate to evaluation are used as a guide in planning the more inten-
the top of the slope or to some major change in lithology sive, specific investigation during which the major part of
or slope angle. The lateral area must encompass sources of the quantitative data are obtained.
groundwater and geologic structures that are aligned with The second stage, which is more intensive and de-
the area of instability. tailed, includes boring, sampling, trenching, and other spe-
The depth of the investigation is even more difficult to cific techniques designed to obtain the data needed to sat-
define in advance. Borings or other direct techniques should isfy the objectives outlined in the above checklist. Because
extend deep enough to identify those materials that have of climatic changes, the intensive investigation is preferably
not been subject to past movement, but that could be in- undertaken during the season that is least favorable for sta-
volved in future movement, and the underlying formations bility. For example, borings to determine the minimum
that are likely to remain stable. The boring depth is some- soil strength or groundwater studies to determine maximum
times revised hourly as field operations proceed. When in. water pressures should be made during periods of snowmelt
strumentation of a landslide yields data on the present or following heavy rainfall. A similar investigation made
depth of activity, planned depths are sometimes found to during the driest and hottest period of the year could give
be insufficient and increases are necessary. The specifica- completely misleading data.
tions should be flexible enough to allow additional depth The third stage is iterative. As new data are obtained,
of investigation when the data obtained suggest deeper they will point to the need for additional data from specific
movements. Longitudinal cross sections should be drawn locations. The investigative plan must provide for additional
through the center of the slide and depict possible toe work that was not a part of the initial scope. Experience in-
bulges and uphill scarps; circular or elliptical failure sur- dicates that the additional work stemming from information
faces sketched through these limits can suggest the maxi- obtained during the planned study will range from 30 to 50
mum depth of movement. Continuous thick hard strata percent more than that which was originally considered ade-
within the slope can limit the depth. However, at least one quate for the intensive investigation.
boring should extend far below the suspected depth of The fourth stage involves continuing surveillance of
shear; sometimes deep slow movements are masked by the any area where activity is suspected or where corrective ac-
greater activity at shallower depths. For a second estimate, tion has been taken. The surveillance period is indeterminate
the depth of movement below the ground surface at the but should extend through at least one cycle of annual cli-
center of the slide is seldom greater than the width of the matic change and, to be most meaningful, should be longer
zone of surface motion. term to include the worst climatic conditions. For example,
84
corrective work may be done during a number of consecu- areas. One distinct advantage of existing photographic
tively dry years, and the area may not be subjected to its and remote-sensing imagery is that it provides information
worst climatic test until 4 or 5 years later (assuming an 11- on ground surface conditions before the particular site be-
year cycle of climatic change). If the climatic changes oc- comes of interest for construction. Some aerial photo-
cur in a 22-year cycle, the period of observation required graphs of the U.S. Department of Agriculture date from
could be correspondingly much longer. Rarely do those the late 1930s,
and imagery of LANDSAT and the Earth
who finance such long-term surveillance have the wisdom Resources Technology satellite dates from the early 1970s.
and foresight to continue the data gathering, because the
period required extends beyond the tenure of most public Color and Color Infrared Photography
officials or private managers. However, public and profes-
sional interests are best served if surveillance is maintained For mapping purposes, black-and-white panchromatic
long enough for the occurrence of the full range of environ- photography generally provides the best detail at the lowest
mental conditions that can reasonably be expected; other- cost. However, in the study of landslides, certain features
wise, the extent of continuing risk may never be adequately become more obvious in color (4.15). Color photography
defined. emphasizes differences in vegetation and wet areas through
changes in soil color and vegetation vigor. Stratification in
SITE TOPOGRAPHY exposed soil and rock is often more easily recognized and
mapped in color. Color infrared photography is particularly
As previously stated, the site topography (surface geometry) helpful because of two properties. First, most water ap-
is the first clue to potential instability and the degree to pears blue in such photography; seeps issuing from bare
which an area has undergone landslide activity. The topog- ground frequently will have a blueish tinge that is far more
raphy is first determined by aerial surveys (photogrammetry), obvious in the color infrared photography than in either
which provide an overall view of the site conditions. How- black-and-white or color photography. Second, color in-
ever, because of the detail required and the masking of the frared photography is useful in mapping hidden cracks and
vegetation and the landslide itself, detailed ground surveys fissures through their influence on growth of vegetation.
must be included as a major tool in landslide investigation. Fissures above the water table act as drains; the color or
image depicts dull red colors of inhibited growth, and
Aerial Survey springs and sheared zones that hold water sometimes fea-
ture more vigorous growth by brilliant reds. However, such
Existing Sources of Photography inferences should be verified on the ground before they are
included on maps.
All of the United States and most of the remainder of the
world have been covered by some form of aerial photog- Thermal Sensing
raphy or remote sensing, including LANDSAT; the principal
sources are given in Chapter 3. Generally, these existing Thermal sensing involves scanning the ground with a detec-
photographs have been made at moderately high altitudes tor that is sensitive to selected portions of infrared radiation.
(typically at a scale of 1:10000 to 1:40000). Probably the The image produced by the shorter wave (2 to 5 urn) is com-
most widely used scales of photography by these agencies parable to the image produced by black-and-white or color
are 1:20000(1 cm = 200 m; 1 in = 1667 ft) or 1: 24000 infrared photography. In the second range (approximately
(1 cm = 240 in; 1 in = 2000 ft). At this scale, most land- 8 to 14 um), the radiation is comparable to the heat radi-
slides appear as minor topographic anomalies and their de- ated from the ground. Thus, for any given texture of ground
tailed features, such as scarps, toe bulges, and seeps, can surface, the level of radiation measured is proportional to•
seldom be identified, even from enlargements. However, the ground temperature. This range of thermal sensing is
these photographs are useful in identifying areas that may highly sensitive to small differences in ground temperature.
be prone to landsliding because they may show the typical For example, if the imagery is obtained during winter,
undulating disturbed topography or arcuate scarps that are seeps and springs issuing from the warmer, deep ground
associated with past landslides. Such existing photography will be depicted as warm points contrasting with the cold
is frequently a good starting place for evaluating the surface ground; similar imagery obtained during summer nights
geometry of the site, but it is seldom of sufficient accuracy will show the seeps and springs as colder than the surround-
to provide useful data for detailed landslide studies. ing ground surface. To be most effective, this sensing must
Smaller scale photography can be obtained by the more be done during the night or early dawn when there is insig-
sophisticated remote-sensing systems of high-altitude nificant infrared reflectance from the sun.
photography (1:60 000 and 1:120000) provided by the Because of changing angles associated with aircraft move-
National Aeronautics and Space Administration. Although ment, the thermal infrared image is seldom a scale represen-
the imagery obtained from LANDSAT with a resolution of tation of the ground feature. Instead, it must be used in
56 x 79 m (185 x 260 ft) is not sufficient to identify the conjunction with aerial photographs or accurate maps so
topography of most landslides, it does provide clues to the that the distorted geometry of the infrared image can be
major geologic structures that sometimes influence land- related to the actual geometry of the ground surface and
slide activity. However, such high-altitude, small-scale data the points depicted on the imagery can be located exactly.
should be considered as supplementary information that The resolution of the imagery is not great because the sens-
provides clues of regional significance rather than as the ing generally encompasses 0.002 to 0.005 rad (0.115° to
primary source of topographic data on landslide-prone 0.286°) of angle from the aircraft or 2 to 5-rn (7 to 16-ft)
85
Figure 4.1. Bench marks and triangulation leveling network.
resolution for 1000-rn (3300-ft) height. At high altitude,
the heat radiation from significant areas is averaged and IIPcInPF
Aerial survey data are used in two ways. First, and most im-
portant, when used in coordination with ground surveys (in-
cluding reference points established on the ground surface),
they provide detailed topographic maps of the area through
photogrammetric processes. This is the procedure with
which organizations obtaining aerial survey data are most
familiar and that requires the least guidance. A second use of enhanced movement. Thermal imagery is useful after pe-
aerial photography and remote-sensing data is in interpreta- riods of unusually dry or wet weather. However, the vari-
tion or terrain evaluation (described in Chapter 3). Because ous forms of thermal imagery are most revealing when the
the technical quality of photographs and imagery for accu- ground surface can be observed (that is, when deciduous
rate mapping is more demanding, the specifications for vegetation is free from leaves). Aerial photographs taken
photography should be directed by topographic require- at intervals for several years after corrective measures are
ments. However, the special needs for interpretation should taken provide an excellent tool for assessing the overall
also be considered so that one photographic mission can benefits achieved by the measures and for obtaining a pre-
serve both purposes. liminary view of the effect of the landslide and associated
corrective measures on the ecosystem of the area. Since
Terrestrial Photogrammetry satisfying all those needs is not always possible, it may not
be feasible to bring all of the tools of aerial photography
Photogrammetric measurements of ground geometry can be and remote sensing into play in any single investigation.
made from oblique photographs obtained at the ground sur-
face. For example, two or more permanent photographic Ground Surveys
sites that overlook a slide area can be used to monitor slide
movement by successive sets of simultaneous photographs. On-the-ground surveys are necessary to (a) establish the
Such observations have been used to measure changing dam ground control for photogrammetric mapping and instru-
deflections. Although the data reduction is more complex mentation, (b) obtain topographic details where the ground
than that for photogrammetric mapping, the technique is surface is obscured by vegetation (this is particularly impor-
useful for determining movement of any selected points, tant because of the accuracy required in mapping landslides),
provided they can be seen in the photograph. and (c) establish a frame of reference against which move-
ments of the ground surface can be compared.
Specifying Scale and Coverage Quality
Ground Control for Instrumentation
The scale of photography and thermal imagery must be de-
termined on the basis of the size of the features to be iden- The first requirement is a system of local bench marks that
tified. In the mapping of springs, seeps, and fissures, a reso- will remain stable during the course of the investigation and
lution circle 0.5 to 1 m (1.5 to 3 ft) in diameter is desirable. as far into the future as movements will be observed. These
A suitable scale for the aerial photography for landslide must be located far enough outside the suspected zone of
studies is of the same order that is required for final high- sliding that they will not be affected by any movements.
way alignment studies (from 1:3000 to 1:6000 on contact Ultimately, the bench marks should be referenced to control
prints). Maps are drawn to a scale of 1: 1000 (1 cm = 10 m; monuments of federal and state survey systems. However,
lin83 ft) tol:5000(lcm50m;lin=4l7 ft) and for convenience, a subsystem of local bench marks should
with contour intervals of 0.5 m (1 to 2 ft) for most slides. be established close enough to the zone of movement that
Good quality aerial photographs for mapping and interpre- they can be used as ready references for continuing surveys.
tation are necessary tools at the beginning of an investiga- At least two monuments of position and elevation should
tion, but additional coverage is desirable after episodes of be established on each side of the zone of sliding or sus-
pected movement; as indicated in Figure 4.1, these should photography may not be sufficiently, accurate or detailed
be as close as possible to the movement zone, but not influ- for landslide studies because vegetation obscures the ground
enced by future enlargement of the slide. Experience sug- 'surface. Therefore, detailed on-site mapping is necessary.
gests that the distance from a bench mark to the closest Major features, such as scarps, bulges, and areas of jumbled
point of known movement should be at least 25 percent of topography, should be defined (Figure 4.2). Because of the
the width of the slide zone. In areas of previous landslides, changing nature of landslides, the surface surveys should be
the minimum distance may be greater. In mountainous conducted at the same time the photography is taken;
areas, adequate outcrops of bedrock can sometimes be otherwise, the movements of the landslide will confuse the
found uphill or downhill from the landslide; in areas of topographic map. Even then, it may not be possible to ob-
thick soil, deep-seated bench marks may be necessary. tain a precise correlation between the surface topography
determined on the ground over a period of days or weeks
Control Network and that obtained at a single instant from aerial surveys.
Differences should be expected, and these should be noted
The bench marks should be tied together by triangulation on the topographic maps that are produced.
and precise leveling loops. If there are enough bench marks,
the movement of any one can be detected by changes in the Cracks, Seeps, and Bulges
control network. Intermediate or temporary bench marks
are sometimes established closer to the zone of movement Although many cracks, seeps, and bulges, as well as other
for use in the more frequent surveys of the landslide area. minor topographic details, can be identified in aerial photog-
However, these should be checked against the permanent raphy, their full extent can seldom be determined unless the
monument grid each time they are used. photographs are taken with an unusually high degree of res-
olution in vegetation-free areas. Therefore, independent
Topography crack and bulge surveys should be made by surface methods.
Developing cracks are often obscured by grass, leaves, and
As previously stated, topography obtained from aerial root mats, particularly at their ends; these cracks should be
U PS L OP E
NOR SCARP
ZONE OF DEPRESSION
OR SUBSIDENCE
,1
/
/ \
DIAC ONAL
TEN5 ION-SHEAR /
CRAI
\
EN ECHELON II,
CRACKS -
/
1' - ZONE OF BULGING
DOWNSLOPE
TOE
SSI BLY
SHEAR DISPLACEMENT DOWNHILL MUD WAVE OR
TOE RIPPLE
TENSION CR ACK ON TOE BU LGE
SPRINGS. SEEPS
87
carefully uncovered so their total extent can be mapped. by a system of traverses or grids across the landslide area
Hidden cracks can be identified by subtle changes in leaf (Figures 4.3 and 4.4). Typically, a series of lines more or
mold patterns, tearing of shrubs, and distortion of trees and less perpendicular to the axis of the landslide, spaced 15 to
tree root systems. Boulder alignments or sliding trajectories 30 in (50 to 100 ft) apart with stakes at intervals of 15 to
should also be noted. Cracks should be staked on both sides, 30 m (50 to 100 ft), should be maintained and referenced
as well as referenced to the movement system, because the to the control bench marks. Grids should be laid out so
entire crack system shifts with continuing landslide move- that the reference points are aligned with trajectories of
ment. maximum slope or apparent movement if sliding is continu-
Seeps and springs are the ultimate exits for water-bearing ing. In addition, where soil and rock weaknesses cause sec-
strata and cracks and thus are clues to the water paths that ondary movements that are skewed to the major slide, in-
influence soil and rock stability. Because seeps often follow termediate points should be established. For small slides
cracks that have been opened by soil or rock movement, or widely spaced areas of suspected movements, single
they can sometimes be traced to sources uphill. The points traverse lines of reference (Figure 4.4) are often used.
of disappearance of surface runoff into cracks and fissures Appropriate reference wands or flags should be placed
should also be mapped. Seeps, springs, and points of water nearby so that the staked points can be found despite se-
loss change with rainfall, snowmelt, and ground movement. vere movement. The elevation and coordinates of each
Thus, meaningful data on their location and shifts cannot be point on the traverse or reference grids should be determined
obtained by a single survey or regular intervals of observa- by periodic surveys. In areas where highly irregular topog-
tion. Instead, they should be located during and shortly raphy suggests rapid differences in movement from one point
after periods of intense rainfall or snowmelt and after epi- to another, reference points should be spaced more closely,
sodes of significant movement. regardless of any predetermined grid pattern. Such closely
spaced stakes help to define the lateral limits of the land-
Movement Grids and Traverses slide, as well as the direction of movement of localized
tongues within the slide. This is particularly important in
The continuing movement of a landslide can be measured the later stages of movement if secondary flows develop
Figure 4.3. Observation grid. ElO. 00 +20 +40 +60 El1+00+20 +40 +60 +80 E12+00 +20 E12 60
N 13+20
N 13+00
+80
+60
+40
A N12+00 A
+60
+20
NI 1+00
+60
+40
+20
r N 10+00 A
A BENCH MARK Grid: 20 to 40-rn spacing
0 POINT FOR MEASUREMENT Note: 1 m = 3.3 ft.
from the weakening of the soil by sliding. Depending on Instrumentation of landslides is discussed in detail in
the rate of movement, these grid points should be checked Chapter 5; however, certain supplementary measurements
at intervals ranging from a few days to several months. In should be a part of the survey program. For example,
addition, they should be observed after periods of unusual crack width changes should be measured directly by taping
weather changes, such as snowmelt, high rainfall, or sharp across stakes set on each side of the crack; vertical offsets
temperature changes. In this way, any relation between on cracks and scarps should be obtained by direct measure-
landslide movement and climatic changes can be established. ment. These direct measurements serve as a check to the
more sophisticated systems (discussed in Chapter 5) that
Crack Measurement determine the in-depth movements. In addition, the sur-
face location and elevations of reference points of the in-
Most earth movements are accompanied by cracking of the strumentation systems must be determined at each time of
ground (Figure 4.2). The principal scarp is most prominent; observation.
it is paralleled by developing scarps (arc- or crescent-shaped
cracks with the points of the crescents pointing downhill Representation of Topographic Data
from the scarps) and secondary or older scarps downhill.
Along the lateral limits of the slide, cracks are formed by The photogrammetric data are correlated with the ground
differential shear between the moving mass and the intact survey controls and detailed topography, and they are used
soil beyond. The shear often generates parallel diagonal ten- to establish two or more maps of the landslide area. The
sion cracks (termed en echelon) in this zone. In the bulge first encompasses the landslide (or suspected landslide) plus
zone near the toe of the slide, there are frequently short ten- the surrounding area; the topography extends uphill and
sion cracks parallel to the direction of movement as well as downhill beyond major changes in slope or lithology. To-
crescent-shaped tension cracks with the points of the cres- pography should be developed on each side for a distance
cents pointing uphill. If the movement extends below the of approximately twice the width of the sliding area (or
toe of the slope, there will be bulge and shear cracks or more when the zone of potential movement is not well de-
subtle ripples in the soil well beyond the slide toe. Survey fined). Typical scales for such mapping of large slides may
points should be set on the more prominent of these fea- be 1:2500 to 1:5000 (1 cm = 25 m to 1 cm = 50 m; I in =
tures and beyond them, provided they are not close to the 208 ft to 1 in = 417 ft).
grid or traverse points. The second topographic map is more detailed and encom-
1+00
DOWN SLOPE
go
passes the observed slide area plus all of the uphill and down- portant that the profiles be selected so as to depict the worst
hill cracks and seeps associated with the slide. Typically, and less critical combinations of slope and movement within
the detailed map extends beyond the landslide uphill and the landslide area. To have at least one additional profile in
downhill for a distance of half the length of the slide or to the stable ground 15 to 30 m (50 to 100 ft) beyond the
significantly flatter slopes. Horizontally, the detailed topog- limits of the slide area on each side of the slide is usually de-
raphy should extend at least half of the width of the slide sirable so that the effect on the movement of ground sur-
area beyond the limits of the slide. Contour intervals in face slope alone can be determined.
such detailed topography should be as close as 0.5 m (1 to Each longitudinal profile of the landslide is generally
2 ft). The horizontal scale is typically 1:1000(1 cm = 10 plotted separately. If there are significant movements, the
m; I in = 83 ft) or larger. successive sets of elevations and consecutive profiles can be
shown on the same drawing to illustrate the changing site
Profiles topography. The original topography should be estimated
In addition to the topographic map, profiles-of the slide area from old maps and shown for comparison, where possible.
are prepared (Figures 4.5 and 4.6). The most useful of these However, to reference old maps precisely to the more de-
follow the lines of steepest slope of the slide area. Where the tailed topography obtained for the landslide investigation
movement definitely is not in the direction of the steepest is difficult. Differences between existing topography and
slope, two sets of profiles are necessary: One set should be preslide topography may represent survey mismatches as
parallel to the direction of movement and the other parallel well as actual changes in the ground surface. Adjustments
to the steepest ground surface slopes. In small landslides, of the preslide profile from old maps to the profile from
three profiles may be sufficient; these should be at the cen- new surveys can be made by comparing old and new topog-
ter and quarter points of the slide width (or somewhat closer raphy beyond the limits of observed slide movements.
to the edge of the slide than the quarter points). For very
large slides, the longitudinal profiles should be obtained at Displacement Vectors and Trajectories
spacings of 30 to 60 m (100 to 200 ft). It is particularly im- The survey grids and other critical points are entered on the
Figure 4.5. Landslide contours and profile location. RIGHT CENTER LEFT
485
485
480
480
475
470 475
465 470
465
460
460
455
455
450
450
445
445
440
C 440
435
0 435
I-
430
430
425
425
420
420
415 415
410
410
0
o 25m - 'LINE
36 +0Q -
all
Figure 4.6. Landslide ground I I
surface profiles.
25 m
HORIZONTAL AND r 100 ft
VERTICAL SCALES I T
475 ARE EQUAL
25m' cfo
I
450
100 It 0
E
- 425
475
CENTER LINE OF SLIDE
OPPOSITE STA. 36+55 ......
400
450
425
400 I I
2+00 1+75 1+50 1+25 1+00 0+75 0+50 0+25 0
STATIONING 1100 m from center line of right-of-wayl
more detailed topographic map of the landslide area. Both Figure 4.7. Movement vectors since beginning of
the topography and the depicted grid points should be ref- measurement.
01
structure, groundwater, and weathering in the prevention formations may have been displaced or distorted by an-
and evaluation of sliding has been emphasized by Philbrick cient landslides or by rapid weathering of less-resistant strata.
and Cleaves (4.9), Zâruba and Mend (4.18), and Deere and Some formations harden on exposure, especially in regions
Patton (4.4). Although data have been published on the of highly seasonal rainfall, and give a false impression of the
major geologic formations and structural features of most strength or stability of the formation. The outcrop data
of the world, they are seldom in sufficient detail for land- should be correlated with deep samples before final conclu-
slide studies. As long as their extent is comparable to that sions are reached.
of the failure zone, small changes in bedding and geologi-
cally minor fractures or irregularities in structure, which Faults and Joints
are seldom shown in published maps and reports, influence
failure as significantly as regional features. The local geol- Many geologic discontinuities, such as unconformities, cavi-
ogy is determined by surface observation (geologic recon- ties, formation contacts, and facies changes, can affect slope
naissance) and by interpretation of the subsurface explora- stability. Two of the most prominent, and perhaps most
tion data. The reconnaissance generates an estimate of the readily identified, are faults and joints. A fault is defined
local geology and aids in selecting locations and techniques as a surface or zone of rock fracture along which there has
for subsequent exploration and in interpreting the results been displacement parallel to the surface of the fracture.
of the overall investigation. Joints are rock cracks or partings without displacement
parallel to the surface. The recognition of faults and joints
Outcrops is particularly important to predesign and site or route se-
lection. The potential for instability is greater in areas with
The principal means for estimating the geology of an area is extensive faulting or close jointing than in areas without
the outcrop, which is defined as "that part of a geologic for- discontinuities. Revisions in route or locations of engineer-
mation or structure that appears at the surface of the earth" ing structures may be imperative if the potential is serious;
(4.1). Some geologic formations can be identified indirectly alternatively, the design can be altered to fit the situation.
by characteristic landforms and others can be deduced from The strike of the fault, angle of dip of the fault plane,
soils that are derived from the weathering of the formation; type and competency of the associated rock, extent of the
however, the primary element used in identification is the fault (especially the thickness of the sheared or gouge zone),
outcrop itself (rock, residual soil, or even old landslide de- and condition of the rocks and materials on either side of
bris). Outcrops can be examined on steep slopes, in river the zone should be determined for both site selection and
channels, in highway and railroad cuts, and in quarries and evaluation of an existing slide. If available, geologic maps
borrow pits. The relation between geologic formations is may show the location of major faults. If detailed maps
determined by correlating structure, lithology, and unit are not available, an experienced interpreter can often iden-
thicknesses with site topography. This is a major technique tify questionable areas from their topographic reflections
for geologic mapping and is described in texts on field geol- on maps, aerial photographs, and LANDSAT imagery.
ogy (4.3, 4.7). However, outcrops can be misleading. The These areas are then located in the field to determine their
.T
ff .
/
• ._'/;.-•
-'1.'.:-:;.
92
Figure 4.9. Bedding-plane slide in hard metasiltstone and
possible significance with regard to site or slide.
metasandstone in Tennessee.
Joint systems are seldom indicated on geologic maps,
but they are often more troublesome than faults as far as
slope stability is concerned. The strike, dip, and spacings
of the various joint sets are determined by field mapping of
outcrops and a supplementary examination of the rock
cores. Statistical depictions of orientation, dip, and spacing
or frequency (average number of joints per meter) can be
correlated with the slope geometry, proposed cutting and
filling, or observed slide features. Close joints in areas iii-
volving several rock types may be especially troublesome.
In Figure 4.8, for example, sandstone overlies shale. The
differences in density, permeability, strength. and rigidity
of these materials in the near-vertical slope have caused
rock falls. I-lad the position of these rock types been re-
versed or had the joints been less frequent, the problem
would likely have been less severe. Had the geology of this
cut been known in advance, a design could have been for- ,-,-
mulated to fit joint patterns and thus reduce the mainte-
,iance now required. Deere and Patton (4.4) have described
slopes in interbedded shale and sandstone as one of the most
common slide problems.
Inclined Bedding
Residual soils, as a result of the processes by which they Drainage should be noted and mapped in as much detail as
were formed, sometimes exhibit the structural characteris- possible at the outset of the investigation. Much of this can
tics of the geologic formations from which they were (IC- be done with the aid of' large-scale topographic maps and
rived. This relict structure often influences slope stability aerial photographs. Localized seeps, poor drainage, and wet-
in soils in a manner similar to that in which bedding joints ground indicators, such as cat tails and willows, cannot al-
and faults influence rock stability, in such cases, the param- ways be seeii in photographs: these can be found through
eters that reflect soil strength, cohesion and angle of friction careful surface reconnaissance and their locations added to
of t he intact soil, may prove misleading. Strengt It tests the detailed topographic maps.
should evaluate the surfaces of the relict structure as well as
the intact soil: analyses must reflect the orientation of Evidence of Past Instability
weaknesses defined by the relict structure.
Many clues can often alert the investigator to past landslides
Landforms and future risks. Some of these are hummocky ground,
bulges. depressions, cracks, bowed and deformed trees,
The study of landt'ornts is an integral part of the study of slumps, and changes in vegetation. The large features can
93
be determined from large-scale maps and aerial photographs; Figure 4.10. Grid of borings in suspect
however, the evidence often is either hidden by vegetation area before sliding commences.
BEYOND SLIDE
Identify the weaker formations that are likely to be
involved in movement;
Identify the stronger formations that offer significant
BULGE
resistance or that might limit the extent of the zone of
movement or provide support for retaining structures;
Locate aquifers, define groundwater levels and pres- TOE
94
Boring and Sampling Techniques plotted as a function of depth. The results may have signifi-
cance in identifying the ion concentration and density of
The boring and sampling techniques that can be employed the strata. These aid in identifying soils and rock directly
are innumerable, and selection depends on the information from the boring data. They are particularly useful where
required, the nature of the soil and rock, the topography there are gaps in the data, such as core loss in broken rock.
and ground trafficability, and the amount of money avail- An even more valuable use of this information lies in estab-
able. ASTM standards, standards of other associations, lishing the correlation of strata between one boring and an-
books on drilling and sampling, and textbooks on geotech- other. For example, it is difficult to compare samples ob-
nical engineering (for example, 4.14, 4.16) describe methods tained from two different holes to determine whether soils
of boring and sampling and their applications and should be or rocks that have similar classification characteristics rep-
consulted for details. Those techniques particularly adapted resent the same stratum. However, by comparing the con-
to landslide studies are summarized in Table 4.1. tinuous borehole logs, one can match the patterns of the
The type of sample obtained depends on the information different properties; similar patterns suggest similar stratifi-
needed. Oriented cores are required to determine the strike cation. Thus, while borehole logging may have limited en-
of inclined bedding or fractures. Samples from test borings gineering significance by itself in one hole, it is a significant
are adequate for identification of formations, but samples tool for boring interpretation and correlation when used in
from thin-wall tubes or rotary coring are required for good adjacent holes.
shear and consolidation tests. Accurate and detailed records
of the conditions encountered are essential. If samples are Test Pits and Trenches
lost, a description of the drilling peculiarities may imply the
soil and rock condition. When samples are obtained, their One of the best ways to sample and determine the structure
value must be assessed from a detailed examination by a and other physical properties of soils and weathered or weak
geologist or geotechnical engineer, guided by the driller's rock foundations is by excavating pits or trenches. Shallow
records. Because the act of sampling changes soils and test pits can be excavated with hand tools. Mechanical
rocks, the factors involved in those changes must be known equipment, such as backhoes, clamshells, draglines, or trac-
to estimate the extent of alteration. Finally, the samples tors equipped with front-end loaders, are required for deep
must be carefully preserved, packed, and handled so they pits or long trenches. The sides of the excavation should be
will not be unduly changed by exposure and shipping. sampled, logged, and photographed in detail to provide a
three-dimensional picture of the materials and structure.
Drilling Equipment Pits and trenches also provide excellent sites for in situ test-
ing. When the observations and tests are complete, the ex-
One of the problems common to most landslide investiga- cavation should be filled or, in some cases, incorporated in
tions is accessibility. The same instability that generates the remedial design by serving as a drainage outlet.
movement makes accessibility difficult. For this reason,
the investigative unit should be equipped with some type Geophysical Studies
of all-terrain vehicle. Too often, materials and conditions
within a particular slide may be underestimated or over- Geophysical exploration uses the changes in certain force sys-
looked simply because of inadequate borings at critical, tems in the earth to define possible boundaries between dif-
but inconvenient, inaccessible points. The drilling industry, ferent materials as well as to estimate some of the engineer-
in collaboration with manufacturers of specialized vehicles, ing properties. The forces include elastic shock waves,
has been able to develop subsurface exploration equipment gravity, and electric current, none of which is directly con-
that is highly versatile and mobile for work in rugged, cerned in landslide behavior. Thus, the methods are some-
steep terrain. times termed indirect exploration; the data of interest are
inferred from the properties measured. Geophysical methods
Borehole Logging do not replace borings and sampling or test pits and trenches.
Rather, they supplement these procedures and greatly re-
Because of the extreme variations in the character of the duce the time and cost and the environmental problems that
soil and rock and perhaps because of inadequacies on the often result from large-scale drilling operations.
part of those who perform the boring and sampling and
test pit work, there may be gaps in the underground pro- Resistivity
files established from any of the borings or geophysical
studies. In some instances, the techniques of borehole log- Resistivity measurements are made by passing electric cur-
ging borrowed from the petroleum exploration industry are rent through the ground and measuring the resistance of
fruitful. All are based on lowering a sensing device into an the various formations to current flow. The current flow
open borehole and measuring the soil and rock character- is largely electrolytic in that it is dependent on moisture
istics at closely spaced intervals of depth. A number of dif- and dissolved salts within the soils and rocks. The Wenner
ferent qualities can be observed, including self.potential of method, which is the simplest and most commonly used,
the ground, electrical resistivity within a relatively short ver- uses four electrodes, spaced equally in a straight line at the
tical distance,nuclear radiation,sound wave or impulse re- ground surface. Current is passed into the ground through
sponse, density based on nuclear absorption, and water con- the outer two electrodes, and the difference in potential
tent based on hydrogen ion reaction. generated by the resistance to current flow is measured be-
The product of this logging is a graph of each property tween the inner two electrodes. The greater the spacings
95
Table 4.1. Boring, core drilling, sampling, and other exploratory techniques (4.2, 4.8, 4.11, 4.14).
Method and
Reference Procedure Type of Sample Applications Limitations
Auger boring, Dry hole drilled with hand Augercuttings; disturbed, In soil and soft rock; to Soil and rock stratification
ASTM D 1452 or power auger; samples ground up, partially dried identify geologic units destroyed; sample mixed
preferably recovered from from drill heat in hard ma- and water content with water below water
auger flutes terials above water table table
Test boring, Hole drilled with auger or Intact but partially disturbed To identify soil or soft rock; Gaps between samples, 30 to
ASTM D 1586 rotary drill; at intervals (number of hammer blows to determine water content; 120cm; sample too distorted
(Figure 4.15) samples taken 36-mm ID for second plus third incre- in classification tests and for accurate shear and con-
and 50-mm OD driven ment of driving is standard crude shear test of sample solidation tests; sample lim-
0.45 m in three 150-mm in- penetration resistance or (N-value a crude index to ited by gravel; N-value sub-
crements by 64-kg hammer N) density of cohesionless ject to variations depending
falling 0.76 m; hydrostatic soil and undrained shear on free fall of hammer
balance of fluid maintained strength of cohesive soil)
below water level
Test boring of 50 to 75-mm ID and 63 to Intact but partially disturbed In gravelly soils Sample limited by larger
large samples 89-mm OD samplers (number of hammer blows gravel
driven by hammers up for second plus third incre-
to 160 kg ment of driving is penetra-
tion resistance)
Test boring Hole advanced by hollow- Intact but partially disturbed In gravelly soils (not well Sample limited by larger
through stem auger; soil sampled (number of hammer blows adapted to harder soils or gravel; maintaining hydro-
hollow-stem below auger as in test for second plus third incre- soft rock) static balance in hole below
auger boring above ment of driving is standard water table more difficult
penetration resistance or
N); N-value may be dis-
torted by auger and should
be compared with ASTM
D 1586
Rotary coring Outer tube with teeth ro- Relatively undisturbed sam- In firm to stiff cohesive soils Sample may twist in soft
of soil or tated; soil protected and pIe, 50 to 200-mm wide and soft but coherent rock clays; sampling loose sand
soft rock held by stationary inner and 0.3 to 1.5 m long in below water table difficult;
tube; cuttings flushed up- liner tube success in gravel seldom oc-
ward by drill fluid (Deni- curs
son—fixed cutter on Outer
tube; Pitcher—spring-
loaded on outer tube;
Acker air-mud core bar-
rel—larger clearances for
viscous driling fluid)
Rotary coring Similar to rotary coring of Soil cylinder 28.5 to 53.2 In soils and soft rocks that Sample smaller; equipment
of swelling rock; swelling core re- mm wide and 600 to 1500 swell or disintegrate more complex
clay or soft tamed by third inner plas- mm long encased in plastic rapidly in air (protected
rock tic liner tube by plastic tube)
Rotary coring Outer tube with diamond Rock cylinder 22 to 100 To obtain continuous core Core lost in fracture or vari-
of rock, bit on lower end rotated to mm wide and as long as 6 in sound rock (percentage able rock; blockage prevents
ASTM D 2113 cut annular hole in rock; m depending on rock of core recovered for dis- drilling in badly fractured
core protected by station- soundness tance drilled depends on rock; dip of bedding and
ary inner tube; cuttings fractures, rock variability, joints evident but not strike
flushed upward by drill equipment, and skill of
fluid driller)
Rotary coring Similar to rotary coring of Rock cylinder, typically To determine strike of joints Method may not be effective
of rock, rock above; continuous 54 mm wide and 1.5 m and bedding in fractured rock
oriented core grooves scribed on rock core long with compass orien-
with compass direction tation
Rotary coring Outer tube with diamond Rock cylinder 36.5 to 85 To recover core better in frac- Same as ASTM D 2113 but
of rock, bit on lower end rotated mm wide and 1.5 to 4.6 tured rock, which has less to lesser degree
wire line to cut annular hole in rock; m long tendency for caving during
core protected by stationary core removl; to obtain
inner tube; cuttings flushed much faster cycle of core
upward by drill fluid; core recovery and resumption of
and stationary inner tube drilling in deep holes:
retrieved from outer core
barrel by lifting device or
"overshot" suspended on
thin cable (wire line)
through special large diam-
eter drill rods and outer
core barrel
92
Table 4.1. Continued.
Method and
Reference Procedure Type of Sample Applications Limitations
Rotary coring 22-mm hole drilled for Continuous core reinforced. To obtain continuous Grout may not adhere in some
of rock, inte- length of proposed core; by grouted steel rod core in badly fractured, badly weathered rock;frac-
gral sampling steel rod grouted into soft, or weathered rock tures sometimes cause drift
method hole; core drilled in which recovery is of diamond bit and cutting
(4.10) around grouted rod low by ASTM 02113 rod
with 100 to 150-mm
rock coring drill (same
asforASTM 02113).
Thin-wall tube, 75to 1250-mm thin-wall Relatively undisturbed sam In soft to firm clays, short Cutting edge wrinkled by
ASTM 01587 tube forced into soil pIe, length 10 to 20 diam- (5-diameterl samples of gravel; samples lost in loose
with static force (or eters stiff cohesive soil, soft sand or very soft clay below
driven in soft rock); rock and, with aid of water table; more distur-
retention of sample drilling mud, in firm to bance occurs if driven with
helped by drilling mud dense sands hammer
Thin-wall tube 75 to 1250-mm thin-wall Relatively undisturbed sam- To minimize disturbance Method is slow and cumber-
fixed piston tube, which has internal ple, length 10 to 20 diam- of very soft clays (drilling some
piston controlled by rod eters mud aids in holding sam-
and keeps loose cuttings ples in loose sand below
from tube, remains sta- water table)
tionary while outer thin-
wall tube forced ahead
into soil; sample in tube
is held in tube by aid of
piston (osterberg-type
activates piston hydrau-
lically; Hong-type by a
ratchet)
Swedish foil Sample surrounded by Continuous samples 50 mm In soft, sensitive clays Samples sometimes damaged
thin strips of stainless wide and as long as 12 m by coarse sand and fine
steel, stored above cutter, gravel
to prevent contact of soil
with tube as it is forced
into soil
Dynamic sound- Enlarged disposable point None To identify significant dif Misleading in gravels or loose
ing (4.6, 4. 12, on end of rod driven by ferences in soil strength saturated fine cohesionless
Figure 4.15) weight falling fixed dis- or density soils
tance in increments of
100 to 300 mm
Static penetra- Enlarged cone, 36-mm N one To identify significant dif- Stopped by gravel or hard
tiOn (4.6, diameter and 60° angle ferences in soil strength seams
4.12, Figure forced into soil; force or density; to identify soil
4.15) measured at regular inter- by resistance of friction
vals sleeve
Borehole Inside of core hole viewed Visual representation To examine stra'tification, Best above water table or
camera by circular photograph fractures, and cavities in when hole can be stabilized
or scan hole walls by clear water
Pits and Pit or trench excavated to Chunks cut from walls To determine structure of Moving excavation equip-
trenches expose soils and rocks of trench; size not limited complex formations; to ment to site, stabilizing
obtain samples of thin excavation walls, and
critical seams such as fail- controlling groundwater
ure surface may be difficult
Rotary or. Toothed cutter rotated or Ground To penetrate boulders, Identifying soils or rocks dif-
cable tool chisel bit pounded and coarse gravel; to identify ficult
well drill churned hardness from drilling
rates
Percussion drill- Impact drill used; cuttings Rock dust To locate rock, soft seams, Drill becomes plugged by
ing (jack ham- removed by compressed or cavities in sound rock wet soil
mer or air air
track)
97
are between the electrodes, the greater the depth of influ- Geologic Maps
ence measured relative to the center of the electrode line
or spread will be. Dense rock with few voids and little As indicated in the earlier section on geologic reconnaissance,
moisture, such as most granites, will have high resistance, geologic maps provide information on physical features,
but saturated clay will have low resistance. Sometimes the lithology, and geologic structure. However, existing maps
failure surface of a landslide can be detected as a zone of seldom have sufficient detail for the evaluation of an actual
low resistance due to the concentration of moisture. The or anticipated problem. A geologic map has as its base a
major advantages of resistivity lie in the portability of the topographic map and depicts formations that immediately
instrument and the fact that large areas can be covered at underlie the ground surface. The projection of each forma-
a relatively small cost. The major disadvantage is that data tion to the ground surface is developed from the outcrop
interpretation is difficult and largely conjectural where mapping and boring data. In problem areas, detailed engi-
strata are not horizontal or uniform in thickness and where neering geologic maps should be prepared on a large scale—
contrasts in the resistivities of the materials are not sharp. the same as that for topographic maps, such as 1:1000 to
1:6000(1cm = 10 in to 1cm = 60 in; I in=83ftto 1 in
Seismic Refraction 500 ft). The engineering geology map should also depict
features such as rock outcrops, depths to bedrock, strike
The seismic refraction method is based on the measurement and dip of beds, faults, and joints, and locations of seeps
of the time required for a shock compression wave to pass and springs. The area covered in the detailed geologic map
from one point to another thiough the earth. The shock will depend on terrain and topography, alignment and
waves are generated by hammer impact or by detonating grade of roadway, anticipated soil and geologic conditions,
an explosive at or just below the ground surface. Some of and extent of the slide, if one has already occurred.
the waves are deflected or refracted by the more rigid,
deeper formations and return to the surface where their Profiles
times of arrival are recorded. In most seismic work involv-
ing landslides, a multichannel seismograph system is used. Centerline profiles of roadway location (made for design)
It includes a number of detectors or geophones that are often include data on soil and rock formations as inter-
placed at the surface at varying distances from the shock preted from geologic maps and borings at representative
source, amplifiers that enhance the signals, and a recording cross sections. In landslide-prone areas, the number of these
oscillograph that produces a time-based record of the sig- routine borings and map interpretations should be increased.
nals received from all the detectors simultaneously. The spacings along the centerlines of routine surveys are
When the shock wave from the explosion reaches each usually 15 to 150 m (50 to 500 ft). In landslide-prone
geophone,it appears on the recording as a pronounced areas, spacings should be on the order of 8 to 30 in (25 to
change in the trace and is termed the first arrival. The time 100 ft). The profile should show each of the materials en-
of first arrival at each geophone is used to compute the countered in each boring and appropriate additional infor-
depth to successively more rigid strata. Seismic measure- mation, such as moisture content, penetration resistance,
ments are like resistivity measurements in that the environ- strength, and rock core recovery.
ment is not disturbed, the equipment is portable, and rather
large areas can be covered at relatively small cost. But, inter- Cross Sections
pretation of seismic measurements, also like that of resis-
tivity measurements, is conjectural where the geology is Cross sections, such as the one shown in Figure 4.12, are
complex and where velocities of the various materials are beneficial in depicting detailed subsurface conditions for
not in sharp contrast. The technique is limited to strata landslide studies. These should show soil and rock in and
that are successively more rigid with depth; it cannot dif- below the slide, dip of the strata, groundwater, moisture
ferentiate softer strata below rigid ones. contents, and sliding surface or zone of rupture. In all
cases, the sections should extend from well into the stable
Gravity ground above the scarp or crown of the slide to some dis-
tance beyond the toe. Geologic cross sections should be
Precise measurements of the earth's gravity field can detect plotted by using an undistorted scale (horizontal scale
areas of low density. Colluvium or old landslide debris is equal to vertical scale); this is essential for a quantitative
usually less dense than the virgin materials. Large bodies evaluation of the subsurface geometry that influences the
of loosened rock can be identified where the density con- slide.
trast is great; boundaries cannot be defined except by bore-
hole logging. Three-Dimensional Diagrams and Models
Wt
Figure 4.12. Cross section through center
PIEZOMETER
of slide.
V PIEZOMETRIC LEVEL IN SILTY FINE SAND
1_X
8-4
-3
)Ot<'
B-i
A AND B HORIZONS
• GRANITE GNEISS
Note: 1 m = 33 ft
approach in developing a model involves the use of peg- tial failure, and the amount of increase can be depicted by
board and string. The board represents some elevation well using a series of transparent overlays on the original base
below the sliding surface. Holes are drilled in the board to map or by simply color coding successive new scarps and
plan scale. Pegs are inserted that represent the borings; the fractures on the original map. Often there is a considerable
top of the peg is at ground level. The pegs are colored to time lapse (months to years) between the failure and the
depict the changes in materials and structure in the boring. correction. In the meantime, the slide sometimes increases
The scale in all cases should be the same in all three direc- in size so that the proposed corrective measures are no
tions. Strings with different colors or thin plastic sheets longer valid. Thus, it is imperative that the slide continue
can then be laced between the pegs to depict the different to be monitored and the displays revised until repair is com-
boundaries, strata, ground surface, and shear surfaces. Such plete. Some observations should be continued after repairs
models can be especially beneficial in the analysis of com- are made so that the effectiveness of the measures selected
plex slides or in exhibits for litigation. A fence diagram is can be evaluated and the risk of future movement detected.
a second form of three-dimensional representation (Figure
4.13). The borings are displayed as vertical lines on an iso- SURFACE WATER AND
metric plot of location. An exaggerated vertical scale can GROUNDWATER
be helpful in visualizing three-dimensional relations, but it
distorts the perspective of the slopes. Importance of Water
Next to gravity, water is the most important factor in slope
Slide Enlargement
instability. Therefore, identification of the source, move-
Landslides usually continue to increase in size after the mi- ment, amount of water, and water pressure is as important
99
as identification of the soil and rock strata. Yet, those who Groundwater
conduct the field investigations often pay insufficient atten-
tion to water for two reasons. First, the investigative tech- Groundwater can be defined broadly as all water below the
niques, particularly wet drilling, often obscure groundwater. ground surface. Generally, however, the term is restricted
Second, water conditions change, depending on the weather to the water that is not restrained in the soil by capillary
at the time of investigation and on the cumulative effects tension, partially immobilized in the stress field surrounding
of rainfall, snowmelt, surface runoff, infiltration, evapora- clay minerals, or linked to the soil or rock minerals. Al-
tion, and transpiration throughout the year and during though soil capillary moisture, absorbed water, and water
long-term climatic cycles. of hydration may not be considered true groundwater, they
are a part of the total ground moisture system. During wet
Surface Water weather, infiltration reduces capillary tension, allowing the
groundwater to rise and encompass part of the capillary
In many landslides and potential landslides, the surface run- zone; during dry weather, transpiration depletes capillary
off is a major factor contributing to groundwater and the moisture, increases capillary tension, and decreases the level
soil and rock pore-water pressures that lead to reduced soil of free groundwater. Capillary tension increases the effec-
strength and movement. Moreover, the diversion or block- tive stresses among soil particles and temporarily increases
ing of intermittent and continuous surface streams by the soil strength. Loss of capillary tension either by saturation
landslide movement can add surface water (which originally or by drying causes loss of soil strength. An increase in ab-
flowed elsewhere) to the groundwater in the slide area. sorbed moisture is a major factor in the decrease in strength
The permanent surface-water streams can be identified of cohesive soils and some weakly cemented rocks. An in-
from aerial photography, particularly with the aid of color crease in the water of hydration of minerals such as anhy-
infrared photography (Chapter 3). Intermittent streams drite is accompanied by expansion, which can destroy the
can be identified as significant water sources only during bonding between soil or rock particles and decrease their
runoff periods in wet weather. However, the suspicion of strengths. Thus, in most cases, an increase in soil or rock
intermittent surface water for streams can be raised by topo- moisture is accompanied by a decrease in strength.
graphic details, such as gullying and surface washes. A sudden moisture increase in a dry soil can produce a
In some arid regions where the surface-water channels pore-pressure increase in trapped pore air accompanied by
are above the general groundwater table, infiltration from local soil expansion and strength decrease. The slaking or
stream beds directly into the ground may be considerable sudden disintegration of hard dry clay or clay-bonded rock
and be aggravated when the stream bed has been loosened, is caused both by an increase in absorbed water and by pore
fractured, distorted, or dammed by landslide activity. When air pressure.
water courses appear uphill or adjacent to a landslide, the Groundwater has been more narrowly defined as that
establishment of flow gauges both upstream and downstream part of the soil-rock-water system that is free to move from
from suspected zones of surface-water loss into the ground point to point under the influence of gravity. The surface
is usually prudent. If a stream is very small, a sheet metal of that body of free water, which is at atmospheric pres-
V-notch weir and a stilling basin upstream (to establish the sure, is the groundwater table. In simple terms, the ground-
appropriate approach velocity) are sufficient. The head is water table is the elevation of zero (atmospheric) water
measured by a ruler, staff gauge, or continuous recorder. pressure; water is in tension above the water table in the
In larger streams, water-measuring flumes or large weirs zone of capillary saturation and in the unsaturated capillary
may have to be established. fringe, and is under pressure below that table. At any level,
In addition, visual observations of surface water, includ- the water pressure is equal to the unit mass of water multi-
ing sheet runoff, during periods of high rainfall are desirable. plied by the distance (z) below the water table. Above the
An engineer or geologist who can adequately describe and water table, z is negative and the computed pressure is nega-
record the flow should make the observations, even though tive.
the task is seldom convenient or pleasant. Evidence of
sheet runoff sometimes can be found by mud lines and de- Aquifers
bris that become lodged in tall grass and shrubs. Experience
shows that the contributions of sheet runoff and intermittent In engineering usage, an aquifer is a soil or rock stratum
streams to groundwater and pore pressures influencing land- that is significantly more pervious than the adjoining strata.
slides are often overlooked in investigations. Securing evi- An aquifer can also be an opening in the soil or rock forma-
dence of surface water and groundwater is a major objective tion, such as an animal burrow, a shrinkage crack, voids
of reconnaissance. left by rotting of vegetation, joints, fracture zones, and
Hints of excessive soil moisture frequently can be found other discontinuities that provide localized ability to trans-
in the character of the vegetation or wildlife in the vicinity mit water. As the term implies, an aquifer contains water,
of the suspect or sliding area. For example, cattails do not the source of which may be infiltration from precipitation,
ordinarily grow in low dry areas, and bullfrogs require infiltration from streams, leaking water pipes, or even the
water for their reproduction and life cycle. The locations upward discharge from a deeper artesian aquifer. Potential
of permanent and intermittent surface flows are recorded aquifers can be identified from detailed records of the soil
on topographic maps of the slide or incorporated on map and rock boring that should describe all of the more per-
overlays. Attention should be paid to the relation of the vious strata or fracture systems that are capable of trans-
changes in these water courses to the continuing changes mitting water. The absolute permeability of the stratum
in the slide and slide area topography. does not determine whether it is an aquifer. Instead, its
110
jelative permeability compared to the strata above and movement associated with landslides. The water table
particularly the strata below is more significant. For ex- changes with rainfall, groundwater flow from outside the
ample, a stratum of silty fine sand could be an aquifer if area, and movement of the landslide. Thus, the changing
it were confined between clay strata, but would be an aqui- nature of a perched water table is one of its most impor-
clude compared to a stratum of coarse clean sand. The iden- tant characteristics.
tification of some discontinuities, such as vertical cracks An artesian aquifer is one in which the level of atmo-
and joints that are potential aquifers, is more difficult be- spheric water pressure is higher than the upper surface of
cause borings have little chance statistically of encounter- the aquifer, but the water is confined by an overlying aqui-
ing them uniess the borings are inclined. Test pits and test clude. If the water pressure level in a confined aquifer coin-
trenches described in the earlier section on subsurface ex- cides with that of the aquifer above, a normal water table is
ploration are far more useful in determining the presence present at that instant. However, drainage of the overlying
and spacing of such localized discontinuities. A thorough aquifer may not necessarily affect an underlying confined
understanding of the geology of the individual formation aquifer; thus, a confined aquifer could be normal on one
also provides clues to potential discontinuities that might day and artesian on the next, without a change in pressure.
act as aquifers. Moreover, it might even become perched with more drastic
An aquiclude is a stratum or discontinuity that is suffi- changes in water distribution below the aquifer.
ciently less pervious than the adjoining strata that it is a
barrier to groundwater. For example, silt washing into a Groundwater Observations
crack in the ground can produce a clastic dike or aquiclude
that will block the flow in a sandy seam. The very move- Piezometers
ment of a landslide can shift pervious strata to align with
impervious strata, generating localized aquicludes. These Essentially, a piezometer is a small-diameter well in which
aquicludes change with the continuing movement of the the water level or water pressure in an aquifer can be ob-
landslide. served. Piezometers have many forms. Simple borings with
Once potential aquifers have been identified, one must slotted or perforated casings become observation wells that
determine whether they transmit water or are subject to reflect normal or perched water levels. Wells penetrating
water pressure. Long-term observations, particularly dur- into a confined aquifer can measure artesian pressure if the
ing periods of wet weather and high general groundwater casing is sealed into the aquicludes above. The design and
levels, are necessary. Moreover, because of changing topog- the installation of an adequate system of piezometers re-
raphy and changing interrelations of the strata with the quire a thorough understanding of the location and perme-
movement of a landslide, an aquifer today can be dry to- ability of the aquifer and of the surrounding aquicludes. If
morrow. Therefore, potential aquifers are evaluated by the volume of water within a piezometer tube or well is
the variations in water pressure as measured by piezometers. large with respect to the flow through an aquifer, the pi-
ezometer will be slow to respond to pressure changes. Thus,
Piezometric Level in aquifers of low permeability or flow, piezometers that re-
quire small water volume changes in order to respond to
The piezometric level at a point is the elevation to which pressure changes are essential. In pervious aquifers, simple
water eventually will rise in a small tube sealed into the ap- holes supported with perforated, screened plastic pipe sur-
propriate aquifer. The water pressure (u) at a point is equal rounded by a filter of sand are adequate. Details of typical
to the piezometric level (hpiez) minus the elevation of the piezometers and an electric water-level detector are shown
point (h 01 ) multiplied by the unit weight of water: in Figure 4.14.
The water pressure within a soil or rock stratum or crack, The changing nature of groundwater is well recognized by
as reflected by the piezometric level, is a major factor in geologists and geotechnical engineers. Changes in rainfall
shear strength and the most significant single factor in land- infiltration and changes produced by groundwater usage are
slide activity. A number of piezometric levels can be de- common. However, the changes produced by a landslide
lined at any location. Although the two types of water are less well recognized. For example, the cracks associated
tables (normal and perched) are well recognized by geol- with the landslide may create a more pervious zone of soil
ogists and geotechnical engineers, the multiplicity of pi- or rock that drains well-established aquifers. The opposite
ezornetric levels present in hillside areas, particularly in may also be true: Aquifers may be blocked by the land-
areas prone to landsliding, may surprise even experienced slide movements, and thus a normal water table becomes
groundwater hydrologists. artesian. Groundwater levels must be observed throughout
A normal water table is the level to which the surface the period of slide investigation. During periods of dry
water infiltrates in the ground or the level at which the weather, in which there is little movement of the landslide,
water pressure is atniospheric. Below this level, the observations at intervals of a week or two may be adequate.
groundwater is more or less continuous and pressure in- When the slide is moving rapidly, and particularly during
creases hydrostatically. A perched water table is one that and following periods of snowmelt and rainfall, daily,
is sustained above an underlying independent body of hourly, or continuous readings by a recorder may be de-
groundwater table by an aquiclude. Normal aquifers are sirable to correlate episodes of ground movement with
sometimes converted to perched aquifers by the rotational groundwater changes.
101
Permeability pressure at the point of exit is approximately atmospheric.
However, the actual water pressure just within the aquifer
A knowledge of the ability of a formation to transmit at the point of exit is usually greater than atmospheric,
water (permeability) is essential in the planning of drain- and such estimates are likely to be high. When several such
age systems to correct landslide activity. However, the ef- determinations are made, they can yield the order of mag-
fects of minor variations in soil texture and particularly the nitude of the permeability of the stratum under its natural
effects of cracks and fissures cannot be easily determined. conditions of flow.
The permeability of the soil probably varies more from
point to point than does any other soil property. Further- Springs and Seeps
more, the order of magnitude of the variation of permeabil-
ity in common soils, ranging from gravels to clays, is greater The intersection of an aquifer with the ground surface pro-
than the variation of the other properties relevant to land- duces either a concentrated flow in the form of a spring or
slide analyses. a diffused flow in the form of a seep. Both of these repre-
The simplest form of in-place permeability tests is a bor- sent the exit or discharge of the aquifer and may be re-
ing that is cased through the various soil strata down into garded as safety valves for the release of groundwater pres-
the aquiclude above the aquifer whose permeability is to sure. So long as springs or seeps flow freely, an unusual
be measured. The hole is then drilled into the aquifer. The buildup of groundwater pressure in the aquifer that sup-
soil permeability is measured by adding water to the hole ports the spring or seep will not be likely. If the rate of
to a predetermined level and noting the rate at which the discharge of the spring or seep is known and if the perme-
water level drops (4.17). The test should be repeated sev- ability of the formation can be estimated, an estimate is
eral times if it is suspected that the aquifer is not saturated; even possible of the pore-water pressure within the aquifer
otherwise, the inflow during the first few trials may merely at varying distances within the hillside. Unfortunately, en-
represent the filling of empty soil voids. Numerous modifi- gineers and geologists who evaluate landslides often regard
cations of this simple approach have been proposed to take a spring or seep as a causative factor (it can be, if the efflu-
into account the penetration of the hole into the aquifer, ent of a spring above a slide infiltrates downhill into the
the hole diameter, and other geometric characteristics. slide zone), but generally the total effect is more beneficial
However, experience shows that highly refined analyses of than detrimental. For example, a sudden stoppage of a
such a simple test are seldom justified. spring or seep may be the precursor of landslide activity.
A better field permeability test involves pumping water Increased flow in a seep or spring frequently indicates that
into the hole and measuring the rate of flow once equilib- an aquifer is draining, the piezometric pressures are reduc-
rium has been established. The-water pressure is measured ing, and the stability is increasing. Springs and seeps sug-
at several different locations and distances within the same gest which potential aquifers contain water and could be
aquifer at points surrounding the hole. Typical distances involved in the buildup of pore-water pressure. Thus, a
from the inflow hole are 5 and 15 m (15 and 50 ft); the pi- spring or seep once located should be identified with the
ezometric holes are aligned in at least four directions. A particular soil stratum or rock formation that produces
more thorough investigation involves three sets of piezom- the flow.
eters at distances such as 5, 10, and 20 in (15, 33, and 65 A significant discharge of a spring or seep should be col-
ft) and at least four different directions from the inflow lected and the quantity monitored by a V-notch weir or
hole (4.14). If the piezometric level in the aquifer is suffi- similar device. The discharge from small springs can be col-
ciently high, a pump-out test can be used instead of a pump- lected by use of 5 to 10-cm (2 to 4-in) plastic pipes em-
in test. The arrangement of observation wells or piezom- bedded in gravel-filled collecting wells. The flow can be
eters is similar. The test hole must be sufficiently large so piped to a buried oil drum, which serves as a catch basin for
that the pump can be placed inside it; otherwise, the water silt; the edge of the oil drum can be cut to form a V-notch
can be lifted only about 8 m (25 ft). Pump-out tests are weir, which serves as a measuring device. It is sometimes
sometimes more reliable than pump-in tests because any helpful to install a continuous water-level recorder to indi-
soil fines that accumulate in the well are flushed out by cate rapid changes in the spring or seep discharge. As pre-
the flowing water. In pump-in tests, those fines can ac- viously pointed out, infrared photography, as well as ther-
cumulate in the soil or rock pores and give a false indica- mal sensing, can show where springs and seeps are located,
tion of low permeability. In pump-in tests, care must be even though they may be partially obscured by vegetation
taken not to contaminate an aquifer that is used for drink- or colluvium cover.
ing water. For example, river water for pump-in tests
should be chlorinated. Correlation
If the groundwater is definitely discharged through seeps
and springs whose flow can be collected and measured, the The location of aquifers and springs or seeps should be
permeability of an aquifer from those discharges and the shown on both the topographic maps of the landslide area
gradient found within the aquifer may possibly be estimated and the various landslide cross sections that are plotted from
by a grid of piezometers. Typically, the flow is downhill topographic data. Since water comingither from runoff
through an aquifer toward the toe of the slope. The pi- or from infiltration or groundwater originating elsewhere
ezometers, therefore, should be in lines parallel to the di- is a major factor in most landslides, the evaluation of the
rection of maximum slope in order to measure the hydraulic groundwater aquifers and the changes in the piezometric
gradient. Of course, a single piezometer might be intro- level are a vital part of the investigation. Unfortunately,
duced into the aquifer on the assumption that the water experience shows that these groundwater changes seldom
102
Figure 4.14. Groundwater observation devices.
(a) SIMPLE OBSERVATION WELL (b) SIMPLE SEALED PIEZOMETER Cc) ELECTRIC WATER LEVEL PROBE
WITH PLASTIC CASING
r
GAUGE
VENT HOLE VALVE = /
MILLIAMMETER
..... .: ..:..•
RAM1ED
CLAY
SEAL
:.
u
0
.-
=(2)
'
..
IMPERVIOUS
STRATUM
COPP
AJ j
0..:. ........
.........:...... 0 100mm
II
777777
SLOTTED
PLASTIC
SAND
. ..
PLASTIC
WITH
LEAD SHOT
h) iA•
•1 Iul r.
PIPE SLOTTED : . .: ... AQUIFER A. . A
PLASTIC
PIPE
BARE WIRE
10mm
are evaluated in sufficient detail to obtain a complete pic- Human Changes Before Construction
ture of the factors leading to landslide activity.
Many areas of the world have been altered by human activ-
ENVIRONMENTAL FACTORS ities, such as terracing for agriculture, diversion of streams,
mining, leveling for housing and industrial construction,
Both natural and human changes in the environment have and cuts and fills for highways, airfields, and railroads. An-
a profound effect on landslides. The history of the area, other human activity is alteration of the groundwater
both during human occupation and during recent geologic table. The piezometric level is lowered significantly by
times, determines the conditions leading to land movement; wells. In areas adjacent to large cities that depend on
moreover, as mentioned previously, the historical behavior groundwater or in areas in which wells are used for irriga-
provides clues to potential instability. Therefore, the total tion, the level has been depressed more than 100 m (300
environment must be investigated to provide both the his- ft); the affected area may extend for many kilometers.
torical background and a key to future changes. This lowering of the piezometric level within the soil and
rock imposes an increased effective stress, which in some
Weather instances contributes to the stability of the slope. Para-
doxically, lowering the groundwater table can also trigger
The climate of the area, as expressed in the various compo- sliding because of the difference in the total and buoyed
nents of weather, is the ultimate dynamic factor influencing weight of the soil. Draining land for agriculture may lower
most landslides. The data ordinarily available for landslide the water table, but this is seldom of significance in land-
investigations from weather stations within 100 to 200 km slides because such drainage is generally in bottomlands,
(60 to 120miles) include rainfall (15 mm, hourly, and which are not subject to sliding.
daily), temperature (daily maximum and minimum and A rise in the water table can ultimately have an adverse
daily and monthly mean), evaporation (daily and monthly), effect on slope stability. Such a rise may occur around
wind (maximum, hourly, daily, and monthly), snowfall small towns that begin to use surface water instead of
(daily and monthly), relative humidity (daily and monthly), groundwater to meet increased water demands. Irrigation
and barometric pressure (maximum and minimum daily). in and regions has significantly elevated the groundwater
Unfortunately, the weather station may be too distant for table and thereby significantly decreased the strength of soils
the data to be fully representative of the site. The effects that are cemented by water-soluble agents such as calcium
of these factors can seldom be evaluated analytically be- carbonate or dry clay. For example, in certain areas of
cause the relations are too complex. Empirical correla- India, hilly land, which for years had been stable, suddenly
tions of one or more of the weather factors (particularly became landslide-prone after irrigation began. In one area
rainfall, snow, and melting temperatures) with episodes of of the Himalayas, a mountain road was destroyed each year
movement or movement rates can point out those environ- by landslides for a distance of about 10km (6 miles) be-
mental influences that must be controlled to minimize cause of irrigation of the terraced mountainsides above.
movements. The irrigation water penetrated through the closely jointed
103
rock and produced flow slides. Similarly, water impound- A detailed study of the daily logs of inspectors, daily
ments can raise the water table and change the stability of reports of project superintendents, records of blast-hole
nearby hillsides. For example, a highway fill across a reser- advance and explosives used, andjournals of the contractor
voir suddenly experienced sliding when the soil was satu- will provide clues to the actual construction conditions.
rated at high water levels, and then stress and localized pore- Sometimes these records contain references to springs that
water pressure increased when the reservoir was rapidly have been covered by embankments, references to unsuit-
drawn down. able materials that have been inadvertently placed in fills,
Changes in slope by excavation and filling in the vicinity and evidence of overblasting or movements that have been
of the study area may give clues as to the long-term effects forgotten in the scramble to complete construction on
of excavation and filling. For example, some rocks, such time. The written records supplemented by personal inter-
as shale, are hard and strong when first excavated. How- views often provide the best clues to the factors that trig-
ever, exposure to air and changing weather causes the shale gered landslide activity.
to break down, soften, and slide. Such failures have been
experienced in the Appalachian Mountains, on the eastern Effect of Ecosystem on Sliding
flanks of the Rocky Mountains, and along the Pacific
Coast. Although older cuts and fills may not be so deep The biological environment of the site plays a part (although
as those currently contemplated, they sometimes exhibit usually minor) in the behavior of a landslide. For example,
symptoms of long-term deterioration. Their overall per- a good vegetation cover promotes infiltration of rainfall
formance, particularly local sliding, constitutes full-scale, and minimizes surface runoff and local gullying. The effect
in-place tests of the physical properties of the soil and of the increase in the local groundwater by infiltration may
rock. Such tests are more reliable indicators of future sta- be less serious than that of the concentrations of stress by
bility than the short-term laboratory tests that are custom- local gullyiñg. A thick mat of vegetation will also reduce
arily made to evaluate potential sliding. the amount of water that becomes groundwater by enhanc-
A reconnaissance of older cuts and fills is recommended ing water loss due to transpiration. Of course, during cold
to evaluate the stability of human changes in topography. periods and periods of snowmelt, the water loss by trans-
Calculations based on information regarding unstable areas piration is negligible.
will provide valuable data on the ultimate strength of the The reinforcing effect of a strong root mat is significant
materials and on the degree and rate of change that can oc- in the scarp area of landslides. In some marginally stable
cur because of weathering. This reconnaissance of existing slopes, the root mat can be the difference between sliding
cuts and fills has been sorely neglected in most landslide and creep. Moreover, a well-developed root system signifi-
studies, yet the data obtained are probably the most reli- cantly reduces retrogression of the scarp up the hill. In
able of all in evaluating the strength of the materials. To highway excavation, the root mat may be affected by fac-
make a valid correlation between the performances of old tors other than the slope. For example, the intercepting
cuts or fills and those proposed requires that the geologic ditch, which is sometimes excavated above a deep highway
similarity of the two areas be established. Moreover, the cut on a hillside, may cut through the root mat and be-
climate at the area of old construction must be correlated come the focal point for future sliding. Haul roads above
with that at the new area if the older area is not in the im- cuts similarly damage the roots and promote local sliding,
mediate vicinity of the site under construction. which leads to retrogression.
Animal burrows may also play some part in small slides.
Changes Brought by Construction For example, the interlacing burrows of rodents can weaken
the soil and provide channels for concentration of surface
New construction in areas in which landslide activity is sus- water and its infiltration into the ground. However, there
pected or has commenced should be monitored to deter- is little documented evidence of such effects.
mine whether changes predicted by studies made before Overgrazing of hillsides reduces the vegetation cover and
construction actually occur, to evaluate the methods used frequently promotes more rapid infiltration and localized
in such studies, and to predict future landslide activity in sliding, which then triggers more profound movements,
an area where movement is just commencing. The major particularly in semiarid regions where vegetation does not
purpose of monitoring, however, is to determine what recover rapidly from the overgrazing and where the rainfall
changes the construction actually produces. Because may be extremely intense during short periods.
changes in the design of slope and drainage features are
often made in the field, the completed project is usually Effect of Sliding on Ecosystem
different from that shown in the plans and specifications.
Moreover, fills may not be compacted as specified or in- Landslides may also change the ecosystem of an area.
tact materials in virgin cuts may be loosened by uncon- Groundwater flows are altered by landslide movement; lo-
trolled blasting. Sometimes, the person responsible for cations of springs and seeps change, and these changes are
changes in the plans and specifications will take no notice reflected in differences in vegetation. For example, wet-
of them because of ignorance of their importance or will area vegetation, such as cattails, often develops within a
hide them because of\fear of criticism or recrimination. few months in the depressed areas commonly associated
Thus, the investigato/ has difficulty in finding out exactly with landslides. Similarly, the cracks and the scarp areas
what changes might have been made to the provisions of above a landslide provide local drainage of the topsoil and
the plans and specifications and is often placed in the po- an associated loss in the ability of this material to sustain
sition of a detective ferreting out information. growth of vegetation during periods of dry weather. Thus,
104
scarps that are hidden by vegetation are revealed by a reduc- changes, cutting, filling, and landslides. Although labora-
tion in the vigor of that vegetation. This can be sensed by tory tests can simulate an almost unlimited range of stress
color infrared photography during dry weather. On the other changes, the sample tested has already been subjected to a
hand, in areas of bare rock, open joints in which moisture significant stress cycle (unloading during sampling and re-
is trapped and roots penetrate provide the only zones loading during testing) that is not necessarily similar to the
that can sustain life. In one particular major rock slide in stress changes involved in a landslide. These inherent stress
a steep mountainous area, the joints were clearly delineated limitations of both in-place and laboratory tests must be
by lines of small shrubs, but the remaining rocks were bare. understood by those who use soil or rock test data in eval-
Those joints became the scarps of multiple rock slides. The uating slides.
continuing enlargement of uphill joints above the rock
slide could be seen in the gaps between the root mats and Borehole Tests
the walls of the cracks. Thus, a study of the vegetation of
an area may provide supplementary clues to the rate of Certain tests have been devised to be performed in the same
earth movement and its focal points, such as seeps, springs, bore holes that are drilled for identifying the soil strata and
cracks, and fissures in the ground. for securing the small-diameter samples. Although bore-
hole tests suffer from the limited volume of material tested,
FIELD TESTING they do allow the soil to be tested without.the disturbance
produced by removing the sample from the ground, taking
To evaluate the potential stability of a slope that has not it to the laboratory, and preparing it for testing. However,
failed or to assess the effectiveness of corrective measures some disturbance and sloughing are caused by stress relief
for a slope that has failed requires that the physical proper- in the bore-hole walls.
ties of the materials involved be measured. Of course, mea-
surements are most conveniently done in laboratory tests Dynamic Penetration Test
of undisturbed samples secured from the site, but obtaining
representative samples is difficult for a number of reasons. Driving a device into the ground by impact measures the
resistance of the soil to rapid or dynamic displacement.
Discontinuous samples with relatively small diameters Thus, it indirectly measures shear strength under the same
can miss thin critical strata (such as the slickensided surface form of impact loading produced in the test. If the dy-
of movement of an ancient slide) that control sliding. namic and static shear strengths are similar, the test can be
The laboratory tests can integrate neither the effects an indicator of the static shear strength, which may also
of the discontinuities, such as cracks, within the soil nor be empirically related to other soil properties, such as the
the effects of localized hard spots, such as gravel, in a clay relative density of sands or the compression index of clays.
matrix. Typical correlations are summarized by Sowers and Sowers
Distortion, disturbance, and moisture and stress (4.14).
changes are always associated with taking a sample out of The standard penetration test (ASTM D 1586) is an ad-
the ground, handling it, transporting it, and preparing it junct to split-tube sampling. A split-tube sampler 37 mm
for laboratory testing. (1.4 in) ID, 50mm (2 in) OD, and 660mm (26.0 in) long
(Figure 4.15) is driven 450mm (18 in) in three 150-mm
To eliminate these difficulties, various in-place tests (6-in) increments into undisturbed soil at the bottom of a
have been devised. An important difference between lab- borehole by blows of a 63.5-kg (140-1b) hammer falling
oratory and in-place tests is their relation to the initial in 760 mm (30 in). The sum of the blows for the second and
situ state of stress. A laboratory test must often reproduce third increments is the standard penetration resistance (N);
this state of stress whereas a field test inevitably begins at it is expressed in blows per 300 mm (blows per 1 ft). Be-
this state of stress. In either case, the in situ state of stress fore the test, the sampler is seated 20 to 40 mm (8 to 16
must be evaluated, which is often difficult and expensive. in) in the hole bottom. Because of cuttings or other
For some soil deposits, such as a deep, normally consoli- weakened material in the bottom of a bore hole, the first
dated clay with a horizontal surface, the horizontal stress 100 to 150mm (40 to 60 in) may not be meaningful.
will be a fraction of the vertical stress, depending on the Therefore, the standard penetration test includes gaps in
coefficient of earth pressure at rest. Unfortunately, such the penetration resistance record. The N-value encompasses
a simple situation is seldom valid for landslide studies be- both hard and soft seams in the 300-mm distance. Al-
cause the ground surface is not level and other significant though some investigators have attempted to drive such
stresses may be imposed by desiccation, artesian water sampling tubes as far as 1.8 m (5 ft), counting blows for
pressures, tectonic forces, residual forces from strains pro- each 150-mm increment, the accumulating skin friction
duced by past earth movements, and changes produced in and the buildup of soil resistance within the samples usu-
any of the above by erosion or construction. ally produce resistances that increase with each successive
The fact that any in-place test must be conducted with increment until the sample is withdrawn and the bore-
reference to the existing in situ state of stress is an impor- hole cleaned out.
tant limitation in itself. Although the stress field is altered A more sensitive dynamic test involves driving a cone
somewhat by boring or introducing some testing device into point that is 25 to 100 mm (2 to 4 in) in diameter and has
the soil, the existing stress field cannot be changed signifi- a point angle of 60° into the soil by means of a weight
cañtly. Thus, in-place testing usually cannot simulate the [typically 50 to 100 kg (110 to 220 ib) falling 0.5 to 1 m
large changes in stress that accompany environmental (1.5 to 3 ft)] ; the drive rod is 25 to 30mm (1 to 1.2 in) in
105
Figure 4.15. Penetrometers. (a) SPLITTUBE SAMPLER FOR STANDARD PENETRATION TEST WATER
DRILL
SHOE SPLIT TUBE PORT ROD
50
(bI DRIVE CONE ON 25-,rnn DRILL ROD cl DUTCH CONE WITH FRICTION SLEEVE FOR
STATIC TEST
CONE ROD
SLEEVE TUBE
FRICTION SLEEVE
100 mm
25..
6O
36 mm
36 mm
Nole: 1 mm = 0.04 in.
diameter. By such a procedure, the penetration resistance ranging from 30° to 90° and diameters from 36 to 50 mm
for each 100 to 300mm (4 to 12 in) can be measured with- (1.4 to 2 in), as shown by Hvorslev (4.6). One form of
out the accumulation of soil friction along the drive rod. cone widely used in the Netherlands and generally termed
In many soils, a 50-mm (2-in) point driven with the 63.5- the Dutch cone employs a cylindrical sleeve that is 100
kg (140-1b) hammer falling 760 mm (30 in) gives penetra- mm (4 in) long and 36 mm (1.4 in) in diameter and is
tion resistances for 0.3 m (1 ft) of driving that are equiva- above a 60°, 36-mm (1.4-in) conical point, as shown in
lent to N. The drive cone on a 25-mm rod is also shown Figure 4.15. The cone is forced ahead slowly by a steady
in Figure 4.15. pressure that is measured. Simultaneously or successively,
Dynamic penetration resistance is generally correlated depending on the design, the frictional resistance of the
empirically with soil properties measured by either labora- sleeve against the soil is measured. The cone directly pro-
tory tests or field tests of the same material. In this way, vides information on the point bearing capacity of the
large numbers of low-cost penetration tests supplement the soil. It can be interpreted in terms of the point bearing
more limited information obtained by more expensive lab- of piles. Although at one time the cone resistance of soil
oratory tests. Although many relations between resistance was believed to be identical to that of a pile, experience
and soil properties such as angle of internal friction in sands has shown that cone resistance may be double that of
and undrained shear strength (cohesion) of saturated clays piles. The cone does provide detailed information on the
have been published (4.14, 4.16), these should not be used relative strength of the soil at small intervals. For example,
indiscriminately. Instead, a new correlation should be es- some cones have an electronic readout that generates a
tablished from the data obtained on the site in question, continuous graph or electromagnetic tape showing both
or the data should be used to verify the accuracy of the cone resistance and sleeve resistance as a function of depth.
published relations (4.5, 4.13). The record is well suited to identifying weak zones, such
as the shear surface of the slide, which may be less than a
Static Penetration Test few centimeters thick. By way of contrast, conventional
sampling might not find such a thin zone of weakness. The
Static penetrometers measure the resistance of the soil to ratio of cone resistance to sleeve resistance is an indicator
displacement of a point by a static or slowly increasing of the type of soil (4.12). The cone penetrometer is an ex-
load. Most tips are in the form of cones having point angles tremely valuable supplement to the more direct boring and
106
sampling techniques. It helps to identify changes in strati Figure 4.16. Menard pressuremeter for borehole dilation
test (not to scale).
fication and to pinpoint weak materials that should be in-
vestigated in more detail by direct methods. GAS PRESSURE
The borehole shear test measures the shear strength of the Vane Shear Test
soil in an annular zone surrounding the boring. The device
consists of an expandable plug with serrations on its outer Vertical blades at the end of a thin rod produce a vertical
surface to grip the soil walls of the hole when pressure of cylindrical surface of shear when rotated (Figure 4.17).
known magnitude is applied internally by a hydraulic sys- The torque required to initiate continual rotation is a mea-
tem. The soil is then sheared by pulling the device upward sure of the peak undrained strength of the soil, and the
through the hole. If several such tests (essentially undrained torque required to maintain rotation after several revolu-
direct shear) are made on the same stratum at varying in- tions measures the residual or disturbed strength. To mini-
ternal pressures, the Mohi failure envelope can be obtained. mize end effects, the length of the vane should be at least
The test is limited because it shears the soil in a different twice its width. The blades should be sufficiently thin that
direction than that involved in the landslide process. There- there is a minimum soil disturbance due to displacement
fore, if the soil has anisotropic properties (usually the axis and sufficiently thick that they do not bendunder load. In
of weakness is parallel to the greatest extent of the surface very soft soils, the vane and its torque rod are forced into
of shear movement and more or less perpendicular to the the soil to each level to be tested. A reference test using
direction of shear in the borehole shear tests), the results the torque rod without the vane is required so that the
may be misleading. There is usually some soil smear in the torque necessary to overcome rod friction can be subtracted
walls of the borehole; thus, the soil involved in the test from the total torque measured when the soil is tested. In
may be partially disturbed. However, the disturbance from firm soils or at great depths, the test is made in undisturbed
this cause is likely to be less than that resulting from rough soil 300 to 760 mm (12 to 30 in) below the bottom of a
handling of soil samples. The size of the device is such that bore hole; hence, the resistance of the torque rod in the
it integrates the effect of soil irregularities over a cylindrical hole generally is negligible. Numerous procedures and
surface with a diameter of 76 mm (3 in) and a length of forms of equipment, ranging from simple torque wrenches
about 300 mm (12 in). to elaborate torque meters that apply a uniform angular
107
strain rate, have been utilized (4.6). Our opinion is that of weakness, the sides of the box should be perpendicular
the increased accuracy of the results does not justify the to that plane, and the plane should lie between the top and
use of elaborate procedures and complex equipment. Cau- bottom halves of the box. Such a test setup is shown in
tion should be exercised when one interprets the peak Figure 4.18. Good contact is secured between the soil and
strength; in some cases, the strength measured in a vane the box either by careful trimming of the soil or by pour-
test has been found to be as much as 30 percent greater ing plaster to fill the space between the box and the soil.
than that measured by other methods. Moreover, safety Because of the difficulties in trimming gravelly soils or
factors computed from such strengths have been found to soft rock, the plaster filling is recommended.
be unrealistically high. Vane data should be correlated A normal load is placed on the block to be tested by
with other shear data for use in analysis and design. means of a plate, the dimensions of which are slightly
smaller than that of the box, fitting just inside the upper
large-Scale Pit Tests half of the box. The load is applied by jacking against a
piece of heavy machinery above the pit or against a heavy
As previously stated, one of the major limitations of lab- steel beam anchored to the ground by earth anchors. The
oratory tests is their inability to integrate the variations anchors must be sufficiently far from the test zone so that
in the soil, particularly in zones with weak or hard spots. the stresses around them do not influence the test. The
This can be overcome by large-scale, in-place tests per- bottom half of the box is anchored securely in place by
formed in pits or trenches excavated to the questionable packed soil, concrete, or plaster in the bottom of the test
strata or zones of slickensiding. Although the range of pit. The top half is then jacked sideways by a calibrated
stresses and particularly the range of groundwater pres- system so that the amount of lateral movement and the
sures that can be evaluated by such tests are limited, the load causing the movement can be measured. The direc-
tests permit large volumes of soil to be evaluated under tion of jacking should be parallel to that which is suspected
the conditions present within the total mass without the in a potential landslide zone or parallel to the movement
problems of sample disturbance and exposure inherent in that has occurred in an actual landslide (as indicated by
small-scale sampling and laboratory testing. slickensides on the shear surface). The same surface can
be tested at several different normal loads if the test for
Load Test each vertical load increment is stopped soon after peak
strength or significant movement develops. At this point,
The oldest form of in-place test is the plate load test (4.14). a larger normal load is applied and the test is resumed.
A pit is excavated to the surface of the stratum in question, Such a direct shear test determines the average shear stress
and a rigid square or circular plate is placed on the ground. required to produce failure on a predetermined failure
Its width should be as great as possible, but no wider than plane. If the peak strength of the soil must be determined,
about two-thirds the thickness of the stratum whose strength separate tests should be performed on fresh sample blocks
is to be evaluated. The plate is loaded incrementally so that for each normal load applied. This will require a large test
at least 10 successively greater loads are applied before the trench or more than one test pit.
plate shears the soil beneath it. The results of such a test The results of a large-scale shear test simulate the shear
can be interpreted in terms of soil bearing capacity to give strength of the soil along an actual failure surface. The re-
the shear strength of the soil along a curvilinear surface sults integrate the effects of both hard and soft zones if the
(which may crudely approximate the failure surface of a test sample is sufficiently large. Unfortunately, it is diffi-
landslide). However, there are many different interpreta- cult to include the effects of changing water pressure.
tions of such tests, all yielding different values for the shear However, our experience shows that meaningful shear test
strength parameters. Therefore, the test has limited value data have been obtained from such tests, particularly if
in determining the strength of the soil involved in the in- done during the wet season. Correlating the results of an
stability of large earth masses. in-place test with those of smaller laboratory tests on sim-
ilar soils makes it possible to extend the data obtained
Large-Scale Direct Shear Test from the in-place tests to include the effects of changing
water pressure introduced under controlled laboratory
A large-scale direct shear test can be performed in a pit at conditions. Thus, a more reliable combination of data for
the level of the suspected weak stratum; in the case of an evaluating the strength of the soil mass involved in the
existing landslide, the test may be performed on the actual landslide is provided by a combination of judiciously se-
failure surface of the soil. The pit is excavated to the level lected large-scale field tests and laboratory tests than by
of the stratum or shear surface to be evaluated and should laboratory tests alone.
be large enough to allow engineers and technicians to work Standard equipment for making such tests is ordinarily
around the sides without disturbing the soil to be tested in not available. Instead, the equipment is fabricated to fit
the center. All the soil within the pit is excavated, except the size of the sample needed for the particular situation,
that to be tested, which is left in the form of a block or the space available within the test pits or trenches, and the
crude stump above the bottom of the pit. The size of the geometry of the shear surface. It can be improvised out of
block is dictated by the engineer's or geologist's evaluation steel angles, channels, and plates at a reasonable cost. The
of the variations of soil strength. It should be large enough loading is provided by calibrated hydraulic jacks or jacks
to be representative of the stratum as a whole and not just with load cells; and the movements are measured by mi-
its weaker or stronger segments. A double box is placed crometer dial gauges. Although such measurements may
around the block in question. If there is a definite plane be characterized as crude, their lack of precision is more
W .
Figure 4.18. Direct shear test of strength along failure surface. a) PLAN VIEW TRACTOR
E SHEAR~BOX TR E NC H
I.
I I
L-------
TRACTOR
CONCRETE
- - THRUST
BLOCK
CALIBRATED
HYDRAULIC
JACK .-- BALL JOINT
NORMAL LOAD PLATE
UPPER SHEAR BOX
SHEAR BOX
- .---
- SHEAR
STRUT SURFACE
Note: Trench shored for safety and support of tractor. GRILLAGE
than compensated for in their realistic representation of calculated. Since the density of various soils and rocks
field conditions. does not vary greatly from point to point, calculations
based on estimates may be sufficiently reliable for esti-
Borehole Dynamics mates of the rigidity of the mass. The seismic compres-
sion wave velocity is particularly valuable as an indication
The effect of a landslide on a borehole, particularly in of discontinuities in rock, because it is reduced by cracks
squeezing or shearing, can be a valuable indicator of rates or microfissures.
of movement and locations of shear surfaces. Although The refraction technique does not allow identification
equipment for measuring such changes is available, much of a weak fractured zone beneath a sounder, higher velocity
meaningful information also can be obtained from stratum. However, this identification can be done by cross-
drillers' observations in those borings that are not instru- hole seismic measurements, utilizing borings from 3 to 12
mented. Among the factors to be observed are loss of in (10 to 40 ft) apart. The seismic impulse is generated at
drilling fluid, gain of water or fluid, gas, squeezing of hole, the depth of the stratum measured by an explosion, and
damage to drilling tools, loss of samples, blocking of cores, the time required for the shock wave to travel through the
drill rod drop, broken core, slickensided samples, and un- material is measured in the second hole by a geophone sus-
usually weak zones. pended at the level of the stratum in question. Although
engineering properties relevant to landslide evaluation,
Geophysical Tests such as shear strength, cannot be determined directly by
such techniques, the suspect stratum can be identified and
The geophysical studies previously described can be used a crude empirical correlation can be developed between
not only as a means to define the stratification but also as seismic velocity and engineering properties. Below the
a direct measure of certain physical properties of the soils. water table the compression wave velocity in materials of
Although these properties may not necessarily be those low rigidity is obscured by the compression wave velocity
needed to evaluate stability, an empirical correlation in water.
among the properties of interest can sometimes be estab- The apparent resistivity of a soil can be related empiri-
lished in the stability analysis of the landslide. cally to soil type and soil moisture. An aquifer with highly
The seismic refraction technique yields the compression ionized water exhibits a far lower apparent resistivity than
wave velocity of the soil or rock. If the density is known, a dry stratum in which the minerals do not ionize readily.
the dynamic modulus of elasticity in compression can be For example, the shear surface within a landslide frequently
ME
Figure 4.19. Correlation
of slide movement with
210
precipitation.
180
150
0
120
C
0,
90
0,
60
30
E
200
0 z
- 6 0
150
F-
F-
0.
100 0
0
50
N D J F 11 A M J J A S 0 N D J F M A M J J
MONTHS
exhibits a marked lower electrical resistance than the sur- maximum water pressure gradient, or to the observed tra-
rounding intact soils, probably because of the disturbance jectories of movement are the beginning point for a stability
of the minerals and the accumulation of water along the analysis of an earth mass. The various strata encountered
shear surface. and the groundwater levels or pressures are depicted on
At best, however, geophysical tests are only indirect these cross sections. In addition, overlays should show the
supplements to the more direct means of evaluating the engineering properties relevant to the stability analyses.
qualities of the soil and rock in place. The tests are simple, The surfaces of failure, as deduced from ground surface ob-
can be performed in a short period of time, and are rela- servations, sophisticated instrumentation, borings, test
tively inexpensive. However, interpretation of the results trenches, and pits, define a body of soil on which various
must be done with extreme caution; too often interpreters force systems act. The external forces depicted on an over-
jump to conclusions and obtain misleading results. lay to the cross sections provide the beginnings of the anal-
yses described in Chapter 7.
CORRELATION OF DATA
Time-Based Observations and
During an investigation, considerable data will be accumu- Correlations
lated. The variations of these data, which may occur ran-
domly or in some definitepattern, should be studied in The significance of the different factors involved in a land-
three dimensions because the mass of soil or rock involved slide can frequently be found from empirical correlations
in a landslide is three dimensional. Some of the ways in between movement and observed forces or environmental
which this can be accomplished are discussed briefly below factors. For example, a time-based graph of both landslide
displacement and rainfall or accumulated rainfall and snow-
Areal Variations melt may show a visual relation (Figure 4.19). Similar
time-based relations can be observed between construction
Some of the data obtained will vary with geographic posi- operations and movement. In some cases, plots of the ob-
tion. These data can be depicted best in the form of maps. served phenomena as functions of the logarithm of time
If the variations in the data are systematic, they can be de- are instructive. For example, stochastic processes, such as
picted as contours, such as lines of equal strength or lines the readjustment of coarse particles through creep and
of equal groundwater pressure. These variations are fre- secondary soil compression, often can be approximated by
quently plotted on overlays to the topographic map of the a straight line on a plot of movement as a function of the
site; several overlays may be desirable. logarithm of time. Thus, the time graph or the logarithm
of the time graph becomes a diagnostic tool in analyzing
Cross Sections the causes of movement.
110
More intensive investigation uncovers the soil and environ- 4.7 Lahee, F. H. Field Geology. McGraw-Hill, New York,
mental factors that produced the movement. The mathe- 6th Ed., 1961, 926 pp.
4.8 Lambe, T. W. Soil Testing for Engineers. Wiley, New
matical analysis of a landslide is based on the field investi- York, 1951, 165 pp.
gation and obviously can be no more accurate than the 4.9 Philbrick, S. S., and Cleaves, A. B. Field and Laboratory
data obtained from the field work; likewise, the corrective Investigations. In Landslides and Engineering Practice
measures based on the analyses can be no more effective (Eckel, E. B., ed.), Highway Research Board, Special
Rept. 29, 1958, pp. 93-111.
than the quality of the data used in the analyses. Unfor- 4.10 Rocha, M. A Method of Integral Sampling of Rock
tunately, our experience has been that field investigations Masses. Rock Mechanics, Vol. 3, No. 1, 1971, pp. 1-12.
are often inadequate and that geologists and engineers 4.11 Sangierat, G. The Penetrometer and Soil Exploration.
sometime jump to conclusionsbefore the data have been Elsevier, New York, 1972, 464 pp.
obtained and evaluated; thus, they merely perpetuate their 4.12 Schmertmann, J. H. Static Cone Penetrometers for Soil
Exploration. Civil Engineering, Vol. 37, No. 6, 1967,
past mistakes. A technically sound solution can be derived pp. 71-73.
only from technically sound data. 4.13 Schultze, E., and Melzer, K. J. The Determination of
the Density and the Modulus of Compressibility of Non-
REFERENCES Cohesive Soils by Soundings. Proc., 6th International
Conference on Soil Mechanics and Foundation Engineer-
ing, Montreal, Vol. 1, 1965, pp. 354-358.
4.1 American Geological Institute. Glossary of Geology. 4.14 Sowers, G. B., and Sowers, G. F. Introductory Soil
AGI, Washington, D.C., 1972, 805 pp. Mechanics and Foundations. Macmillan, New York,
4.2 American Society for Testing and Materials. Annual 3rd Ed., 1970, 556 pp.
Book of ASTM Standards: Part 19—Soil and Rock; 4.15 Sowers, G. F. Remote Sensing: A New Tool in Site In-
Building Stones; Peats. ASTM, Philadelphia, 1977, vestigation. Law Engineering Testing Co., Marietta, Ga.,
494 pp. Bulletin G-5, 1973.
4.3 Compton, R. R. Manual of Field Geology. Wiley, New 4.16 Terzaghi, K., and Peck, R. B. Soil Mechanics in Engineer-
York, 1962, 378 pp. ing Practice. Wiley, New York, 2nd Ed., 1967, 729 pp.
4.4 Deere, D. U., and Patton, F. D. Slope Stability in Resid- 4.17 U.S. Bureau of Reclamation. Design of Small Dams.
ual Soils. Proc., 4th Pan-American Conference on Soil U.S. Bureau of Reclamation, Denver, 1973, 816 pp.
Mechanics and Foundation Engineering, San Juan, 4.18 Zâruba, Q., and Mend, V. Landslides and Their Con-
American Society of Civil Engineers, New York, Vol. 1, trol. Elsevier, New York, and Academia, Prague, 1969,
1971, pp. 87-170.
205 pp.
4.5 Fletcher, G. F. A. Standard Penetration Test: Its Uses
and Abuses. Proc., Journal of Soil Mechanics and Foun-
dations Division, American Society of Civil Engineers,
New York, Vol. 91, No. SM4, 1965, pp. 67-75.
4.6 Hvorslev, M. J. Subsurface Exploration and Sampling of PHOTOGRAPH CREDITS
Soils for Civil Engineering Purposes. U.S. Army Engineer
Waterways Experiment Station, Vicksburg, Miss., 1949, Figure 4:8 Courtesy of Tennessee Department of Transportation
521 pp. Figure 4.9 Courtesy of Tennessee Department of Transportation
Chapter 5
Field
Instrumentation
Stanley D. Wilson and P. Erik Mikkelsen
The development of field instrumentation has had a signifi- replace field observations (5.38) and investigative procedures
cant impact on geotechnical engineering in the past 15 discussed in other chapters. Instead, it augments other
years. Of particular importance is the notable contribution data by providing supplementary information and by warn-
of instrumentation to landslide-oriented problems (5.50). ing of impending major movements. Typical situations for
The usefulness of field instrumentation for the identifica- which various instruments have been used are
tion of landslide movements, for the monitoring of slides
that have been remedially treated, and for other geotech- To determine the depth and shape of the sliding sur-
nically oriented measurements has been described in the face in a developed slide so that calculations can be made
literature (5.1, 5.8, 5.18, 5.36, 5.45, 5.48, 5.50, 5.56). Al- to determine the available soil strength parameters at fail-
though the earlier Highway Research Board Special Report ure and so that remedial treatments can be designed;
on landslides (5.20) mentioned the use of simple instru- To determine absolute lateral and vertical move-
mentation (which is by no means obsolete and is described ments within a sliding mass;
in this chapter), its publication generally preceded the in- To determine the rate of sliding (accelerating or de-
troduction of modern instruments. The economic impact celerating movements) and thus warn of impending dangers;
of slope failures created a demand for better instrumenta- To monitor the activity of marginally stable natural
tion and led to its general acceptance and use. This chap- slopes or cut slopes and the effects of construction activity
ter describes the types of instrumentation currently avail- or precipitation on them;
able, their limitations, and their specific applications to To monitor groundwater levels or pore pressures nor-
slope instability problems. In addition, traditional ap- mally associated with landslide activity to enable effective
proaches to the determination of slide movements and stress analyses to be performed;
groundwater levels are reviewed. To provide remote digital readout or a remote alarm
The need to gather certain quantitative data to analyze system that would warn of possible dangers; and
slope-stability problems and to design remedial measures To monitor and evaluate the effectiveness of various
is discussed in other chapters. Topographic mapping, geo- control measures.
logic mapping, subsurface soil and groundwater investiga-
tions, and laboratory strength testing are normally per- In the last situation, savings are often realized in remedial
formed to aid in determining the cause of the slide, the treatment by a planned and monitored sequence of con-
mode of failure, and the physical and engineering character. struction. For example, drainage might be initially installed
istics of the soil and rock involved. Such investigations are and its effect monitored to determine whether a planned
necessarily performed on already developed landslides, the buttress is actually necessary.
characteristics of which may be observed and noted. How-
ever, small movements of a soil mass prior to or even at in- INSTRUMENTATION PLANNING
cipient failure are usually not visually evident; so the value
of information that can be obtained at the ground surface Adequate planning is required before a specific landslide is
is limited. However, instrumentation can provide valuable instrumented. The steps are (a) determine what types of
information on incipient, as well as fully developed, land- measurements are required, (b) select the specific types of
slides. In this respect, instrumentation is not intended to instruments best suited to make the required measurements,
112
(c) plan the location, number, and depth of instrumentation, probe pipes will suffice to determine the depth. If the
and (d) develop the recording techniques. movements are large and rapid, accuracy is not an essential
Initially the planning process requires the development requirement and even relatively crude inclinometers may
of ideas on the causes of the landslide and the probable suffice. When the rate of movement is small and the depth
limits of the depth and outer boundaries of the movements. and distribution are not known, more precise instrumenta-
Of course, to have the answers before the system is planned tion is required. Carefully installed precision inclinometers
would be helpful. Reconnaissance of the area, study of are best in such instances, although there may be cases in
the geology, review of rainfall records, and observation of which extensometers or strain meters can be used to advan-
topographic features, especially recent topographic changes, tage.
will often provide clues. Unfortunately, no two slides are The Casagrande type of piezometer is the most useful
alike in all details, and experience alone without the appli- general purpose pore-pressure measuring device, but may
cation of judgment may lead to erroneous concepts. give too slow an initial response in fine-grained soils; there-
An instrumentation system in which the instruments do fore, pneumatic or electric types may be preferable. High-
not extend below the zone of movement, were installed at air-entry, low-flow piezometers should be used in clays or
the wrong locations, or are unsuitable is a waste of time clay-shales in which permeabilities are low or suctions may
and money. Loss of time may mean that corrective treat- be present because of unloading (5.53).
ment is started too late to save the project. The types of instruments, layout, and monitoring
schedules are usually determined by the specific needs of
Types of Measurements Required a project. Several basics, however, should be thoroughly
evaluated for any system. Instruments should be reliable,
Landslides, by definition, involve movement, and the mag- rugged, and capable of functioning for long periods of time
nitude, rate, and distribution of this movement are gener- without repair or replacement. They must also be capable
ally the most important measurements required. Equally of responding rapidly and precisely to changes so that a
important in many slide problems, however, are measure- true picture of events can be maintained at all times. High
ments of pore-water pressures within the slide area, partic- sensitivity is usually a prerequisite when performance is
ularly in layered systems in which excess hydrostatic monitored during construction, since it is often the rate of
pressures may exist between layers. change rather than the absolute value that provides the key
If the depth of sliding is readily apparent from visual to proper interpretation. The location of instruments re-
observations, surface measurements may be sufficient for quires a thorough understanding of the geologic and sub-
obtaining the rate of movement. The surface measure- surface conditions if meaningful data are to be obtained.
ments should extend beyond the uppermost limit of visual This is particularly true of pore-pressure recording devices
movement so that possible extension in advance of crack- that are intended to measure pressures in specific zones of
ing can be monitored. Vertical and horizontal measure- weakness or potential instability. Since most measure-
ments of movement of the ground surface at various loca- ments are relative, a stable base or datum must be provided
tions within the slide area should be obtained. Vertical so that absolute movements can be determined.
offsets, widening of cracks, and toe heave should be mon- In this chapter, the primary emphasis is on measurement
itored. The direction of movement can often be inferred techniques and instruments for landslides; less emphasis is
from the pattern of cracking, particularly by the matching placed on conventional surface surveying techniques. Con-
of the irregular edges of the cracks. If the depth and thick- ventional surveying methods are fundamental to displace-
ness of the zone of movement are not apparent, inclinom- ment measurements at the surface (5.23); this has been dis-
eters or similar devices that can detect the movement with cussed more fully in Chapter 4. Modern deformation in-
depth must be used. Pore pressures at or near the sliding struments for landslides have augmented the methods of
surface must be measured to enable an effective stress anal- surface measurements, but more important they have ex-
ysis to be performed and to assess the adequacy of drain- panded capabilities in terms of accuracy, simplicity, and
age measures. Rapid-response piezometers are advantageous, ease or convenience of operation. In addition, an ever-
particularly in impervious soils. expanding range of instruments makes possible the mea-
surements of subsurface deformations (5.13, 5.17, 5.21,
Selection of Instrument Types 5.25, 5.44). Table 5.1 gives summaries of various survey
methods, instruments for measuring subsurface deforma-
Many types and models of instruments are available for tions, and piezometers.
measuring the changing conditions in a landslide; they vary
in degree of sophistication, particularly in regard to read- SURFACE SURVEYINQ
out capabilities. Instruments have been developed to mea-
sure vertical and horizontal earth movements, pore-water Conventional Surveying
pressures, in situ stresses and strains, dynamic responses,
and many other parameters. However, in most landslide In an active slide area, surface movements are normally
problems, the measurement of horizontal movement of monitored to determine the extent of slide activity and the
the foundation soil or rock and the measurement of pore- rate of movement (5.21, 5.32). Optical instrument surveys
water pressures are of primary concern. Instruments com- and tape measurements are used to determine lateral and
monly used for these purposes are described in this chapter vertical movements. Bench marks and transit stations, lo-
If the movement is known to be along a well-defined cated on stable ground, provide the basis for which subse-
shear plane, such as a bedding plane or fracture, simple quent movements of hubs can be determined optically and
113
Table 5.1. Surveying methods, horizontal movement devices, extensometers, and piezometers (5.13)
Method or
Measurement Limitations and Reli-
Instrument Type Range Accuracy Advantages Precautions ability
Method Chaining Variable ±1/5000 to Is simple and inexpensive; Requires clear, relatively Excellent
Ordinary, 1/10 000 of has direct observation flat surface between
third order distance points and stable refer-
Precise. ± 1/20 000 to ence monuments
first order 1/200 000 of Corrections for tempera-
distance ture and slope should be
applied and a standard
chain tension used
Electronic dis- 20 to 3000m ± 1/50 000 to Is precise, long range, fast, Accuracy is infl'jenced by Good
tance mea- 1/300 000 of usable over rough terrain atmospheric conditons;
surement distance accuracy at short ranges
(EDM) (<30 to 90 m) is lim-
ited for most instruments
Optical leveling ±3 to 6 mm Is simple, fast, particularly Has limited precision; re- Excellent
Ordinary, with self-leveling instru- quires good bench mark
second and ments nearby
third order
Precise, first ±0.6 to 1.2 mm Is more precise Requires good bench Excellent
order (paral- mark and reference
lel plate points and careful ad-
micrometer . herence to standard
attachment, procedures
special rod)
Off sets from a
baseline
Theodolite 0 to 1.5 m ±0.6 to 1.5 mm Is simple; has direct observa- Requires baseline unaf- Excellent
and scale tion fected by movements
and good monuments;
accuracy can be im-
proved by using a tar-
get with a vernier and
by repeating the sight
from the opposite end
of the baseline
Laser and 0 to 1.5 m ±1.5 mm Is faster than transit Is seriously affected by Good
photocell atmospheric conditions
detector
Triangulation ±0.6 to 12 mm Is usable when direct mea- Requires precise measure- Good
surements are not possible; ment of base distance
is good for tying into points and angles; requires
outside of construction good reference monu-
area ments
Photogram- ± 1/5000 to Can record hundreds of po- Weather conditions can Good
metric 1/50 000 of tential movements at one limit use
distance time for determination of
overall displacement pat-
tern
Horizontal Long base 150mm 0.3mm Is precise; can be used to Has risk of electrical Good
movement strain meter check horizontal move- failure; has limited ap-
instrument with electrical ment at top of other de- plication for tunnels
readout vices, such as inclinom-
eters
Fixed multi- ±20 arc 5a Is available in portable Is complex; does not
point bore- version and double pivot measure continuous
hole deflec- version to measure move- prof ile
tometers ment along two axes
Fixed multi- ±0.03 mm in 3 m Is precise; can be removed Is complex; does not
point bore- for repairs or reuse; uses measure continuous
hole inch- standard inclinometer , profile
nometers casing Accuracy is lost when
removed and replaced
Portable bore-
hole inch-
nometers
Wheatstone ±12° , op- ±20mm in 30 m Has long experience record; Requires lengthy calcula- Very
bridge tional to is not sensitive to tempera- tions; reads one axis at good
pendulum ±25° ture a time; has no provi-
-
sions for automatic
readout
114
Table 5.1. Continued.
Method or
Measurement Limitations and Reli-
Instrument Type Range Accuracy Advantages Precautions ability
Acceler- ±300 op- ±5 mm in 30 m Reads two axes at a time; Requires lengthy calcula- Good
ometer tional to has automatic readout tions without automatic
±900 and recording provi- readout; requires manual
sions check of data for errors
with automatic readout
Vibrating ±15° or 20° ±10mm in 30 m Is available in single or Requires lengthy calcula- Good
wire double axis models tions
Errors due to zero drift
are possible
Bondedre- ±20° ±10mmin30m Has adjustable range on Requires lengthy calcula- Fair
sistance some models; uses ordi- tions; reads one axis at
St rain nary square tubing for a time
gauge casing on one model Errors due to zero drift,
temperature, or electrical
connections are possible
Extensometer Tape ito 30 m ±0.03 to 0.3mm Is simple, precise, portable; Has accuracy limited by Excellent
is good for measuring tension adjustment; re-
tunnel diameter changes quires temperature cor-
rection
Portable rod i to 8 m ±0.03 to 0.3 mm Is simple, precise, portable Has limited span; has ac- Excellent
curacy limited by sag;
invar tubes can be
used to minimize tem-
perature corrections
Weight- Variable ±5 to 20 mm Is simple Has creep in wire that Fair
tensioned leads to errors; is vul-
wire nerable to damage in
tunnel
University of 150mm ±0.03to0.13mm Is simple, precise, and large Is not adaptable to re- Good
Illinois rod range (some models can be mote reading; has two
reset and extended through anchor units only
linings placed later); can be
quickly installed; is more
blast and damage resistant;
has easily adjusted anchors
Interfels rod Variable Is simple, precise; can have Has projecting head that Good
multiple anchors; can ac- is vulnerable to damage
cept remote readout trans-
ducers
Variable- 15 to 90 mmb ±0.05 to 0.13mm Has multiple anchors, up to - Has variable tension that Fair;
tensioned 6 or 8 (some models are de- requires varying calibra- short
wire signed for remote readings tion factors, wire fric- term
with transducers using tion and hysteresis that
bonded resistance or can seriously affect ac-
vibrating-wire strain curacy, risk of electrical
gauges) failure, and projecting
head that is vulnerable
to damage
Constant- 50 mmb ±0.05 to 013 mm Has multiple anchors; is de- Has wire friction and hys- Fair;
tensioned signed for remote reading teresis that can seriously short
wire using potentiometers; has affect accuracy, risk of term
constant calibration factor electrical failure, and
projecting head; is com-
plex mechanically
Piezometer Open system ±3mm head Is simple, inexpensive, and Central observation sys- Excellent
adequate for most earth tem cannot be used
problems
Well point Is simple, inexpensive, and Has large time lag in low
universally available; can porosity materials
be driven in place (k < 1 .zm/s) and metal-
lic elements that may
corrode; cannot mea-
sure negative pore
pressures
Casagrande Is simple, inexpensive; has Cannot measure negative
no metallic elements, long pore pressures; requires
service life, and provisions borehole and carefully
for offset riser pipe and placed bentonite or
flushing tip grout seal
115
Table 5.1. Continued.
Method or
Measurement Limitations and Reli-
Instrument Type Range Accuracy Advantages Precautions ability
by tape measurement. As shown in Figure 5,1,transit lines ticularly higher order surveys, can define the area of move-
can be established so that the vertical and horizontal dis- ment, more accurate measurements may be required in
placements at the center and toe of the slide can be observed. many cases. Terzaghi (5.47) stated that "if a landslide
Lateral motions can be detected by transit and tape mea- comes as a surprise to the eyewitness, it would be more ac-
surements from each hub. When a tension crack has opened curate to say that the observers failed to detect the phenom-
above the top of a slide simple daily measurements across ena which preceded the slide," The implication is that the
the crack can be made between
' two hubs driven into the smallest movements possible should be measured at the
ground. In many cases, the outer limit of the ground move- earliest possible time. Terzaghi (5.47) also describes the
ments is not known, and establishing instrument setups on movements that precede a landslide.
stable ground may be a problem. The detection of small surface movements when crack-
Various techniques and accuracy achieved in optical ing is not apparent requires a trained observer. If the ground
leveling, offset measurements from transit lines, chaining surface is covered with rocks, or if there is a rock embank-
distances, and triangulation have been discussed extensively ment, horizontal stretching will result in local instability of
in the literature (5,8, 5.23), particularly for dams, embank- individual rocks such that walking over the slope gives one
ments, and buildings. Although conventional surveys, par- a sense of insecurity. Overturned rocks can be detected by
a change in coloring or surface weathering. Trees inclined EDM can be used to monitor large slides with large move-
at the base but changing to vertical trunks a meter or so ments and provide a rapid way to survey many points on
above the ground may indicate old slide movements. In- the mass from a single, readily accessible location. An ex-
clined but straight trunks indicate recent movements. ample of such an installation involves an ancient landslide
Cracks covered over with leaves or surface duff can be de- in Washington along the Columbia River, where the boun-
tected by walking over the area and noting the firmness daries of the active slide are more than 0.6 km (1 mile) in
of support. Frequently animals will avoid grazing in a po- width and 4.8 km (3 miles) in length. Yearly movements
tential landslide area because of uncertain support or hid- vary from only 1 m to 6 to 9 m (0.3 ft to 20 to 30 ft) and
den fissures. Small openings on the downhill side of struc- depend on the time of year and the rainfall. A permanent
tures, or next to tree trunks, may indicate creep. Overtaut station, readily accessible all year, has been set up on the
or excessively sagging utility lines or misalignment of fence opposite side of the river, and monthly distance readings
posts or utility poles are excellent indicators of ground are taken to 14 points on the slide area and to 2 points lo-
movements. Such movements, when accurately monitored, cated outside the slide zone. The distances involved vary
serve as an important tool in assessing the potential hazard from about 1500 to 6000 m (5000 to 20000 ft). Figure
to structures, nearby residents, and the public. 5.2a shows the movements (changes in distance) recorded
by the EDM (electrotape) during a 1-year period for 2
Other Types of Surface Surveying selected 'points at the Columbia River slide, based on
monthly readings. At the end of the year, the points were
There are three rapidly developing surveying techniques resurveyed by triangulation. The discrepancy is about 10
today, and these will undoubtedly find increasing use in cm (4 in), which, although larger than anticipated, is quite
field measurements. Some are already in extensive use satisfactory considering the total movements. Figure 5.2b
(e.g., electronic distance measurement and lasers); others shows recorded changes for 2 points believed to be on
are in limited use or are in the experimental stage (e.g., stable ground. The variation of monthly readings is seen
terrestrial photogrammetry). Electronic distance measure- to be ± 0.06 m (± 0.2 ft); variation is no greater for a 4765-
ment (EDM) has changed surveying practices more than m (15 630-ft) length than for a 1844-rn (6050-ft) length.
anything else in the last 100 years (5.15). EDM devices Laser instruments are already widely used for setting
have proved particularly suited for use over rugged terrain; alignments, and they are well suited for setting a reference
they perform more accurately and much faster than ordi- line for offset measurements to surface monuments. Laser
nary surveying techniques and require fewer personnel. beams are also used with some EDM instruments. It should
Lightweight EDM instruments can be used efficiently be possible to measure offsets with an error no greater than
under ideal conditions for distances as short as 20 m (66 0.003 to 0.006 m (0.01 to 0.02 ft) (5.23).
ft) and as long as 3 km (2 miles); errors are as small as Terrestrial photogrammetry has been used in some cases
±0.0032 m (± 0.010 ft) (5.28, 5.43). Larger instruments on dam projects (5.35) and in mines, but no landslide mea-
using light waves or microwaves can be used at much surements using this method have been found in the litera-
longer distances. The accuracy of EDM is influenced by ture. Phototheodolites are used to take successive stereo-
weather and atmospheric conditions; comparative read- photographs from a fixed station along a fixed camera
ings with three different instruments are described by Pen- axis; movements are identified in a stereocomparator, and
man and Charles (5.40). accuracies of 0.006 to 0.009 m (0.02 to 0.03 ft) have been
- --
- BARKER eo*io
clAck •ONITORING POINTS
A163
BEIIC
i7?'
0
- FESET (0
/ -
117
Figure 5.2. Microwave measurements of slide (a) POINTS LOCATED ON SLIDE
MICROWAVE
-0.6
RIANCU AT ION
——
--------
-MICROWAVE
2.4
TRIANGULATION
-3.0
z
C,
I
0
(CAT PA Hr 4760m 0
2
C-
CO
a
iw
reported. Moore (5.35) reports average errors of about one so that both shear and opening or closing of the joint can
order of magnitude greater than those from precise ground be monitored. Another simple device is a hardwood wedge
surveying. lightly forced into the joint crack with a mark placed on it
at the level of the rock. If the joint opens, then the wedge
Crack Gauges will fall to a deeper level in the rock. However, if the joint
is closing or subject to shear movement, such a wedge indi-
When geologic mapping is considered adequate to describe cator may not be helpful.
the area affected by a landslide, simple qualitative measure-
ments can provide knowledge of the activity. Movements Tiltmeters
on cracks, particularly those uphill and those downhill
from the well-defined zone of movement, give clues to the Tiltmeters can be used to detect tilt (rotation) of a surface
increasing size associated with many landslides. Therefore, point, but such devices are relatively new and have had
it is desirable to monitor the change in width, as well as fairly limited use. They have been used mostly to monitor
the change in elevation, across the cracks. This can easily slope movements in open pit mines and highway and rail-
be done by direct measurement from hubs set on both way cuts, but they have potential application in landslide
sides of the cracks. Crude, simple gauges can be con- areas. They may be used in any area where the failure
structed in the field to provide a more accurate and con- mode of a mass of soil or rock can be expected to contain
tinuing indication of crack movement. Periodic tape mea- a rotational component. One type of tiltmeter is shown
surements can detect movements (and rates) not visually in Figure 5.3, and sample tiltmeter data from a mine slope
apparent. are shown in Figure 5.4. The same types of servo acceler-
In the areas of shear along the sides of the sliding area, ometers are used with the tiltmeter as with sensitive mdi-
a similar crack indicator system is used, except that three- nometers (a fairly rugged transducer with a similar range
point stations are often installed. However, because of the and sensitivity). The prime advantages are light weight,
greater width of the shear zone, such simple devices may simple operation, and compactness at a relatively low
not be satisfactory. In jointed rock, the change in joint cost.
spacing can be determined by scribing marks or bonding INCLINOMETERS
washers on the rock surface on both sides of the joint.
These indicator marks should be joined by a straight line The development of inclinometers has been the most im-
'U
portant contribution to the analysis and detection of land- a function of depth below the ground surface and as a func-
slide movements in the past 2 decades. Although it has tion of time (Figure 5.6). The application of the inclinom-
been used most extensively to monitor landslides, the in- eter to landslides is readily apparent, namely, to define the
clinometer has gained widespread use as a monitoring in- slip surface or zones of movement relative to the stable
strument for dams, bulkheads, and other earth-retaining zones. Inclinometers have undergone rapid development
structures and in various areas of research (5.1, 5.13, 5.17, to improve reliability, provide accuracy, reduce weight and
5.34, 5.44, 5.55, 5.59). Since the introduction in the early bulk of instruments, lessen data acquisition and reduction
1950s of the pendulum-actuated inclinometer operating in time, and improve versatility of operations under adverse
grooved plastic casing (5.45, 5.50), the same basic concepts conditions. Automatic data-recording devices, power cable
have been applied to instruments manufactured by at least reels, and other features are now available.
12 U.S. and foreign firms. Although the principle of opera- An inclinometer system has four main components.
tion has remained unchanged (Figure 5.5), inclinometers
have been improved in accuracy and ease of operation and A guide casing is permanently installed in a near ver-
have undergone numerous modifications to adapt them to tical borehole in the ground. The casing may be made of
individual projects. plastic, steel, or aluminum. Circular sections generally
The inclinometer measures the change in inclination (or have longitudinal slots or grooves for orientation of the
tilt) of a casing in a borehole (Figure 5.5) and thus allows sensor unit, but square sections are used with some types.
the distribution of lateral movements to be determined as A portable sensor unit is commonly mounted in a
Figure 5.3. Portable tiltmeter. Figure 5.4. Tilt at ground surface due to advance of longwall mining face.
+2.0
'- 30
+1.5
0' -200
I- LONGITUDINAl
AXIS
z
D + 0.5
0
x TAN AS VEX S 0
0 I AXIS
1016
Guide :heeis
Probe
119
Figure 5.6. Measurement of slide movements with inclinometer. Figure 5.7. Inclinometer.
SKEAR
ZONE S
ON ABLE
FROM INCLINOMETER> WELL WELL HIlTS
BULCE —
A B
OF SLIDE
BATTERY
PRECISION
10
Note. 1m331t;lcn,-04r. GALVANOMETER
INDICATOR
carriage designed for operation in the guide casing (often
referred to as the probe or torpedo).
A control cable raises and lowers the sensor unit in
the casing and transmits electrical signals to the surface.
For accurate depth control of the sensor unit, the cable is
usually graduated or lowered on a separate surveyor's steel
L —_' T - -
WATERTIGHT
- SLE
tape.
A portable control and readout unit at tile surface
supplies power, receives electric signals, and displays read. PENDULUM
120
4. The photographic type photographically records the As originally conceived, the first modern inclinometers
projection of a pendulum with two degrees of freedom on used pendulum-activated resistors to convert inclinations
a target oriented by a magnetic compass; no orientation is into electrical measurements; instruments of this type have
needed within the casing or borehole. This type is not had long and successful experience records (5.45). A typical
often used because its accuracy is generally much lower inclinometer of this type consists of a pendulum, the tip of
than that of previous types and data reduction is much which makes contact with a resistance coil, subdividing the
more time consuming. Measurements with this instrument coil into two resistances that form one-half of a Wheatstone
are described by Hanna (5.25). bridge (Figure 5.8). The other half of the bridge, contained
in the portable control box, includes a precision potenti-
With the exception of the photographic type, most incli- ometer. The potentiometer readings are proportional to
nometers measure the inclination of the casing in two mu- the inclination of the torpedo. The system has a sensitivity
tually perpendicular near-vertical planes. Thus, horizontal of about 3 min of arc and generally demonstrates a preci-
components of movement, both transverse and parallel to sion of±l3 to 25mm (±0.50 to 1.0 in) over a 33-rn (100-
any assumed direction of sliding, can be computed from ft) casing (1: 1000). The instrument is adaptable to most
the inclinometer measurements. Deviations from this ideal measurement programs; however, the data acquisition time
occur because of limitations and compromises in manufac- is at least twice that of the more recent accelerometer
turing and because of installation circumstances. Instru- types. Cornforth (5.14) and Green (5.24) describe the per-
ments of different manufacture cannot generally be used in- formance of this type of inclinometer in more detail.
terchangeably. In the interest of accuracy, the interchange- The most sensitive transducer commercially available
able use of instruments (probes) of the same manufacture is the servo accelerometer. A servo accelerometer is com-
also should be avoided. posed of a pendulous "proof mass," that is free to swing
The function of the inclinometer is to detect the change within a magnetic field (Figure 5.9). The proof mass is
in. inclination of the casing from its original near-vertical provided with a coil or "torquer," which allows a linear
position. Readings taken at regular preestablished depths force to be applied to the proof mass in response to cur-
inside the casing allow the change in slope at various points rent passed through the coil. A special sensing unit, called
to be determined; integration of the slope changes between a pick-off, detects movement of the proof mass from a ver-
any two points yields the relative deflection between those tical position. A signal is then generated and converted by
points. Repeating such measurements periodically provides a restorer circuit, or servo, into a current through the coil,
data on the location, magnitude, direction, and rate of cas- which balances the proof mass in its original position. In
ing movement. The integration is normally performed from this manner, the current developed by the servo becomes
the bottom of the boring, since the bottom is assumed to an exact measure of the inertial force and thus of the trans-
be fixed in position and inclination (Figure 5.5). ducer's inclination. The proof mass consists of either a
All too often, confusing or unexplainable results appear pendulum with jewel-bearing support or a flexure unit that
because of instrumentation problems. These problems are operates on the cantilever principle. Jewel-bearing acceler-
common to all inclinometers; therefore, they are discussed ometers are fragile and subject to frictional interference at
in some detail so that, even without a detailed understand- their bearings. Flexure accelerometers generally offer the
ing of the inner workings of the instruments, one will be same precision and have greater durability. Accelerometer
able to recognize these anomalies, the limits of accuracy, systems have a sensitivity of approximately 18 s of arc in
and the meaning of the results. In simple terms, the person ranges from 30° to 90°, and they generally demonstrate a
closest to the job should be able to answer the basic ques- precision of ± 1.3 to 2.5 mm (± 0.05 to 0.10 in) over a 33-
tions: Is the slide active? How fast is it moving? How deep m (lOO-ft) casing (1:10 000). One excellent example of
is it? the precision of such measurements is shown in Figure
5.10. These data were taken with a servo-accelerometer
Inclinometer Sensors sensor by the U.K. Transport and Road Research Labora-
tory during the advancement of a tunnel (5.6).
Before a description is given of some of the sophisticated Temperature has only a negligible effect on readings
types of inclinometers, one device deserves mention. The taken with pendulum or accelerometer transducers. How-
borehole probe pipe, also called the slip indicator or poor ever, a marked variation of reading with temperature has
boy, is one of the crudest of the subsurface measuring been demonstrated for transducers using bonded resistance
devices. It typically consists of a 25-mm (1-in) diameter strain gauges (5.24), and errors resulting from temperature
semirigid plastic tube in a borehole. Metal rods of increas- changes, as well as variations in zero drift, have been re-
ing length are lowered inside the tubing in turn, and the ported for inclinometers using vibrating-wire transducers
rod length that is just unable to pass a given depth gives a (5.17). Even so, Burland and Moore (5.11) used a strain-
measure of the curvature of the tubing in the vicinity of gauge-based inclinometer to obtain some precise results
that point. Tonis and Bartlett (5.49) describe the use of for a diaphragm wall.
this technique for the location of slip surfaces in unstable A recently developed inclinometer that is also used in
slopes. This type of measurement can easily be performed a square casing (45-mm, 1.8-in, internal dimension) is de-
in the riser pipe of an observation well or open piezometer. scribed by Phillips and James (5.41). The unit has a canti-
If there are several shear planes, or if the shear zone is thick, levered pendulum device with a four-arm resistance gauge
a section of rod hung on a thin wire initially can be left at bridge. Over a 5° range, the repeatability under ideal con-
the bottom of the pipe; subsequently, the rod is pulled up ditions is reported to be ± 5 s of arc. However, the rela-
to detect the lower limit of movement. tively small range may render it impractical for use in a
121
Figure 5.9. Servo accelerometer for inclinometer SEALED HOUSING
measurements (5.13).
1 --------------------1
DC
POWER
TILT
ANGIE
FLEXURE OAMPING
NETWORK
&
PAGO LE
TORQUE
TOR
STO PS E
POSITION SENSOR
_______ -I - - - - - - - - - - —J
TRUE VERTICAL
3ROUNO 1 0
LEVEL r'
45
9-6
1-8
Ii
12
13
14
IS
-16
CROSS SECTION
PUP
- UI_jiL LIZ CL • U
landslide area where boreholes often have significant in- ing for the sensor unit, proper casing installation is of para-
clinations. mount importance.
It is important for the observation accuracy that the
Casing Installation original installation be as close to vertical as practicable.
The error in inclinometer surveys is proportional to the
Inclinometer casings generally may be installed for the full product of the casing inclination and angular changes in
depth normally encountered in landslides. Some known in- the sensor alignment. For inclined casings, angular changes
stallations are close to 300 m (1000 ft) deep (5.7). Since in sensor alignment on the order of 1° to 2° may produce
the measurement is referenced with respect to the bottom errors of several centimeters per 10 m (several inches per
of the casing, the casing bottom must be extended about 100 ft) of casing, as shown in Figure 5.11. Sensor-alignment
6 m (20 ft) or more into soil that will not undergo lateral change occurs from time to time because of one or several
displacement. If any doubt arises as to the stability of the factors, such as wheel play in the groove, wear of the sen-
casing bottom, movement of the casing top should be sor carriage (particularly wheel assemblies), internal change
checked by precise surveying methods. The accuracy of in the sensor itself, and change in the alignment between
the observations is generally limited not by the sensitivity sensor and carriage.
of the inclinometer, but by the requirement that successive The casing for the inclinometer is usually installed in
readings be made with the same orientation of the instru- 1.5 and 3.0-m (5 and I 0-ft) lengths, which are joined to-
ment at the same point in the casing. Because the casing gether with couplings and either riveted or cemented or
must provide a reliable orientation and a continuous track- both to ensure a firm connection (Figure 5.12). Each
122
coupling represents a possible source of leakage for grout grooves. During installation, the casing can become even
or mud, which can seep into the casing and be deposited more twisted or spiraled so that, at some depth, the casing
along the internal tracking system. For this reason, cou- grooves may not have the same orientation as at the ground
plings should be sealed with tape or glue. The bottom sec- surface. Spiraling as great as 18° in a 24-rn (80-ft) plastic
tion of the casing is closed with an aluminum, plastic, or casing has been reported (5.24); 1°/3-rn (10-ft) section is
wooden plug, which also should be sealed. If the drill hole not uncommon. Because significant errors in the assumed
is filled with water or drilling mud, the inclinometer casing direction of movement may result, deep inclinometer cas-
must be filled with water to overcome buoyancy. Some- ings should be checked after installation by using spiral in-
times, extra weight (sand bags, drill stem) may be needed. dicators available from the inclinometer manufacturer. For
The annular space between the boring wall and the in- a particular installation, groove spiraling is generally a sys-
clinometer casing may be backfilled with sand, pea gravel, tematic error occurring for each section in the same amount
or grout. The selection of backfill depends mostly on soil, and the same direction. In any event, spiraling, whether it
rock, and groundwater conditions (i.e., whether the bore- is due to manufacture or to torque of the casing during in-
hole is dry, wet, caving, stable without mud). The type of stallation, can be measured and is thus not really an "error"
drilling technique (e.g., rotary, hollow-stem, cased holes) is in inclinometer measurement. Spiraling is important only
also important. Poorly backfilled casings can introduce a when the true direction of the movement is determined and
scatter in initial measurements, and, if maximum precision at that time can and should be measured and the results ad-
is required, great care should be taken in the selection of justed accordingly.
the proper backfilling material. Grouting is generally pre- Small irregularities in the tracking surface of the casing
ferred, but may not always be possible, particularly in per- can lead to errors in observation, especially if careful repe-
vious geologic materials, such as talus. Grouting may be tition of the depth to each previous reading is not exer-
facilitated by use of a 20 to 25-mm (3% to I-in) diameter cised. If plastic casing has been stored in the sun before
plastic tube firmly attached to the casing bottom through installation, each section may be locally warped (opposite
which grout is pumped from the ground surface until the grooves not parallel) and large measurement errors can oc-
entire hole is filled. In a small clearance, drill-hole grout cur with minor variations at the depths at which the read-
can be tremied through drill rods via, a one-way valve at ings are taken (5.23).
the inside bottom of the casing. Aluminum casings should be used with some caution
The as-manufactured casing may have some spiral to the because severe corrosion may occur if the casing is exposed
0.06
0.05
11
0 ABS PLASTIC
0.04
- GROOVES TO ALIGN -
INCLINOMETER SENSOR
00.03
z
152 rn,n (6 in)
ALUMINUM CASING ALUMINUM
0.02 COUPLING
BUTT JOINT AT
CASING END
POP RIVETS
0.0) 305 ,nm (12 in) AT END OF
ALUMINUM COUPLING COUPLING
POINT OF CONTACT
FOR SETTLEMENT STANDARD
15 PROBE WHEN CASING
Jo• MAKING SETTLEMENT LENGTHS
50 ISO I 50 MEASUREMENTS 1.5 n, IS H) - ALUMINUM
CASING DRIFT (in) or3rn(lOft) CASING
NOTE CASING DRIFT OVER ANY DEPTH
INCREMENT IS MEASURED AS HORIZONTAL
C AS IN ALUMINUM
BOTTOM PLUG
-
DEVIATION FROM TRUE VERTICAL AXIS IN 'O.P RIVETS AT
POP RIVETS
TWO MUTUALLY PERPENDICULAR PLANES. END OF COUPLING
AT Y. P0 IN TS
DEFLECTION ERROR IN ONE PLANE IS RESULT
OF DRIFT IN OTHER PERPENDICULAR PLANE.
123
to alkaline soil, corrosive groundwater, or grout. Epoxy Figure 5.13. Installation and detail of multiposition in-place
coatings help to minimize this problem. Burland and Moore inclinometer.
(5.11) recommend cleaning the casing with a stiff brush be- 8.4mm (3.34 in) O.D. ABS PLASTIC
INCLINOMETER CASING JUNCTION
fore taking readings. Frequent flushing with water is also Box
helpful. The top of the casing is generally capped with a
tight-fitting plug to prevent intrusion of debris. In addition, ij
the inclinometer casing should be protected and padlocked
at the ground surface. When selecting protective casing or FIXED
00 SPRING
monument covers for use around the top, one should keep WHEEL ''-/il-LOADED ,
WHEEL
in mind the casing-mounted pulley or similar device.
TOP
ANCHOR . S;
In-Place memometers (DUMMY
HOUSING);,
fixed unit is removed for repairs, the overall accuracy will JOINT
--
be reduced to that of a portable unit or less. Also,acceler- SENSOR
ASSEMBLIE
TUBE 4'
ometers may have long-term drift. With portable units, these
and other effects are canceled by taking readings in oppo-
site directions. Fixed units can be monitored remotely or MIN 115-mm
..
(4.5-in)
connected to alarm systems, which are additional advan- BORING
tages. In some systems, a standard grooved inclinometer PORTLAND CEMENT
casing is used so that a portable inclinometer can be used GROUT BACKFILL—
/,.
to check readings of the fixed system.
Because of its accuracy, the fixed borehole inclinometer
GROOVED PLASTIC CASING
CAST IN PLACE OR GROUTED
', ''I
IN PLACE IN BOREHOLE.'.:
:11 I
t1'6
may be used to measure small movements in rock. Its adapt-
ability to various remote-monitoring systems allows for a
continuous record of displacement. As designed, most sys-
tems are not intended to obtain a continuous profile of de-
formation. Rather, they measure critical movements of a
few sections within the borehole inclinometer. This instru-
ment, therefore, is not necessarily a replacement for the
portable units, but rather has its own purpose and is of the accelerometer and multiplying by the gauge length
worthy of consideration for continuous monitoring of or spacing between sensors. This gives the relative displace-
movements during critical construction stages. ment of each sensor, and these relative displacements can
A recently developed in-place inclinometer (Figure be summed to determine the total displacement at each
5.13) employs a series of servo-accelerometer sensors; it sensor. Since the grooves in the casing may spiral, the
is available with both one- and two-axis sensors (5.13). spiral should be checked after the casing is installed so that
The sealed accelerometer packages are spaced along stan- the exact direction of displacement measurement is known.
dard grooved inclinometer casing by a series of rods. The Any number of sensors at any spacing can be used in one
rods and sensors are linked by universal joints so that they borehole. Maximum deflection range is ±300, and sensitiv-
can deflect freely as the soil and the casing move. The sen- ity is reported to be ± 0.01 mm in I m (± 0.001 in in 10 ft),
sors are aligned and secured in the casing by spring-loaded but this is not usually attained because of long-term drift.
fins or wheels, which fit the casing grooves. Use of the Monitor and alarm consoles and telemetry systems are
grooved casing and guide wheels allows removal of the in- available.
strument for maintenance, adjustments, and salvage. Since
the casing is standard inclinometer casing, portable inch- EXTENSOMETERS AND STRAIN
nometers can be used in the casing when the fixed instru- METERS
ment is removed.
Readings are obtained by determining the change in tilt Extensometers and strain meters measure the increase or
124
decrease in the length of a wire or rod connecting two be interconnected by the borehole so that the recorded
points that are anchored to the soil in the borehole and water level may be of little significance.
whose distance apart is approximately known. One com-
mercially available device is shown in Figure 5.14. When Piezometers
gauge lengths are on the order of a rneteror less, these de-
vices are often referred to as strain meters rather than as Pore pressures and groundwater levels in a slide area can be
extensometers. When they are used as extensometers, the measured by a variety of commercially available piezom-
accuracy and repeatability depend on the type of sensing eters (5.25). The selection of the best type for a particular
element and its range of travel and also on the type of con- installation involves several considerations.
necing wire or rod and the methods used to control the
tension. Dead weights are best for maintaining constant Piezometers are usually installed in a difficult en-
tension in wires; if these cannot be used, constant tension vironment and may become inoperative because of pinch-
springs are acceptable, although there may be some hyster- ing off of the tubes, blockage by air bubbles, electrical
esis. Relatively inexpensive wire-wound or conducting plas- short circuits, and malfunctions of buried moving parts.
tic linear potentiometers are often used as sensors, and rel- Therefore, reliability and durability are often of greater
atively simple, battery-operated Wheatstone bridge cir- importance than sensitivity and precision. For example,
cuits are used for manual readout. Sensitivity is on the in some cases it does not matter that the actual head may
order of 0.1 percent of the range of travel, but repeatabil- be in error by 0.3 in (1 ft) or so as a result of time lag,
ity and accuracy may be no better than 0.5 mm (0.02 in), provided the piezometer is functioning properly. If a mal-
depending on the type of anchor and connecting member; function occurs, the fact that the apparent head can be re-
this is usually sufficient, however. corded to a millimeter is of little importance.
When the devices are used as strain meters, the repeat- The basic problem with a piezometer is that the en-
ability and accuracy are essentially the same as the sensi- ergy required to operate it prevents the instrument from
tivity. Thus, for a grouted-in-place assembly 3 in (10 It) recording a pore-pressure change immediately. For a given
long with an invar rod and a range of 25 mm (1 in), unit pressure change, this energy is proportional to the volume
strains as low as 0.000 01 may be detected with relatively of pore water that must flow into the instrument. A pi-
inexpensive instrumentation. Horizontal stretching of em- ezometer with an open standpipe, for example, requires a
bankments has been observed by installing anchors at vari- much greater volume of water than one with a stiff dia-
ous positions at a given elevation and attaching horizontal phragm and therefore requires a correspondingly greater
wires to dead weights on the downstream lace; this was time to respond to a change of pore pressure. Hvorslev
done at Oroville Dam (5.55). Care is required to ensure (5.27) discusses time-lag effects for open-standpipe piezom-
that the wires (or rods) do not get pinched off if localized eters; Brooker and Lindberg (5.10) evaluated these effects
vertical shear movements occur. Other cases of application for closed types; and Penman (5.39) and Vaughan (5.52)
are'described by Dutro and Dickinson (5.19) and Heinz studied the response times of various types of piezometers.
(5.26). In slides in which piezometers are placed in boreholes, it is
usually possible to provide a large collecting volume of
PORE-PRESSURE AND porous material (sand) around the tip and thus to reduce
GROUNDWATER the time lag in open standpipes.
MEASUREMENTS Partially saturated soils (fills, soils above the water
table, and soils in which gas is generated from organic mat-
Observation Well ter) pose particular problems because the gas in the voids
The most common water-level recording technique, despite exists at'higherpressures than the water as a result of sur-
more sophisticated methods, is the observation of the water face tension (capillarity). Thus, one must know whether
level in an uncased borehole or observation well. A particu- the piezometer is reading the pore-air pressure or the pore-
lar disadvantage of this system is that a perched water table water pressure. To prevent the entry of air into the pi-
or artesian pressure can occur in specific strata that may ezometer tip and thus to ensure that pore-water pressure
FLANGED PIPE
r 2.5.mm (1-in) TUBING
13.mm (0.5-in) GROUT TUBES7
WIRES
EXTENSOMETER
TO REMOTE READOUT
125
is being measured, fine ceramic tips, with high air-entry Figure 5.15. Open-standpipe SCREW CAP OR
WOODEN PLUG
values, can be used (5.2). Even with a high air-entry value piezometer.
filter, air will slowly enter the piezometer cavity by diffu-
sion and, if reliable long-term readings are required, deair-
CEMENT/SAND
ing facilities must be provided (5.52). OR BENTONITE
4. Additional problems resulting from diffusion of either BACKFILL
vation well in which the elevation to which the water rises APPROX.
ln,LONG
is measured directly by means of a small probe. In this
case, the static head is the average head that exists over the 0.3. :..t Note: 1 n, = 3.31t;
1cm0.4in.
depth of the inflow part of the well below the water table.
This measured head may be higher or lower than the free
water table and, in the case of moderately impervious soils,
Figure 5.16. Casagrande
may be subject to a large time lag. Although the open borehole piezometer.
standpipe is not satisfactory in impervious soils because of
time lag or in partially saturated soils because the signifi-
cance of the measured head may be difficult to evaluate,
its simplicity, ruggedness, and overall reliability dictate its
use in many installations. 10.,,,,,,
PLAST I
BACKFILL
TUBING
Casagrande
126
Figure 5.17. Pneumatic piezometer. (a) PORTABLE PORE-PRESSURE INDICATOR
OUTPUT PRESSURE
GAUGE -
-
NYLON TUBING
(ORATING JIRE
ENSING TUBE
E
E
I APORAGN
127
vice consists of a tube placed in the drilled hole extending SYSTEMS FOR MONITORING
to the ground surface. The bottom of the tube should be ROCK NOISE
slotted or made of a porous material or a sufficiently fine
screen that it will restrict movement of soil into the piezom- Use of the rock-noise detection method for observing the
eter but at the same time permit free access of water. Only stability of soil and rock materials is a relatively new con-
that part of the tube that is actually within the stratum in cept, and most practical applications so far have been in
question should be pervious. When the tube is centered in mines (5.4, 5.5, 5.30, 5.60). More work has been done in
the drilled hole, a known volume of clean sand is placed basic research than in practical application of the technique,
around the piezometer; the purpose is to create a sand fil- although sufficient laboratory and field tests have been per-
ter between the soil and the piezometer tip. formed to demonstrate the importance and usefulness of
Construction of an impervious barrier above the piezom- the method. The ability to detect the occurrence of dis-
eter tip and sand pocket is essential. A well-established tress in a rock mass before the development of measurable
procedure is to drop balls of soft bentonite into the hole movement is obviously a significant technological advance.
and then tamp those balls around the piezometer tubes by Experimental work that has led to the development of
using an annular hammer. Prepared bentonite in ball or techniques and instrumentation for measuring microacous-
pellet form is now available commercially; it has a specific tic, microseismic, or subaudible rock noise has been car-
gravity that is sufficiently large to allow it to sink through ried on in the United States since the early 1940s (5.5).
the water in the piezometer hole so that a hammer is not All of this work, regardless of terminology, has been re-
necessary. Alternatively, a cement-bentonite grout can be lated to measuring transient noise disturbances in earth
tremied into the hole above the sand filter. Such a seal (soil and rock) materials for the purpose of establishing
can be pumped through a 1.3 to 1.9-cm (½ to 3/4 -in) diam- the relative stability and, in some instances, locating zones
eter pipe adjacent to the piezometer tube. Installing only of weakness in earth materials (5.29, 5.31, 5.42, 5.60). In
one, or at most two, piezometers in a borehole has gener- the mid-l960s, Goodman, Blake, and others (5.4, 5.22,
ally been found preferable because bentonite seals are 5.31) adapted the techniques developed by the U.S. Bureau
sometimes difficult to construct and may leak slightly de- of Mines to civil engineering applications and the study of
spite precautions. However, special-purpose, multiple- soil and rock slides. Particular emphasis was placed on the
point piezometers (with four tips) have been used success- determination of relative stability, location of failure planes,
fully by Vaughan (5.51) and merit consideration. determination of the epicentral location of specific distur-
Both pneumatic and electric piezometers can be sealed bances, and improvement of field equipment. Rock-noise
in a similar manner to an open-standpipe or a Casagrande detection systems can be divided into three basic elements:
piezometer in a borehole. Since the piezometer tip cannot sensors, signal-conditioning equipment, and recording and
be deaired after installation, it should be soaked in deaired data-acquisition equipment.
water beforehand and kept in the water until it is lowered
into the borehole. A low air-entry filter tip can be placed Sensors
in a saturated sand pocket, but a high air-entry tip should
be pushed into the soil beyond the base of the hole or sur- The frequency response of the sensors is a critical design
rounded by a porous grout, such as plaster of paris. This parameter. Rock noise is reported to occur over a broad
will ensure more rapid equalization, which may be of con- range of frequencies between about 50 and 10 000 Hz
siderable value in obtaining reliable pore-pressure measure- (5.4); however, disturbances at lower frequencies (20 to
ments soon after installation in low-permeability clay. Both 40 Hz) have been observed at Downie slide, British Colum-
electric and pneumatic piezometers should be checked for bia, and are believed to be important. The occurrence of
malfunction before and during installation, and particular rock noise at frequencies above 10 kHz has been observed
care is needed in driving electric piezometers to prevent in the laboratory when rock specimens are placed under
overpressure. high compressive loads (5.30). Because the frequency range
In soft and medium-stiff soils, it may be preferable to of noise phenomena at a particular site is unknown, the
push the piezometer directly into the ground. Flush- frequency response of the sensors must be sufficiently
coupled heavy steel water pipe will be required, and the broad to ensure that the system will respond properly to
piezometer tip must be robust enough to resist damage to any meaningful noises that occur. In addition, the sensors
the well point and should be designed to minimize distur- must be capable of (a) producing a high-level signal that is
bance around the tip (5.34). It may be of the Casagrande, proportional to the amplitude and frequency of the excit-
pneumatic, or, less commonly, electric type. This drive-in ing noise and is without spurious response and (b) transmit-
piezometer is self-sealing and rapidly installed and, if pneu- ting the signal through long electric cables. For this rea-
matic, has a rapid response. Recently, an improved method son, the output impedance of the sensor must be relatively
of sealing piezometers in boreholes was described by low to prevent the cable from acting as an antenna that
Vaughan (5.51). Instead of compacting bentonite balls could pick up radio, television, and ignition interference.
just above the sand filter, the borehole is completely
grouted with cement-bentonite grout. Even if the perme- Signal-Conditioning Equipment
ability of the grout is significantly higher than that of the
surrounding soil, little error will result because of the rela- Signal-conditioning equipment includes amplifiers and fil-
tively large grouted length. In many cases, a sand filter ters for the enhancement of the rock-noise signals and re-
need not be included, and the piezometer tip can be jection of unwanted noise or interference. It also includes
grouted directly with little error. the alteration of the received signal to whatever form is re-
128
quired to operate data acquisition equipment, usually ana- be actuated when the piezometric level exceeds a certain
logue or event pulse form. Rock-noise events are minute elevation.
bursts of energy; therefore, high amplification is required One problem with such warning systems is to determine
by the signal-conditioning equipment to make the signals in advance the boundary between tolerable and intolerable
useful. Further, the amplifiers must be of a type that gen- change. Should the extensometer be set for 0.25, 2.5, or
erates little internal noise and responds accurately to rock- 25 mm (0.01,0.1, or 1.0 in)? Should each instrument be
noise signals over the entire frequency range specified for monitored individually, or do only a few key instruments
the sensors. Sharp cutoff filtering for both high and low need such signals? Probably no computer or automatic
frequency must be incorporated into the signal-conditioning warning system will ever replace engineering judgment in
equipment to reject unwanted noise that might interfere the evaluation of data from field instrumentation systems.
with the rock-noise signals. With respect to advance warning, impending failure is nor-
mally signaled by a long-term change in rate rather than by
Recording and Data-Acquisition a short-term localized change. The data must be reviewed
Equipment periodically by a geotechnical engineer experienced both in
field measurements and in the design requirements of that
After rock-noise signals are received, amplified, and condi- particular project.
tioned, they must then be recorded or analyzed or both.
A rock-noise signal is generally a sharp burst of acoustic DATA ACQUISITION AND
energy that attenuates rapidly. It typically sounds like a EVALUATION
snap, click, or grind and, in analogue form, has the appear-
ance of the record of a miniature earthquake of short du- Data-Acquisition Methods
ration. To be able to record the signals in analogue form
is important in their identification as rock noise rather The data obtained from the instrumentation described con-
than as noise generated from other sources. Recording sist of a sequence of numbers that are converted by means
in analogue form is usually not necessary on a continuous of a calibration chart or other relation to length, volume
basis, but can be done only periodically to ascertain that change, pressure, or other parameter. These numbers, plus
interfering signals are not being received. other information, such as date, depth, or instrument num-
Data recorded in analogue form, which is the form in ber, must be recorded in the field in some form that can
which rock noise data are usually presented, are generally later be properly identified and retrieved for analysis.
inconvenient to use for routine analysis of noise rate. A There are several ways in which this can be accomplished.
convenient method of determining noise rate is to electron-
ically convert each rock noise burst to a pulse signal, which Manual Recording at Sensor Location
is counted mechanically or electronically on a unit-of-time
basis. The number of pulses or events per unit of time can Each individual sensor or cluster of sensors can be connected
then be recorded to document the time-rate history of to its own gauge or panel at a nearby location (Figure 5.20a).
rock-noise activity during a long period of time. Although The field crew carries a portable control bOx to that specific
time-rate data may need to be accumulated continuously location, connects it to the individual sensor, obtains and
in active areas, they are generally sampled during selected records the pertinent data, and forwards the data sheets to
intervals of time, such as several minutes or an hour, and the field office or home office for reduction and analysis.
sampled periodically rather than continuously. Ambient This procedure is time consuming on large jobs and requires
background rock-noise data from a specific area should be a substantial amount of labor, and the resulting data sheets
studied before a reasonable estimate of recording duration are voluminous and subject to errors in reading, recording,
and intervals can be recommended. Time-rate data can be and subsequent retrieval and analysis. However, the initial
accumulated and recorded by a variety of methods. The capital investment is low.
selection of the method depends on the cost and the re-
quired convenience to the user. Automatic Recording at Sensor Locations
AUTOMATIC WARNING AND Significant reduction in the time required for data acquisi-
ALARM SYSTEMS tion and subsequent savings in analysis of the data can be
effected by use of automatic recording devices that can be
Installing a slide-warning system may be desirable in some transported to the sensor location or a nearby panel. For
instances. Such systems vary from simple slide fences, example, inclinometer surveys require that the sensing tor-
sometimes used by railroads (5.9), to more complicated pedo be transported to the hole and lowered manually to
in-place inclinometers, extensometers, and piezometers. the desired depths. However, if the site is accessible by
Their purpose is to provide automatic warning in the event panel truck, the cable, reel, and associated readout equip-
of a sudden change that could be indicative of an impend- ment can be transported readily to the site and the data
ing earth movement. The mechanics of such systems are can be recorded on magnetic or punched tape in response
relatively simple to devise. An in-place inclinometer or ex- to manual command (Figure 5.20b).
tensometer can be used to actuate a red light in a central
location when the movement exceeds a certain threshold, Manual Recording at Central Terminal
which, for example, may occur along a specific shear plane
in soil or a bedding plane in rock. A similar alarm could If the sensors are of a type whose output signal can be
129
Figure 5.20. Data acquisition systems. several holes (5.26). Similar systems can readily be de-
vised to record, for example, the readings of individual
MANUAL READOUT AT SENSOR LOCATION
extensometers.
DATA SHEET I
READOUT BOX (MANUALLY I Inclinometer Observations
E~_ RECORDED)
130
ponent of the anticipated deformation as a positive change Figure 5.21. Sample field data sheet for transcribing inclinometer
For example, in an area suspected of landslide activity, the data to punched cards.
first set of readings is taken by orienting the fixed wheels
of the torpedo in the casing groove closest to the downhill INCLINOMETER FIELD DATA
position. For a complete survey, the torpedo is reversed I 2 3 4 5 6 1 8 9 IC 1112 13 II 151611 1619 2021 22 2324252621782030
1800 after the first set and the readings are repeated. The
i Ri 1E ii1•i 1' , A1v1,i 1C11•1
algebraic difference of the two sets of readings is used to I i
1. JOB TITLE
compute the profile of the casing or, more important, to
determine any change compared to other surveys; using
the algebraic difference is, in effect, the same as using the i ,s-'ti i
131
processing are mainly useful in reducing the time and labor Figure 5.22. Schematic diagram of magnetic tape recording of
involved in office computation of the data. inclinometer data.
Automatic data recording contributes to the complexity
of the measurement operation and introduces an additional
source of error. With manual data recording, the technician FREIDUENCY
IITEN
can scan the data for face errors and make corrections or
reread the.casing on the spot. Most data-recording systems
do not allow this advantage; the data must be scanned for
errors after being printed out in the office and before com-
PORTABLE
GENERATOR H VOLTAGE
CONTROL
132
Figure 5.23. Geologic profile and inclinometer observations EXAMPLES
at failure plane of landslide on 1-94, Minneapolis (5.57).
249
This section briefly describes several projects that were suc-
cessfully instrumented to measure pore pressures and to
detect the depth and rate of movement before corrective
treatment was designed. The emphasis is on selection of
instrumentation and analysis of data rather than on correc-
tive treatment.
244-
Minneapolis Freeway
In 1967, a landslide developed along a 335-rn (1 100-ft) sec-
0 __1 tion of 1-94 on the east side of Minneapolis (5.56, 5.57).
Ls The movements were evidenced by upheaval of the sub-
243-
SEAN
Ls grade and by cracking and slumping of the cut slope. Al-
NEASUREO
POINTS though the area had been investigated before construction,
iLINES70NE no instability problems had been anticipated. Therefore,
242_ determining the depth of movement was necessary before
investigating the cause of the movements could begin. Fif-
LINESTONE(Ls) teen borings were drilled to serve the dual purpose of pro-
viding subsurface soil information and holes for the instal-
241 __ lation of field instrumentation. Two sections in the vicinity
' 8ENTONIT(
SEAl 0 10 20 3Q of the slide were instrumented rather extensively. To mea-
- ft
' DEFLECTION torn)
Note: 1 to - 3.3 It; 1 tore = 0.04 in. sure the depth and rate of movement, 10 inclinometers
were installed, and to evaluate groundwater conditions, 7
Casagrande piezometers were installed.
Figure 5.24. Section showing failure plane and movement Several of the inclinometers detected movement within
distribution of landslide on 1-94 (5.57).
a few days, and within a month it was apparent that slope
failure was occurring as a result of sliding along a near-
212 0 212 E horizontal plane approximately at elevation 243 m (798
_'_
710] 2p— ft). Figure 5.23 shows the geologic profile and the de-
$6 216 tailed movements recorded by one of the inclinometers,
and Figure 5.24 shows the distribution of those movements
760 - NIILGA
along a cross section and the location of the failure plane.
GLACIAL DRIFT The piezometric data obtained during 2 months of observa-
tion showed that there were several perched water tables,
250. ()
IN but none with high uplift pressures. Water levels also were
11101
SNALE 11111 LIut.1T observed in the 10 inclinometer casings: However, this is
STRING(R FAILURE SURFACE
not always a reliable means of groundwater observation be-
_._. _. _. ._._•_• _.
cause grout may tend to seal water either in or out of the
aluminum well casing.
lION
JAR The data from the inclinometers demonstrated conclu-
sively that slope failure was occurring along a near-
Note: 1 n, 3.31t; 1 roe,- 0.04 in 18mm horizontal plane only 3.3 to 4.5 m (10 to 15 ft) below
the bottom of the subcut elevation, and, since the borings
had failed to detect any unusual material at this depth, a
the ground itself will expand and contract as its temperature decision was made to undertake a test-pit program to ob-
varies. At a hillside stability project in Montana, rock out- serve the materials along the failure surface. Eight test
crops were found to expand and contract seasonally by as pits were dug, varying in depth from 3.3 to 5.5 m (10 to
much as 1.5 mm (0.06 in). Electric lead wires also change 18 ft). During test-pit inspection, the soil within the fail-
their electric resistance with temperature and will cause ure zone could be observed, sampled, and tested. The slide
erroneous readings unless such a change is properly taken was found to be taking place on a thin seam of potassium
into account. bentonite, averaging less than 2.5 cm (1 in) in thickness;
this material had been removed by the wash water during
Piezometers sampling and therefore had not been detected previously.
Stability computations were conducted on three repre-
Piezometric heads should be plotted on time graphs show- sentative cross sections where the limits of sliding were
ing rainfall and other data that may influence the pore pres- clearly defined by inclinometer data and by field observa-
sure. If drainage has been installed, the quantity of seepage tion of the heaving and cracking. In addition, the geology
should also be recorded and plotted. If possible, the re- was more completely mapped at those locations. Since
sponse of each piezometer should be checked periodically the movement developed along a horizontal plane surface,
to determine its recovery rate. a sliding wedge analysis was used. Hydrostatic pressures,
133
based on available piezometric data, were assumed. After a massive H-beam was inserted to provide the necessary
extensive analysis and review, the remedial scheme finally structural strength and resistance to bending. Next, a cur-
recommended and adopted was reinforced concrete but- tain wall was hung on the exterior of the piles to provide
tresses cast in narrow slit trenches excavated normal to the a finished surface. These cylinder pile walls, which act as
roadway center line (5.57). cantilever beams, were designed to limit the maximum de-
flection of the top of the wall to approximately 5 cm (2 in).
Potrero Tunnel Movements Figure 5.26 shows the complexity of modern freeway
construction in urban development. After the material
The Potrero Tunnel of the Southern Pacific Railroad was was partially removed, as shown in the upper portion,
constructed in 1906 and underlies a ridge that crosses the movement occurred in the lower slope and cracks were de-
only rail entrance into San Francisco (5.46, 5.56). Excava- tected in the apartment house. Inclinometers were then in-
tion for a freeway above the tunnel in 1967 initiated move- stalled, as shown in the lower portion. Most of the inch-
ments that cracked the lining and caused the two walls of nometers detected horizontal offsets in various layers well
the tunnel to move closer together. However, visual obser- below the bottom of the excavation. To stabilize this sec-
vation and surface measurements did not indicate clearly tion and complete the project required the installation of
(a) whether a massive landslide was involved and (b) what cylinder piles below the lower retaining wall, which had
the relation was between freeway construction and tunnel already been completed, and two rows of cylinder piles
movements. Those questions had to be answered in order below the upper retaining wall, which also had been com-
to resolve legal responsibilities and to design and construct pleted. The cylinder piles penetrated well below the move-
remedial measures. The following types of instruments ment zones. The effectiveness of the cylinder piles in stop-
and measuring systems were installed: ping the movements is shown by the detailed records of
movement of inclinometer 5 in Figure 5.27. During Janu-
Inclinometers to measure horizontal ground move- ary and February 1963, the two rows of cylinder piles were
ments, installed and had an immediate influence in slowing down
Extensometers to measure ground extension adjacent the rate of movement of inclinometer 5. After completion
to the tunnel, of the pile installation, the movement stopped and the sec-
Portable extensometers to measure closing of the tun- tion has since been stable (5.37).
nel sides and heave of the roof, At the Tukwila interchange south of Seattle, inclinom-
Bench marks inside the tunnel to measure changes in eters were installed in a completed wall to verify its perfor-
alignment and elevation of the track, and mance after construction. Similar installations elsewhere
Bench marks to measure surface movement. showed that, in general, wall deflections seldom exceeded
the design criteria of 5 cm (2 in). In this wall, however,
The distribution and magnitude of the ground movements small but gradually increasing horizontal movements were
at a typical section are shown in Figure 5.25. Analysis of
the data indicated that the ground downhill from the tun-
nel was stable and that the movements uphill were evenly Figure 5.25. Movements at Potrero Tunnel, San Francisco (5.56).
distributed over a 9-rn (30-ft) thick zone. The increased
lateral earth pressure against the sides of the tunnel re-
sulted from the freeway construction and was the cause 40 -
of the tunnel distress. Corrective treatment consisted of ORIGINAl GROUND
placing two rows of heavily reinforced cast-in-place con- 35
ISrn,,,/n,o. I 15 r,,n,/n,o.I
-
crete piles alongside the tunnel with a connecting strut Si RESULIANI NORENENI
—
above the tunnel; these were installed from the subgrade JO I — I TIED- ANON
REUININC STRUCTURE
I
elevation of the freeway. The same instruments that were
\
used to determine the displacements were also read during 75
ZONE)OF EXCAVATION
// 7 WAIL/ANCOR (1961)
and after construction to verify the effectiveness of the re- I \ /
medial measurements. / \ /
10 / SUNGRAGE-
1 >
/
/ \
Seattle Freeway I
I ,/
\ \_- - I5/o.,'
I I
R I /
In early 1960, a freeway constructed through downtown I I
'
lA
Seattle required the excavation of cuts into a gently sloping
sidehill. A discovery was made at an early stage that even
relatively shallow cuts, extending some distance along the nt
'I'M
hillside, could initiate serious ground movements. Exten- / TUNNEL
sive measurements, obtained primarily from inclinometers, 0 I FRANCISCAN /
(1906)
demonstrated that the slides were progressive (5.56). After I FORMATION
detailed study of the problem, the Washington State High- -5 ISHALES, SAHOSTONU.
SERPENTINE. RADII FRACTUREDI
way Department developed the concept of cylinder-pile re-
taining walls, which consist of large-diameter concrete cais- -ID
sons cast in prebored, closely spaced holes placed uphill
NOTe: 1 re = 3.3 IT; I 100, = 0.04 in.
from the proposed cut before excavation. In each caisson,
134
Figure 5.26. Seattle freeway problem (5.56).
bRICK APT,
I NOOSE
(a) CONSTRUCTION STAGES
ORIGINAL GROUND SURFACE -.._ I _L
_) -iRACK
60 -- -
CRACK
DEC. 62
DEC. 02
CCR AN.
Z
01
I /
/
,b oct.62
L50
/ DEC 62
MOVEMENT IAN. 63
5O
OACOFFILL
0I
<I © 33mm
MOVEMENT ZONE -
1 Lj
1TH MOVEMENT ZONES
1111
t.JI Li
CYLINDER
CYL NDEI
I3O D PILES
Notes: 1 m 3.3 It; f mm 0.04 in. =
Circled numbers refer to inclinometer locations.
135
Figure 5.28. Section through Fort Benton slide. Montana (5.58).
SLOPE BEBRIS
INCLINOMETER S—I flUE
20
APPRUX. 11111 OF
OBSERVABLESURFACE
INCLINOMETER S—S INCLINOMETER S—U CRACBS BEHIND TRACUS
0
E TRACUS
LIMIT OF HUIMOCRY 2650
AREA AT TOE OF SLOPE GROUNDWATER SURFACE 2
0
I-.
00 COVIEMOED BUTTRESS
\ IGRABEN/ ,"
>
190
-- - - --- 2600
U
z
0
.5
CL 1.1
SURFACE OF HARO 0
CLAY I
SHALE I-
0
C___
Figure 5.30. Movement of top of inclinometer S-4, Fort Benton slide (5.58).
1.0 .1.8
6.0
- ABSOLUTE VAL(S FTER 6/I /18 ABE
WIHELIABLESIRCE HV(MEHTS CAASEB -
5.2
U
• z
I-
2
1 5.0
LOOSENING OF UPPER SECTION OF
CASING
p.1
, s I-
>
2
0 I / .0
>
4.0 10.2
z
F- AFTER 5,1 5/10 SLOPE
—j
0
N
3.0
INDICATOR INSRUMENT
MOULD NUT PHSS BEND I .
I I i — MO EMENT DETERNIHED FROM
0
6
I TR NSIT SIGHT LINE
.5
2.0 - --- ---- - -,--
IF INSTALLED OR 5/15/1
MOVEMENT DETERMINED
FNOI SLOPE INOICATOR
-\ 14
1.0
REHOINGS -.
ou -__ I ii I I 11I I II II II II fi II
DECEMBER JANUARY FEBRUARY NARCA APRIL NAY JUNE 1010 AUGUST SEPTEMBER -
136
records of the magnitude and direction had been kept, the 5.7 British Columbia Hydro and Power Authority. Indlinom-
eter Data, Downie Slide. British Columbia Hydro and
predominant track movement had been vertical and accu-
Power Authority, Vancouver, job file, 1976.
mulated settlement during the years amounted to nearly a 5.8 British Geotechnical Society. Field Instrumentation in
meter. In addition, the movement rate varied seasonally, Geotechnical Engineering. Wiley, New York, 1974,
being greatest in spring and summer and least during late 720
fall and winter. By mid-February 1970 (installation was 5.9 Broms, B. B. Landslides. In Foundation Engineering
Handbook (Winterkorn, H. F., and Fang, H. Y., eds.),
in November 1969), the inclinometer at the crest of the Van Nostrand-Reinhold, New York, 1975, pp. 373401.
slope indicated about 1.5 cm (0.5 in) of lateral (southward) 5.10 Brooker, E. W., and Lindberg, D. A. Field Measurement
movement; however, no change was recorded at the toe of of Pore Pressure in High Plasticity Soils. In Engineering
the slope during this 3-month period. Movement of md- Effects of Moisture Changes in Soils (Aitchison, G. D.,
ed.), Proc., International Research and Engineering Con-
nometer S-4 continued, accelerating somewhat in early May ference on Expansive Clay Soils, Texas A&M Univ.,
and decreasing in late July and August. Data from this College Station, 1965, pp. 57-68.
same installation showed that movement was occurring at 5.11 Burland, J. B., and Moore, J. F. A. The Measurement of
a depth of 18 m (59 ft) and, because the bend in the casing Ground Displacement Around Deep Excavations. In
was relatively abrupt, was occurring within a relatively thin Field Instrumentation in Geotechnical Engineering
(British Geotechnical Society), Wiley, New York, 1974,
zone, probably only a few centimeters thick. Although re- pp. 70-84.
liable data were being obtained from the installation at the 5.12 Cooling, L. F. Second Rankine Lecture: Field Measure-
top of the embankment, no changes were recorded at the ments in Soil Mechanics. Geotechnique, Vol. 12, No. 2,
toe of the slope, and it became apparent that the zone of 1962, pp. 75-104.
movement beneath the toe of the slope was passing below 5.13 Cording, E. J., Hendron, A. J., Jr., Hansmire, W. H.,
Mahar, J. W., MacPherson, H. H., Jones, R. A., and
the 9-rn (30-ft) depth of the casing. Hence, a second, deeper O'Rourk, T. D. Methods for Geotechnical Observations
casing (S-6) was installed, and Instrumentation in Tunneling. Department of Civil
Soon after the additional installation was made, the data Engineering, Univ. of illinois, Urbana-Champaign; Na-
showed definite displacement at the toe of the slope; move- tional Science Foundation, Vol. 2, 1975, pp. 293-566.
5.14 Cornforth, D. H. Performance Characteristics of the Slope
ment was occurring within a relatively narrow zone some Indicator Series 200-B Inclinometer. In Field Instrumen-
11 m (37 ft) below the ground surface. These data are tation in Geotechnical Engineering (British Geotechnical
shown in Figures 5.29 and 5.30. The zone of movement Society), Wiley, New York, 1974, pp. 126-1 35.
correlated with the surface of the hard clay-shale layer that 5.15 Dallaire, E. E. Electronic Distance Measuring Revolution
Well Under Way. Civil Engineering, Vol. 44, No. 10,
was encountered at the same depth. Once the failure
1974, pp. 66-7 1.
mechanism had been established from the instrumentation
5.16 Dibiagio, E. Discussion: Session 2—Equipment. In
data, the design of corrective measures could be started. Field Instrumentation in Geotechnical Engineering
Buttressing was selected as the most practicable means of (British Geotechnical Society), Wiley, New York, 1974,
stabilizing the hillside. (Figure 5.28). Had the field data pp. 565-566.
revealed movement was occurring within a cohesive zone 5.17 Dunnidiff, C. J. Equipment for Field Deformation Mea-
surements. Proc., 4th Pan-American Conference on Soil
(e.g., through a thin seam of bentonite clay), some other
Mechanics and Foundation Engineering, San Juan, Ameri-
approach to stabilization would have been necessary. can Society of Civil Engineers, New York, Vol. 2, 1971,
pp. 319-332.
REFERENCES 5.18 Durr, D. L. An Embankment Saved by Instrumentation.
Transportation Research Board, Transportation Research
Record 482, 1974, pp. 43-50.
5.1 American Society of Civil Engineers. Proc., Specialty Dutro, H. B., and Dickinson, R. 0. Slope Instrumenta-
5.19
Conference on Performance of Earth and Earth- lion Using Multiple-Position Borehole Extensometers.
Supported Structures, Purdue Univ., Lafayette, Ind., Transportation Research Board, Transportation Research
American Society of Civil Engineers, New York, Vol. 1,
Record 482, 1974, pp. 9-17.
Part 1, 883 pp., Part 2, 1555 pp.; Vol. 2, 154 pp.; Vol.
5.20 Eckel, E. B., ed. Landslides and Engineering Practice.
3, 385 pp., 1972.
Bishop, A. W., Kennard, M. F., and Penman, A. D. M. Highway Research Board, Special Rept. 29, 1958, 232 pp.
5.2
Pore-Pressure Observations at Selset Dam. In Pore Pres- 5.21 Franklin, J. A., and Denton, P. E. The Monitoring of
sure and Suction in Soils, Butterworths, London, 1961, Rock Slopes. Quarterly Journal of Engineering Geology,
pp. 91-1 02. Vol. 6, No. 3, 1973, pp. 259-286.
Bishop, A. W., Kennard, M. F., and Vaughan, P. R. De- 5.22 Goodman, R. E., and Blake, W. Rock Noise in Landslides
5.3
velopments in the Measurement and Interpretation of and Slope Failures. Highway Research Board, Highway
Pore Pressure in Earth Dams. Proc., 8th International Research Record 119, 1966, pp. 50-60.
Congress on Large Dams, Edinburgh, Vol. 2, 1964, pp. 5.23 Gould, J. P., and Dunnidiff, C. J. Accuracy of Field De-
formation Measurements. Proc., 4th Pan-American Con-
47-7 2.
Blake, W., and Leighton, F. Recent Developments and ference on Soil Mechanics and Foundation Engineering,
5.4
Applications of the Microseismic Method in Deep Mines. San Juan, American Society of Civil Engineers, New
In Rock Mechanics: Theory and Practice, Proc., 11th York, Vol. 1, 1971, pp. 313-366.
Symposium on Rock Mechanics, American Institute of 5.24 Green, G. E. Principles and Performance of Two Indi-
Mining, Metallurgy and Petroleum Engineers, New York, nometers for Measuring Horizontal Ground Movements.
In Field Instrumentation in Geotechnical Engineering
1970, pp. 429-443.
Blake, W., Leighton, F., and Duvall, W. I. Techniques (British Geotechnical Society), Wiley, New York, 1974,
5.5
for Monitoring the Behavior of Rock Structures. U.S. pp. 166-179.
Bureau of Mines, Bulletin 665, 1974, 65 pp. 5.25 Hanna, T. H. Foundation Instrumentation. Trans Tech
Boden, J. B., and McCauI, D. Measurements of Ground Publications, Clausthal, Germany, Series on Rock and Soil
5.6
Movements During a Bentonite Tunneling Experiment. Mechanics, Vol. 1, No. 3, 1973, 372 pp.
U.K. Transport and Road Research Laboratory, Crow- 5.26 Heinz, R. A. In Situ Soils Measuring Devices. Civil En-
thorne, Berkshire, England, Rept. 653, 1974, 8 pp. gineering, Vol. 45, No. 10, 1975, pp. 62-65.
137
5.27 Hvorslev, M. J. Time Lag and Soil Permeability in ing of Soft Ground Rapid Transit Tunnels. Urban Mass
Groundwater Observations. U.S. Army Engineer Water- Transportation Administration, U.S. Department of
ways Experiment Station, Vicksburg, Miss., Bulletin 36, Transportation, Vols. 1 and 2, 1974; NTIS, Springfield,
1951,50 pp. Va., PB 241 536.
5.28 Kern and Company, Ltd. The Mekometer, Kern ME3000, 5.45 Shannon, W. L., Wilson, S. D., and Meese, R. H. Field
Electro-Optical Precision Distance Meter. Kern and Co., Problems: Field Measurements. In Foundation Engineer-
Ltd., Aarau, Switzerland, Technical Information 10, ing (Leonards, G. A., ed.), McGraw-Hill, New York, 1962,
1974,4 pp. pp. 1025-1080.
5.29 McCauley, M. L. Microsonic Detection of Landslides. 5.46 Smith, T. W., and Forsyth, R. A. Potrero Hill Slide and
Transportation Research Board, Transportation Research Correction. Journal of Soil Mechanics and Foundations
Record 581, 1976, pp. 25-30. Division, American Society of Civil Engineers, New York,
5.30 Malone, A. W. Elastic Wave Measurement in Rock Engi- Vol. 97, No. SM3, 1971, pp. 54 1-564.
neering. Imperial College, Univ. of London, PhD thesis, 5.47 Terzaghi, K. Mechanism of Landslides. In Application of
1968, 299 pp. Geology to Engineering Practice (Paige, S., ed.), Geological
5.31 Mearns, R., and Hoover, T. Subaudible Rock Noise Society of America, Berkey Vol., 1950, pp. 83-1 23.
(SARN) as a Measure of Slope Stability. Transportation 5.48 Tice, J. A., and Sams, C. E. Experiences With Landslide
Laboratory, California Department of Transportation, Instrumentation in the Southeast. Transportation Re-
Sacramento, Final Rept. CA-DOT-TL-25 37-1-73-24, search Board, Transportation Research Record 482,
1973, 16 pp. 1974, pp. 18-29.
5.32 Merriam, R. Portuguese Bend Landslide, Palos Verdes 5.49 Toms, A. H., and Bartlett, D. L. Applications of Soil
Hills, California. Journal of Geology, Vol. 68, No. 2, Mechanics in the Design of Stabilizing Works for Embank-
1960, pp. 140-153. ments, Cuttings and Track Formations. Proc., Institution
5.33 Mexico Ministry of Public Works. Failures at the Tijuana- of Civil Engineers, London, Vol. 21, 1962, pp. 705-711.
Ensenada Highway. Mexico Ministry of Public Works, 5.50 Transportation Research Board. Landslide Instrumenta-
Mexico City, 1975, 220 pp. tion. TRB, Transportation Research Record 482, 1974,
5.34 Mikkelsen, P. E., and Bestwick, L. K. Instrumentation 51 pp.
and Performance: Urban Arterial Embankments on Soft 5.51 Vaughan, P. R. A Note on Sealing Piezometers in Bore-
Foundation Soil. Proc., 14th Symposium on Engineer- holes. Geotechnique, Vol. 19, No. 3, 1969, pp. 405-413.
ing Geology and Soils Engineering, Boise, Idaho, Idaho 5.52 Vaughan, P. R. The Measurement of Pore Pressures With
Department of Highways, Univ. of Idaho, and Idaho Piezometers. InField Instrumentation in Geotechnical
State Univ., 1976, pp. 1-18. Engineering (British Geotechnical Society), Wiley, New
5.35 Moore, J. F. A. The Photogrammetric Measurement of York, 1974, pp. 411-422.
Constructional Displacements of a Rockfill Dam. 5.53 Vaughan, P. R., and Walbancke, H. J. Pore Pressure
Photogrammetric Record, Vol. 7, No. 42, 1973, pp. Changes and the Delayed Failure of Cutting Slopes in
6 28-64 8. Overconsolidated Clay. Geotechnique, Vol. 23, No. 4,
5.36 Mu?ioz, A., Jr., and Gano, D. The Role of Field Instru- 1973, pp. 531-539.
mentation in Correction of the "Fountain Slide." Trans- 5.54 Wilson, S. D. The Use of Slope Measuring Devices to De-
portation Research Board, Transportation Research Rec- termine Movements in Earth Masses. In Field Testing of
ord 482, 1974, pp. 1-8. Soils, American Society for Testing and Materials, Phila-
5.37 Palladino, D. J., and Peck, R. B. Slope Failures in an delphia, Special Technical PubI. 322, 1962, pp. 187-197.
Overconsolidated Clay, Seattle, Washington. Geotech- 5.55 Wilson, S. D. Investigation of Embankment Performance.
nique, Vol. 22, No. 4, 1972, pp. 563-595. Journal of Soil Mechanics and Foundations Division,
5.38 Peck, R. B. Observation and Instrumentation: Some American Society of Civil Engineers, New York, Vol. 93,
Elementary Considerations. Highway Focus, Vol. 4, No. SM4, 1967, pp. 135-156.
No. 2, 1972, pp. 1-5. 5.56 Wilson, S. D. Observational Data on Ground Movements
5.39 Penman, A. D. M. A Study of the Response Time of Related to Slope Instability. Journal of Soil Mechanics
Various Types of Piezometer. In Pore Pressure and Suc- and Foundations Division, American Society of Civil En-
tion in Soils, Butterworths, London, 1961, pp. 53-58. gineers, New York, Vol. 96, No. SMS, 1970, pp. 1521-
5.40 Penman, A. D. M., and Charles, J. A. Measuring Move- 1544.
ments of Embankment Dams. In Field Instrumentation 5.57 Wilson, S. D. Landslide Instrumentation for the Minne-
in Geotechnical Engineering (British Geotechnical Society), apolis Freeway. Transportation'Research Board, Trans-
Wiley, New York, 1974, pp. 341-358. portation Research Record 482, 1974, pp. 30-42.
5.41 Phillips, S. H. E., and James, E. L. An Inclinometer for 5.58 Wilson, S. D., and Hilts, D. E. Application of Instrumen-
Measuring the Deformation of Buried Structures With tation to Highway Stability Problems. Proc., Joint
Reference to Multitied Diaphragm Walls. In Field Instru- ASCE-ASME National Transportation Engineering Meet-
mentation in Geotechnical Engineering (British Geotech- ing, Seattle, American Society of Civil Engineers, New
nical Society), Wiley, New York, 1974, pp. 359-369. York, and American Society of Mechanical Engineers,
5.42 Piteau, D. R., Mylrea, F. H., and Blown, J. G. The Downie New York, 1971.
Slide, Columbia River, British Columbia. In Natural 5.59 Wilson, S. D., and Squier, R. Earth and Rockfill Dams.
Phenomena: Developments in Geotechnical Engineering— 7th International Conference on Soil Mechanics and
Rockslides and Avalanches (Voight, B., ed.), Elsevier, New Foundation Engineering, Mexico City, State-of-the-Art
York, Vol. 1, 1978. Vol., 1969, pp. 137-223.
5.43 St. John, C. M., and Thomas, T. L. The N.P.L. Mekometer 5.60 Wisecarver, D. W., Merrill, R. H., and Stateham, R. M.
and Its Application in Mine Surveying and Rock Mechan- The Microseismic Technique Applied to Slope Stability.
ics. Trans., Institution of Mining and Metallurgy, London, Trans., Society of Mining Engineers, American Institute
Vol. 79, Sec. A, 1970, pp. A31-A36. of Mining, Metallurgical and Petroleum Engineers, New
5.44 Schmidt, B., and Dunnicliff, J. C. Construction Monitor- York, Vol. 244, 1969, pp. 378-385.
138
Chapter 6
Strength
Properties and
Their
measurement
Tien H. Wu and Dwight A. Sangrey
The methods of limiting equilibrium are frequently used to tion to landslides has been limited (see later section on com-
analyze the stability of a soil or a rock mass (Chapter 7). In mon states of stress and stress change).
such analyses, the shear strength of the material is assumed
to be fully developed along the slip surface at failure. This Effective Stress Versus Total Stress
chapter outlines the basic principles that goyern the shear Analysis
strength and the methods that may be used for its measure-
men t. Since the shear strength of soils and rocks is strongly influ-
enced by the drainage conditions during loading, those con-
ditions must be properly accounted for in the use of shear
GENERAL PRINCIPLES strength. A fundamental principle in soil engineering is the
use of effective stress (a'), which is defined as
Mohr-Coulomb Failure Criterion
a'a - u [6.3]
The Mohr-Coulomb criterion is most widely used to define
failure; it states that the shear strength (s) is where
sc+atan4 [6.11 a = total stress, and
u = pore pressure.
where
The shear strength can be expressed consistently in terms of
a = normal stress on slip surface, effective stress, or
c = cohesion, and
= angle of internal friction. sc'+a' tan'c'+(a-u)tan [6.4]
In terms of principal stresses, the Mohr-Coulomb criterion where c' and 0' are the strength parameters for effective
becomes stress. The use of the effective stress parameters requires
that the pore pressure be known so that a' may be evaluated.
0 1 =a3 tan2 [(/4)+(/2)] In general, pore pressure consists of the hydrostatic pore
+ 2c tan [(7r/4) + (/2)] [6.2] pressure related to groundwater level and the excess pore
pressure due to applied loads. When soils are loaded under
where undrained or partially drained conditions, the tendency to
change volume results in pore-pressure change. The excess
a1 = major principal stress, and pore pressure may be either positive or negative, depending
on the type of soil and the stresses involved. Under the
03 = minor principal stress.
fully drained, long-term condition, the excess pore pressure
Other failure criteria, particularly the modified Tresca and is zero, and pore pressure due to groundwater flow can usu-
Von Mises, are sometimes used for soils, but their applica- ally be evaluated without serious difficulty. Hence, analysis
139
with the effective stress description of shear strength (Equa- Figure 6.1. Common states of stress in soil.
tion 6.4) is most useful.
(a) INITIAL ATREST STATE )b) BENEATH A LOADED AREA
For partially drained and undrained conditions, the evalu- P
140
0
f
Figure 6.3. Typical stress-strain cUrves for soils.
A
Figure 6.4. Simple test methods for determining soil strength.
(a) UNCONFINED
COMPRESSION
(b) CONE Ic) VANE SHEAR
plan
profile
60
increases as the rate of loading increases; however, this ef- is obtained by equating the torque measured at failure to the
fect depends on the specific material and varies over a wide moment produced by theshear stresses along the cylindrical
range (6.22, 6. 73). surface. According to Cadling and Odenstad (6.21), the
shear strength for vanes with a diameter-to-height ratio of
LABORATORY MEASUREMENT 1:2 is
OF SHEAR STRENGTH
s = (6/7) (M/7rD3 ) [ 6.6]
A variety of methods is available for laboratory measure-
ment of shear strength. The simple methods are designed to where
determine the shear strength of a sample in a particular con-
dition, such as the water content or void ratio of the soil in M = torque, and
situ. These methods are most often used to determine the D = diameter of the vane.
undrained shear strength (se) of saturated cohesive soils.
More elaborate laboratory tests are able to establish the shear In the application of the results of these simple tests to
strength relation defined by Equation 6.4. These methods the analysis of.slopes, consideration should be given to the
allow combinations of normal and shear stresses to be em- type of soil and loading conditions in situ. The application
ployed and pore pressures to be measured or controlled. The of these test results is commonly limited to saturated co-
more elaborate tests allow more accurate simulation of the hesive soils under undrained conditions. The results are all
field stress or deformation conditions. For example, triaxial expressed in terms of total stress because pore pressures are
compression tests simulat Rankine's active state, and tri- not measured. When the soil is brought to failure rapidly
axial extension simulates Rankine's passive state. The plane- under undrained conditions, the shear strength is defined by
strain and simple shear tests may be used to provide a better 5 = s. If the tests are run slowly or if the soil drains during
simulation of the actual deformation conditions in a slope. shear, the results are generally not applicable.
It is usually assumed that the measured strength is equal
Simple Tests to the in situ strength; however, a major uncertainty is the
effect of sampling disturbance on strength. Several studies
Three types of simple tests are discussed below. (6.15, 6.51) show that even "good" samples may suffer
strength losses as great as 50 percent. The effect of sample
The unconfined compression test is usually performed disturbance is most severe in soft sensitive soils and appears
on a cylindrical sample with a diameter-to-length ratio of 1:2 to become more significant as depth of the sample increases.
or slightly more. The sample is compressed axially (Figure Other factors to be considered include the state of stress and
6.4a) until failure occurs; the shear strength is taken as one- deformation. The directions of the principal stresses and the
half the compressive strength. orientations of failure surfaces in each of these tests are not
In the cone test, a cone with an angle of U is forced the same. They may also be quite different from the direc-
into the soil (Figure 6.4b) under a force (Q), which may be tions along the actual slip surface in a slope (6.50). Hence,
its own weight. The shear strength is obtained from the re- caution should be exercised when the results of these simple
lation strength tests are applied to slope stability problems (6.15).
141
Figure 6.5. Triaxial test. This will depend on the type of soil and the level of stress in-
(a) EQUIPMENT volved. The water content at failure (wf) will usually be dif-
Trio,i CCII ferent from w.
Gouge
Since the excess pore pressure is zero in the drained test,
Cap effective stresses are known throughout the test and partic-
ularly at failure. In a compression test, the effective radial
VoI:n
C
stress (as) is equal to the cell pressure, and the measured
h..ge ffPorous
e slonQ
load on the ram can be used to evaluate the effective axial
stress (Uj). The results of a series of drained tests can be
Pore pressure used to evaluate the effective stress strength parameters in
transducer
Equation 6.4.
In the consolidated-undrained test with pore pressure
)b) STRESS (c( STRESS-STRAIN CURVES measurement (sometimes called consolidated-quick test),
CONDITIONS
.300 the drainage valves are closed after the initial consolidation
of the sample to w. Stress changes are applied through the
LO3 ram, and the excess pore pressure is measured. The pore
2 00
: pressure is subtracted from the total axial and radial stresses
'I00 to give the effective stresses. The effective stresses at failure
from a series of tests are used to define the failure criterion,
as in the drained test. Since the test is run in the undrained
condition after consolidation, the water content throughout
the test and at failure is w,. Excess pore pressures developed
during the test can be either positive or negative, depending
the use of these measurements. Pore pressures in the sample
on the type of soil and stress level.
are also measured by a device connected to the porous stone.
Several equations have been proposed to describe the
Several devices may be used. Electric pressure transducers
magnitude of the excess pore pressure developed as a result
are becoming more common and are replacing the tradi.
of stress changes in an undrained soil. For soils tested in the
tional null indicator and hydraulic cylinder system. triaxial apparatus, or loaded so that Lar2 = L0 3 , Skempton
The axial stress (a') may be increased by application of
(6.67) proposed that the excess pore pressure is given by
a load through the loading ram. From the known stresses
at failure (cr = a and 03 = 02 = ar), Mohr circles or other
stress plots can be constructed. Several triaxial tests, each AuB[Aa3 +AA(a1 - (73)] [6.7]
using a different value of cell pressure (3), are usually per-
where
formed on the same material for the definition of the failure
envelope. The principal stresses at failure are then used to con-
struct Mohr circles or other stress plots from which a failure B = empirical coefficient related to the soil's compressi-
envelope can be obtained. Typical plots of the principal bility and degree of saturation, and
stress difference versus axial strain are shown in Figure 6.5c. A = empirical coefficient related to the excess pore pres-
sure developed because of shear of soil.
The stress-strain behavior is influenced by the confining pres-
sure, the stress history, and other factors. Analytical repre-
sentations of normalized stress-strain relations have been sug- General relations between pore pressure and applied stresses
gested for some soil types. The sample can also be loaded to have been suggested. For example, Henkel (6.37) proposed
failure in extension by increasing the radial stress while main-
taining the axial stress constant; then, a, = 02 = °r and 03 Au = B(Aa + aA7-0 ) [ 6.81
= a. These two methods of loading simulate the stress
states shown in Figures 6.1b and c respectively. - where
The following tests are commonly performed with the
triaxial apparatus. a= empirical coefficient similar to A;
TOC t = octahedral shear stress, equal to
I. In the consolidated-drained test (sometimes called a i/(a 2)2 + ('92 - 0)2 + ((73- )2. and
drained test or slow test), the soil is allowed to consolidate = octahedral normal stress, equal to (ai+ a + (73)/3.
completely under an effective cell pressure (as) so that at
the end of consolidation the excess pore pressure in the soil The consolidated-undrained test is sometimes performed
is zero. The water content of the specimen after consolida- without pore-pressure measurement. Obviously, effective
tion is w. In the triaxial compression test, the axial stress stresses are not known during this test or at failure. The ap-
is increased at a slow rate, and drainage is permitted. The plication of shear strength measured in this test to any field
rate should be slow enough so that water can drain through problem involves assumptions of excess pore pressure that
the soil, and no excess pore pressure should be allowed to are of questionable validity in most cases; thus, this test is
build up. A drained test of soils of low permeability often not recommended. To relate strength parameters obtained
requires several days. from consolidated undrained tests without pore-pressure
The volume change during shear in a drained test can re- measurement to field conditions is difficult.
sult in either an increase or a decrease in the water content. In the unconsolidated-undrained test (sometimes
142
called undrained test or quick test), no drainage is allowed Figure 6.6. Direct shear test.
during any part of the test. When the cell pressure is applied,
(a) SHEAR BOX (b( STRESS STATE
a pore-water-pressure change (Au g ) occurs in the soil. When AT POINT b
Plane-Strain Test
The geometry of many geotechnical problems can be approx- The failure envelope isobtained from several tests, each
imated by the condition of plane strain, in which the inter- using a different effective normal stress, performed on speci-
mediate principal strain (€2) is zero. To simulate this condi- mens of the same soil. The values of r, at failure are plotted
tion, plane-strain tests have been developed (6.30, 6.39). In against the values of or.. The loading is carried but slowly,
plane-strain tests, the sample is consolidated anisotropically so that no excess pore pressure develops; hence, the drained
with zero lateral strain (ex = c,, = 0). After this, the sample condition is obtained.
is loaded to failure by increasing either a or a, and main- In saturated clays, the direct shear test can be performed
taining e,, = 0. The two methods of loading can be used to at a rapid rate so that the time duration is too short for any
simulate the stress conditions at points c or a of Figure 6.2. appreciable amount of water to flow into or out of the sam-
Plane-strain tests can be conducted under undrained, ple. This is an undrained condition, and excess pore pres-
consolidatedundrained, or drained conditions in manners sures of unknown magnitude are usually developed in the
similar to those described for triaxial tests. soil. Consequently, this is essentially a simple test, and the
shear stress at failure represents the undrained shear strength
Direct Shear Test (Su).
The direct shear test is shown in Figure 6.6. The soil speci- Simple Shear Test
men is enclosed in a box consisting of upper and lower
halves; porous stones on top and bottom permit drainage of Several simple shear tests have been developed, but the one
water from the specimen. The potential plane of failure is described by Bjerrum and Landva (6.17) is most commonly
a-a. A normal stress (ui) is applied on plane a-a through a used for testing undisturbed samples. The cylindrical speci-
loading head, and the shear stress is increased until the speci- men is enclosed in a rubber membrane reinforced by wire.
men fails along plane a-a. A stress-deformation curve is ob- This allows the shear deformation to be distributed fairly
tained by plotting the shear stress versus the displacement. uniformly through the sample, as shown in Figure 6.7b. In
Because the thickness of shear zone a-a is not precisely the test, the sample is consolidated anisotropically under a
known, the shear strain cannot be determined. The test vertical stress (Figure 6.7a) and sheared by application of
gives the value of 7- 5 at failure. The vertical stress (a) and stress r,. (Figure 6.7b). The simple shear test can be per-
the shear stress (r5) at point b (Figure 6.6b) are known, but formed under undrained, consolidated-undrained, and
a, is not. The directions of the principal stresses are approx drained conditions. Zero volume change during shear in
imately as shown in Figure 6.6c. Assuming that point a an undrained test can be maintained by adjusting the ver-
(Figure 6.6d) represents the conditions at failure, a Mohr tical stress (at ) continuously during the test. In the simple
circle can be constructed. The foregoing- represents the com- shear test, the principal axes are in the vertical and hori-
mon interpretation of the direct shear test. More elaborate zontal directions initially. At failure, the horizontal plane
analyses have been presented by Hill (6.41) and Morgenstern becomes the plane of maximum shear strain. This condi-
and Tchalenko (6.57). tion approximates that at point b of Figure 6.2.
143
SHEAR STRENGTH PROPERTIES Figure 6.8. Typical failure envelope for cohesionless soils.
OF SOME COMMON SOILS
soil of the same average particle size. The same is true when TYPICAL STRESS AND WATER CONTENT RELATION
a soil composed of angular grains is compared with one made
up of rounded grains. The effect of moisture on ' is small
and amounts to no more than 10 or 20 (6.42).
The failure envelope, which is a straight line at low pres-
sures, cannot be extended to high confining pressures. Tests
with effective normal stresses above 700 kPa (1460 lbf/in2 )
indicate that the failure envelope is curved, as shown in
Figure 6.8 (6. 7, 6. 77). The high normal stresses apparently
Water content
cause crushing of grain contacts and result in a lower friction
angle. Another important factor is the difference in the val-
ues of t ' as measured by different types of tests. The ' fine-grained soils, undrained or partially drained situations
measured in triaxial tests, which permit change in the radial are common. This is a most important difference between
strain, is as much as 40 to 50 smaller than the ' measured in
cohesionless soils and cohesive soils. Another important dis-
plane-strain tests (6.33). This difference has also been ob-
tinction between normally consolidated or lightly overcon-
served in field problems.
solidated clays and heavily overconsolidated clays is based
In ordinary construction situations, sandy and gravelly
on the kind of excess pore pressures developed in these soils
soils of high permeability can be considered to be loaded in during shear. The general characteristics of normally con-
the drained condition. Volume changes occur rapidly, and solidated clays will be discussed first. Extremely sensitive
no excess pore pressures are sustained. Without excess pore
normally consolidated clays are not discussed here but in the
pressures, effective stresses can be estimated from the ground-
later section on sensitive soils.
water levels. Stability analyses can be performed by using the
A clay soil is considered to be normally consolidated if
effective stress strength parameters.
the consolidation pressures before shear are equal to or
For silty soils, the permeability may be sufficiently low
greater than the preconsolidation stress (p). When a series
that excess pore pressures will develop during construction.
of drained triaxial tests is conducted on a normally consoli-
When this is the case, the pore pressures must be measured
dated soil, the failure envelope is a straight line that passes
or estimated if an effective stress analysis is to be performed.
through the origin (Figure 6.9a); thus, c' = 0. A relation be-
The undrained response of sands and gravels is required
tween strength and water content is shown in Figure 6.9b. If
for only a few situations. Saturated loose sand may fail so
consolidated-undrained tests are performed on a normally
rapidly that excess pore pressures are sustained. Similarly,
consolidated soil, positive excess pore pressures develop.
under rapid loading the undrained shear strength may be ap-
The effective stresses at failure will define the same failure
plicable (see the later section on soil behavior under re-
envelope in consolidated-undrained tests as in drained tests
peated loads).
(c' = 0 and 0' = As a result of the positive excess pore
pressures, however, the undrained strength (a1 - 03 ) will be
Soft Saturated Clays and Clayey Silts less than the drained strength of a sample initi4lly consoli-
dated under the same stresses. This characteristic can be ap-
Soils containing significant amounts of clay and silt are plied as a design principle.
called cohesive soils. Because of the low permeability of If the load or stress change in the field induces positive
144
excess pore pressures (as in the case of a fill), the undrained term failures in soft clay soil and the calculated safety fac-
strength will be lower than the drained strength. An initially tors.
stable design can usually be expected to increase in stability The examples of field investigations given in Table 6.1 for
with time as the excess pore pressure dissipates, the water short-term conditions of soft clay soils all involve undrained
content decreases, and the strength increases. On the other behavior of normally consolidated or lightly overconsoli-
hand, if negative excess pore pressures are induced (as in the dated clays. To date, more than 50 slope failures and founda-
case of an excavation), the undrained strength will be larger tion failures in such soils have been investigated and reported.
than the drained strength. Failure may occur sometime after For more than 90 percent of those, the discrepancy between
construction, even though the slope is stable in the undrained calculated and observed safety factors is less than 15 percent.
state. Bishop and Bjerrum (6.9) have described several ex- Since most of the clays investigated are fairly uniform de-
amples. posits, the accuracy would be less for nonuniform clay de-
A good source of information on the reliability of theoret- posits. Many factors contribute to this uncertainty, but
ical models is the failure of real slopes. In a number of strength anisotropy and rate of loading are probably two of
careful investigations, the factor of safety of the slope that the most important. Bjerrum (6.14, 6.15) reviewed notable
failed was compared with the measured shear strengths. If cases in which large discrepancies between prediction and
the theory and soil properties used are correct, the safety performance were observed in highly plastic and organic
factor of a slope at failure should be unity. The results of clays. For those cases, the use of undrained shear strength,
these studies show that, for normally consolidated or lightly as measured by the unconfined compression or vane shear
overconsolidated homogeneous clays of low sensitivity, tests, tends to overestimate the safety factor under undrained
analysis using the undrained shear strength is reasonably ac- conditions. Available case studies of drained failure in nor-
curate for immediate stability. For long-term stability, the mally consolidated clays are too few to support a statement
effective stress analysis is also consistently accurate. Several about the accuracy of predictions. However, the reliability
studies on bearing capacity failure likewise show reasonable of these predictions appears to be about the same as that for
agreement. Summarized results of several case studies are undrained failure of normally consolidated clays.
given in Table 6.1. Eight cases are given of short-term failure
immediately after or during construction. These are un- Heavily Overconsolidated Clays
drained conditions, and the analyses were made by using the
undrained shear strength (se). The computed factors of
Geological and stress histories are important considerations
safety are all close to unity and thus show that failure should
in the behavior of heavily overconsolidated clays. The pres-
have occurred according to theoretical predictions.
When a slope is made by excavation, there is a simulta- ence of fissures, which may be due to passive failure under
neous increase in shear stress due to the slope and a decrease high values of K0 (6.11, 6.20, 6.68) or other causes such as
shrinkage, has an important influence on the strength of
in mean normal stress due to the general unloading of the
soils. The characteristics of some fissures and the strengths
excavation (6.9). In a saturated normally consolidated soil,
along the fissures are described by Skempton and Petley
the increase in shear stress produces a positive excess pore
pressure, and the decrease in normal stress produces a nega- (6. 75). The shear strength of laboratory specimens of fis-
sured clays is strongly dependent on the number, shape, and
tive excess pore pressure. The net excess pore pressure in
inclination of fissures in the specimen (6.54). The presence
various parts of the slope depends on the relative values of
of fissures is less likely in small specimens, which are often
thesetwo effects. If the excess pore pressure is negative, the
trimmed from intact soil between the fissures. Hence, the
soil will decrease in strength with time and drainage. In this
measured strength tends to increase as the size of test speci-
case, the long-term or drained stability will be critical for a
men decreases (6.52, 6. 78). Thus, to extrapolate from the
normally consolidated clay. An example of this situation is laboratory shear strength to the in situ shear strength is often
shown in Figure 6.10. Table 6.1 gives four cases of long- difficult, and frequently in situ load tests must be conducted.
Most heavily overconsolidated clays show stress-strain re-
lations that suggest general strain softening (Figure 6.3,
Figure 6.10. Changes in pore pressure and factor of
curve A). Several concepts may be used to explain this
safety during and after excavation of a cut in clay (6.9). strain-softening behavior. Consider the test results from a
series of heavily overconsolidated clay specimens. If the
t-------Oiginol GWI. peak strength is used to describe failure, an effective stress
Skemptons pore-pressure coefficient.
0 A
failure envelope as shown by curve A in Figure 6.1la is ob-
Al Final GWL tained. The failure envelope is approximately a straight
cc line and, if extrapolated to the axis of a' = 0, there is a co-
Time hesion intercept (c'). If the effective stresses at failure are
used, results from both drained and undrained tests describe
& 0 nO Method Applicoble Here the same envelope. Laboratory tests using normal stresses
that are close to the normal stresses in the field should be
performed because research (6.11, 6.66) has shown that
Factor Of Sotety (C,Ø',nethod) the failure envelope for the peak strength of heavily over-
consolidated clays is curved in the low stress region and
passes through the origin.
Pee flu,e
As time and drainage increase and the negative pore pres-
Equilibriwe sure dissipates, a heavily overconsolidated clay will absorb
145
Table 6.1. Examples of stability prediction in soft and overconsolidated clays.
146
Table 6.1. Continued.
= =
Note: LL liquid limit; PL plastic limit; wc = = =
water content; 5u undrained shear strength; c' cohesion intercept in terms of effective
=
stress; 0' angle of internal friction in terms of effective stress; Ø =
residual angle of internal friction; R residual factor =
(peak =
- -
strength average strength at failure)/lpeak strength residual strength); and 1 kPa 0.145 lbf/in 2. =
aAverage bPeak. C
Residual.
Figure 6.11. Shear strength levels developed by by curve B in Figure 6.1 Ia. This fully softened strength is
heavily overconsolidated clays.
typically the same as the strength of the same soil in the
(a) TYPICAL MOHR FAILURE ENVELOPES normally consolidated condition. The strain necessary to
develop fully softened strength in a heavily overconsolidated
clay varies from one soil to another, but is on the order of
10 to 100 percent. The significance of using the fully soft-
ened strength in the long-term design of slopes has been dis-
cussed by several authors (e.g., 6.24, 6. 70).
When much larger shear displacements take place within
a narrow zone, the clay particles become oriented along the
direction of shear, and a polished surface or slickenside is
formed (6.46, 6.57). In natural slopes, slickensides may be
developed along surfaces of old landslides, bedding planes,
or zones of deformation caused by tectonic forces. Along
these surfaces, the shear strength may approach the residual
)b) TYPICAL STRESS.STRAIN CURVE
Peas strength strength (6.69); this concept has been the subject of exten-
sive study in the field and in the laboratory. The failure
envelope, curve C in Figure 6.1 1a is typical for the residual
,oftened strength
strength. The straight line passing through the origin defines
k I the residual angle of internal friction ('r). Kenney (6.46),
bi
147
Figure 6.12. Residual shear strengths'of soil minerals (6.46). most dramatic landslides in sensitive soils are the flow slides
occurring in Pleistocene marine clays in the Scandinavian
20
countries of northern Europe and in the St. Lawrence River
00 U rr. valley of eastern North America. The sensitivity of these
soils is due to either leaching or natural cementation.
80
The effective stress failure envelope for both leached and
-. 60
naturally cemented sensitive clays is distinctly different from
that of other soft clays, and the strengths of both types of
40 sensitive soil are dominated by structure. Special testing
techniques must be used to define these failure envelopes.
20
As shown in Figure 6.13a, the strength envelope for a na-
0--
0 20 50 00 150 200 250 turally cemented sensitive clay is a unique curve in stress
Norma lot,,,, (Cr;).kPO
space (6.59). The failure envelope encloses a low stress
Note: 1 kPa = 0.145 IbI/:rr2 region within which the cemented structure remains intact.
When stress changes in a slope exceed the limits of the fail-
ure curve, the structure is destroyed. Then high positive
sample. As water content increases, strength decreases. Un- excess pore pressures are produced, and the soil behaves as a
der field conditions, the long-term or drained conditions are remolded soil. Thus, the failure envelope for the remolded
critical and there is no assurance that an initially stable slope soil (Figure 6.13a) governs the strength only after rupture of
will remain stable in the long term. the cemented structure and has no influence on initial failure.
The behavior of heavily overconsolidated clays is obvi- The cemented structures of these sensitive soils can be
ously much more complicated than that of normally consol- broken down by consolidation, particularly during isotropic
idated soils. In the application of shear strength concepts to consolidation of a triaxial test specimen.. The natural state
a slope stability analysis, the most difficult problem is to of stress for a sensitive cemented soil lies within its domain
select the appropriate operating strength. A further compli- of natural cementation. In the design of slopes in these soils,
cation is that the strength changes are strain-softening it is usually desirable to keep the stresses within these same
(Figures 6.3 and 6.1 1). As noted in the earlier section on limits. Therefore, stresses used in laboratory tests should
stress-strain characteristics, it is unreasonable to expect all also lie within these limits. A common mistake in testing
of the soil along a failure surface to reach peak strength at these materials is to use a high cell pressure, which destroys
the same time. If the soil at some parts of a failure surface the structure during consolidation (6.15). In this case the
has not yet reached the peak or has passed the peak, the op- failure envelope for remolded soil is obtained, and it may
erating strength must be chosen to represent the average of not be applicable to the actual stress conditions in the
the strengths along the entire failure surface (see Chapter 7). field.
The use of the fully softened strength may be too conserva-
tive (6.25).
Another factor that may affect the stability of slopes in Figure 6.13. Strength envelopes for sensitive
heavily overconsolidated clays is the effect of residual stress. soils.
will become less stable with time, any design using the un-
drained strength must be considered temporary and the un-
drained strength should only be used if the clay is intact.
Fissured clays soften and drain so rapidly that undrained
conditions should not be assumed in design. The long-term LEACHED SOILS
Sensitive Soils
148
For a leached sensitive soil (Figure 6.13b), the strength during an undrained triaxial test. In an unsaturated soil,
envelope falls entirely below the failure envelope for the menisci exist at the air-water interfaces and the water is un-
remolded soil. For both the leached and naturally cemented der capillary tension. Hence, the initial pore-water pressures
sensitive soils, the failure envelope is a unique characteristic have a negative value equal to u0 . After the all-around pres-
for each soil. A separate set of tests must be used to deter- sure (cr3) is applied, the compression of these air spaces
mine the strength of each soil. Laboratory studies also show brings about a volume change and a rise in the air and water
that the strength of these soils is highly anisotropic and that pressures; these are shown as points a and b in Figure 6.14a.
compression, extension, and simple shear tests commonly The difference between the pore-air pressure (u a ) and the
give different strength envelopes (6.16, 6.56, 6.59). pore-water pressure (u) is equal to the capillary tension.
The difficulties in sampling and testing sensitive soils have As a1 is increased, both ua and uw change with strain.
resulted in a great deal of emphasis on field measurement of In a system with both air and water under pressure, the
their strengths. The vane test is used extensively; corrections effective stress may be written as (6.8)
to account for anisotropy and strain rate have been suggested
by Bjerrum (6.15). O'UUa +X(Ua _Uw) [6.9]
149
eluvium. It may develop to great depths, and natural or recognize the potential for change in the strength of a joint
human-made slopes may be cut into the eluvium. Natural or joint system during the lifetime of an engineering project.
rock slopes also weather to produce residual soil, which Experimental measurements of joint strengths can be made
will be acted on by forces of gravity. The soil and rock de- by the use of laboratory or field direct shear tests.
bris that moves down the face of a slope is called colluvium. A stress-displacement curve of a rough joint is similar to
Residual soils have a wide range of properties, depending curve A in Figure 6.3. The peak strength (point a) is ob-
on the parent material and the degree of weathering (6.29, tained when the large projections along the joint are sheared
6.53). In the initial stages of weathefing, coarse rock frag- off. Beyond this point, the strength decreases and ap-
ments are produced; the ultimate product of weathering is proaches the residual strength. The residual strength repre-
clay. Between these extremes is usually a mixture of grain sents the strength of the joint after the projections have been
sizes, and the behavior of residual soils is similar to that de- sheared off.
scribed in previous sections for the soils with similar particle Typical results of drained tests in which effective stresses
size distributions. Some residual soils are also cemented. could be measured are given in Table 6.2. The strength is
More detailed information about residual soils and their almost always frictional in character, even when large
properties is given by Deere and Patton (6.29). amounts of debris are present in the joint.
Colluvium presents some particular slope stability prob-
lems (6.27, 6.53). The extremely heterogeneous nature of
the material makes sampling and testing difficult. In some Figure 6.15. Strength of rock masses with closed block joints
cases, loose structure and large grain size result in high values (6.32).
of permeability. This characteristic is often accentuated by (a) JOINTED ROCK MASS (b( FAILURE ENVELOPE
the presence of less permeable layers immediately below the I I II I I
colluvium. Under this condition, infiltrating surface water peak strength
flows through the colluvium, generally parallel to the slope, criterion for joint
150
SOIL BEHAVIOR UNDER to repeated loading have been the subject of extensive re-
REPEATED LOADS search in recent years (6.3, 6.23, 6.58, 6.60, 6.62). In some of
those studies, pore pressures and effective stresses were de-
Ground motion during earthquakes subjects slopes to re- termined (6.23, 6.60, 6.65). In most cases, however, only
peated loading. Consider again the slope shown in Figure total stress analysis was possible because rapid rates of load-
6.2. The stresses shown are those acting under static condi- ing did not permit the measurement of pore pressures. The
tions. When subjected to earthquake motions, additional present state of knowledge is based on contributions from
stresses of a cyclic nature are induced in the soil. The na- both types of study.
ture of the stresses at point b is shown in Figure 6.16. For
simplicity, it is assumed here that the ground motion during Repeated Load Tests
an earthquake consists only of shear waves propagating ver-
tically through the soil. The cyclic stress (Te) during an Laboratory tests to measure soil strength under repeated
earthquake usually consists of a series of irregular pulses, as loads can be made with the triaxial cell or the simple shear
shown by the stress-time plot in Figure 6.16c. apparatus (6.63). Because of difficulties in interpretation
Several methods can be used to calculate the response of and analysis of random loads, such as those shown in Figure
a slope to repeated loading (see Chapter 7). Some require 6.16c, current laboratory tests usually employ regular stress
use of a constitutive relation for the soil, and others use the pulses. The shape of the stress pulse inay be square or tri-
soil strength under dynamic loading. When most natural angular or sinusoidal. In the simple shear test, the stress con-
soils are subjected to earthquake or other types of repeated ditions shown in Figure 6.16a and b can be simulated. The
loading, the resulting fluctuations in stress produce irreversi- sample is consolidated under the static stresses (a and r),
ble changes in pore pressures. These produce long-term and and the peak shear stress during an earthquake is Te Hence,
short-term changes in soil strength, and this must be recog- the cyclic stress (Te - r) is applied (Figure 6.16b).
nized in the design of slopes to resist earthquakes and other The triaxial test cannot simulate the rotation of principal
kinds of dynamic loads. Strength changes in soils subjected axes under earthquake loading; therefore, this phenomenon
must be ignored. The test procedures are similar to those
for static tests discussed in the section on the triaxial test.
Figure 6.16. Stresses under dynamic loading.
The principal stresses under static loading are (71 and 03 .
The sample is consolidated first under the static stresses
al INITIAL STATE
OF STRESS
(bi EARTHQUAKE
STRESS
Ic) EARTHQUAKE STRESS
VERSUS TIME
(Figure 6.17a), after which the cyclic stresses (01 - 01 and
- 03 ) are applied (Figure 6.17b). Because of the relatively
short duration of earthquakes, the cyclic stresses are usually
applied in the undrained condition. If the soil is saturated,
the effective stress does not change under an applied hydro-
static stress. Thus, the loading can be simplified to fluctua-
tion in axial stress only (Figure 6.17c).
Model tests of soil slopes and embankments loaded by
means of a shaking table have also been used in design (6.3).
Time
Stress-Strain Characteristics
Figure 6.17. Dynamic triaxial test conditions. Under repeated loading, the strain produced by a given peak
stress is usually different from that produced by a static
(a) INITIAL STATE (b( CYCLIC OR (c) CYCLIC OR
OF STRESS EARTHQUAKE EARTHQUAKE stress of the same magnitude. Strain continues to increase
STRESS STRESS
with successive cycles and depends on several factors, but
particularly on the duration of the load, the magnitude of
the stresses, and the number of load cycles. A typical stress-
strain curve is shown in Figure 6.18. At small strain, the
cyclic stress produces the hysteresis ioop ab. The shear
modulus and damping are equal to the slope of ab and the
area enclosed by the ioop respectively. At large Strains, the
hysteresis ioop is cd. The shear modulus decreases with
strain, and the damping increases with strain. Estimates of
the shear modulus and damping factor can be made on the
Figure 6.18. Stress-strain basis of available empirical data (6.35, 6.65). For dry soils
relation for cyclic loading. and soils with low degrees of saturation, the modulus tends
to increase with cycles of loading. Figure 6.19 compares
soil behavior under repeated loading (curves B and C) with
that under a monotonically increasing stress (curve A). The
stress-strain relation and pore pressure under repeated load-
ing depend on stress level, stress history, type of loading,
number of stress cycles, and degree of saturation.
Shear strain When a saturated soft clay or loose sand is subjected to a
151
high level of stress (for example, or, in Figure 6.19a), posi- a particular strain is shown in Figure 6.19d. This is a com-
tive pore pressures develop (curve C in Figure 6.19b); After mon and useful relation, particularly when only total stresses
a sufficient number of load cycles, the accumulated pore are known. At lower levels of cyclic stress, failure does not
pressure will lead to failure, and the stress-strain curve C occur, even under a large number of loading cycles. This is
(Figure 6.19a) is obtained. Failure occurs as the effective shown as the asymptote in Figure 6.19d; it is called the criti-
normal stress is reduced (Figure 6.19c). This failure condi- cal level of repeated loading (6.60). Curves such as these
tion is described by several different terms (6.23, 6.62). If make it possible to choose a design stress corresponding to
the sample is loaded to some stress level lower than a, the the anticipated number of loading cycles. Alternatively, the
pore pressure may build up to a certain value and remain at critical level of stress may be used for design. In terms of
that level (curve B, Figure 6.19b) and the strain will ap- effective stress, the critical level and the corresponding void
proach a limiting value (curve B, Figure 6.19a). In this case, ratio have been equated (6.60) to the "critical state" of the
no failure occurs. soil, as defined by Schofield and Wroth (6.61). Soils other
than soft clays and loose sands can also experience strength
Failure Under Repeated Loading changes as a result of earthquakes or other repeated loading.
The fundamental phenomena that control the strength
Failure caused by high pore pressure, as shown in Figure changes and the states that define the critical level of stress
6.19, is called liquefaction (6.62), particularly when applied have been considered by Sangrey (6.59), and a summary is
to cohesionless soils. The relation between cyclic stress level given in Table 6.3.
(S) and number of cycles (N) necessary to achieve failure or
REFERENCES
Figure 6.19. Behavior of soils under repeated loading. 6.1 Aitchison, G. D. Problems of Soil Mechanics and Construc-
STRESS-STRAIN RELATIONS FOR (c) FAILURE DUE TO INCREASED PORE tion on Soft Clays and Structurally Unstable Soils. Proc.,
MONOTONIC AND CYCLIC LOADING PRESSURES 8th International Conference on Soil Mechanics and Foun-
dation Engineering, Moscow, Vol. 3, 1973, pp. 161-190.
6.2 American Society of Civil Engineers and U.S. Committee
on Large Dams. Design Criteria for Large Dams. American
Society of Civil Engineers, New York, 1967, 131 pp.
6.3 Arango, I., and Seed, H. B. Seismic Stability and Deforma-
tion of Clay Slopes. Journal of Soil Mechanics and Foun-
dations Division, American Society of Civil Engineers, New
York, Vol. 100, No. GT2, 1974, pp. 139-156.
6.4 Barden, L., McGown, A., and Collins, K. The Collapse
Mechanism in Partly Saturated Soil. Engineering Geology,
O'3 Vol. 7, No. 1, 1973, pp. 49-60.
6.5 Barton, N. Review of a New Shear-Strength Criterion for
PORE PRESSURE ACCUMULATION (di CYCLIC STRESS LEVER VERSUS Rock Joints. Engineering Geology, Vol. 7, No. 4, 1973,
UNDER CYCLIC LOADING NUMBER OF CYCLES TO FAILURE pp. 287-332.
6.6 Barton, N. Estimating the Shear Strength of Rock Joints.
Proc., 3rd Congress, International Society of Rock Me-
chanics, Denver, Vol. 2, Part A, 1974, pp. 2 19-220.
6.7 Bishop, A. W. The Strength of Soils as Engineering Mate-
rials. Geotechnique, Vol. 16, No. 2, 1966, pp. 91-128.
6.8 Bishop, A. W., Alpan, I., Blight, G. E., and Donald, I. B.
Critical Level of Factors Controlling the Strength of Partly Saturated Cohe-
Loading sive Soils. Proc., Research Conference on Shear Strength
of. Cohesive Soils, Boulder, American Society of Civil En-
gineers, New York, 1960, pp. 503-53 2.
Table 6.3. Design recommendations for strength of soils subjected to earthquake or other repeated loading (6.59).
152
6.9 Bishop, A. W., and Bjerrum, L. The Relevance of the Tri- struction in Rock. In Failure and Breakage of Rock
axial Test to the Solution of Stability Problems. Proc., Re- (Fairhurst, C., ed.), Proc., 8th Symposium on Rock Me-
search Conference on Shear Strength of Cohesive Soils, chanics, American Institute of Mining, Metallurgy and
Boulder, American Society of Civil Engineers, New York, Petroleum Engineers, New York, 1967, pp. 237-302.
1960, pp. 437-501. 6.29 Deere, D. U., and Patton, F. D. Slope Stability in Residual
6.10 Bishop, A. W., and Henkel, D. J. The Measurement of Soil Soils. Proc., 4th Pan-American Conference on Soil Me-
Properties in the Triaxial Test. Edward Arnold Publishers, chanics and Foundation Engineering, San Juan, American
London, 1962, 227 pp. Society of Civil Engineers, New York, Vol. 1, 1971, pp.
6.11 Bishop, A. W., Webb, D. L., and Lewin, P. I. Undisturbed 87-170.
Samples of London Clay From the Ashford Common 6.30 Duncan, J. M., and Seed, H. B. Strength Variation Along
Shaft. Geotechnique, Vol. 15, No. 1, 1965, pp. 1-3 1. Failure Surfaces in Clay. Journal of Soil Mechanics and
6.12 Bjerrum, L. Progressive Failure in Slopes of Overconsoli- Foundations Division, American Society of Civil Engi-
dated Plastic Clay and Clay Shales. Journal of Soil Me- neers, New York, Vol. 92, No. SM6, 1966, pp. 81-104.
chanics and Foundations Division, American Society of 6.31 Esu, F. Short-Term Stability of Slopes in Unweathered
Civil Engineers, New York, Vol. 93, No. SM5, 1967, pp. Jointed Clays. Geotechnique, Vol. 16, No. 4, 1966, pp.
3-49. 32 1-3 28.
6.13 Bjerrum, L. Closure to Main Session 5. Proc., 7th Interna- 6.32 Goodman, R. E. Methods of Geological Engineering in Dis-
tional Conference on Soil Mechanics and Foundation En- continuous Rocks. West Publishing, St. Paul, 1976,472 pp.
gineering, Mexico City, Vol. 3, 1969, pp. 410-414. 6.33 Green, G. E., and Bishop, A. W. A Note on the Drained
6.14 Bjerrum, L. Embankments on Soft Ground. Proc., Spe- Strength of Sand Under Generalized Strain Conditions.
cialty Conference on Performance of Earth and Earth- Geotechnique, Vol. 19, No. 1, 1969, pp. 144-149.
Supported Structures, Purdue Univ., Lafayette, Ind., 6.34 Hansbo, S. A New Approach to the Determination of the
American Society of Civil Engineers, New York, Vol. 2, Shear Strength of Clay by the Fall-Cone Test. Proc., Royal
1972, pp. 1-54. Swedish Geotechnical Institute, No. 14, 1957, 47 pp.
6.15 Bjerrum, L. Problems of Soil Mechanics and Construction 6.35 Hardin, B. 0., and Drnevich, V. P. Shear Modulus and
on Soft Clays and Structurally Unstable Soils. Proc., 8th Damping in Soils: Measurement and Parameter Effects.
International Conference on Soil Mechanics and Founda- Journal of Soil Mechanics and Foundations Division,
tion Engineering, Moscow, Vol. 3, 1973, pp. 111-159. American Society of Civil Engineers, New York, Vol.
6.16 Bjerrum, L., and Kenney, T. C. Effect of Structure on the 98, No. SM6, 1972, pp. 603-624.
Shear Behavior of Normally Consolidated Quick Clays. 6.36 Henkel, D. J. Investigations of Two Long-Term Failures in
Proc., Geotechnical Conference on Shear Strength Proper- London Clay Slopes at Wood Green and Northolt. Proc.,
ties of Natural Soils and Rocks, Norwegian Geotechnical 4th International Conference on Soil Mechanics and Foun-
Institute, Oslo, 1967, Vol. 2, pp. 19-27. dation Engineering, London, Vol. 2, 1957, pp. 315-3 20.
6.17 Bjerrum, L., and Landva, A. Direct Simple-Shear Tests on 6.37 Henkel, D. J. The Shear Strength of Saturated Remolded
a Norwegian Quick Clay. Geotechnique, Vol. 16, No. 1, Clays. Proc., Research Conference on Shear Strength of
1966, pp. 1-20. Cohesive Soils, Boulder, American Society of Civil Engi-
6.18 Blight, G. E. Effective Stress Evaluation for Unsaturated neers, New York, 1960, pp. 533-554.
Soils. Journal of Soil Mechanics and Foundations Division, 6.38 Henkel, D. J., and Skempton, A. W. A Landslide at Jack-
American Society of Civil Engineers, New York, Vol. 93, field, Shropshire, in a Heavily Overconsolidated Clay.
No. SM2, 1967, pp. 125-148. Geotechnique, Vol. 5, No. 2, 1955, pp. 131-137.
6.19 Brand, E. W., Muktabhant, C., and Taechathummarak, A. 6.39 Henkel, D. J., and Wade, N. H. Plane Strain Tests on a
Load Tests on Small Foundations in Soft Clay. Proc., Spe- Saturated Remolded Clay. Journal of Soil Mechanics and
cialty Conference on Performance of Earth and Earth- Foundations Division, American Society of Civil Engineers,
Supported Structures, Purdue Univ., Lafayette, Ind., New York, Vol. 92, No. SM6, 1966, pp. 67-80.
American Society of Civil Engineers, New York, Vol. 1, 6.40 Henkel, D. J., and Yudhbir. The Stability of Slopes in the
Part 2, 1972, Pp. 903-928. Siwailk Rocks in India. Proc., 1st Congress, International
6.20 Brooker, E. W. and Ireland, H. 0. Earth Pressures at Rest Society of Rock Mechanics, Lisbon, Vol. 2, 1966, pp.
Related to Stress History. Canadian Geotechnical Journal, 16 1-16 5.
Vol. 2, No. 1, 1965, pp. 1-15. 6.41 Hill, R. The Mathematical Theory of Plasticity. Oxford
6.21 Cadling, L., and Odenstad, S. The Vane Borer: An Ap- Univ. Press, London, 1950, 355 pp.
paratus for Determining the Shear Strength of Clay Soils 6.42 Holtz, W. G., and Gibbs, H. J. Triaxial Shear Tests on
in the Ground. Proc., Royal Swedish Geotechnical Insti- Pervious Gravelly Soils. Journal of Soil Mechanics and
tute, No. 2, 1950, 88 pp. Foundations Division, American Society of Civil Engi-
6.22 Casagrande, A., and Wilson, S. D. Effect of Rate of Load- neers, New York, Vol. 82, No. SM!, 1956, pp. 1-22.
ing on the Strength of Clays and Shales at Constant Water 6.43 Ireland, H. 0. Stability Analysis of the Congress Street
Content. Geotechnique, Vol. 2, No. 3, 1951, pp. 25 1-263. Open Cut in Chicago. Geotechnique, Vol. 4, No. 4, 1954,
6.23 Castro, G. Liquefaction and Cyclic Mobility of Saturated pp. 163-168.
Sands. Journal of Geotechnical Engineering Division, 6.44 Jaeger, J. C. Friction of Rocks and Stability of Rock
American Society of Civil Engineers, New York, Vol. 101, Slopes. Geotechnique, Vol. 21, No. 1, 1971, pp. 97-134.
No. GT6, 1975, pp. 55 1-569. 6.45 Kankare, E. Failures at Kimola Floating Canal in Southern
6.24 Chandler, R. J. Lias Clay: The Long-Term Stability of Finland. Proc., 7th International Conference on Soil Me-
Cutting Slopes. Geotechnique, Vol. 24, No. 1, 1974, pp. chanics and Foundation Engineering, Mexico City, Vol. 2,
21-3 8. 1969, pp. 609-616.
6.25 Chandler, R. J., and Skempton, A. W. The Design of 6.46 Kenney, T. C. The Influence of Mineral Composition on
Permanent Cutting Slopes in Stiff Fissured Clays. Geotech- the Residual Strength of Natural Soils. Proc., Geotechni-
nique, Vol. 24, No. 4, 1974, pp. 457466. cal Conference on Shear Strength Properties of Natural
6.26 Conlon, R. J., Tanner, R. G., and CoIdwell, K. L. The Soils and Rocks, Norwegian Geotechnical Institute, Oslo,
Geotechnical Design of the Townline Road-Rail Tunnel. 1967, Vol. 1, pp. 123-129.
Canadian Geotechnical Journal, Vol. 8, No. 2, 1971, pp. 6.47 Kjaernsli, B., and Simons, N. Stability Investigations of
299-3 14. the North Bank of Drammen River. Geotechnique, Vol.
6.27 D'Appolonia, E., Alperstein, R., and D'Appolonia, D. J. 12, No. 2, 1962, pp. 147-167.
Behavior of a Colluvial Slope. Journal of Soil Mechanics 6.48 Kwan, D. Observations of the Failure of a Vertical Cut in
and Foundations Division, American Society of Civil Engi- Clay at Welland, Ontario. Canadian Geotechnical Journal,
neers, New York, Vol. 93, No. SM4, 1967, pp. 447-473. Vol. 9, No. 3, 1971, pp. 283-298.
6.28 Deere, D. U., Hendron, A. J., Jr., Patton, F. D., and 6.49 Ladd, C. C. Test Embankment on Sensitive Clay. Proc.,
Cording, E. J. Design of Surface and Near-Surface Con- Specialty Conference on Performance of Earth and Earth-
153
Supported Structures, Purdue Univ., Lafayette, md., 6.64 Sevaldson, R. A. The Slide in Lodalen, October 6, 1954.
American Society of Civil Engineers, New York, Vol. 1, Geotechnique, Vol. 6, No.4, 1956, pp. 167-182.
Part 1, 1972, pp. 101-128. 6.65 Silver, M. L., and Seed, H. B. Deformation Characteristics
6.50 Ladd, C. C., and Foott, R. New Design Procedure for Sta- of Sands Under Cyclic Loading. Journal of Soil Mechanics
bility of Soft Clays. Proc., Journal of Geotechnical Engi- and Foundations Division, American Society of Civil Engi-
neering Division, American Society of Civil Engineers, New neers, New York, Vol. 97, No. SM8, 1971, pp. 1081-1098.
York, Vol. 100, No. GT7, 1974, pp 763-786. 6.66 Singh, R., Henkel, D. J., and Sangrey, D. A. Shear and K0
6.51 Ladd, C. C., and Lambe, T.W. The Strength of "Undis- Swelling of Overconsolidated Clay. Proc., 9th International
turbed" Clay Determined From Undrained Tests. Proc., Conference on Soil Mechanics and Foundation Engineering,
Symposium on Laboratory Shear Testing of Soils, Ottawa, Moscow, Vol. 1, Part 2, 1973, pp. 367-376.
American Society for Testing and Materials, Philadelphia, 6.67 Skempton, A. W. The Pore-Pressure Coefficients A and B.
Special Technical PubI. 361, 1964, pp. 342-371. Geotechnique, Vol.4, No.4, 1954, pp. 143-147.
6.52 Lo, K. Y. The Operational Strength of Fissured Clays. 6.68 Skempton, A. W. Horizontal Stresses in an Over-
Geotechnique, Vol. 20, No. 1, 1970, pp. 57-74. Consolidated Eocene Clay. Proc., 5th International Con-
6.53 Lumb, P. The Properties of Decomposed Granite. Geotech- ference on Soil Mechanics and Foundation Engineering,
nique, Vol. 12, No. 3, 1962, pp. 226-243. Paris, Vol. 1, 1961, pp. 351-357.
6.54 Marsland, A. The Shear Strength of Stiff Fissured Clays. 6.69 Skempton, A. W. Long-Term Stability of Clay Slopes.
In Stress-Strain Behavior of Soils (Foulis, G. T., ed.), Geotechnique, Vol. 14, No. 2, 1964, pp. 77-101.
Proc., Roscoe Memorial Symposium, Cambridge Univ. 6.70 Skempton, A. W. First-Time Slides in Over-Consolidated
Press, 1971, pp. 59-78. Clays. Geotechnique, Vol. 20, No. 3, 1970, pp. 320-324.
6.55 Mitchell, J. K., and Houston, W. N. Causes of Clay Sensi- 6.71 Skempton, A. W., and Brown, J. D. A Landslide in Boulder
tivity. Journal of Soil Mechanics and Foundations Divi- Clay at Selset, Yorkshire. Geotechnique, Vol. 11, No.4,
sion, American Society of Civil Engineers, New York, Vol. 1961, pp. 280-293.
95, No. SM3, 1969, pp. 845-871. 6.72 Skempton, A. W., and Golder, H. Q. Practical Examples of
6.56 Mitchell, R. J., and Wong, K. K. The Generalized Failure the 0 = 0 Analysis of Stability of Clays. Proc., 2nd Inter-
of an Ottawa Valley Champlain Sea Clay. Canadian Geo- national Conference on Soil Mechanics and Foundation En-
technical Journal, Vol. 10, No. 4, 1973, pp. 607-616. gineering, Rotterdam, Vol. 2, 1948, pp. 63-70.
c ci iviorgenstern, N. R., and Tchalenko, J. S. Microscopic 6.73 Skempton, A. W., and Hutchinson, J. N. Stability of Nat-
Structures in Kaolinite Subjected to Direct Shear. Geo- ural Slopes and Embankment Foundations. Proc., 7th
technique, Vol. 17, No.4, 1967, pp. 309-328. International Conference on Soil Mechanics and Founda-
6.58 Sangrey, D. A. Changes in Strength of Soils Under Earth- lion Engineering, Mexico City, State-of-the-Art Vol.,
quake and Other Repeated Loading. In Earthquake Engi- 1969, pp. 291-340.
neering (Cherry, S., ed.), Proc., 1st Canadian Conference 6.74 Skempton, A. W., and La Rochelle, P. The Bradwell Slip:
on Earthquake Engineering, Univ. of British Columbia, A Short-Term Failure in London Clay. Geotechnique, Vol.
Vancouver, 1972, pp. 82-96. 15, No. 3, 1965, pp. 221-242.
6.59 Sangrey, D. A. Naturally Cemented Sensitive Soils. Geo- 6.75 Skempton, A. W., and Petley, D. J. The Strength Along
technique, Vol. 22, No. 1, 1972, pp. 139-152. Structural Discontinuities in Stiff Clays. Proc., Geotechni-
6.60 Sangrey, D. A., Henkel, D. J., and Esrig, M. I. The Effec- cal Conference on Shear Strength Properties of Natural
tive Stress Response of a Saturated Clay Soil to Repeated Soils and Rocks, Norwegian Geotechnical Institute, Oslo,
Loading. Canadian Geotechnical Journal, Vol. 6, No. 3 Vol. 2, 1967, pp. 29-46.
1969, pp. 241-252. 6.76 Vargas, M., and Pichler, E. Residual Soil and Rock Slides
6.61 Schofield, A., and Wroth, P. Critical State Soil Mechanics, in Santos (Brazil). Proc., 4th International Conference on
McGraw-Hill, New York, 1968, 310 pp. Soil Mechanics and Foundation Engineering, London, Vol.
6.62 Seed, H. B. Landslides During Earthquakes Due to Soil 2, 1957, pp. 394-398.
Liquefaction. Journal of Soil Mechanics and Foundations 6.77 VesIc, A. S., and Clough, G. W. Behavior of Granular Ma-
Division, American Society of Civil Engineers, New York, terials Under High Stresses. Journal of Soil Mechanics and
Vol. 94, No. SM5, 1968, pp. 10554122. Foundations Division, American Society of Civil Engineers,
6.63 Seed, H. B., and Peacock, W. H. Test Procedures for Mea- New York, Vol. 94, No. SM3, 1968, pp. 661-668.
suring Soil Liquefaction Characteristics. Journal of Soil 6.78 Ward, W. H., Marsland, A., and Samuels, S. G. Properties
Mechanics and Foundations Division, American Society of the London Clay at the Ashford Common Shaft: In
of Civil Engineers, New York, Vol. 97, No. SM8, 1971, Situ and Undrained Strength Tests. Geotechnique, Vol.
pp. 1099-1119. 15, No.4, 1965, pp. 321-344.
154
Chapter 7
Methods of
Stability
Analysis
N. R. Morgenstern and Dwight A. Sangrey
The functional design of a slope requires that the deforma- a method of analysis that predicts deformations to an ac-
tions of the earth mass constituting the slope be limited to ceptable degree of accuracy would likely have to include
amounts governed by the use of the ground on or adjacent
to the slope. If there is no particular development near the Representative stress-strain relations, including be-
slope, the allowable deformations can be large, provided that havior from peak to residual shear strengths;
the earth mass does not fail with resulting uncontrolled de- Anisotropy;
formations. On the other hand, if there are structures close Variable pore pressure distributions;
to the slope or services buried within the soil beneath the Nonhomogeneity arising from variation of material
slope, smaller deformations may result in unacceptable per- properties with depth, layering, and discontinuities;
formance, and functional design would require that the aver- Influence of initial stress; and
age stress level in the earth mass be lower in this case than in Construction sequence.
the case in which no development is near the slope.
Analysis is an important component of design, which also Most of these factors have a major influence on the behavior
includes other equally important components, such as drain- of a slope; and, even if proven analytical techniques were
age considerations and construction control. In analysis, the readily available to predict deformations, the determination
mechanical properties of the earth materials are evaluated of all of these factors in a manner suitable for a deformation
quantitatively to arrive at a configuration consistent with analysis would seldom be a practical undertaking. Appro-
the performance requirements of the slope. In this chapter, priate analytical techniques are not yet at hand, although im-
the relative roles of limit equilibrium and deformation anal- pressive progress is being made by means of finite element
yses are discussed. The principles of limit equilibrium anal- methods.
ysis are outlined, and details of the techniques commonly The situation is not so pessimistic with regard to fills.
used in practice are given with examples. Both soil and rock Representative stress-strain relations within the working
slopes are discussed. The chapter concludes with a brief dis- range can be specified with a greater degree of reliability for
cussion of deformation analysis as it is currently used. Other embankments than for cuts or natural slopes. Moreover,
aspects of design are discussed in Chapters 8 and 9. since the soil mass is a processed material, it is generally
more uniform and homogeneous than naturally occurring
ROLES OF LIMIT EQUILIBRIUM deposits. Hence, deformation analysis has a role in the de-
AND DEFORMATION ANALYSES sign of embankment slopes, and that role will be reviewed in
greater detail later in this chapter.
Although the performance of a slope often is dictated by The stabilities of natural slopes, cut slopes, and fill slopes
allowable deformations, quantitative prediction of displace- are commonly analyzed by limit equilibrium methods. These
ments of the slope is seldom undertaken routinely. Instead, methods take into account all of the major factors that in-
analyses are made by the use of limit equilibrium methods, fluence the shearing resistance of a soil or rock mass; this is
and the performance of a slope is evaluated in terms of its one of their significant advantages. In addition, they are
factor of safety (F). There are several reasons for this. simpler than deformation analyses. However, because the
For realistic configurations commonly encountered in actual stress-strain relations are not used in the analysis,
design, particularly in the design of cuts and natural slopes, limit equilibrium methods do not result in calculation of ex-
155
pected deformations. Deformations are controlled by de- According to this definition, the factor of safety relates to
signing for an appropriate factor of safety. At least to this the strength parameters and not to the strength itself, which
extent, the use of limit equilibrium methods of analysis is in the case of an effective stress analysis depends on the ef-
semiempirical. The factor of safety cannot be measured ex- fective normal stress as well. Moreover, this definition im-
cept when the slope is failing, in which case it is known to plies that the factor of safety is uniform along the entire slip
be unity. As noted previously, there are many instances in surface.
which the precise deformations in a slope are of little con- Other definitions of the factor of safety are also used in
cern, provided the material stays in place. In such cases, it is geotechnical engineering. For example, the factor of safety
entirely appropriate to undertake analyses with limit equilib- is commonly defined as the ratio of a disturbing moment to
rium methods; however, regardless of the complexity of a restoring moment when the moment equilibrium about the
the method employed, the empirical nature of the design center of rotation of a slip circle is under consideration. This
criteria should be borne in mind. definition is not convenient when slip surfaces are noncircu-
lar and therefore is of limited applicability. It is equivalent
LIMIT EQUILIBRIUM ANALYSIS to the definition given above in the case of a 0 = 0 analysis.
When the bearing capacity of a foundation is calculated, the
Limit equilibrium methods for slope stability analysis have
factor of safety is usually d'etermined as the ratio of the max-
been used for several decades, and numerous techniques have
imum bearing capacity to the allowable bearing capacity.
been developed. Limit equilibrium analysis is used in design
Used in this way, the factor of safety is a ratio of loads and
to determine the magnitude of the factor of safety. When a
differs from the factor of safety applied to strength param-
slope has failed, however, the factor of safety is unity, and
eters. This is one reason why the magnitude of the factor of
the analysis can then be used to estimate the average shear-
safety commonly adopted in design for bearing capacity dif-
ing resistance along the failure surface or along part of the
fers substantially from the values used in slope stability anal-
failure surface if the shearing resistance is assumed to be
yses. Occasionally, the factor of safety is also taken as the
known along the remainder. Regardless of the specific pro-
ratio of the calculated stress at a point to the allowable stress.
cedure for carrying out the computations, the following
This definition can only be applied when the stress distribu-
principles are common to all methods of limit equilibrium
tion in the slope is evaluated in detail, and it bears no simple
analysis.
relation to the definition adopted in limit equilibrium anal-
A slip mechanism is postulated. This is done without ysis. As noted by Barron (7.2), even the definition given
any major kinematic restriction except that the mechanism here is restrictive because it is based on an assumption for
be feasible and sensible. In the simpler configurations, the the likely stress path to failure. However, the implications
slopes are assumed to fail along planes or circular sliding sur- of whether failure might occur under drained or undrained
faces. When conditions are not uniform, more complex conditions and whether the soil may have a metastable struc-
shapes are known to be appropriate, and analyses have been ture can be taken into account by adjusting the factor of
developed to handle surfaces of arbitrary shape. safety accordingly.
The shearing resistance required to equilibrate the as- Partial safety factors are sometimes used in stability anal-
sumed slip mechanism is calculated by means of statics. The ysis. A simple case is the application of separate factors of
physical concepts used here are that the potential slip mass safety to two strength parameters, for example, c'/F1 and
is in a state of limiting equilibrium and that the failure crite- tan '/F2 , where F1 and F2 have different values. Uncer-
rion of the soil or rock is satisfied everywhere along the pro- tainty about other terms in the stability equation might also
posed surfaces. Various methods differ in the degree to be included by application of partial safety factors or coef-
which the conditions for equilibrium are satisfied, and some ficients to loads applied to the slope, water pressures, and
common methods of analysis violate conditions of static other parameters. This practice has been common in some
equilibrium. This is an important factor in evaluating the parts of Europe (7.29) and recently has been used in proba-
rigor of any method. bilistic methods of design.
The calculated shearing resistance required for equilib- An understanding of the role of the factor of safety is
rium is compared with the available shear strength. This vital in the rational design of slopes. One well-recognized
comparison is made in terms of the factor of safety, which role is to account for uncertainty and to act as a factor of
will be defined more precisely below. ignorance with regard to the reliability of the items that en-
The mechanism with the lowest factor of safety is ter into analysis. These include strength parameters, pore-
found by iteration. For example, if it is assumed that the pressure distribution, and stratigraphy. In general, the lower
slip surface is circular, a search is made for the critical slip the quality of the site investigation is, the higher the factor
circle. When the position of the slip surface is dictated by of safety should be, particularly if the designer has only li-
a dominant weakness, such as sheared clay at residual mited experience with the materials in question. Another
strength, other trials are unnecessary. use of the factor of safety is to provide a measure of the
average shear stress mobilized in the slope. This should not
The factor of safety should be defined as clearly as pos- be confused with the actual stresses. A realistic stress distri-
sible and its role understood. The definition used here is as bution cannot, of course, be calculated from a limit equilib-
follows: rium analysis because stress-strain relations are not used in
the analysis. As indicated earlier, a major role of the factor
The factor of safety is that factor by which the shear strength of safety is that it constitutes the empirical tool whereby de-
parameters may be reduced in order to bring the slope into a formations are limited to tolerable amounts within economic
state of limiting equilibrium along a given slip surface. restraints. In this way, the choice of the factor of safety is
156
greatly influenced by the accumulated experience with a par- Figure 7.1. Geometry used in slip circle analysis
for 0 = 0 case.
ticular soil or rock mass. Since the degree of risk that can be
taken is also much influenced by experience, the actual mag-
nitude of the factor of safety used in design will vary with
material type and performance requirements. These and re-
lated topics influencing the design of stable slopes are dis-
cussed in Chapter 8.
157
pressed as the ratio of the undrained strength (su ) to the ver- analysis. Although the strength of the embankment mate-
tical effective stress (p) under which the material has been rial is neglected, this assumption is often realistic for reasons
consolidated. Although the ground close to the surface may of strain compatibility between the fill and the foundation
have been overconsolidated by desiccation, this will be ne- materials. In any case, the assumption leads to conservative
glected in the analysis presented here. A linear increase in results. The final design charts are too numerous to be in-
undrained strength with depth can be treated in analysis by cluded here.
assuming that the soil consists of a sequence of layers chosen A comprehensive set of design charts for the stability of
to represent the variation of strength with depth to an ap- an embankment on a uniform homogeneous purely cohesive
propriate degree of accuracy. Since this is cumbersome, foundation was prepared by Pilot and Moreau (7.62). In this
Gibson and Morgenstern (727) sought an analytical solution case the strength of the embankment material is taken into
to the problem. They showed that the factor of safety for a account, but the foundation is characterized by a uniform
cut in a normally consolidated soil may be found from the undrained strength. The depth to a rigid stratum is varied
following equation, which is analogous to Equation 7.2: as well. An extensive analysis over the range of all the param-
eters of interest has been undertaken, and the factor of safety
F = N (s. /P)n [1 - Iy)] [7.31 is presented directly in terms of these parameters.
where
Anisotropic
= unit weight of water, and
The undrained strength of a soil is commonly anisotropic.
N = stability factor that depends only on the slope
angle (0).
Figure 7.2. Variation of stability number
with slope angle for cuts in normally
The variation of N with 0 is shown in Figure 7.2. The fac- consolidated clays (7.27).
tor of safety is independent of the height of the slope. In
theory, failure occurs along an infinite number of slip circles
that emerge above the toe, and the need to consider a depth
factor is obviated. In practice, the position of the actual slip
circle will be dictated by some small variation from the as-
sumed strength-depth relation. Further theoretical support
for the use of slip circles for this class of problems has been L====iMWM
.....u•.
provided by Booker and Davis (7.6), who show, from a
more exact analysis based on the theory of plasticity, that
the slip circle analysis provides acceptable results for slopes
steeper than 50 , i.e., within the usual working range.
Hunter and Schuster (7.38) considered the stability of a
iiu••iu
cut in soil with a strength profile shown in Figure 7.3. The
undrained shear strength increases with depth, but there is a
finite intercept at the surface. Both finite and infinite depths
SLOPE ANGLE Idegl
of soil were taken into consideration and showed that
F = (se /p) ('y'/y)N [7.4] Figure 7.3. Idealized strength-depth relation for
normally consolidated clay (7.38).
where y' is the submerged unit weight of the soil and N de-
pends on 0 and M. M is a parameter used to specify the
strength of the soil at the surface of the ground and is given
by
UNDRAINED SHEAR STRENGTH (Si,)
\
M = (h/H)(y/'y') [7.5]
158
This can arise from the development of an anisotropic fabric where N depends on C2/Cl and the slope angle (0). Lo
during deposition and subsequent consolidation. Even if (Z53) undertook the evaluation of N for uniform soils and
anisotropic structure can be ignored, anisotropic states of for profiles where the strength increases with depth. For
stress during consolidation result in anisotropy in the un- soils conforming to Equation 7.6, the factor of safety ob-
drained strength (7.30). Casagrande and Carrillo (7.8) pro- tained by taking anisotropy into account is significantly
posed that the undrained strength varies with direction ac- lower than that from conventional analysis, except for steep
cording to slopes. The design charts presented by Lo allow this to be
assessed readily.
C, = C2 + (Cl - C2 ) cos2 i [ 7.6]
Method of Slices
where
When the surface profile or the stratigraphy is irregular, it is
= angle of deviation from vertical of major principal not convenient to seek an analytical solution. Instead, Equa-
stress at failure, tion 7.1 is evaluated numerically by the method of slices.
C1 = directional strength in i direction, Figure 7.6 shows the geometry of a circular surface of slid-
C1 = strength of vertical specimen, and ing as used in that method. F is found from
C2 = strength of horizontal specimen.
F = ( s 1 Q)/(Wsin a) [7.8]
There is some experimental support for Equation 7.6 and,
since the major principal stress rotates from the vertical to- where i denotes each slice in turn, and Z is the summation
ward the horizontal around a slip circle, design charts for the of all slices. By use of Equation 7.8, an almost arbitrary de-
stability of slopes in anisotropic soils are of practical interest gree of complexity can be considered, including anisotropy
Based on Equation 7.2, and variation of strength with depth. It is common to un-
dertake these analyses with a digital computer and many
F = N(C1/'yH) [7.7] programs exist. For illustrative purposes, the factor of safety
is calculated below for a sample problem. The assumptions
for the problem are that
Figure 7.4. General solution for cuts in unlimited
depths of normally consolidated clays (7.38). The soil is saturated,
No volume change occurs,
N or F (s/p)(-I-j)N -
= (h/HU/)
iN
Figure 7.6. Geometry pertaining to a particular circular
surface of sliding as used in the method of slices.
159
No dissipation of pore pressure occurs, The critical slip circle and the minimum F must be found by
The mode of failure is slip circle, and iteration. Convergence isfastest if 0 is moved horizontally.
The shear strength along the slip surface is the only
factor contributing to resistance. Effective Stress Analysis of Soil Slopes
Figure 7.7 shows the cross section of the failure and the Planar Slip Surface
force diagram.
The factor of safety is determined by the use of Equa- When a soil mass moves predominantly in a translational
tion 7.8; i.e., F is the resisting moment divided by the driv- manner, there is little internal distortion and an infinite
ing moment. The resisting moment is the sum of the mo- slope analysis is often representative. In this analysis, the slip
bilized strength (sn) in each layer times the radius of the surface is assumed to be a plane parallel to the ground sur-
circle, and the driving moment is the sum of the weight of face and the end effects can be neglected. When sliding
each slice times its moment arm from center of rotation. takes place in the mantle of weathered material with small
F can be defined this way only because all terms in the re- ratios of depth to length, this type of analysis is often appro-
sisting moment have s. priate. The mechanics of sliding along a planar slip surface
The calculated dimensions of the slices shown in Figure will be presented in detail, for they are instructive with re-
7.7 are given in Table 7.1. The remaining calculations are gard to all effective stress analyses.
Consider the slope and slip surface shown in Figure 7.8.
Sum of the driving moments = 29 507 kN-m; In an effective stress analysis the pore pressure distribution
Resisting moments = 24(25 x AB + 15 x BC + 35 is assumed to be known, and in this case steady seepage is
xCD)=24(25 x7.5+ 15x4+35 x34)=34500kNm; assumed to be parallel to the surface with the groundwater
and table as indicated. The forces acting on a slice of width b
For this example, F = 34 500/29 507 = 1.17. are shown. Since there is no internal distortion and end ef-
fects are neglected,
Figure 7.7. Cross section and force diagram of example slip
surface failure.
dE=dX=0
(a) SCALED CROSS SECTION
The weight of a slice W is given by
/*
W=7bd [7.10]
5 kPa
8.9 kN/m3 The total normal force (N) is related to the effective force
5 kPa;,' = 19.0 kNIm3 (N') and pore pressure (u) by
5 kPa
0.4 kN/m3
N=N'+uQ [7.11]
)h( FORCE DIAGRAM (typical slice of width b)
Center of Rotation
0 and
1wn
LR uQ('yh)bseca [7.12]
0\
(b
Sc = nrobilized.strength
[1 Slice where h is the piezometric head acting on the slip surface.
The shear force acting along the base of the slice (S) is
on failure surface
Analyses using Equation 7.17 or 7.18 are readily performed Substituting this relation into the equation for overall mo-
by hand. Several cases exist to illustrate the utility of this ment equilibrium results in the following expression for F:
analysis in practice (7.39, 7.66).
F=[c'+(W cos a-uQ) tan ']/(Wsina) [7.24]
Circular Slip Surface
Rotational slides occur in many types of soil, and to analyze Figure 7.8. Forces acting on idealized slice of infinite slope.
slopes in these materials by assuming a circular slip surface is
common practice. Taylor (7.68) presents stability charts b
for uniform slopes characterized by both friction and cohe-
sion. However, since these charts are in terms of total
stresses, they are of limited value. For heterogeneous condi-
tions, the friction circle method of analysis used by Taylor Flow
is less practical than methods that use slices. Therefore, E+dE
x_--
only various methods of slices will be discussed here.
One method of analysis that is accurate for most purposes
is that advanced by Bishop (73). The forces acting on a
T f 1W
rx+dx
typical slice are shown in Figure 7.9. Moment equilibrium
about the center of rotation gives
Wx=SR [7.19]
where
IN
Figure 7.10. Tabular form for computation of factor of safety Figure 7.13. Idealized cases of seepage in slopes
if surface of sliding is circular and interslice forces are for which chart solutions are available (7.37).
neglected (7.71).
I 2 3 4 5 6 7 8
Figure 7.11. Values of m0 for use- in calculation of factor of safety Surface water 4x slope height
if surface of sliding is circular and interslice forces are neglected behind toe of slope
(7.71).
/4
,,,0 'cosa * 5,"saF tO,,
1.2
Surface water 2s slope height
behind toe of slope
' /0
08
06
F=m- nr [7.26]
where m and n are called stability coefficients and are pre-
sented as functions of the parameters and ratios noted above.
This method has been called the Fellenius method, the U.S. The stability coefficients were evaluated over a wide range of
Bureau of Reclamation method, the common method of values for (c'/'yH), ', , and D, and the results were pre-
slices, and the ordinary method. As confirmed in practice sented in both chart and tabular form for ease of use. For a
(7.64), it results in factors of safety that are too low. Al- given problem, reference to the appropriate chart or linear
though simplicity may have been a reasonable argument at interpolation between charts allows one to extract the ap-
one time for using Equation 7.24, the widespread use of propriate values of In and n. The factor of safety is given
computers precludes algebraic complexity as an excuse for directly by Equation 7.26. In many cases that arise in prac-
not adopting methods of analysis that are superior from the tice, ru is not uniform, and an average value has to be evalu-
point of view of mechanics. ated in order to use the stability coefficients. Guidance in
Bishop and Morgenstern (7.5) sought to develop charts calculating an average value of ru is given (7.5). When ru is
that would facilitate the use of Equation 7.21 for problems zero, stability coefficient m gives the same factor of safety
formulated in terms of effective stress. For the uniform as Taylor's charts for a material without cohesion.
slope shown in Figure 7.12, they showed that Equation 7.21 Other stability charts that may be found useful have been
could be cast into the dimensionless form prepared by Hoek and Bray (7.37) for each of the cases
162
Figure 7.14. Slip surface dimensions and slice forces for use in Figure 7.15. Examples of variation of Fm and Ff
Spencer's analysis (7.67). with 0 in Spencer's analysis (7.67).
(a) E c'yH0.02
(0
4 4O O
rO.5
Dotted line passes through
I -.. points of action of
T
interslice forces.
Cli
I Note: 1 m = 3.3 ft.
b/2 b/?
F.I.070
---- -----
-
Fm
0.9
15
E,:
i 0
eo
15 20 25
shown in Figure 7.13. These cases represent failure along a For overall equilibrium, the sum of the horizontal compo-
circular slip surface in a soil mass subjected to steady seepage nents and the vertical components of the interslice forces
and containing a tension crack. The magnitudes of(c'/'yH), must each be zero; that is,
', and 13 are needed to enter a particular chart representing
one of the cases shown in Figure 7.13. The factor of safety [Qcos0] =0 [7.281
is then readily found. The charts are based on a form of fric.
tion circle analysis that tends to underestimate the factor of and
safety. Although they are convenient to use for preliminary
design, final design should be checked with a more rigorous [Q sin 0] =0 [7.29]
analysis.
As pointed out previously, the simplified Bishop method Moreover, if the sum of the moments of the external forces
of analysis is adequate for circular slip surfaces, but it as- about the center of rotation is zero,the sum of the moments
sumes that no shear is mobilized between slices and it does of the interslice forces about the center of rotation is also
not satisfy the condition of force equilibrium in the horizon- zero. For a circular slip surface,
tal direction. Why the method works can be demonstrated
by reference to the method presented by Spencer (7.67). [Qcos(a-0)] = 0 [7.30]
Spencer followed the usual procedures in the method of
slices (Figure 7.14) and argued that the action of the internal If 0 is assumed constant in order to make the problem de-
forces acting on a slice can be replaced by their resultant (Q) terminate, Equations 7.28 and 7.29 reduce to
inclined at 0 to the horizontal and passing through the mid-
point of a thin slice in order to satisfy moment equilibrium. YIQ = 0 [7.31]
In general, both Q and 0 will vary from slice to slice, and
from force equilibrium it can be shown that In either Equation 7.30 or 7.31, the value of F will depend
on the assumed constant value of 0. For all but one value
Q = [(c'b sec a/F) + (tan Ø'/F) of 0, the safety factor determined (Equation 7.31) by force
equilibrium (Ff) will not equal the safety factor determined
x(W cos a-ub seca)-W sinai (Equation 7.30) by moment equilibrium (Fm). This is
I cos(a-0)[l+(tan'/F)tan(a-0)] [7.27] shown in Figure 7.15 by the intersection of curves of F f and
163
Fm plotted against values of 0. This intersection gives the A convenient approximate method of analysis suitable
pair of values of 0 and F that satisfy both equations simul- for hand calculations and sufficiently accurate for many
taneously. If 0 is set equal to zero and substituted into practical purposes is described by Janbu, Bjerrum, and
Equation 7.30, the equation governing Bishop's simplified Kjaernsli (7.43) and Janbu (7.42). This method follows
method (Equation 7.21) is recovered. Therefore, Bishop's the method of slices and is based on a more elaborate pro-
simplified method represents the solution for F m with 0 cedure described in detail in the latter reference. The factor
equal to zero. In general, the equation governing moment of safety is given by
equilibrium is rather insensitive to variations in 0, and errors
of only a few percent result. For the assumption that 0 F = f0 Z [[c'b + (W - ub) tan çb']/cos a ma ]]
equals zero used in the Bishop simplified method, these er- (Wtana) [7.33]
rors are not conservative with respect to actual values of 0
greater than zero. where in,,, is defined in Equation 7.22 and used as shown in
Figure 7.10. The term f0, which is a correction factor for
Noncfrcular Slip Surface the role of the internal forces, is a function of the curvature
of the slip surface and the type of soil; recommended values
When the distribution of shearing resistance within an earth for this factor are given in Figure 7.16.
mass becomes nonuniform, slip can occur along surfaces Another approximation procedure for analyzing slip along
more complex than a circle. Since the shape of the failure a noncircular sliding.surface is the wedge analysis. In this
surface will be controlled to a large degree by the departure
method the potential sliding mass is separated into a series of
from uniformity, one may wish to analyze the stability along wedges, and the equilibrium of each wedge is considered in
surfaces of arbitrary shapes. Noncircular analysis has proved turn. Only the conditions of horizontal and vertical equilib-
useful in a number of cases of actual slides (7.22, 7.40). rium are used in the analysis, which is often performed in
Morgenstern and Price (757) developed a method of anal- a graphical manner. Moreover, an assumption must be made
ysis that treats a slip surface of arbitrary shape and satisfies regarding the inclination of the force transmitted across the
all equilibrium requirements. To make the analysis statically interface between any two wedges. The forces utilized in
determinate, a relation between the internal forces is as- this method of analysis are shown in Figure 7.17 for a rock
sumed of the form mass separated into two wedges; the notation is defined be-
low.
X=Xf(x)E [7.32]
W1 ,W2 = weight of wedge,
where U1 ,U2 = resultant water pressure acting on base of wedge,
N1 ,N2 = effective force normal to base,
X = factor to be determined in the solution, and T1 ,T2 = shear force acting along base of wedge,
f(x) = arbitrary function concerning the distribution of L1 ,L2 = length of base,
internal forces. a1 ,a2 = inclination of base to horizontal,
Pw12 = resultant water pressure at interface,
The choice of f(x) is limited by conditions of physical ad- P12 = effective force at interface, and
missibility, which require that no tension be developed in 6 = inclination of P12 to horizontal.
the earth mass and that the failure criterion not be violated.
Subject to these conditions, the factor of safety is rather in- Although the rock mass shown in Figure 7.17 is separated
sensitive to variations in f(x). A digital computer is needed into two wedges, the analysis can be extended readily to
to solve resulting Equations 7.30 and 7.31. The Spencer any number of wedges. The factor of safety is varied so
analysis of a slip circle discussed previously is equivalent to that the force polygons constructed for each of the wedges
setting f(x) equal to unity. satisfy horizontal and vertical equilibrium. A value of 6 is
a2-f 2LJ2
W2
WI
Ratio d/L—
164
assumed, and an initial value of F is taken; this allows the Steady seepage through a soil is governed by the Laplace
force polygon for the first wedge to be constructed, as equation, for which a variety of methods are available to
shown by the dashed line in Figure 7.17. With the value of find solutions subject to appropriate boundary conditions
P12 obtained in this manner, the force polygon for the sec- (7.9, 7.31). These techniques are equally useful in the de-
ond wedge is readily constructed. This will not close, in sign of drainage measures if the influence of the design on
general, and the value of F must be varied until it does. The the magnitude and distribution of the pore pressures must
final solution also is given in Figure 7.17. be ascertained. In some cases, analytical methods will have
The factor of safety is sensitive to the assumed value of the capability to provide solutions in closed form, but more
6. Conservative values of the factor of safety may be found generally the problems that arise in practice are analyzed by
by setting 6 equal to zero. The maximum possible value of constructing flow networks graphically by the use of elec-
6 is that value which is compatible with failure in the direc- tric analogue techniques or numerically by means of the
tion of the interface, but this usually leads to an overesti- finite element method (7.33, 7.46, 7.70).
mate of the factor of safety. Experience indicates that put- An understanding of hydrogeologic controls is important
ting 6 between 100 and 150 usually gives reasonable results. in the solution, either by analysis or by observation, of any
steady seepage problem. In particular, it is essential to rec-
Pore-Pressure Distribution ognize the control that regional groundwater systems have
on the local pore-pressure distribution in a natural slope or
To perform a stability analysis in terms of effective stress re- excavation. The recognition of groundwater discharge areas
quires that the pore pressures be specified. A distinction is and an identification of the recharge areas are a starting
made between problems of undrained and drained loading. point for almost all seepage studies, either computational or
In the first case, the magnitude of the pore pressure is influ- observational. High pore pressures are often found in dis-
enced to a large degree by the changes in total stress to charge areas associated with topographic lows, and they may
which the earth mass is subjected; in the second case, the have a controlling influence on the stability of slopes. Anal-
pore-pressure distribution is controlled by steady seepage yses of regional groundwater systems have been made by
conditions. When an analysis of a soil loaded under un- Toth (7:72) and Freeze and Witherspoon (7.25, 7.26), and
drained conditions is made in terms of effective stress, the their significance for slope stability problems is noted by
pore pressures must be either predicted or measured. The Patton and Deere (7.61).
confidence level for predicting pore pressures in cuts is not Heterogeneity of soil deposits has a marked influence on
high; however, it is somewhat better in the case of founda- the pore-pressure distribution under steady seepage condi-
tions beneath embankments because the total stress change tions. Kenney and Chan (7.47) describe a detailed field
is less influenced by the initial state of stress in the ground. study of the anisotropy of the permeability in a varved soil.
For a clay soil, the initial stresses in the ground are not usu- The ratio of the permeability of flow parallel to the varves to
ally known to a high degree of accuracy. In general, the that perpendicular to the varves was found to be less than
pore pressure (u) that is used in a stability analysis is given five. Although this is less than might be expected, it still
by exerts a significant influence on the pressure distribution
that develops during steady seepage conditions. Eigenbrod
[7.341 and Morgenstern (7.22) report a striking example of the
influence of geologic detail on the pore-pressure distribu-
where tion within a slope about 20 in (70 ft) high in Cretaceous
bedrock overlain by till. This slope was shown to possess at
u0 = pore pressure before any stress change, and least three water tables, the upper two of which were
Au = change in pore pressure due to change in total perched on bentonitic layers. Moreover, no pore pressures
stress. acted along much of the failure surface, since it lay on a
fractured free-draining coal layer. Careful field observations
The change in pore pressure (Au) can be estimated approxi- guided by an appreciation for the role of geologic details are
mately if the pore-pressure coefficients and the changes in needed in cases like this if misleading results are to be
total stresses are known. In turn, the change in total stress
avoided.
at a point may be estimated from stress analyses. Linear
elasticity is commonly assumed; therefore, the results are of
limited applicability, particularly if the factor of safety ap- Analysis of Rock Slopes
proaches unity and a substantial volume is permitted to
yield. In the case of cuts, the total stresses decrease so that The analysis of the stability of a rock slope is normally un-
with time the pore pressures increase from their end-of- dertaken in terms of effective stress because the permeability
construction values, thereby resulting in reduction of the of rock masses is usually so high that undrained loading does
factor of safety. not arise. To perform an analysis requires that the shear
Although it is not common to predict pore-pressure strength parameters, the pore-water pressures, and the slid-
changes in slopes due to changes in total stress, it is common ing mechanism be specified. In hard rocks, the sliding mech-
to measure them on major projects and to use the data di- anism is controlled by structural features, and a detailed
rectly in stability analyses. Instrumentation for measuring study of those features is a prerequisite for rational design
pore pressures is discussed in ChapterS. In the case of em- of a rock slope. Further guidance on the evaluation of rock
bankments on soft impervious foundations, monitoring pore structure characteristics is given in Chapter 9.
pressures is almost routine. However, this monitoring may The analysis of a rock slope in terms of a factor of safety
not be adequate to avert a slip (7.1). is a subordinate activity to achieving a clear understanding
of the controlling geology and water-pressure configuration. condition for toppling is dictated by the position of the
In the case of slopes in soil, the confidence limits attached weight vector of the block relative to the base of the block.
to the application of the total process of investigation, test- If the weight vector passing through the center of gravity of
ing, and analysis are determined by the successful explana- the block falls outside the base of the block, toppling will
tion of well-documented case histories (7.66). However, in occur; this is shown in Figure 7.18 (7.37).
the case of rock slides, few such case histories report com- A common case of instability in rock slopes is sliding
pletely the geologic configuration, give good quality test along a planar surface. The analysis of this case is a simple
data, and have been subjected to an acceptable analysis. extension of the stability of a rough rigid block on an in-
Moreover, the shearing resistance in rocks is often sensitive clined plane, and, for the case shown in Figure 7.19a, we
to movements. Small movements can result in substantial have
decreases in available shearing resistance and hence in the
factor of safety. Therefore, the confidence limits of design F=(2c/yH)P+[Q cot i-.R(P+S)]tanØ
for a given factor of safety are generally lower for rock
slopes than for soil slopes. In other words, rock slopes de- - (Q+RS cot [7.35]
signed for a given factor of safety have a higher degree of
risk associated with them than most soil slopes designed for where
the same factor of safety. If the degree of risk is to be the
same, it is prudent to design rock slopes for higher factors of P = [1 -(Z/H)] cosec Op [7.36]
safety, unless only the residual shearing resistance is being
mobilized. However, design for a higher factor of safety is [7.37]
not always economically feasible.
The shearing resistance of rock is more variable than that R (yw/'y) (Z/Z) (Z/H),and [7.38]
of soil, and the same is true of the water-pressure distribu-
tion. Both the strength and the water-pressure distribution S = (Z/Z) (Z/H) sin [7.39]
are dominated to a large degree by the pattern of discontin-
uities in the rock mass. In the case of strength, the usual But if the tension crack is in the surface above the slope
(Figure 7.19h),
procedures for investigation involve in situ testing or sam-
pling and testing, as discussed in Chapter 4. The water-
pressure distribution can only be evaluated by field investi- Q=[[1 - (Z/H)]2 cos Op
gations coupled with an appreciation of the hydrogeologic (cot O p tan of - 1)) [7.40]
constraints.
Morgenstern (7.55) reviews the influence of ground- should be used. All other terms are as shown in the figure.
water on the stability of rock slopes and shows how flow Hock and Bray (7.37) undertook a systematic study of the
through a discrete network of discontinuities can be re- influence of the various parameters that enter into Equation
placed by flow through an imaginary medium of equivalent 7.35 and prepared design charts to facilitate its use. If a
permeability. If the crack spacing is small compared to the planar mechanism is appropriate, it is also possible to assess
size of the discontinuity, conventional procedures for solving the influence of drainage and other stabilizing measures,
problems of flow through porous media are applicable. If such as berms or cables, by using simple extensions of Equa-
this is not the case, the attitudes and hydraulic conductivi- tion 7.35.
ties of the actual discontinuities should be considered (Z54, Often two or three sets of discontinuities intersect, and
7. 74). This is difficult to do, and considerable judgment is sliding of a wedge of rock becomes possible. In general, dif-
required to properly evaluate the water pressure distribution. fering amounts of friction and cohesion can act on the
Patton and Deere (7.61) stress the large local differences in planes, and the water pressure acting normal to the surfaces
water pressures that can arise in a jointed rock mass and the bounded by the planes should be taken into consideration.
large fluctuations that can develop as a consequence of rain- A comprehensive analysis of the mechanics of wedge failure
fall. They also note the significant effects that faults can is given by Hendron, Cording, and Aiyer (7.32). Figure 7.20
have on groundwater conditions. Those effects can be either shows a tetrahedron bounded by two base planes, which
beneficial or deleterious, depending on the attitude and char- may be intersecting joint sets. Failure may occur by sliding
acteristics of the fault gouge. Groundwater flow around along the line of intersection of the two planes or by sliding
faults and other similar features requires special study in on either one of the two planes. General procedures have
slope-stability analyses. The evaluation of water pressures in been developed for determining the factor of safety graph-
a rock mass is made more difficult by the sensitivity of the ically by vector analysis by the use of stereo networks and
hydraulic conductivity to small deformations. A small numerically by hand or by computer (7.32, 7.37). The
amount of slip along a discontinuity can result in a dispro- following steps are included in the analysis.
portionate change in hydraulic conductivity (7.65). Although
methods of analysis can be developed to take these effects The intersections of the various joint sets with one an-
into consideration (7.7, 7.56), the information needed to other and with the slope face are inspected to determine the
conduct these analyses is only rarely available. tetrahedra that may be potential failure wedges; these are
Stability analyses are usually concerned with sliding. then analyzed in detail.
However, discrete blocks of rock can also rotate outward The forces tending to disturb the equilibrium of the
or topple. A tendency to topple depends on the shape of wedge are added vectorially to give the resultant driving
the block and is usually readily discernible in the field. The force. These disturbing forces arise from the weight of the
Figure 7.18. Conditions wedge, any external load applied to the wedge, and the pore-
for sliding and toppling water forces acting on the various faces.
of rock block on
The mode of failure is then determined. For example,
inclined surface (7.37).
a wedge supported on two base planes can either slide along
the line of intersection of the two planes, slide on either
plane, or rotate on either plane. The kinematics of failure
will depend on the orientation of the disturbing force in re-
lation to the orientation of the supporting planes. Various
kinematic tests can be defined (7.32).
After the mode of failure is determined, the maximum
shearing resistance that can be mobilized in the direction of
movement is compared to the shearing forces necessary for
-0I---- - equilibrium in order to compute a factor of safety.
Stable blocb
Hendron, Cording, and Aiyer (7.32) point out that, in
—:— SlngOMy - the common case of a wedge resting on two base planes and
b>Tand
Th acted on only by its own weight, sliding will occur along the
-}-- - -
line of intersection of the two planes if a line drawn down
b/h > Tan , the dip in both planes tends to intersect the line of intersec-
tion. If in either one of the planes a line drawn down the
dip is directed away from the line of intersection, sliding will
_ 4
ding & toppling
occur on that plane only and the wedge will move away.
from the line of intersection. If a wedge is acted on only by
___ Topplrng only
b/h<Tan its own weight, it will slide down the maximum dip if sliding
0
occurs on one plane, and the factor of safety can be readily
0 0 20 30 40 50 60 70 80 90
BASE PLANE ANGLE (deg)
computed. If sliding occurs along the line of intersection in
the absence of pore pressures, the angle of shearing resistance
required for equilibrium will always be equal to or less than
Figure 7.19. Rock-slope failure due to sliding along single the slope of the line of intersection. The actual factor of
discontinuity (7.37). safety will depend on the relative attitudes of bounding
planes. Stability charts for some frictional wedges without
Tension crack in slul,e face water-pressure loading and with sliding along the line of in-
tersection have been prepared by Hoek and Bray (737).
cx~~
Stereographic projection is a most useful tool for the anal-
ysis of the stability of rock wedges. Both friction and cohe-
sion as well as water pressures can be taken into account.
Designers concerned with rock slopes are well advised to be-
come familiar with these techniques, which are described in
the references cited earlier, as well as by John (7.44), Heuze
and Goodman (734), and 1-loek (7.36).
(b) Some rock slides can be treated as two-dimensional prob-
57— surface of slope
lems on noncircular surfaces. Figure 7.21 shows the failure
slope
face
surface of the famous Frank slide (7.13), an example of this
type of slide; the Vaiont slide is another example (759).
Pictorial view of
wedge failure
167
These cases are amenable to analysis by the use of the same ited knowledge of the in situ stresses before excavation.
methods adopted for noncircular slides in soil. Under special conditions, particularly in analysis of dams and
embankments of compacted earth fill, the method is being
DEFORMATION ANALYSIS used successfully (7.51); but the deformation analysis of nat-
ural soils is primarily a research activity at present and has
The use of deformation analysis in the design of slopes is an little practical application. A review of the principles of the
attractive concept; however, practical applications of defor- finite element method is nevertheless appropriate because
mation analysis are limited at present. Until the last decade, some practical applications are being made at present and
this limitation was imposed by the absence of suitable analyt- progress in developing practical applications in the future is
ical methods to perform deformation analysis. Few elastic anticipated.
solutions were available for boundary value problems simi- The finite element method has been applied to geotech-
lar to slopes, and the solutions available were not well suited nical engineering problems since the late 1960s, having been
to practical design. The development of the finite element developed a decade earlier for applications in structural en-
method has largely eliminated this impediment. As engineers gineering and continuum mechanics. Since that time, ex-
have taken advantage of this analytical tool, however, other tensive literature has developed, including several texts. An
equally vexing problems have been recognized, and these important and useful reference and summary of this work
still severely limit the practical application of the finite ele- was compiled by Desai (7.16); this volume is recommended
ment method in most slope design problems. as a starting point for those interested in the method.
In principle, a deformation analysis, particularly an anal- The underlying principle of the finite element method is
ysis utilizing the finite element method, must include the fol- that the behavior of a complex continuum can be approxi-
lowing characteristics. mated by the collective behavior of parts representing the
continuum. The parts are selected so that their individual
The stress field must satisfy equilibrium at every point. behaviors can be described simply. By requiring adjacent
The prediction of this equilibrium stress field is usually done elements to behave similarly at selected points of contact
by use of elastic theory to describe the stresses and deforma- or nodes, the overall continuum is modeled.
tions, but plasticity or other material models might also be In a typical application (Figure 7.22a), an embankment
used. To predict a stress field requires that the stress-strain deformation analysis begins by using a group of simple ele-
relation for the soil be known. ments to approximate the fill. In a two-dimensional anal-
Boundary conditions of stress and deformation must ysis, either triangular or quadrilateral elements are commonly
be satisfied. These characteristics can be contrasted to the used. For a two-dimensional analysis, equations describing
principles of limit equilibrium noted in the earlier section on the behavior of these elements in plane strain are used in con-
principles of limit equilibrium analysis. trast to the plane stress formulation often used for problems
in structural engineering (7J8). A three-dimensional prob-
The lack of success in applying deformation analysis in lem can be modeled by using solid elements. For each ele-
the design of slopes can be attributed directly to the difficul- ment in the model, a series of equations and an appropriate
ties associated with the stress-strain relation noted above. stress-strain relation, such as elasticity, are used to relate the
Soil is generally nonlinear, nonuniform, inelastic, and aniso- application of loads to deformation at the nodes. In the di-
tropic. Each of these characteristics must be idealized and rect method of element formulation (Figure 7.22b), a tri-
used in a deformation analysis, and the difficulty in describ- angular element with nodes only at the corners can be
ing natural soil deposits in these terms is the major factor used. Other elements in the problem are formulated in a
that limits application of the finite element method in the similar fashion. The deformation of adjacent elements is
analysis of slopes. The problem is aggravated by our lim- then constrained to be identical at the nodal points. There
Figure 7.22. Finite element idealization of (a) FINITE ELEMENT REPRESENTATION OF EMBANKMENT CROSS SECTION
embankment.
II >
FZ2
is no requirement that deformations or stress be compatible instability in some programs. When volume changes do oc-
between other than nodal points on adjacent elements in cur (for example, during drainage or in partially saturated
this example. As a result, this solution would be only an ap- soils), the variation in Poisson's ratio can be related experi-
proximation of the actual system behavior. A more accurate mentally to stress changes (7.50) or to the coefficient of
approximation might result by adding nodes at the midpoint earth pressure at rest (K0). Consolidation effects during
of each side of the element. Variational methods of ele- loading can also be considered (7.10, 7.41). Some advan-
ment formulation also add to the power, accuracy, and tages have been shown (7.11) for using the shear modulus
flexibility of the procedure (7.1 7). (G) and bulk modulus (K) instead of Young's modulus
After a model of the embankment is formulated, appro- (E) and Poisson's ratio (v) for formulating elastic finite
priate loading and deformation boundary conditions are im element problems in soil.
posed. In geotechnical engineering problems, one of the During deformation analysis, it is common to exceed the
important loads is the self-weight of the material, which can available strength of soil at specific points long before over-
be represented by applying a vertical force at each node all collapse of the slope is imminent. This presents the prob-
(7.21). Several studies have shown, however, that this lem of modeling local yield. Since many soils exhibit strain-
method has limitations in predicting deformations. An al- softening behavior after yield (Chapter 6), it is appropriate
ternate approach is to perform a sequential analysis in which to model this by use of finite element methods. Höeg (7.35)
the problem is solved in successive stages to represent a series describes one method using an elastic-plastic model, and
of steps in construction or excavation (7.20). several other techniques have been proposed (Z17).
The procedures described above lead to the assembly of Modeling natural deposits of rock and heavily overcon-
a large number of simple equations, and the solution of solidated clay requires consideration of joints and similar
these equations requires use of a digital computer. Finite discontinuities. These features can experience shear trans-
element programs applicable to geotechnical engineering lation along the joints, as well as dilatancy and opening of
problems are readily available from several sources and the the joints during movement of the slope. The application
literature. Because continuing improvements are made in of the finite element method to rock (7. 75) and particularly
these programs, no specific references are given here. to dilating jointed rock (7.28) has relevance to slope-stability
problems.
The use of finite element methods to predict the stability
Other characteristics of the constitutive behavior of soil
of slopes does not offer any improvement over accurate lim-
and rock have been considered for application to finite ele-
it equilibrium methods (7.19). Its usefulness lies in defor- ment deformation analysis. These include anisotropy (7.14),
mation analysis, in which accuracy is directly related to the
creep, and tensile cracking (7.23). As greater and greater
stress-strain behavior of the soil. Active research continues
complexity is added to the modeling parameters, the cost of
to be directed toward this subject. Modeling the complex
the finite element analysis increases, often beyond practical
behavior of soil has progressed from initial efforts using
limits. However, the major problem limiting the application
linear elasticity to include many of the more complicated
elements of the constitutive relation. Linear elasticity is al- of deformation analysis to natural slopes and cuts remains
most always a poor model of actual soil behavior, but it does the heterogeneous and complex nature of the materials in-
have the advantages of low cost and simple application volved.
(7.19). In some cases, simple linear models can be used to
determine qualitatively the form of particular slope or em- REFERENCES
bankment deformations. However, this method can lead to 7.1 Al-dhahir, Z., Kennard, M. F., and Morgenstern, N. R.
significant errors and should be used with caution. Another Observations on Pore Pressures Beneath the Ask Lagoon
application of linear elasticity is described by Cole and Embankments at Fiddler's Ferry Power Station. In In
Situ Investigations in Soils and Rocks, British Geotechni-
Burland (7.12), who used finite element analysis to extrap- cal Society, London, 1970, pp. 26 5-276.
olate from measured field performance. They used field 7.2 Barron, R. A. Correspondence. Geotechnique, Vol. 14,
measurements to calibrate their analysis and by trial and er- No.4, 1964, pp. 360-361.
ror selected the best values of the linear stress-strain relation 7.3 Bishop, A. W. The Use of the Slip Circle in the Stability
to match their measurements. These constitutive relations Analysis of Slopes. Geotechnique, Vol. 5, No. 1, 1955,
pp. 7-17.
were then used to predict subsequent loads and deforma- 7.4 Bishop, A. W., and Bjerrurn, L. The Relevance of the
tions during further construction involving the same mate- Triaxial Test to the Solution of Stability Problems. Proc.,
rial. Research Conference on Shear Strength of Cohesive Soils,
Boulder, American Society of Civil Engineers, New York,
Several other forms of the basic stress-strain relation have 1960, pp. 437-501.
been proposed (7.19). Most methods deal with nonlinear 7.5 Bishop, A. W., and Morgenstern, N. R. Stability Coeffi-
elastic forms, such as bilinear (7.20) or hyperbolic (7.49). cients for Earth Slopes. Geotechnique, Vol. 10, No. 4,
Multilinear and spline functions (7.15, 7.23) have been used 1960, pp. 129-150.
.7.6 Booker, J. R., and Davis, E. H. A Note on a Plasticity So-
to define the constitutive relation in even greater detail. The lution to the Stability of Slopes in Horn ogenous Clays.
theory of plasticity (7.63) offers great potential through the Geotechnique, Vol. 22, No.4, 1972, pp. 509-513.
use of yield functions to describe nonlinear effects, together 7.7 Brekke, T. L., Noorishad, J., Witherspoon, P. A., and
with the coupling of volume changes to stress changes. In Maini, Y. N. T. Coupled Stress and Flow Analysis of Frac-
the use of elastic theory, volume changes are modeled by a tured Dam Foundations and Rock Slopes. Proc., Sympo-
sium on Percolation Through Fissured Rock, International
variable Poisson's ratio. Because of the relative incompressi- Society of Rock Mechanics and International Association
bility of undrained saturated soils, Poisson's ratio has a con- of Engineering Geology, Stuttgart, Deutsche Gesellschaft
stant value of approximately 0.5. This number may be für Erd- und Grundbau, Essen, West Germany, 1972, pp.
changed slightly (e.g., to 0.49) to eliminate computational T4-J, 1-8.
169
7.8 Casagrande, A., and Carrfflo, N. Shear Failure of Aniso- 7.26 Freeze, R. A., and Witherspoon, P. A. Theoretical Anal-
tropic Materials. In Contributions to Soil Mechanics: ysis of Regional Groundwater Flow: 3—Quantitative In-
1941-1953, Boston Society of Civil Engineers, 1944, pp. terpretations. Water Resources Research, Vol. 4, No. 3,
122-135. 1968, pp. 581-590.
7.9 Cedergren, H. R. Seepage, Drainage, and Flow Nets. 7.27 Gibson, R. E., and Morgenstern, N. R. A Note on the
Wiley, New York, 1967, 489 pp. Stability of Cuttings in Normally Consolidated Clays.
7.10 Christian, J., and Boehmer, J. W. Plane Strain Consolida- Geotechnique, Vol. 12, No. 3, 1962, pp. 212-216.
tion by Finite Elements. Journal of Soil Mechanics and 7.28 Goodman, R. E., and Dubois, J. Duplication of Dilatancy
Foundations Division, American Society of Civil Engineers, in Analysis of Jointed Rocks. Journal of Soil Mechanics
New York, Vol. 96, No. SM4, 1970, pp. 143 5-1457. and Foundations Division, American Society of Civil En-
7.11 Clough, R. W., and Woodward, R. J. Analysis of Embank- gineers, New York, Vol. 98, No. SM4, 1972, pp. 399-422.
ment Stresses and Deformations. Journal of Soil Mechanics 7.29 Hansen, J. B. The Philosophy of Foundation Design: De-
and Foundations Division, American Society of Civil Engi- sign Criteria, Safety Factors, and Settlement Limits. In
neers, New York, Vol. 93, No. SM4, 1967, pp. 529-549. Proc. (Vesic, A. S., ed.), Symposium on Bearing Capacity
7.12 Cole, K. W., and Burland, J. B. Observations of Retaining and Settlement of Foundations, Duke Univ., Durham,
Wall Movements Associated With A Large Excavation. N.C., 1965, 1967, pp. 9-14.
Proc., 5th European Conference on Soil Mechanics and 7.30 Hansen, J. B., and Gibson, R. E. Undrained Shear Strengths
Foundation Engineering, Spanish Society for Soil Me- of Aniso tropically Consolidated Clays. Geotechnique, Vol.
chanics and Foundations, Madrid, 1972, pp. 327-342. 1, No. 3, 1949, pp. 189-204.
7.13 Cruden, D. M., and Krahn, J. A Reexamination of the 7.31 Harr, M. E. Groundwater and Seepage. McGraw-Hill, New
Geology of the Frank Slide. Canadian Geotechnical Jour- York, 1962, 315 pp.
nal, Vol. 10, No. 4, 1973, pp. 581-591. 7.32 Hendron, A. J., Jr., Cording, E. J., and Aiyer, A. K. Ana-
7.14 D'Appolonia, D. J., and Lambe, T. W. Method for Predict- lytical and Graphical Methods for the Analysis of Slopes
ing Initial Settlement. Journal of Soil Mechanics and in Rock Masses. U.S. Army Engineer Nuclear Cratering
Foundations Division, American Society of Civil Engineers, Group, Livermore, Calif., Technical Rept. 36, 1971.
New York, Vol. 96, No. SM2, 1970, pp. 523-544. 7.33 Herbert, R., and Rushton, K. R. Ground-Water Flow
7.15 Desai, C. S. Nonlinear Analyses Using Splice Functions. Studies by Resistance Networks. Geotechnique, Vol. 16,
Journal of Soil Mechanics and Foundations Division, Amer- No. 1, 1966, pp. 53-75.
can Society of Civil Engineers, New York, Vol. 97, No. 7.34 Heuze, F. E., and Goodman, R. E. Three-Dimensional
5M1, 1971, pp. 1461-1480. Approach for Design of Cuts in Jointed Rock. In Stability
7.16 Desai, C. S., ed. Applications of the Finite Element of Rock Slopes (Cording, E. J., ed.), Proc., 13th Sympo-
Method in Geotechnical Engineering: A Symposium. U.S. sium on Rock Mechanics, Univ. of Illinois at Urbana-
Army Engineer Waterways Experiment Station, Vicksburg, Champaign, American Society of Civil Engineers, New
Miss., 1972, 1227 pp. York, 1972, pp. 397-404.
7.17 Desai, C. S. Theory and Applications of the Finite Ele- 7.35 Hdeg, K. Finite Element Analysis of Strain-Softening Clay.
ment Method in Geotechnical Engineering. In Application Journal of Soil Mechanics and Foundations Division, Amer-
of the Finite Element Method in Geotechnical Engineering: ican Society of Civil Engineers, New York, Vol. 98, No.
A Symposium (Desai, C. S., ed.), U.S. Army Engineer SM1, 1972, pp. 43-58.
Waterways Experiment Station, Vicksburg, Miss., 1972, pp. 7.36 Hoek, E. Methods for the Rapid Assessment of the Stabil-
3-90. ity of Three-Dimensional Rock Slopes. Quarterly Journal
7.18 Desai, C. S., and Abel, J. F. Introduction to the Finite of Engineering Geology, Vol. 6, No. 3, 1973, pp. 243-255.
Element Method: A Numerical Method for Engineering 7.37 Hoek, E., and Bray, J. W. Rock Slope Engineering. Insti-
Analysis. Van Nostrand-Reinhold, New York, 1972, 477 tution of Mining and Metallurgy, London, 1974, 309 pp.
pp. 7.38 Hunter, J. H., and Schuster, R. L. Stability of Simple Cut-
7.19 Duncan, J. M. Finite Element Analysis of Stresses and tings in Normally Consolidated Clays. Geotechnique, Vol.
Movements in Dams, Excavations, and Slopes. In Appli- 18, No. 3, 1968, pp. 372-378.
cation of the Finite Element Method in Geotechnical En- 7.39 Hutchinson, J. N. A Landslide on a Thin Layer of Quick
gineering: A Symposium (Desai, C. S., ed.), U.S. Army Clay at Furre, Central Norway. Geotechnique, Vol. 11,
Engineer Waterways Experiment Station, Vicksburg, Miss., No. 2, 1961, pp. 69-94.
1972, pp. 267-326. 7.40 Hutchinson, J. N. A Reconsideration of the Coastal Land-
7.20 Dunlop, P., and Duncan, J. M. Development of Failure slides at Folkestone Warren, Kent. Geotechnique, Vol. 9,
Around Excavated Slopes. Journal of Soil Mechanics and No. 1, 1969, pp. 6-38.
Foundations Division, American Society of Civil Engineers, 7.41 Hwang, C. T., Morgenstern, N. R., and Murray, D. W. Ap-
New York, Vol. 96, No. SM2, 1970, pp. 47 1-493. plication of the Finite Element Method to Consolidation
7.21 Dunlop, P., Duncan, J. M., and Seed, H. B. Finite Element Problems. In Application of the Finite Element Method
Analyses of Slopes in Soil. Department of Civil Engineer- in Geotechnical Engineering: A Symposium (Desai, C. S.,
ing, Univ. of California, Berkeley, Rept. TE-68-3, 1968. ed.), U.S. Army Engineer Waterways Experiment Station,
7.22 Eigenbrod, K. D., and Morgenstern, N. R. A Slide in Vicksburg, Miss., 1972, pp. 739-765.
Cretaceous Bedrock at Devon, Alberta. In Geotechnical 7.42 Janbu, N. Slope Stability Computations. In Embankment-
Practice for Stability in Open Pit Mining (Brawner, C. 0., Dam Engineering (Hirschfeld, R. C., and Poulos, S. J.,eds.),
and Milligan, V., eds.), Society of Mining Engineers, Amer- Wiley, New York, 1973, pp. 47-86.
ican Institute of Mining, Metallurgical and Petroleum En- 7.43 Janbu, N., Bjerrum, L., and Kjaernsli, B. Soil Mechanics
gineers, New York, 1972, pp. 223-23 8. Applied to Some Engineering Problems. Norwegian Geo-
7.23 Eisenstein, Z., Krishnayya, A. V., and Morgenstern, N. R. technical Institute, Oslo, PubI. 16, 1956, pp. 5-26.
An Analysis of the Cracking at Duncan Dam. Proc., Spe- 7.44 John, K. W. Engineering Analysis of Three-Dimensional
cialty Conference on Performance of Earth and Earth- Stability Problems Utilizing the Reference Hemisphere.
Supported Structures, Purdue Univ., Lafayette, md., Proc., 2nd Congress, International Society of Rock Me-
American Society of Civil Engineers, New York, Vol. 1, chanics, Belgrade, Vol. 3, 1970, pp. 385-391.
Part 1, 1972, pp. 765-778. 7.45 Kankare, E. Failures at Kimola Floating Canal in Southern
7.24 Esu, F. Short-Term Stability of Slopes in Unweathered Finland. Proc., 7th International Conference on Soil Me-
Jointed Clays. Geotechnique, Vol. 16, No. 4, 1966, pp. chanics and Foundation Engineering, Mexico City, Vol. 2,
321-328. 1969, pp. 609-6 16.
7.25 Freeze, R. A., and Witherspoon, P. A. Theoretical Anal- 7.46 Kealy, C. D., and Busch, R. A. Determining Seepage Char-
ysis of Regional Groundwater Flow: 1—Analytical and acteristics of Mill-Tailings Dams by the Finite Element
Numerical Solutions to the Mathematical Model. Water Method. U.S. Bureau of Mines, Rept. RI 7477, 1971, 113
Resources Research, Vol. 2, No. 4, 1966, pp. 641-656. pp.
170
7.47 Kenney, T. C., and Chan, H. T. Field Investigation of Per- Open Pit Mining (Brawner, C. 0., and Milligan, V., eds.),
meabifity Ratio of New Liskeard Varved Soil. Canadian Society of Mining Engineers, American Institute of Min-
Geotechnical Journal, Vol. 10, No. 3, 1973, pp. 473488. ing, Metallurgical and Petroleum Engineers, New York,
7.48 Kenney, T. C., and Uddin, S. Critical Period for Stability 1971, pp. 23-48.
of an Excavated Slope in Clay Soil. Canadian Geotechni- 7.62 Pilot, G., and Moreau, M. La Stabllité des Remblais Sur
cal Journal, Vol. 11, No. 4, 1974, pp. 620.623. Sols Mous. Eyrolles, Paris, 1973.
7.49 Kondner, R. L., and Zelasko, J. S. A Hyperbolic Stress- 7.63 Prevost, J-H., and Hdeg, K. Effective Stress-Strain-Strength
Strain Formulation for Sands. Proc., 2nd Pan-American Model for Soils. Journal of Geo technical Engineering Di-
Conference on Soil Mechanics and Foundations Engineer- vision, American Society of Civil Engineers, New York,
ing, Brazil, American Society of Civil Engineers, New York, Vol. 101, No. GT3, 1975, pp. 259-278.
Vol. 1, 1963, pp. 289-324. 7.64 Sevaldson, R. A. The Slide in Lodalen, October 6, 1954.
7.50 Kulhawy, F. H., and Duncan, J. M. Stresses and Movements Geotechnique, Vol. 6, No. 4, 1956, pp. 167-1 82.
in Oroville Dam. Journal of Soil Mechanics and Founda- 7.65 Sharp, J. C., and Malni, Y. N. T. Fundamental Considera-
tions Division, American Society of Civil Engineers, New tions on the Hydraulic Characteristics of Joints in Rock.
York, Vol. 98, No. SM7, 1972, pp. 653-655. Proc., Symposium on Percolation Through Fissured Rock,
7.51 Lefebvre, G., Duncan, J. M., and Wilson, E. L. Three- Stuttgart, International Society of Rock Mechanics and
Dimensional Finite Element Analysis of Dams. Journal of International Association of Engineering Geology,
Soil Mechanics and Foundations Division, American Deutsche Geseilschaft für Erd- und Grundbau, Essen, West
Society of Civil Engineers, New York, Vol. 99, No. 5M7, Germany, 1972, pp. Ti-F, 1-15.
1973, pp. 495-507. 7.66 Skempton, A. W., and Hutchinson, J. N. Stability of Nat-
7.52 Little, A. L., and Price, V. E. The Use of an Electronic ural Slopes and Embankment Foundations. Proc., 7th
Computer for Slope Stability Analysis. Geotechnique, International Conference on Soil Mechanics and Founda-
Vol. 8, No. 3, 1958, pp. 113-1 20. tion Engineering, Mexico, State-of-the-Art Volume, 1969,
7.53 Lo, K. Y. Stability of Slopes in Anisotropic Soils. Journal pp. 291-340.
of Soil Mechanics and Foundations Division, American 7.67 Spencer, E. A Method of Analysis of the Stability of Em-
Society of Civil Engineers, New York, Vol. 91, No. SM4, bankments Assuming Parallel Inter-Slice Forces. Geotech-
1965, pp. 85-106. nique, Vol. 17, No. 1, 1967, pp 11-26.
7.54 Louis, C. Stiomungsvorgànge in klüftigen Medien und 7.68 Taylor, D. W., Stability of Earth Slopes. Journal of Boston
ihre Wirkung auf die Standsicherheit von Bauwerken und Society of Civil Engineers, Vol. 24, No. 3, 1937, pp. 197-
Boschungen im Fels. Institut Bodenmechanik und 246.
Felsmechanik, Universitat Fridericiana Karlsruhe, Vol. 30, 7.69 Taylor, D. W. Fundamentals of Soil Mechanics. Wiley,
1967, 121 pp. New York, 1948, 700 pp.
7.55 Morgenstern, N. R. The Influence of Groundwater on 7.70 Taylor, R. L., and Brown, C. B. Darcy Flow Solutions
Stability. In Stability in Open Pit Mining (Brawner, C. 0., With a Free Surface. Journal of Hydraulics Division,
and Milligan, V., eds.), Society of Mining Engineers, Amer- American Society of Civil Engineers, New York, Vol. 93,
ican Institute of Mining, Metallurgical and Petroleum En- No. HY2, 1967, pp. 25-33.
gineers, New York, 1971, pp. 65-81. 7.71 Terzaghi, K., and Peck, R. B. Soil Mechanics in Engineer-
Morgenstern, N. R., and Guther, H. Seepage Into an Ex- ing Practice. Wiley, New York, 1967, 729 pp.
7.56 16th, J. A Theoretical Analysis of Groundwater Flow in
cavation in a Medium Possessing Stress-Dependent Permea- 7.72
bility. Proc., Symposium on Percolation Through Fissured Small Drainage Basins. Proc., 3rd Hydrology Symposium,
Rock, International Society of Rock Mechanics and Inter- National Research Council of Canada, Ottawa, 1962, pp.
national Association of Engineering Geology, Stuttgart, 75-96.
Deutsche Gesellschaft für Erd- und Grundbau, Essen, West 7.73 Whitman, R. V., and Bailey, W. A. Use of Computers for
Germany, 1972, pp. T2-C, 1-15. Slope Stability Analysis. Journal of Soil Mechanics and
Morgenstern, N. R., and Price, V. E. The Analysis of the Foundations Division, American Society of Civil Engineers,
7.57
Stability of General Slip Surfaces. Geotechnique, Vol. 15, New York, Vol. 93, No. SM4, 1967, pp. 475-498.
No. 1, 1965, pp. 79-93. 7.74 Wittke, W., Rissler, P., and Semprich, S. Three-Dimensional
Morgenstern, N. R., and Price, V. E. A Numerical Method Laminar and Turbulent Flow Through Fissured Rock Ac-
7.58
for Solving the Equations of Stability of General Slip Sur- cording to Discontinuous and Continuous Models. Proc.,
faces. Computer Journal, Vol. 9, No. 4, 1967, pp. 388- Symposium on Percolation Through Fissured Rock, Inter-
393. national Society of Rock Mechanics and International As-
Muller, L. The Rock Slide in the Vaiont Vailey. Felsme- sociation of Engineering Geology, Stuttgart, Duetsche
7.59
chanik und Ingenieur Geologie, Vol. 2, No. 3-4, 1964, pp. Gesellschaft für Erd- und Grundbau, Essen, West Germany,
148-212. 1972, pp. Ti-H, 1-18 (in German).
7.60 Nakase, A. Stability of Low Embankment on Cohesive 7.75 Zienkiewicz, 0. C., Best, B., Dullage, C., and Stagg, K. G.
Soil Stratum. Soils and Foundations, Vol. 10, No. 4, 1970, Analysis of Nonlinear Problems in Rock Mechanics With
pp. 39-64. Particular Reference to Jointed Rock Systems. Proc., 2nd
7.61 Patton, F. D., and Deere, D. U. Geologic Factors Con- Congress, International Society of Rock Mechanics, Bel-
trolling Slope Stability in Open Pit Mines. In Stability in grade, Vol. 3, 1970, pp. 501-509.
171
Chapter 9
Engineering of
Rock Slopes
Douglas R. Fiteau and F. Lionel Peckover
In transportation corridors, the objective of rock-slope engi- precedent with the art of estimation and judgment. The
neering is to maintain slopes for maximum safety and effi- engineer-geologist must obtain quantitative information on
ciency. Minimizing rock excavation and predicting the safety those factors that are necessary for making calculations of
and ultimate behavior of rock slopes, whether for highway, the probable stability of the slope. Those factors include
railway, spiliway, quarry, dam site, or opencut mine, are com- structural geology, local topography, drainage, hydrogeology,
mon objectives of civil, geology, and mining engineers. The tectonic history, and other environmental features that may
rational design of rock slopes is particularly important if add to or detract from the stability of the slope.
slopes are steep, if safety is important, and if slope design Whether a slope will be stable or unstable will depend on
significantly affects project costs. how the forces that tend to resist failure compare with those
The present empirically based, cut-and-try methods and that tend to cause failure. This concept defines the factor of
techniques used to design rock slopes are inadequate. New safety for the slope as the ratio of the sum of the resisting
principles and improved technological capability are needed, forces that act to prevent failure to the sum of the driving
particularly in mining operations, in which the trend is to- forces that tend to cause failure. To a considerable extent,
ward open-pit mines, and in transportation facility construc- the problem of rock-slope design and related aspects is one
tion, which may be increasingly required in steep terrain. of applied mechanics, and the necessary margin of safety to
That rock slopes must be treated differently from soil be provided in any particular case is a question of judgment.
slopes cannot be emphasized enough. In the analysis of rock The terms of reference and related slope-design problems
slopes, one must recognize the differences in the basic char- of an open-pit excavation, for example, may be entirely dif-
acteristics and behavior of soil and rock. Unlike a soil mass, ferent from those of highway and railway cuts. For that
which is a relatively homogeneous and continuous medium matter, slope-design problems and requirements for highways
composed of uncemented particles, arock mass is a hetero- and for railways can also differ markedly. Highways can usu-
geneous and discontinuous medium composed essentially of ally have a greater degree of slope instability on their rights-
partitioned solid blocks that are separated by discontinuities. of-way than railways. Unlike automobiles, trains cannot
The geometry of the spaces between the solid materials and steer or brake readily and can be derailed by rocks no larger
the interlocking properties of the components of soil and than 30 cm (1 ft). Therefore, a remedial measure that may
rock are totally different. Failure in soil tends to occur be used on a highway slope may be entirely unacceptable on
within the soil mass, and the direction of the surface of fail- a railway slope. No particular attempt has been made here,
ure tends not to depend on variations of soil properties. The however, to separate railway and highway rock-slope engi-
surface of failure in hard rock masses, however, tends to fol- neering techniques.
low the preexisting discontinuities and not to occur through Rock-slope engineering is concerned not with large land-
the intact rock to any great extent unless the rock is quite slides but with rock falls of individual blocks, translation of
soft. The shear strength of a rock mass is determined largely small rock masses, and occasional slides of accumulated de-
by the presence of the discontinuities, and the result is that bris from gullies, talus slopes, and postglacial slide areas.
the rock mass is anisotropic in its strength and deformational This chapter discusses the aspects of rock-slope engineering
properties.. relevant to designing cut slopes and maintaining the long-
The design of rock slopes involves both engineering and term efficiency and safety of existing slopes. Discussed are
geology and, in addition, a combination of the knowledge of the significant factors in rock-slope design, the procedures
192
Figure 9.1. Well-developed, steeply dipping set of joints along and mechanical properties of the rock mass are a function
highway at Slocan Lake, British Columbia. Joints control slope of the attitude, geometry, and spatial distribution of these
stability almost entirely.
defective surfaces. The basic principles of rock-slope design
are based on
193
Infiuing Materials and Openness the shear strength of the discontinuity is a function of the
shear strength of the infilling materials and the rock mate-
Infilling materials are those materials that occur between the rials that form the asperities.
walls of the discontinuities in the mass. The three most im-
portant characteristics of infilling materials are thickness, Origins
type, and hardness. Jaeger (9.45) notes that, if infilling is
sufficiently thick, the walls of the discontinuity will not The origins of discontinuities will affect their engineering
touch and the strength properties will be those of the infill- significance in the slope. Faults, as compared to joints, for
ing material. example, have different origins and accordingly different
geometry, spatial distribution, weathering and infilling char-
Spacing acteristics, and seepage characteristics.
When a slope consists of several rock types, their combined Before one can completely comprehend the particular prob-
mechanical behavior may differ considerably from that of lenis of stability, one must understand the lithology of the
the constituent units themselves. Hence, each particular physical properties. not only of the rock mass itself but of
rock type may require individual assessment for its behavior all the materials in the mass. Usually a slope is made up of
in the slope. Different rock types and the products that re- a complex of rocks of diverse geologic origins. It may have
sult from their alteration have inherently different weaknesses markedly different sequences of sediments, may be intruded
and strengths as a result of their origins and compositions. by bodies of igneous rocks,or may be partly metamorphosed.
Hence, the properties of different rock and infilling materials The mass represents an association of several lithologic units
can vary within wide limits (9.107). The characteristics of whose mechanical behavior is that of an integral whole,which
each rock type significantly influence the friction angle, na- may differ considerably from the individual lithologic units
ture of asperities, and hardness of the walls of the discontin- themselves.
uities. A sedimentary rock sequence, for example, is markedly
different from an igneous series or a metamorphic complex.
Rock Hardness Each particular type is characterized by a certain texture,
fabric, bonding strength, and macro and micro structures.
There is a relation between rock hardness and unconfined The most important rock properties are the nature of the
compressive strength. According to Jennings (9.46), an in- mineral assemblage and the strength of the constituent min-
crease in hardness has a corresponding increase in the shear erals; a rock material cannot be strong if its mineral constit-
strength. Infilling material and second-order asperities are uents are weak or if the strength of the bonds between the
sheared through during shear failure along a discontinuity; minerals is weak.
194
The body of rock or the host rock in which the discon- Figure 93. Hypothetical possible positions of groundwater table in
tinuities occur directly influences the strength characteris- jointed rock slope (9.108).
Climatic Conditions
TEMPERATURE --
-60
The effects of climate on the stability of rock slopes in trans-
portation corridors and the various remedial measures that
must be taken to accommodate these conditions are impor-
tant in rock-slope engineering. Daily temperature variations, 10.0
precipitation, snow, and freeze-thaw conditions, acting either
independently or in combination, often cause significant sta- 7.5
bility problems.
Terzaghi (9.108) noted that the groundwater conditions
and hence the effective hydrostatic pressures can vary within
BER
LLS
195
occur in the spring and fall when the mean temperature is principal stresses act in the vertical and radial directions re-
about 0°C (32°F) and frequent freeze-thaw cycles are occur- spectively. For the convex slope, the converse is the case,
ring. When temperatures are above freezing, the frequency of and horizontal tangential stresses are tensile. The maximum
rock falls is a function of degree of rainfall. The number of principal stress is still vertical, but the minimum principal
rock falls originating on the steep rock faces in the Fraser stress acts in a horizontal direction tangent to the slope, and
Canyon is probably high because of the absence of snow, the slope material is in tension. Cohesion that would nor-
vegetation, and soil, the presence of which would insulate mally have occurred is reduced. Since a rock mass is relatively
the rock from unusually low temperatures and from changes weak in tension, tensile stress concentrations in the slope in-
in temperature. duce instability, causing unrestrained blocks to slide out.
Frost action probably directly or indirectly accounts for For an infinite slope in rock (normally specified in rock
more rock falls than all other factors combined. Water un- mechanics when rock strength includes cohesion), a slope
dergoes about a 9 percent volume increase when it freezes can be steeper than that indicated by the angle of friction.
and exerts tremendous pressure when it freezes in a confined Normal stress increases downslope, and the effect of cohe-
space. According to Reiche (9.97), "Water-filled cracks or sion relative to that of friction decreases. The slope thus
joints which terminate downward and which are narrow and flattens to the angle of internal friction. The profile of such
perhaps irregular may be converted into essentially closed a slope, therefore, is theoretically concave from top to bot-
systems by preliminary freezing of the water in the superfi- tom. The variation of slope angle with slope height has a
cial parts. In such cases the combination of expansion and small but significant influence on the stress distribution in
low compressibility may exert a disruptive force which, if the slope according to Yu and Coates (9.111) and, therefore,
the temperature continued to fall and rock pressure per- can be expected to have some influence on the stability.
mitted, would approach 30,000 pounds to the square inch
at -22°C (i.e., -7.6°F)." Freezing temperatures of this order Time Factor and Progressive Failure
are not uncommon in temperate zones.
Piteau and others (9.93) correlated mean annual move- Natural rock slopes undergo progressive failure in time by
ments of an overturning failure with average monthly rain- the processes of creep and flow. Hence, it is of considerable
fall, snowfall, and temperature. Although they found that importance that one recognize whether the analysis and ulti-
a direct relation existed between movement and rainfall mate design of the slope meet the requirements of short-
(i.e., groundwater pressure buildup), they found that snow- term or long-term stability. Also, because of progressive
fall and freezing retarded movement. The average monthly failure, one must consider potential maintenance problems
rainfall peaked in January, but movement decreased in De- and design accordingly. For practical purposes, allowances
cember, when freezing and snowfall conditions started. The must be made for some reduction in the strength properties
snow blanket prevented rainfall from infiltrating into the of the rock mass with time (9.76).
slope, and freezing temperatures reduced the availability of Murrell and Misra (978) note that time-dependent strains
free water on the slope. occur when rock material is subjected to relatively high
stresses for long periods. Most of the forces involved in such
Slope Geometry in Plan and Section deformations are indeterminate functions of time, being de-
pendent on the effects of the excavation, regional stresses,
Most current theories of slope stability consider the slope to alteration processes in the mass, physical and cheinical action
be two dimensional (i.e., a unit length of an infinitely long of groundwater, and seasonal variations of temperature and
slope is considered to be in plane strain) and the plan radii rainfall. These lead to fatigue and opening of cracks with ir-
of the crest and toe of the slope to be infinite. However, reversible deformations and progressive weakening of the
this latter condition is not normally encountered in practice; mass. Therefore, continued movement of a rock slope is
slopes that are concave in plan tend to be more stable than cause for concern. Movements are important, for relative
those that are convex. displacements along defects in the rock mass tend to reduce
Highway or railway cut slopes in mountainous terrain are the resistance along these defects and may bring about failure.
often convex and therefore have a greater tendency toward A slope maybe stable when first excavated but, because
instability. In open-pit mines, on the other hand, slopes are of gradual deterioration and adjustments toward equilibrium,
usually concave and therefore more stable. In a study of may become unstable with the passage of time. The time re-
slopes of diamond pipes in South Africa, Piteau and Jennings quired for deep-seated failure in hard rocks is almost impos-
(9.92) and Piteau (9.89) found that the plan radius of curva- sible to evaluate. Near-surface failure, such as raveling or de-
ture of the slopes has a marked effect on slope stability. taching of rock segments, however, may develop only a few
They applied this finding to the DeBeers Mine to predict the years after the excavation has been completed. This kind of
final breakback position, subject to its slope geometry. failure should be considered in the slope design and remedial
Lorente de No (9.65) and Förster (9.27) made similar findings. measures proposed. Soft rocks, such as shale, mudstone,
Horizontal tangential stress concentrations in such a slope and other types of argillaceous materials, can undergo niag-
can be either compressive or tensile, depending on the slope nitudes of deformation that may lead to failure within sig-
geometry (9.64). Horizontal stresses tangent to the slope are nificantly shorter periods, sometimes within several days.
beneficial in a concave slope, for they create an archlike ef-
fect whereby the blocks forming the partitioned rock mass Residual and Induced Stress
tend to be squeezed together. Compressive stresses substan-
tially improve the shearing strength. All three principal The cut slope created by an excavation affects the stresses
stresses are compressive, and the maximum and minimum in a rock mass at the boundary of the excavation. However,
IM
predictions of the magnitude of these stress concentrations ture, lithology, groundwater, river hydraulics) were made to
and their effects on the stability of the slope are complex. assess those factors controlling slope stability. More than two-
To date, results of studies of stress distribution in slopes and thirds of all incidents occurred where the river had been de-
the manner in which stresses affect the stability are largely flected into the bank and had undermined the slope. Severe
hypothetical. No mathematical or physical models are avail- lateral erosion by the river was the result of either general di-
able to predict the effects of varying the slope excavation rectional changes in the river or the development of alluvial
geometry and the ultimate variation in stress concentration fans that forced the river into the opposite bank.
(9.73). In this regard, probably the most significant advances Until failure mechanisms of rock slopes are better under-
have been made in using finite element procedures to consider stood,consistently reliable predictions of rock-slope behavior
the stress-strain compatibility of the slope. are not possible. Because natural slopes may provide some
Contrary to earlier views, bedrock is not a predictable clues, as much attention as possible should be given to ex-
platform in which the only acting forces are vertical due to amining the way in which local slides and deformations de-
the weight of the rock. Rocks may also be subject to signifi- velop. Analysis of rock slides in an area can provide excel-
cant horizontal residual stresses that under certain circum- lent information on the mechanics and patterns of their for-
stances could have important influences on the behavior of mation.
rocks at excavations. Hast (9.32) notes that this influence Coates and Gyenge (9.17) make use of information from
has been proved to be quite substantial in deep excavations. the performance and characteristics of existing or previously
Regional stresses, denudation, tectonic uplift, glacioisostatic existing slopes and have developed the principle of incre-
rebound, and other conditions might also affect the stability mental design for slopes. This is the "process of extrapolat-
of surface excavations in rock, even though less dramatically ing from the known to the new, or predicting the conditions
that result from a change in the present operations." Al-
Existing Natural and Excavated Slopes though this work is basically applied to open pits and the in-
cremental predictions may not be of high accuracy, some of
Slope design should take into account past experience with the basic concepts developed can be applied to rock-slope de-
both stable and unstable slopes. KIey and Lutton (9.51), sign for transportation routes.
Lutton (9.66), and Shuk (9.104) show that analyses of both
natural and excavated slopes will provide valuable background Dynamic Forces
information for proposed excavation design, particularly those
for mountainous terrain. The significance of earthquake vibrations is well documented
Usually the angles of both natural and excavated slopes elsewhere; effects of blasting are considered later in this chap-
provide conservative estimates of slope angles that can be ter. The brief discussion here is of the effects of vibrations
achieved in slope design. Humans can invariably improve on along transportation corridors due to vehicles.
the slope angles provided in nature by giving careful atten- There is some belief that traffic vibrations, particularly
tion to drainage, artificial stabilization methods, and control from trains, may be a significant factor with respect to slope
of natural slope-forming processes. Whether one is evaluating stability. However, a comparison of relative masses shows
profiles from surface excavations or natural slopes, the prob- that traffic-induced vibrations are extremely small, even for
lem is in estimating the degree of the conservatism inherent slopes close to the right-of-way. Therefore, such vibrations
in the population of slope profiles being considered. This can be considered insignificant. According to Peckover
problem is properly resolved by the engineer and the geolo- (9.84), the incidence of slope failures along railways when
gist (primarily geomorphologist) complementing each other. the train is passing, for example, is no greater than would be
The safety factor of natural slopes is commonly not much expected from the proportion of total time that the railway
greater than unity (9.76). The value of unity, however, ap- is occupied.
plies specifically to that situation in which the most adverse
groundwater conditions develop naturally. A hillside will FUNDAMENTAL PROCEDURES IN
normally have a safety factor greater than one most of the ANALYSIS OF ROCK SLOPES
time, but it may become one when the water conditions or
the natural disturbing forces, such as recurring earthquakes, This section describes the basic approach to analysis of rock
are as severe as are likely to occur. slopes as well as the theoretical and analytical process. The
Slopes should not be compared if the general modes of process can be summarized as follows: Discontinuities in the
their formation (i.e., types of excavation and slope-forming rock mass are systematically measured and statistically ana-
processes) differ. When slope processes are similar, stable slope lyzed to determine their nature and distribution. Estimates
case histories can be relied on to predict a lower bound to the are made of the strength properties of the discontinuities.
design slope angle. The use of slope case histories requires These factors are quantitatively described and theoretically
that factors such as slope and failure geometry, geology, and applied to determine the strength along any potential failure
material properties be obtained. Slope monitoring can also plane. Shear strength parameters are assessed, thus allowing.
be helpful when case histories are used in slope design. the factor of safety of the slope to be calculated and slope
In highway and railway slope problems, the most impor- design to be considered.
tant factor relating to case history analyses is probably the in-
cidence of failure. This is shown by Piteau (9.90) in a regional Determination of Structural and Other
slope stability study of the Fraser Canyon. Comparative anal- Relevant Geologic Characteristics
yses of incidence of slope failures (rock falls, landslides, de-
bris slides) and of geological factors (geomorphology, struc- In the analysis of a high rock slope, structural discontinuities
197
of the rock mass are mapped in detail and each feature is Determination of Structural Domains
quantitatively characterized. The geologic survey is aimed and Design Sectors
at measuring a sufficient number of joints to allow the data
to be analyzed statistically. The statistical analyses and judg- Because both the geologic conditions and the bearing of the
ment indicate whether the best estimate has been made for proposed cut face can vary from one location to another
the whole population. The entire geologic survey must be along a cut, different slope designs and remedial measures
conducted so that, from the joint characteristics recorded, may be required at different locations. Therefore ,one should
the shear strength in the direction of the joints can be nu- analye individually those parts of the proposed cut slope
merically assessed by comparison with similar features tested that are similar in terms of both orientation and physical and
in the laboratory. mechanical characteristics. Areas of similar geologic charac-
Physical access to all discontinuities in a rock mass is not teristics are designated structural domains. Samples of the
possible. Therefore, maximum information must be ex- geologic structural properties within a structural domain
tracted from all locations where access is possible. For other will not differ significantly from one part to another. There-
locations, information is obtained by a variety of means in- fore, the slope stability characteristics and design parameters
cluding exposure mapping, tunnels, trenches, drilling (core of the entire structural domain can be determined from sam-
logging), terrestrial photogrammetry, aerial photograph in- pling only part of it.
terpretation, and various geophysical methods. In exposure Boundaries of straight slope segments that have similar
mapping, either some variation of detail line mapping or orientations are determined and superimposed on structural
fracture set mapping is generally best to use. domain boundaries to form design sectors. Characteristics
Line mapping methods are discussed by Jennings (9.46), of the various design sectors are selected for stability analy-
Piteau (9.87), and Haistead, Call, and Rippere (9.30), and sis and slope design. Within each design sector, typical joints
fracture set mapping by Call (9.14), Mahtab, Bolstad, an I or design joints, which represent the mean characteristics of
Kendorski (9.67), Da Silveira and others (9.20), and the relevant joint sets, are selected for use in the design cal-
Herget (9.36). culations(9.94). Discussions relating to definition of prop-
Sources of errors in joint surveys are discussed by Terzaghi erties of design joints are given by Steffen and Jennings
(9.109), Robertson (9.99), and Piteau (9.88). (9.106) and Piteau (9.88).
Boundaries of structural domains usually coincide with
General collection and processing of geologic data are dis-
major geologic features, such as faults, shear zones, dikes,
cussed by Knill (9.52), the Canada Department of Energy,
sills, geologic contacts, and unconformities (9.88). The ana-
Mines and Resources (9.94), and the International Society
lyst's ability and experience in assessing the structural geol-
of Rock Mechanics (9.44).
ogy and the variations in structural characteristics will deter-
Geologic conditions vary from project to project, and
mine the accuracy and usefulness of the structural domain
thus a geologic survey at one site may be entirely different
determinations. Comparative analyses of joint populations
from that at another. Features that should be considered in
within structural domains or between structural domains
the survey are coordinates; elevation; rock type and hard-
should be subsequently performed.
ness; type of geologic structure; strike (direction of structure
surface) and dip (angle of structure surface); dip and strike Development of a Rock Mass Model
continuity; thickness, type, and hardness of infilling mate- Depicting Geologic Structure
rials and the proportion of voids and presence of water in the
materials; roughness; and waviness (wave shape or interlimb After the characteristics of the geologic structural population
angle). Coordinates, elevation, strike, and dip are used to de- in each structural domain are defined, a model of the rock
line position and orientation of the discontinuity in space. mass is developed to depict the three-dimensional relations
These features, plus dip and strike continuity and bearing of of the geologic structure. Some workers refer to the rock
the sample line, serve to define intensity. Rock type and hard- mass model as a schematic concept or structural picture of
ness, characteristics of in filling material, presence of voids the rock mass. Since each structural domain is similar in a
and water, roughness, and waviness are used for assessing statistical sense, a rock mass model is developed for each
frictional and cohesive strength and deformation properties. structural domain. The model can be of a graphical, physi-
The survey data are processed, and the attitude, geometry, cal, or mathematical nature or a combination of these.
and spatial distribution of the jointing are determined by An essential requirement of the model is that it accurately
computer analysis to yield structural domains, joint sets and represent the actual geologic structural population in a statis-
their average properties, and major discontinuities. tical sense and that it apply to the entire design sector. A
After the regional and local geology is assessed, a geo- graphical model using the stereographic projection (9.48) is
logic iiap is constructed to show the major and minor struc- usually used, and extensions are often made to mathematical
tural features and the general lithologic distribution. Ste- or physical models to determine whether the first boundaries
reographic projections are easy to use and are of great bene- selected are adequate or should be changed. A typical rock
fit in evaluations of geometric relations and in showing struc- mass model using a stereographic net depicting the angular
tural populations. These projections can also be shown on relations between faults that form a potential wedge failure
the general geologic map of the area. and a proposed cut slope is shown in Figure 9.6. A spatial
Stereographic projection principles and the use of equal- diagram of these relations is shown in Figure 9.7.
area, equal-angle, and polar projections are documented by Once the boundaries of the various structural domains
Donn and Shimer (9.22), Phillips (9.86), Terzaghi (9.109), are defined, attention is given to delineating joint sets within
and John (9.48). each domain and determining the characteristics of each joint
198
Figure 9.6. Typical graphical model or schematic concept of failure modes usually investigated are plane, wedge, stepped,
mass on Wuiff stereographic net. Angular relations between and circular. (Discussions of different failure modes are
faults, which form potential wedge failure, and proposed cut
presented in Chapters 2 and 7.) A tension crack for each
slope are shown. *
case can be assumed to exist at the surface. Methods of
slope-stability analysis for various failure modes are dis-
RUlE 20,
P. cussed by Jennings (9.47), Hamel (9.31), Heuze and Good-
man (9.37), Hoek and Bray (9.42), Hendron (934), and
Goodman (9.29).
The analyses consist of the following operations:
FAULT F2-
TRIKE 60° Estimations of continuity ofjointing on potential
)IP 70° SE
failure planes;
CUT Assessment of the strength of the intact rock;
SLOPE -
STRIKE 90°,
Determination of the effects of the joint characteris-
DIP 55°S. tics on the strength along joints;
E OF
Development of the necessary equations of limit equi-
'TSR SECTION librium for the possible modes of failure;
F F AND F2
Use of various potential failure planes singly or in com-
bination to test for these failure modes; and
Determination of the factor of safety of the slope.
NORTI-4 —SOUT
JOINT SET ROAD
T$
\\•.
\., \...
B0 0.
FAULTS 'OLIAT
2O°
amined in their various combinations until the worst situa- Figure 9.9. Graphical model in stereographic projection showing
tion is found (i.e., the case that gives the minimum factor of angular relations between joint sets, which form potential wedge
failure, and proposed cut slope location shown in Figure 9.8.
safety for the particular slope angle examined). The proce-
dure is repeated for different slope angles, and the respective
factor of safety is determined for each case.
200
Figure 9.10. Order in which slope treatment methods should be
A mechanical stability analysis in this case is somewhat
considered for selection.
redundant because of the steepness of the potential wedges.
For illustrative purposes, however, a simple three-dimensional
analysis (9.13) can be carried out to show the minimum
SITE CONDITOINS
angle of residual joint friction (r),which would be required ENGINEERING
JUDGMENT
ANGLE, HEIGHT, AND CONDITION OF SLOPE
to hold up such a wedge if undercut. Consider the following SIZE AND CONDITION OF ROCKS
PATH OF ROCK FALL TO RIGHTOFWAY
values determined from the rock mass model shown in Fig- MAINTENANCE REQUIRED
ure 9.9:
3= 740 , average dip of the north-south joint set,
a= 840 , angle measured in the plane of the north-south
PROTECTION __ WARNING
joint set between the strike of this plane and the STABILIZATION
METHODS METHODS METHODS
line of intersection, and
w= 1270 ,angle measured in a plane perpendicular to the
line of intersection of the two joint sets. EXCAVATION RELOCATION ELECTRIC WIRE
ELECTRIC FENCE
SCALING AND TRIMMING INTERCEPTING SLOPE,
DITCHES, AND BERMS COMBINATION WITH
r would have to be a minimum of 57.20 in order to
DRAINAGE
SHOTCRETE DITCH SHAPING PROTECTION
201
time because of strength changes, rock deterioration, fluc- the basis of proposals or bids invited from contractors known
tuating groundwater, and other environmental factors, to have up-to-date knowledge of the equipment and tech-
which generally produce less stable slopes. The most eco- niques required. The proposal or bid should be for a unit
nomical overall design, therefore, accounts for this decrease price or cost-plus-fixed-fee contract and not for a lump sum.
in stability with time and includes consideration of a main- The U.S. National Committee on Tunneling Technology
tenance and remedial program. Remedial measures for rock (9. 79), although addressing mainly underground work, is a
slopes along transportation corridors in steep terrain must be good source of information on sound contracting practices
integrated with other maintenance tasks. If the route runs in construction engineering and contract law for rock con-
for some distance through mountainous terrain, work will struction work.
probably be sufficient to keep personnel on rock-slope main-
tenance work full time. Unlike most soil slopes, on which Selection of Priorities
weathering is not an important factor, rock slopes require
continual attention if serious problems are to be avoided. The condition of all rock slopes along transportation routes
A consistent long-term remedial program is required that in steep terrain should be thoroughly inspected once a year
contains carefully selected priorities. Experienced people or more frequently depending on the severity and implica-
are needed for both engineering and construction. Judgment tions involved. Inspection can be done by (a).a helicopter
must be based on a detailed knowledge of occurrences at survey of inaccessible locations and of the overall conditions
each dangerous location. of the slopes and (b) a ground survey of hazards at specific
locations to determine the need for detailed studies. Based
Personnel on annual surveys and case histories of the slopes, detailed
studies and alternative treatments are planned. Factors to
In setting up a rock-slope remedial and maintenance program be considered in deciding on locations for remedial work
for a major transportation route, a project engineer should are
be assigned full time to the work. The project engineer
should have available an experienced engineering geologist Maintenance costs, including patrols required;
and geotechnical engineer and the capability to supervise Costs of remedial measures and expected benefits to
contract work. If necessary, consultants should be used for be gained;
highly specialized parts of the work and to provide some
Degree of risk to route, determined by considering the
personnel on a temporary basis, depending on the magnitude
amount of traffic, records of past events, measured rock
and frequency of the problems. Continuity of personnel
movements, and frequency of clearing rocks from the vicin-
should be encouraged, for experience in rock-slope engineer- ity;
ing is a most important attribute. An engineering geologist
Occurrence of accidents, rock falls, washouts, land-
on the team should be able to translate information clearly slides, and the like;
into engineering terms and integrate information with engi-
Conditions downhill from the right-of-way, which
neering requirements.
would indicate how serious an accident or derailment might
The project engineer must keep in touch with all develop-
be; and
ments relating to the overall project. The geotechnical engi-
Views of maintenance and patrol personnel.
neer and engineering geologist should make detailed inspec-
tions, perform slope-stability analysis and design, exercise
judgment on specific rock conditions, and help in planning The annual program should emphasize stabilizing or pro-
and supervising remedial work. It is particularly important tecting the greatest number of locations where rock-fall
that the staff of the highway, railway, or other transporta- hazards exist in order to obtain the maximum economical
tion agency concerned with rock-slope problems develop improvement in safety. Large problem areas can often be
sufficient expertise within the organization to critically eval- divided into smaller areas for treatment. The tendency to
uate work performed by private companies or other agencies. overprotect the route at locations of recent accidents should
Some of the principal functions of the project engineer and be resisted in the interest of obtaining a balanced overall
engineering geologist in planning, design, and maintenance program. Conditions at each particular location may vary
are described by McCauley (9.70). significantly, and the application of different remedial mea-
sures should be considered accordingly. Details of remedial
Work Arrangements work must be developed section by section along the route
and be based on detailed inspections, existing records, expe-
To estimate the cost of remedial measures in advance is dif- rience, and sound judgment. However, the most careful
ficult, for unforeseen conditions are often revealed as the choosing of hazardous locations and the most thorough anal-
work proceeds. For this reason, decisions on an overall bud- ysis in selecting remedial measures will never entirely prevent
get required for the improvement of rock-fall hazards are unexpected rock falls from occurring.
often based on judgment as well as analysis. All concerned
should understand that some flexibility is needed in the Records
total budget approved for an annual program so that adjust-
ments can be made in expenditures required at different lo- For the efficient planning of rock-slope engineering work, a
cations as well as in contract provisions to deal with changed data storage and retrieval system should be organized and
conditions during the work. Because major revisions often maintained for permanent reference. The following types of
are required during the work, rock work should be done on records should be maintained:
202
Figure 9.11. Rock slope before (top) and after (bottom) removal of overhang.
zro
203
Figure 9.12. Oblique photographs of cut bench at Hell's Gate Bluffs, British Columbia, above which excavation was proposed to remedy
overturning failure.
............- -. . .
-. . .-. •
#
A)' . . - -. .. . ,, '..•...'• •--
Aerial photographs can be useful for studying the condi- graphs can be used to examine slopes three-dimensionally
tion and con figuration of slopes in steep terrain and for giv- for purposes of tentative slope-excavation design. Aerial
ing important cities to the causes and potential sources of' oblique photographs can be taken with cameras mounted
slope instability. High-level photographs are generally of on the ends of a 5-ni (15-it) boom carried by a helicopter.
minimal use for local slope-desiii purposes. If possible, Photographs can be taken at any angle from vertical to hori-
low-level photographs at a scale of I cm equals 50 to 120 zontal and contours can be plotted to a fixed datum. Cracks
in (1 in equals 400 to 1000 it) should be obtained. Uses and features on the order of 5 to 8 cm (2 to 3 in) wide can
of aerial photographs for engineering purposes are described be detected in this way. Photographs are particularly useful
in Chapter 3 and by Norman (9.81) and Mollard (9.74). in evaluations of rock slopes that are in steep terrain and not
Aerial oblique or terrestrial oblique single photographs easily accessible from the ground (Figure 9J2). Use of ter-
or stereoscopic photographs are frequently used for rock- restrial photogramniet ry in rock-slope engineering is dis-
slope engineering work (Figure 911). Stereoscopic photo- cussed by Ross-Brown and Atkinson (9J01).
204
Monitoring and Inspection Figure 9.13. Forty-three-meter
(140-ft) presheared through cut
involving 130 000 m3 (170 000
Planning for rock-slope engineering work should include a yd3 ) near Britannia, British
program of monitoring slope movements and identifying Columbia, and including cut
characteristics that indicate changing stability. Detection bench to protect highway from
rock falls. Photo is toward
of general creep or slow translation of highly fractured or
south, but sketch is toward
ft slope-forming material is important in decisions regard- north.
ing remedial work. Horizontal and vertical movements of
points plotted against time or depth or both provide impor-
tant information concerning the behavior of a rock slope.
Graphs provide it clear indication of the onset of slope fail-
tire when plots representing the change in position do not
remain linear. When such accelerated movements are evident,
the slope must be approaching failure, and measures should
be taken to analyze and remedy the situation.
For studies of complex stability problems, the advice of
a specialist is needed to determine the most suitable slope- j -'
monitoring program. A wide variety of commercial instru-
ments is available for measuring both surface and subsur-
face movements and for recognizing the characteristics in-
dicative of potential instability. However, the simplest
methods and mechanical equipment are not only the most
practical but also invariably the most reliable. The various
methods of instrumentation are discussed in Chapter 5, in
the June 1972 issue of Highway Focus (9.39), and by Frank-
liii and Denton (9.28). Benson (9.8), and Medley (9.33).
The simplest and generally most efficient method of mon-
itoring recently excavated slopes is to measure both horizon-
tal and vertical displacements of protected metal survey pins Incorporating benches in the slope, and
or hubs set in concrete or grout along a straight line at the Excavating in a manner that minimizes damage to the
crest and toe of the slope. Main control reference points rock mass.
should be located outside the area where movement may oc-
cur. If. in the first year. movements appear negligible (other Excavations made for remedial purposes should provide a per.
than what might be expected from normal elastic rebound nianent solution to the slope-stability problem so that addi-
due to unloading), the number of surveys in succeeding years tional excavation work in the future need not be necessary.
may be decreased until the slope becomes stable. In this The principal problems and disadvantages associa ted with
work, high precision is essential so that the onset of nonlin- excavation methods lie in their cost. Accessibility after con-
ear movement can be determined as early as possible and re- struction may be difficult and, since the slope usually must
medial work can be planned accordingly. be excavated from the top downward, mobilization and setup
Individual cracks, which may be indicative of ceneral in- costs can be prohibitive. Also, disposal sites often are limited
stability in an area, can be monitored simply by bridging the with the result that waste rock may have to be transported
crack with a wad ofgrout or shotcrete or by inserting a wedge for some distance, unless it can be used for local construction.
in the crack. A regular part of the work program for evaluat-
ing and monito ring should involve inspect ion and niainte- Benches
nance of remedial installations, such as rock bolts, walls,
nets, and drains, already in place. Regular inspection is an The overall slope angle and different slope angles that may
essential precaution in protecting a substantial capital invest- be designed for different parts of the slope are determined
ment. by slope-stability analysis methods (Chapter 7). Each cut
should be designed to suit the prevailing rock conditions.
Methods of Stabilization Slopes containing materials with strength properties that
vary significantly in section may require that some parts be
Excavation and Related Design Aspects flatter than other parts if rock falls and maintenance costs
are to be minimized. The use of' variable slopes, benches be-
For purposes of improving the stability of rock slopes, exca- low layers of rapidly weathering rock, and St rue t oral support
vation is used either to reduce the driving forces contributing and correct blasting techniques can practically eliminate rock
to failure or to remove unstable or potentially unstable sec- falls in new cuts.
tions of the slope that may lead to failure (i.e.. rock falls and Benches are used in slopes to minimize rock falls onto the
slides). Stability, therefore, can be achieved by roadway (Figures 9.13 and 9.14). For soft rocks (Figure
9.14), such as shale, mudstone. and other argillaceous rocks.
- Removing unstable or potentially unstable material. benches tend to reduce excessive weathering and erosion and
Flattening the slope. provide rock-fall catehments. Erosion due to groundwater
Removing weight from the upper part of the slope. runoff is also controlled since the energy of surface flows is
205
Figure 9.14. Benches that follow stratigraphy in Niagara Escarpment Figure 9.16. Use of multiple benches and controlled blasting
sedimentary rocks on Highway 403 near Hamilton, Ontario. Design techniques at Mica Dam, British Columbia, to maintain rock slope in
also includes overburden being stripped back 3 m (10 ft) from top soundest possible condition and to prevent rock falls.
of rock and light fence barrier on edge of highway.
:'-
____ - çs..'-.•
'..
. '
4: '.
- :.
, - : -. •... 141
..
.. :-.- - ...... •
\ -
'4
206
Figure 9.17. Idealized cross section showing stepped slope design in soft rock based on SLOPE ROUNDING
-
- FT ROCK LINE
CENTERLINE
erosion. Seeding and mulching or other suitable methods on the rocks to be removed. Scaling and trimming work re-
of slope stabilization can be readily applied; however, for dluires specialized experience, and the performances of differ.
rapidly raveling slopes, about half of the bench or step width ent contractors should be compared when additional work
should be filled before seeding is done to prevent smothering is considered. Depending on site conditions, potentially use-
of the seed. ful tools are bencher drills, gas jackhammers, air-operated
scaling tools, suspended powered platforms (spiders), and
Scaling and Trimming hydraulic boom cranes (giraffes) for access to low and inter-
mediate slopes. Fquipment for scaling and other related ac-
Scaling of loose, overhanging, or protruding blocks is a basic tivities in lower and higher reaches of slopes is shown in Fig-
maintenance operation on rock slopes of all sizes along trans- tires 9.19 and 9.23 respectively.
portation routes in steep terrain. Scaling on the upper
reaches of high faces is usually carried out by workers on Blasting Procedures
ropes with hand pry bars, hydraulic splitters or jacks, and
explosives(Figure 9.18). Of necessity. Ihis work is slow and Rock should be preserved beyond excavation lines and grades
intermittent. Mechanical scaling equipment is more efficient in the soundest possible condition. The effects of tension,
and safer, but may have limitations because of severe access compression, and shear stresses developed in a rock mass as
problems. Trimming involves drilling, blasting, and scaling a result of blasting damage are documented by Bauer (9.6),
to remove small ragged or protruding rock in overhang areas Lang and Favreau (9.57), Langefors and Kihlstrom (9.59),
where repetitive scaling would otherwise be required. the U.S. Bureau of Public Roads (9.12), Larocque (9.60),
Postconstruction scaling and trimming should be carried and Lamhooy and Espley.J ones (9.55). Uncontrolled
out on a regular basis. In temperate regions. this is usually blasting results in rough uneven contours, overhreak, over-
started in the spring after the frost leaves the rock. If hangs, excessive shattering, and extensive tension cracks
thorough scaling is performed during excavation, subsequent in the crest of the slope. BlasI ing damage, therefore,
maintenance and remedial work can be greatly reduced. Be- can lead to significantly higher scaling, excavation, re-
fore work begins, an engineering geologist and rock foreman medial treatment, and maintenance costs. The results
should thoroughly inspect each location and make decisions of blast shock waves and gases along faults, joints, bed-
,. leJ.i .
Figure 9.18. Typical rock-scaling operation high above highway Figure 9.19. Typical hydraulic crane (giraffe) and b.icket used for
scaling lower reaches of slopes.
Wf
0.
/ 4
• I •
— :
207
ding, and discontinuities, althougj1 not readily apparent on Figure 9.20. Probability of damage versus
the blasted face, can lead to loosening of the rock. This charge and distance (9.82).
sometimes occurs well behind the face, allowing easier infil- kg I b
000
tration of surface water, which may lead to unfavorable
groundwater pressures and unnecessary frost action.
1
Blast-hole patterns and powder loads must be properly POSSIBLE
208
or subsurface drainage conditions and, hence, increase the Figure 9.21. Temporary cover of polyethylene sandwiched between
two layers of mesh to prevent precipitation infiltration into slope at
stability of the slope should be given high priority in the pro- Hell's Gate Bluffs, British Columbia.
posed work. Drainage measures, as compared with other pos-
sible measures, frequently result in substantial benefits at
significantly lower cost. Often large failures, involving sev-
eral thousand cubic meters of material, cannot be controlled
within practical limits by any means other than some form
of drainage. The application of surface and subsurface drain-
age as part of the general slope design or for stabilization pur-
poses should be considered at the outset of the project be-
cause substantial benefits may result at relatively low costs.
209
Figure 9.22. Shotcrete applied to steep high slope that is prone to Figure 9.24. Example of drainage behind shotcrete (9.1).
minor slides and rock falls.
WA I F H lEAR ING
DISCONTINUITY
PERFORATED
POLETHYLENE TullE
FINE
REINFORCEMENT
POLYURETHANE.
CLOSED CELL FOAM
Figure 9.23. Shotcreting dangerous overhang in extremely steep replaced gunite, a similar material that contains smaller
terrain along tracks of Canadian National Railways in Fraser aggregate. Rock slopes should be thoroughly scaled to
Canyon, British Columbia: (a) steep highly fractured rock face provide the soundest rock condition before shotcrete is
with serious overhang; (b) shotcrete equipment and buildup of
applied. An extensive shotcrete application in an area prone
shotcrete at top of picture and under overhang; (c) start of
shotcreting operation at tunnel portal: and (d) close-up of shotcrete to slides and rockfalls is shown in Figure 9.22. Figure 9.23
buildup in overhang shown above. shows the use of shotcrete for an overhang in extremely
(a) (b) steep terrain.
Specifications and discussions of the application of shot-
crete arc given by the American Concrete Institute (9.2).
When shotcrete is applied to an irregular rock surface, the
resulting surface colt iiguratio ii is smoother. The shotcrete
helps to maintain the adjacent rock blocks in place by
means of its bond to the rock and its initial shear and ten-
sile strength acting as a membrane. The result is that a
composite rock-shotcrete structure is developed on the sur-
face of the rock. There is no transfer of load from the rock
mass to the shotcrete. In that the interlocking quality of
the surface blocks is improved (9.10), shotcrete acts as re-
inforcement and not as support. The more quickly the shot-
crete is applied after excavation, the more effective the re-
suits are.
Deterioration of shotcrete can result from frost action,
- oundwater seepage, or rock spalling due to lack of shot-
crete bond. Theretrc, unfavorable groundwater flows
should be drained for long-term stability of the shotcrete
cover. Weep holes should be drilled or installed through the
hardened shotcrete and into the rock to prevent building of
water pressure behind the shotcrete. In Norway and Sweden,
semirounded plastic pipes are glued to the rock surface to
form surface drainage channels. Small volumes of water are
protected against freezing by rock wool, plastic foil, or even
tcrials, such as straw, sawdust, gravel, or crushed rock, have heating cables. More universally used are short flexible plas-
been used for this purpose. Electric current has also been tic pipes, which are placed in cracks or holes drilled into
used to keep the drain-hole pipe warm enough to prevent water-bearing broken rock: an example of this technique,
ice buildup. as described by Aiherts (9.1), is shown in Figure 9.24.
Initially dry rock surfaces are preferred in the shotcrete
Shoterete process. although careful control of setting admixtures and
nozzle water can give successful applications on wet sur-
Shotcrete is a concrete that consists of Inortar with aggre- laccs. Where alteration products, such as clay or mud, ex-
gate as large as 2 cm ( in) in size and that is projected by air ist on joint or fault planes, care should be taken to clean
jet directly onto the surface to be treated. It is one of the such surfices by air or water jet to ensure a good bond be-
basic methods for treating unstable sections of rock slopes. tween shotcrete and rock. As a rule of thumb, weak mate-
It is used to prevent weathering and spalling of rock sur- rial should be removed to a depth at least equal to the
faces and to provide surface reinforcement between blocks. width of the weak zone before shotcrete is applied. These
The force of the jet compacts the mortar in place. Shot- areas are where the shotcrete will do the most good and
crete is usually applied in 8 to 10-cm (3 to 4-in) layers, and where extra attention will be required in its application.
each layer is allowed to set before successive layers are ap- Shotcrete can be used in combination with steel wire
plied. Shotcrete on rock slopes generally appears to have mesh and rock bolts to give structural support and also to
210
form buttresses for small loads. Where shotcrete is applied Figure 9.25. Cast-in-place reinforced concrete buttress
to mesh, all loose material should be removed from the rock approximately 15 m (50 ft) high providing stabilization protection
of large overhanging slope.
surface and the mesh fabric tightened. Shotcrete can also he
used behind anchor beams to provide a uniform contact with
the rough or uneven rock surface and,if applied across cracks,
may provide a simple means of indicating where movement
is occurring. An extensive application of steel fibrous shot-
crete to stabilize potential rock falls in basaltic rocks along
a railway is described by Kaden (9.49).
The most important advantage of shotcrete in treating
rock slopes is that it offers a rapid, mechanized, and often
uncomplicated solution to rock-fall problems. Various oilier
materials, such as polymers, fiber glass, epoxies, plastic, and
rubberized and asphalt compounds, have been either tried
or considered to protect rock slopes from the effects of cli-
mate and infiltration of groundwater. However, none of
these has yet proved entirely successful on a general basis.
Polyvinyl chloride, butyl, and neoprene sheeting have been
used with limited success in isolated cases, and spray-applied
rubberized bitumen coatings have been used with success in
retarding slaking of shales (9.50).
Retaining Walls manent protection. Basic design and use of retaining walls
are documented by Peck, Hanson, and Thornburn (9.83).
For purposes of rock-slope engineering, retaining walls are The space along railways and highways is often too nar-
used to prevent large blocks in the slope from failing and to row for normal gravity or cantilever types of walls, but an-
control or correct failures by increasing the resistance to chors or rock-bolt tiebacks may be used to overcome this
slope movement. Retaining walls have the advantage of less- problem. Tied-back walls need only have the strength required
ening weathering of the rock slope and of, thus, offering per- for bending and shear resistance between rock bolts. The use
211
Figure 9.27. Cast-in-place reinforced concrete pillar buttresses to fully grouted, untensioned rock bolts.
supporting overhang on road in Austria. Each buttress has two 9-rn
Steel reinforcement of the rock mass, whether considered
(30-ft) long. 356-kN (40-tf) capacity rock bolts.
as rock bolts or rock anchors, has been carried out with
-..--P.- many types of bars, bar bundles, cables, and cable tendons.
0 Both active and passive steel reinforcement systems have
been successfully used for stabilizing rock slopes. The ac-
- OIL .I, - .. tive system involves tensioned or prestressed bars or cables
• that have been anchored at one end within the rock mass
by mechanical means or by cement or chemical grouts or
- by both. The passive system involves untensioned bars that
have been fully grouted throughout their length by cement
or chemical grouts. The active and passive systems are often
#-,t-.;;.1 .._?
compared to prestressed and reinforced concrete, but, since
-. '.•i 1. - - :' the rock mass is a discontinuous medium, the mechanical
properties of the rock mass are different from controlled,
man-made concrete.
' The main advantage of the active or prestressed anchor
•"
- .. IIlI,vll :_ system over the passive system is that no movement has to
: .ffr take place before the prestressed anchor develops its full ca-
pacity; thus, deformation and possible tension cracking of
the slope are minimized. A further advantage of the active
system is that a known anchor force is applied, and proof
loading can be accomplished during installation of each an-
- . . . chor. Also, the tensioned rock-bolt member can be re-
checked, if necessary. at any subsequent time to determine
. -. -. whether the load is being maintained. A greater degree of
confidence in the anchor design is thus provided.
I N General considerations regarding rock-bolt and rock-
JW anchor reinforcement of rock masses are given by Hock and
Londe (9.43) and the U.S. Army Corps of Engineers (9.110).
The reinforcement is usually designed by using the limit equi-
librium met hod of analysis. This method co nsiders only the
ultimate failure condition of the potentially unstable rock
of rock bolts in combination with a concrete facing wall as a mass, which is assumed to fail along a probable failure sur-
gravity section to stabilize a vertical rock face in sedimentary face or surfaces that have been defined by geologic investi-
rock is described by Redlinger and Dodson (9.96). gations. For the active or prestressed system, the anchor
Various types of free-standing and tieback retaining walls force required for stability is applied partly as an increased
include solid cast-in-place concrete walls (9.35), prefabri- compressive normal stress on the failure plane. thus increas-
cated concrete slabs with anchored waling, vertical concrete big frictional resistajice, and partly as a force resisting the
ribs to hold precast concrete panels (9.53), and steel sheet driving force that is causing inst ability of the slope. The
piling with anchored waling. Figure 9.28 shows an interest- value of the component forces will depend on both the de-
ing application of a tieback retaining wall formed of galva- sign geometry and orientation of the anchors and the char-
nized steel members for protection on it high sedimentary acterist ics of the assumed failure surface - For the passive
rock cut requiring face protection. Anchor bolts were system, the maxirnuni available anchor force is determined
grouted in place at I 5-ni (5-fl) verticalintervals to a mini- from the tensile and shear resistance of the steel cross sec-
mum depth in the rock of 1.5 m (5 fi) - Vertical U-channels tion at the assumed failure surface crossed by the anchor.
of 8-gauge galvanized steel were then mounted on the bolts Both systems require adequate anchorage of the steel re-
at 3-in (10-ft ) intervals, and horizontal lagging was placed inforcing member beyond the assumed failure surface; ade-
between them. The void behind the wall was backfilled with quate anchorage involves good bond of the steel to the grout
free-draining material; the water was collected and conducted and good bond of the grout to the rock. Long-term corro-
to a storm sewer. sioii protection of' the anchors must also be provided; this
sometimes requires special sheathing and grouting for the
Steel Reinforcenent prestressed anchors. Grout mixes, installation procedures,
and testing procedures must be carefully considered by the
Steel reinforcement such as rock bolts and rock anchors re- engineer in design and specifications. In this regard the abil-
inforce or tie together the rock massso that the stability ity to test load each anchor after installation is a valuable
of a rock cut or slope is maintained - Rock bolts are corn- field control that is available when prestressed anchors are
nionly used to reinforce the surface or near-surface rock of used. The various types of slope-failure modes that are
the excavation or natural slope, and rock anchors are used judged to be kinematically possible should be considered
for supporting large masses of unstable rock. Short reinforc- in the design of the artificial support system. Actual slope-
ing bars fully grouted into the rock mass are commonly failure conditions in the field may be complex and diflicult
called dowels. Their act ion - however, is somewhat similar to analyze; therefore, it considerable degree of engineering
212
Figure 9.28. Galvanized sheet steel retaining wall that is anchored by shallow rock bolts to prevent failures of high cut slope in sedimentary
rock above highway in Hamilton, Ontario.
- ••'S
DA
-
-
-
I r
e
: L
judgment and experience will continue to be necessary for relaxation or exfoliation by keeping the rock in its original
anchor design. position. The rock-bolt system should be designed in such
Analysis of rock-slope deformation is not generally re- a manner that shear resistance along discontinuities is itii-
quired for transportation routes, except for those involving proved: the rock mass will then acquire significant tensile
support of adjacent structures. Theoretical procedures using strength as a result of the operation.
the finite element met hod are dependent on the assumed To minimize the decompression or loosening effects as-
boundary conditions and properties ofthe rock mass. At the sociated with recently excavated rock slopes, rock bolts
present time, only cases involving small deformations can should be installed and tensioned as soon as possible after
be analyzed wit Ii any reliability. Otherwise, the only proce- each lift of an excavation and preferably before the next
(lure is to use as high a factor of safety as possible against lift is blasted to inhibit dilation effects of the near-surface
ultimate shear failure to minimize undesirable deformation. section of the excavation. The sooner the rock bolts are in-
This problem is discussed by Morgenstern and Fisenstein stalled, the less the subsequent postexcavation movements
(9.75). of the rock mass will be. Figure 9.29 shows a group of large-
diameter rock bolts (980) concentrated in one area at the
Rock Bolts base of a slope to maintain the integrity of large fault blocks.
Figure 930 shows numerous small-diameter rock bolts that
The use of rock bolts is discussed by Lang (9.58) and have been applied over the entire slope in an attempt to se-
Lancaster-Jones (9.56). Rock bolts are basically used to cure smaller individual blocks in the slope.
reinforce the surface and near-surface rock of the excavated The method used for transferring load t'rom the head of
or nat ural slope. In that it is essentially a tension member, the rock bolt to the rock depends largely on the condition
the rock bolt exerts a compressive force, which tends to pre- of the rock. Broken rock may weather away from the bolt
vent elastic rebound, frost action phenomena, and general head and cause the bolt to lose tension. A choice must be
213
Figure 9.29 Concentration on bank of 3.5-cm (1 3/8 -in) diameter madc between protecting the rock face around the head with
14-rn (45-it) long rock bolts at base of section at Hell's Gate Bluffs,
wire mesh and shotcrete, retaining the rock in place by steel
British Columbia. Each rod was cold stretched and stress relieved to
provide minimum ultimate strength of 1.1 MPa (160 000 lbf/in2).
strapping, or distributing the point load by use of concrete
They were tensioned to 712 kN (80 tf) and dropped to 623 kN pads. Usually the tentative design load does not exceed ap-
(70 tf) and then grouted. proximately 60 percent of the ultimate strength of the bolt.
If grouted bolts are used, extreme care should be taken
"e• •' ' ...... .. -
' v. to ensure that the grout does not spread. Excessive grout
spread will reduce the permeability of the rock mass and re-
tard drainage by plugging joints through which groundwater
\Th would normally have free access. The grout must be pumped
4 I
- . • .. into the borehole under low pressure, possibly not exceeding
S -, '..• c. '. 3 ' -,. .. k...
- -.
..
103 kPa (15 lbff in 2), and other measures must be taken to
. retard grout spread. If grouted bolts are used in the lower
k reaches of the slope, free drainage must be maintained.
4 .g.
••\) Rock bolts should be installed by a carefully controlled
Ia procedure. The design of the rock-bolt system should be
•
checked and tested during installation for bolt tension and
adequate anchorage to ensure reliability and performance of
the rock bolts. However, the most important responsibility
of the site engineer is to regularly inspect the slope after the
rock-bolt installation to guarantee that areas that may be-
Figure 9.30. Broadly distributed 1.6-cm (/-in) diameter by 1.2 to
come unstable in the future have been suffIciently supported.
3-rn (4 to 10-ft) long rock bolts at Hell's Gate Bluffs, British
Columbia. These rock bolts were posttensioned to about 53.4 kN
(6 tf) to maintain integrity of small individual blocks in slope. Dowels and Perjb bolts
214
Figure 9.31. Rock dowels used to provide support in sedimentary Figure 9.32. Retaining wall (top) that is partly constructed and
strata that dip toward Rhaetische Railway line in Switzerland (note being cast in place on sound bedrock and finished retaining wall
fence in upper part of slope) (9.84). (bottom) with stone facing. In top photo, note broad base of rock
dowels that provide for added shear strength at toe of retaining
wall as well as excellent bond between concrete and rock.
..;..
..., 'E% . . ....
: ''•' -'.
. •,, . 'v
3 of
.-,.
'1T--
MAJOR FISSURE
Figure 9.33. Geological section of Windy
Point Slide, Australia, showing location of
166.9-kN (375 000-lbf) anchors (average
working load) and drain holes (9.18).
MAJOR FISSURE
.Al. HOLES
/
OCEAN ROAD
20
ANCHORS
No, I ,n 331
215
Figure 9.34. Schematic drawing of trial anchor /
support system including cable anchors,
An interesting anchorage system, shown in Figures 9.34
horizontal stringers, and welded mesh between / and 9.35, was carried out by Barron, Coates, and Gyenge
anchor heads (9.4). 7 (9.4) at a Canadian mine. The system consisted of deep
ANCHOR LOADING1
HORIZONTAL
2 preloaded cable anchors to support the tentatively unstable
STRINGERS ground between the slope face at angle a and some prede-
termined potential failure plane at angle i. Welded wire
mesh was placed over the surface, and horizontal stringers
were installed between the cable terminations to prevent
COVERING
SURFACE
local rock falls and to provide bench support.
A practical guide for the use and installation of anchors
\% is given by Coates and Sage (9.18), and other useful aspects
Cr BOLTS
relating to design and quality control of rock anchors are de-
On CABLES
scribed by Seegmiller (9.102).
Anchored Beams
Anchored beams can be used to distribute the support of
rock bolts over a wide area of the slope and, hence, mini-
mize the number of bolts required. The beams can be made
of concrete or steel and applied in any direction across a
rock face. Anchored beams are particularly useful where
Figure 9.35. Installation of system shown in Figure 9.34: (a) location of two benches where the trial anchor installation is conducted,
(b) inserting the cable in the borehole, (c) formwork for horizontal concrete stringer on lower bench and 66.44 welded wire mesh,
(d) formwork for anchor pads on top bench and steel bar horizontal stringer, (f) cable anchor tensioned and locked, and (f) completed
trial anchor installation on lower bench (9.4).
':.
(c)
or
216
Figure 9.36. Cast-in-place reinforced concrete beams used to Figure 9.38. Cable lashing of a large, potentially hazardous. 2-rn
distribute point loads from rock bolts seated in the beams (9.84). (6-ft) diameter block about 90 m (300 ft) above highway in Fraser
Location is above main line of Austrian Federal Railways. Canyon, British Columbia.
Cable Lashing
217
Methods of Protection ing or sliding rocks are easier to intercept since they are in
constant contact with the slope; in gullies, their paths are
Rock-Fall Characteristics Affecting the even more predictable.
Design of Protection Measures A rock-fall model developed by Piteau and Clayton (9.91)
for a computer program can simulate several hundred rock
Loose rocks can be kept from reaching the transportation falls for a slope of specified geometry in a matter of seconds.
route right-of-way by retaining the rocks on the slope, in- A coefficient of restitution is used for, the rock fall in the
tercepting falling rocks above the roadway, or directing fall- bouncing mode, and increased friction is used for the rolling
ing rocks to pass over or under the transportation route mode to determine the paths of the rock falls for different
without causing harm. Before the best methods can be input velocities and heights. The slope profile is divided into
chosen for a particular location, the characteristics of the straight segments or slope cells (Figure 9.40), which are num-
rock falls and the general geometry of the slope should be bered consecutively from the top to the bottom of the slope.
evaluated. The rate at which fallen rock accumulates is also Rock falls are introduced at different locations above the
an important factor. Without encroaching on the right-of- slope as desired. Probability factors can be assigned to these
way, storage must be provided for rock falls that are likely locations to recognize areas that are either more likely or
to arrive during a reasonable maintenance interval. Protec- less likely to be the source of the rock fall. Catch walls at
tion measures must be designed to protect against the vari- various positions and heights can be assumed in the model
ous ways in which rock falls reach the right-of-way (9.85). to evaluate their effectiveness. In the same manner, the ef-
A rock can arrive at the base of a slope by free falling, fect of varying the slope geometry by including ditches,
bouncing, rolling, or sliding down the slope. For free-falling benches, and berms can be evaluated.
rocks, the only means of protection are to move the roadway This rock-fall model was used to relocate a concrete catch
away from the slope and prevent the rocks from bouncing wall at the base of an active rock-fall area (Figure 9.51). It
and rolling after landing or to protect the roadway with a was also used to determine whether large blocks located at
rock shed or tunnel. The paths of bouncing rocks are diffi- the crest of a major slide area would reach facilities at the
cult to predict, and interception measures above the right- base of the mountainside located some 1500 to 1800 m
of-way require high structures, such as a wall or net. Roll- (5000 to 6000 ft) away from the crest of the slide.
218
Figure 9.41. Intercepting slope ditch or cross ditch, north of Hope Figure 9.42. Shaped berm that is at toe of high rock face and cut in
in Fraser Canyon, British Columbia, to catch rolling rocks originating top of high talus slope above Hope.Agassiz highway in British
up slope on steep rock faces (note that ditch is shaped partly in talus Columbia.
and makes use of blocks of rock).
Relocation a blanket draped over the rock surface to guide falling rock
into the ditch at the base of the slope. The same arrange-
Relocation is most effective where rocks are free falling ment can be used on stony overburden slopes to prevent dis-
from steep rock faces in proximity to the roadway or where lodged stones from rolling down the slope. This practice is
stabilization or some other protection measure is not feasi- commonly used on talus slopes in steep mountainous terrain.
ble. Relocation is feasible when space is available and the Mesh can be combined with long rock bolts to provide a gen-
design criteria are not affected. When combined with proper erally deeper reinforcement. Mesh in combination with both
ditch shaping and possibly with other ditch-level protection, shotcrete and rock bolts provides general reinforcement and
relocation can be the most economical solution and, in some surficial support and retards the deleterious effects of
cases, one of the few solutions to the problem. Unless relo- weathering.
cation practically eliminates the possibility of accidents, Conditions for the use of mesh are particularly suitable
other measures also should be included along with relocation. if no individual rocks are larger than 0.6 to 1 in (2 to 3 ft)
and if the slope is uniform enough for the mesh to be in al-
Intercepting Slope Ditches and Shaped Berms most continuous contact with the slope. If large falling
blocks in certain areas are likely to dislodge or tear the mesh
and present a hazard, rock bolts should be used to reinforce
Slope ditches are used to intercept rock falls partway up the
these particular areas. For anchoring mesh at the top of
slope, and shaped berms are used at the top of the slope.
overburden slopes, posts are cast in concrete blocks in hand-
These methods ensure that rock falls get caught either be-
dug holes. For bedrock slopes,grouted rock bolts or dowels
fore they start to roll or while they are on their way down
are used. A cable is slung between the anchorages: the mesh
the slope. In some locations, ditches can be made to inter-
is then fastened to the cable and rolled down the slope, and
cept rocks and guide them laterally into disposal areas. A
the vertical seams of the mesh are wired together. For wire
typical intercepting slope ditch is shown in Figure 9.41 , and
mesh blankets, the bottom end of the mesh is usually left a
a shaped berm at the base of a steep slope is shown in Figure
meter or so above ditch level (Figure 9.44) and only a nar-
9.42. row ditch is required. The mesh normally used is 9 or II -
Depending on the nature of the terrain, these methods are
gauge galvanized, standard chain-link or gabion wire mesh.
generally inexpensive, simple to construct, and easy to main-
Gabion mesh appears to have an advantage over the standard
tain. An intercepting slope ditch should only be installed on
chain-link materials in that the gabion mesh has a double-
a slope that can accept the introduction of a ditch without
twist hexagonal weave that does not unravel when broken.
the stability of the entire slope being impaired. The ditch
must be designed and built carefully so that the upper slope
is not at any time steeper than planned. The ditch should Protection Methods at Ditch Level
be suitably located to facilitate periodic cleaning by mechan-
ical equipment. Shaped Ditches
Anchored Wire Mesh Depth, width, and steepness of the inside slope and storage
volume of the ditch are important factors in the design of
Wire mesh is a versatile and economical material for use in ditches to contain rock falls. The choice of ditch geometry
protecting the right-of-way from small rocks. Layers of should take into consideration the angle of the slope that in-
mesh are often pinned onto the rock surface to prevent fluences the behavior of falling rocks. Ritchie (9.98) evalu-
small loose rocks from becoming dislodged (Figure 9.43). ated the mechanicsof rock falls from cliffs and talus slopes
Figure 9.44 shows that mesh can also be used essentially as and developed design criteria for ditches. The criteria, which
219
Figure 9.43. Anchored double-twist, hexagonal wire mesh being Figure 9.44. Wire (No. 9) mesh blanket used to control rock falls
Fastened on a high rock face to prevent rock falisonto highway. near Kelso, Washington (9.3). Falling rocks roll down surface of
slope under mesh and drop into ditch.
.:t-
involve the height and angle of slope, depth of ditch, and Figure 9.45. Path of
width of fallout area, are given below (where I In = 33 It) rock trajectory for
various slope angles
and in Figu re 9.45. which also shows the nature of rock tra-
and design criteria
jectories for different slope angles. An example of a typical for shaped ditches
shaped ditch is shown in Figure 9.46. (9.98).
Fallout
Rock Slope
Area Ditch
Height Width Depth
Angle (m) (m) (in)
220
Figure 9.47. Shaped ditches, wire mesh catch fences, and wire Figure 9.48. Special mobile catch fence mounted on flat deck to
protect motorists from rock falls during excavation work and
mesh catch nets (9.84).
scaling in extremely steep terrain.
SHAPED DITCHES WITHOUT CATCH FENCE
(a) (b) T.
ORIGINAL
01TC
l-
-
•-
GABION LEXCAVATED
CATCHMENT AREA
SHAPED DITCHES WITH CATCH FENCE
(c) (dl
-•--.
(-1-
PENCE OR
jNCEVESH
ORIGINAL
(I
--CABLE TIE8ACI(
CA BLE CK
FENCEMH
:i:..
E E :
cated at or near ditch level. The wire mesh (chain-link or
gabion) forming the catch fences is hung on cables supported
CATCH FENCES
on posts or strung between posts or trees. Wire-mesh catch
tg) (h)
fences are usually located on the roadway side of the ditch
.RE MESH ANCHORED
- WIRE .11511 -
ANCHORED NE CABLE or at the base of the slope and can be used with or without
CABLE a shaped ditch. The fence should be suitably situated so
that accumulated rocks can be removed easily. Figure 9.31
E LI VAT ON shows a catch fence, and Figure 9.48 shows a special appli-
CATCH FENCES CATCH NET cation of a mobile catch fence for use in scaling operations.
ACROSS SLOPE OVER GULLY
Catch Walls
catchment area, as shown in Figure 9.47b (9.84). by form-
ing a windrow of material or by installing a low barrier Catch walls can be used to form a rigid barrier to stop roll-
formed of standard highway guardrail, precast concrete ele- ing or bouncing rocks as large as 1 .5 to 2 ni (S to 6 ft) from
ments, or gabions at the shoulder of the roadway. as shown reaching the right-of-way. If effective, they usually increase
in Figure 9.47a (9.84). Bedrock should not remain exposed the storage capacity of (lie ditch so that maintenance inter-
in the bottom of ditches. but should be covered with small
vals can be extended. In many locations, large ditches them-
broken rock or loose earth to keep falling rocks from bounc selves are not effective for intercepting large rolling rocks
ing or shattering. Mearns (9.71) describes how a ditch at and the use of catch walls is advised.
grade was filled with a 0.3-in (I -ft) layer of sand, which To achieve maximum protection and storage capacity,
acted as an energy absorber to prevent lolling rock falls the catch walls should be located oil the side of the ditch
from reaching the road. Several workers have used a simi- closest to the road. In steep terrain, catch walls are commonly
lar technique to successfully dissipate energy of rock falls. used where the right-of-way cuts across postglacial slide areas.
Many postglacial slide areas that have been disturbed by ex-
Wire-Mesh catch iVets and Fences cavation are in a constant state of sloughing and readjust-
ment. If broad areas require wall protection, gaps should be
Wire mesh can be effective in intercepting or effectively left in the catch wall to allow access of maintenance equip-
slowing bouncing rocks as large as 0.6 to I in (2 to 3 ft) ment required to remove rock.fall debris.
when the mesh is mounted as a flexible catch net rather
Concrete ('arc/i Walls
than as I standard, fixed wire fence. If suspended oil a ca-
ble. the mesh will absorb the energy of flying rocks with a Concrete catch walls are the most widely used type of catch
minimum of damage to the wire catch net. Catch nets can wall in steep mountainous terrain. Concrete walls may be
consist of standard chain-link wire mesh or gabion mesh. cast in place or precast in short sectioisand assembled on
This type of barrier has economic potential in some trouble- the site. An efficient precast concrete wall installation is
sonic locations, as shown in Figure 9.47li (9.84). It is best shown in Figure 9.49 (9.84). Figures 9.50 and 9.51 show
to locate catch nets al the lower end of steep gullies where
rocks tend to bounce down to the right-of-way. The cable TIES AT 40.6cm c-to c. TO REDUCE
Figure 9.49. PrecaSt
supporting the catch net is anchored to sound rock OII either IMPACT AND INCREASE HEIGHT OF
concrete wall that WALL FOR GREATER PROTECTION
side of the gully with the result that rocks hit the catch net could be used for \.
ELECTRIC WARNING
fwIRE IOPTIONALI
and fall harmlessly at the base of the structure. If there is a track protection
LI
steady accumulation of rock, a catch wall on the shoulder of (9.84). ORIGINAL
- DITCH -I-
the roadway also may be required. -
CABLE
09m
STRAPPING-. I.
The principle ofthe catch fence is similar to that of a -
09m
catch net. Its purpose is to form a flexible barrier to dissi- EXCAVATED
CATCH MENT ELL COMP ACTED ~ ~-PRECAST
pate the energy of rapidly moving rocks. Various arrange- AREA
BACKFILL ----i 0.9-M-1 cONCRETE
\_ CATCH WALL
ments are shown in Figure 9.47 (9.84) for catch fences to- NOte. 1 r,, 3.3 It; 1 ccc -0.4 ,n.
221
Figure 9.50. Reinforced cast-in-place concrete wall approximately Figure 9.51. Close-up of concrete catch wall shown in Figure 9.50
2 m (7 ft) high and 1 m (3 ft) wide across a major postglacial slide (note breaks in wall due to large blocks from postglacial slide debris
area along edge of Trans-Canada Highway in Fraser Canyon, British rolling down slope).
Columbia (note break in retaining wall where large blocks have hit
wall).
- --
if ..'y
fY
1:
E•,
,•
.-
:i
-
- .. ,•F t'1.'!J
-
TTI -,. --
The box gabion catch wall is a rectangular basket divided Cable Wa/Is
by diaphragms into smaller rectangles that are filled with
slones. The basket is formed of woven hexagonal steel gal- Cable walls consist of steel posts set in concrete, cables
vanized wire mesh. Baskets can be placed to be filled indi- strung horizontally between, and a smaller cable or coarse
vidually or wired logether in groups and filled accordingly. wire woven between the cables to form a crude net or niesh.
The wire mesh tends to reinforce the stone in tension. The This type of installation is sometimes used in Europe.
gabion is a flexible structure that, upon settling or being hit
by impact. tends to deflect and deform instead of break. Rock Sheds and Tunnels
Because gabion walls are highly deformable, differential
settlement is not important. Like the gabion mesh, the ga- Rock sheds and tunnels can be used for protection against
bion does not unravel if broken. Gabions make good catch- rock falls and slides when warranted and when other forms
wall structures, but only recently have they been recognized of stabilization and protection are not effective. Although
in North America as a feasible alternative to more rigid con- expensive, they can give complete protection and should be
crete walls for protecting the right -of-way from rolling rocks. considered in areas with serious problems. Maintenance
They prove particularly useful where the right-of-way crosses costs are normally negligible. The methods of design and
postglacial slide areas. They can be used efficiently to stop construct ion of tunnels are dealt with thoroughly in the
rolling stones as large as 0.6 to I iii (2 to 3 ft). When ade- literat tire. but the design of rock sheds is not so adequately
qua te filter materials are used as backfill, they provide long- covemed, and experience is required to decide on the most
term, free-draining walls. A gabion catch wall is shown in suitable type of structure and the loads to be carried. A
Figure 9.52.
rock shed should be able to resist the energy transmitted by
the largest rock mass likely to pass over it during its life;
Rail Walls
therefore, probability analysis should be involved. The en-
ergy transmitted will depend on whether rocks are falling
Rail walls or rail-and-tie walls consist of vertical posts and hee, bouncing, or rolling. High stress concentrations iii the
horizontal members that are extended between the vertical structure can be reduced by the provision of a thick cover
posts. The vertical posts are usually scrap steel rails set in of loose sand.
holes, which are hand dug or blasted and backfilled with If foundation conditions on the outer side of the road-
concrete. The horizontal members are either ties, as shown bed are not suitable for footings, heavy rock anchors cx-
tending into the upper slope may be used to support a patrol. Patrols have the advantages of being reliable and flex-
cantilevered shed of the type shown in Figure 9.55. Tun- ible, and their frequency can be adjusted to the demands of
nel portals can also act as sheds to protect the roadway or traffic and weather conditions. The disadvantages are that
track, as shown in Figure 9.56. When a rock shed is to be they incur continuing costs and require personnel who are
located at the lower end of a gully, wing walls are usually willing to work in uncomfortable and often hazardous con-
used above the structure to channel material onto it (Fig- ditions.
ure 9.57). Wing walls should be sufficiently high because
the debris will tend to clog and build up when its path is Electric Fences and Wires
restricted by the structure. For this reason, the slope angle
of the roof of the shed should be steeper than the angle of There are several variations of electric warning methods.
repose of the material to be conveyed over the roadway. Electric fences are based on the principle that a falling rock
large enough to endanger traffic will break or pull out one
Methods of Wanfing of the wires and thus actuate a signal to warn approaching
traffic. This principle is particularly adaptable to railways
Although warning systems do not prevent rock falls, they
on which a signal system to control traffic is already in use.
are necessary on transportation routes where other measures
The standard electric warning fence used on railways con-
are too expensive or impractical or where a new hazard has
sists of a row of poles spaced along the uphill ditch line and
developed. In North America, warning methods have been
wires strung between them at a vertical spacing of 25 cm
used on railways in mountains to detect rock falls on tracks
(10 in). Overhead wires, which are supported on members
SO that trains can stop before hitting the material.
cantilevered out from the top of the poles, as shown in the
Patrols installation in Figure 9.58, are often required where rock
The simplest type of warning method is provided by human faces are steep and close to the right-of-way.
Figure 9.52. Gabion catch wall along edge of high slope in Figure 9.54. Rail wall with vertical posts seated in cast-in-place
unconsolidated material on main highway. footings (note top of wall anchored by cables to bedrock for
support) -
Figure 9.53. Rail and tie wall founded on masonry wall to prevent Figure 9.55. Cantilevered rock shed supported by deep anchors
rolling rock from reaching track of Austrian Railway. inserted in upper slope protecting rail line in Switzerland (9.84).
)':ic-•- ,.'______.
• t '.,
.•4 J
223
Figure 9.56. Tunnel in steep terrain with extended portal shedlike structure to protect Canadian National Railway line from rock falls
generated in a slide area (note use of retaining walls of rock blocks below track).
I 4. .
L'• :-
A
.
14 :
-i
Figure 9.57. Rock sheds to carry rock debris originating in deep Figure 9.58. Standard type of railway electric warning fence with
gullies over main line of Canadian National Railway in White wires strung between upright poles and horizontal members that
Canyon in Thompson River Valley, British Columbia. are cantilevered out from top of poles along Canadian National
Railway line in Fraser Canyon, British Columbia.
p.,
Electric warning fences have some advantages, and the Supporting overhead wires on a canopy type of frame
design of each facility can be made to suit the individual bolted to the rock face on the uphill side and supported on
sites. One disadvantage is low efficiency: in some instances poles on the downhill side.
80 percent of the alarms have been found to be false. An-
other disadvantage is that snow-clearing operations. if re- A particularly effective type of electric warning system
quired. are often impeded and maintenance of the fences consists of a single wire, anchored at both ends and linked
proVeS difficult. These difficulties may be reduced or elim- to a warning signal (9.85). Such a wire may be fastened
inated by
around a large unstable rock or across a rock slope above
the right-of-way, across a gully where large rocks roll down,
I. Choosing the spacing of wires according to previous or on top of a protective catch wall.as shown in Figure 9.49.
experience at a site. The installation is simple, economical, and efficient.
Providing a catchmcnt ditch behind the warning fence.
Eliminating the lower wires where the lower slope can Other Methods
be scaled and stabilized, and Warning methods that are dependable under all conditions
224
are undergoing continued study. At this writing, the follow- Guide for Their Utilization and Installation. Mines Branch,
ing are being tried: (a) geophones or vibration meters buried Canada Department of Energy, Mines and Resources, Ot-
tawa, Research Rept. R224, 1973.
at intervals along the roadway shoulder to pick up vibrations
9.19 Crimmins, R. S., Samuels, R., and Monahan, B. P. Con-
from falling rocks by the Canadian National Railways and struction Rock Work Guide. Wiley, New York, 1972,
Swedish State Railways; (b) television monitoring by the 235 pp.
Federal Highway Administration; (c) guided radar by the 9.20 Da Silveira, A. F., Rodrigues, F. P., Crossman, N. F., and
Canadian Institute of Guided Ground Transport and the Mendes, F. Quantitative Characterization of the Geomet-
nc Parameters of Jointing in Rock Masses. Proc., 1st Con-
Japanese National Railways and (d) laser beams by the gress, International Society of Rock Mechanics, Lisbon,
Radio Corporation of America. However, none is known Vol. 1, 1966, pp. 225-233.
to have been sufficiently developed to be recommended 9.21 Deere, D. U., Hendron, A. J., Patton, F. D., Jr., and
for general use. Cording, E. J. Design of Surface and Near-Surface Con-
struction in Rock. In Failure and Breakage of Rock
(Fairhurst, C., ed.), Proc., 8th Symposium on Rock Me-
REFERE,ES chanics, American Institute of Mining, Metallurgical and
Petroleum Engineers, New York, 1967, pp. 237-302.
9.1 Alberts, C. Bergförstrkning genom betonsprutning och 9.22 Donn, W. L., and Shinier, J. A. Graphic Methods in Struc-
injektering. Swedish Academy of Engineering Sciences, tural Geology. Appleton-Century-Crofts, Englewood Cliffs,
Stockholm, IVA Pub!. 142, 1970, pp. 231-240. N.J., 1958, 180 pp.
9.2 American Concrete Institute. Use of Shotcrete for Under- 9.23 Duncan, N. Geology and Rock Mechanics in Civil Engi-
ground Structural Support. ACI, Detroit, Pub!. SP-45, neering Practice. Water Power, Jan. 1965, pp. 25-32.
1974, 467 pp. 9.24 Eckert, 0. Consolidation de massifs rocheaux par ancrage
9.3 Baker, R. F., and Marshall, H. C. Control and Correction. de cables/Consolidation of Rock Masses by Cable Anchors.
In Landslides and Engineering Practice (Eckel, E. B., ed.), Sols-Soils (Paris), Vol. 5, No. 18, 1966, pp. 33-40 (in
Highway Research Board, Special Rept. 29, 1958, pp. English).
150-187. 9.25 Ehrlich, M. E., Scharon, H. L., and Mateker, E. J., Jr. Vi-
9.4 Barron, K., Coates, D. F., and Gyenge, M. Artificial Sup- bration Analysis and Construction Blasting. In Legal As-
port of Rock Slopes. Mines Branch, Canada Department pects of Geology in Engineering Practice, Geological
of Energy, Mines and Resources, Ottawa, Research Rept. Society of America, Engineering Geology Case Histories,
R228, 1971, 145 pp. No. 7, 1969, pp. 13-20.
9.5 Barton, N. Review of a New Shear-Strength Criterion for 9.26 Fookes, P. G., and Sweeney, M. Stabilization and Con-
Rock Joints. Engineering Geology, Vol. 7, No. 4, 1973, trol of Local Rock Falls and Degrading Rock Slopes.
pp. 287-33 2. Quarterly Journal of Engineering Geology, Vol. 9, No.
9.6 Bauer, A. The Status of Rock Mechanics in Blasting. In 1, 1976, pp. 37-55.
Status of Practical Rock Mechanics (Grosvenor, N. E., and 9.27 Frster, W. The Influence of the Curvature of Open Cuts
Paulding, B. W., Jr., eds.), Proc., 9th Symposium on Rock on the Stability of Slopes in Open Work Mining. Proc., 1st
Mechanics, American Institute of Mining, Metallurgical Congress, International Society of Rock Mechanics, Lisbon,
and Petroleum Engineers, New York, 1967, pp. 249-262. Vol. 1, 1966, pp. 193-200.
9.7 Behr, H., and Klengel, K. J. Stability of Rock Slopes on 9.28 Franklin, J. A., and Denton, P. E. Monitoring for Rock
the German State Railway. Deutsche Eisenbahntechnik, Slopes. Quarterly Journal of Engineering Geology, Vol.6,
Germany, 1966, pp. 324-328. No. 3, 1973, pp. 259-286.
9.8 Benson, R. P. Experience With Instrumentation in Rock 9.29 Goodman, R. E. Methods of Geological Engineering in
Slopes. Acres Consulting Engineers, Niagara Falls, N.Y., Discontinuous Rocks. West Publishing, St. Paul, 1976,
1971,49pp. 472pp.
9.9 Bjerrum, L., and J$rstad, F. A. Stability of Rock Slopes 9.30 Halstead, P. N., Call, R. D., and Rippere, K. H. Geologic
in Norway. Norwegian Geotechnical Institute, Oslo, Publ. Structural Analysis for Open Pit Slope Design, Kimberley
79, 1968, 11 pp. Pit, Ely, Nevada. Paper presented at 97th Annual Meet-
9.10 Brekke, T. L. Shotcrete in Hard Rock Tunneling. Bulletin, ing, American Institute of Mining, Metallurgical and Pc-
Association of Engineering Geologists, Vol. 9, No. 3, 1972, troleurn Engineers, New York, 1968, 25 pp.
pp. 241-264. 9.31 Hamel, J. V. Stability of Slopes in Soft, Altered Rocks.
9.11 Bukovansky, M., and Piercy, N. H. High Road Cuts in a Univ. of Pittsburgh, PhD thesis, 1969.
Rock Mass With Horizontal Bedding. In Design Methods 9.32 Hast, N. The State of Stress in the Upper Part of the
in Rock Mechanics (Fairhurst, C., and Crouch, S. L., eds.), Earth's Crust. Engineering Geology, Vol. 11, No. 1, 1967,
Proc., 16th Symposium on Rock Mechanics, Univ. of pp. 5-17.
Minnesota, Minneapolis, American Society of Civil Engi- 9.33 Hedley, D. G. F. Design Parameters for Borehole Exten-
neers, New York, 1977, pp. 47-52. someter Measuring Systems. Mines Branch, Canada De-
9.12 Bureau of Public Roads. Presplitting: A Controlled Blast- partrnent of Energy, Mines and Resources, Ottawa, Inter-
ing Technique for Rock Cuts. U.S. Department of Corn- nat Rept. MR67/59-LD, 1967.
merce, 1966, 36 pp. 9.34 Hendron, A. J. Analytical and Graphical Methods for the
9.13 Calder, P. N. Slope Stability in Jointed Rock. Bulletin, Analysis of Slopes in Rock Masses. Department of Civil
Canadian Institute of Mining, May 1970. Engineering, Univ. of Illinois at Urbana-Champaign, July
9.14 Call, R. D. Analysis of Geologic Structure for Open Pit 1971, 178 pp.
Slope Design. Univ. of Arizona, Tucson, PhD thesis, 9.35 Hcnnequin, M., and Cambefort, H. Consolidation du
1972, 201 pp. remblai de Malherbe/Stabilization of Fill at Malherbe.
9.15 Cedergren, H. R. Seepage, Drainage and Flow Nets. Revue General des Chernins de Fer (Pails), Vol. 85, Jan.
Wiley, New York, 1967, 490 pp. 1966, pp. 78-86.
9.16 Coates, D. F. The Stability of Slopes in Open Pits. Proc., 9.36 Herget, G. Surveys of Geological Discontinuities in Con-
8th Commonwealth Mining and Metallurgical Congress, nection With the Mechanical Behavior of Rock Masses.
Australian Institute of Mining and Metallurgy, Melbourne, Mines Branch, Canada Department of Energy, Mines and
Vol. 6, 1967, pp. 543-550. Resources, Ottawa, Internal Rept. MR70/38, 1970, 28 pp.
9.17 Coates, D. F., and Gycnge, M. Incremental Design in Rock 9.37 Heuze, F. E., and Goodman, R. E. Three-Dimensional Ap-
Mechanics. Mines Branch, Canada Department of Energy, proach for Design of Cuts in Jointed Rock. In Stability of
Mines and Resources, Ottawa, Monograph 880, 1973. Rock Slopes (Cording, E. J., ed.), Proc., 13th Symposium
9.18 Coates, D. F., and Sage, R. Rock Anchors in Mining: A on Rock Mechanics, Univ. of Illinois at Urbana-Champaign,
/1 225
American Society of Civil Engineers, New York, 1971, pp. 9.60 Larocque, G. A Proposed Method of Establishing Bench
397-441. Blast Patterns. Mines Branch, Canada Department of En-
9.38 Highway Focus. Vol. 1, No. 3, Aug. 1969. ergy, Mines and Resources, Ottawa, Internal Rept. MR71/
9.39 Highway Focus. Vol. 4, No. 2, June 1972. 93-I1), 1971.
9.40 Highway Research Board. Erosion Control on Highway 9.61 Legget, R. F. Cities and Geology. McGraw-Hill, New
Construction. HRB, Synthesis of Highway Practice 18, York, 1973, 624 pp.
1973, 52 pp. 9.62 Leighton, F. B. Landslides and Hillside Development. In
9.41 Highway Research Board. Soil Erosion: Causes and Mech- Engineering Geology in Southern California, Association
anisms, Prevention and Control. HRB, Special Rept. 135, of Engineering Geologists, Special Publ., 1966, pp. 149-
1973, 141 pp. 207.
9.42 Hoek, E., and Bray, J. W. Rock Slope Engineering. Insti- 9.63 Lewis, H., McDaniel, A. H., Peters, R. B., and Jacobs,
tution of Mining and Metallurgy, London, 1974, 309 pp. D. M. Damages Due to Drainage Runoff, Blasting and
9.43 Hock, E., and Londe, P. Surface Workings in Rock. Proc., Slides. National Cooperative Highway Research Program,
3rd Congress, Denver, International Society of Rock Me- Rept. 134, 1972, 23 pp.
chanics, Vol. 1, Pt. A, 1974, pp. 613-654. 9.64 Long, A. E., Merrill, R. H., and Wisecarver, D. W. Stability
9.44 International Society of Rock Mechanics. Suggested of High Road Bank Slopes in Rock: Some Design Con-
Methods for the Description of Rock Masses, Joints and cepts and Tools. Highway Research Board, Highway Re-
Discontinuities. ISRM, 1976. search Record 135, pp. 10-26.
9.45 Jaeger, J. C. Brittle Fracture of Rocks. In Failure and 9.65 Lorente de No, C. Stability With Curvature in Plane View.
Breakage in Rock (Fairhurst, C., ed.), Proc., 8th Sympo- Proc., 7th International Conference on Soil Mechanics and
sium on Rock Mechanics, American Institute of Mining, Foundation Engineering, Mexico City, Vol. 2, 1969, pp.
Metallurgical and Petroleum Enginners, New York, 1967, 63 5-6 3 8.
pp. 3-57. 9.66 Lutton, R. J. Rock Slope Chart From Empirical Slope
9.46 Jennings, J. E. A Preliminary Theory for the Stability of Data. Trans., Society of Mining Engineers, American In-
Rock Slopes Based on Wedge Theory and Using Results stitute of Mining, Metallurgical and Petroleum Engineers,
of Joint Surveys. Univ. of Witwatersrand, South Africa, Vol. 247, June 1970, pp. 160-162.
1968. 9.67 Mahtab, M. A., Bolstad, D. D., and Kendorski, F. S. Anal-
9.47 Jennings, J. E. A Mathematical Theory for the Calcula- ysis of the Geometry of Fractures in San Manuel Copper
tion of the Stability of Slopes in Open Cast Mines. In Mine, Arizona. Bureau of Mines, U.S. Department of the
Planning Open Pit Mines, Johannesburg (van Rensburg, Interior, Rept. RI 7715, 1973, 24 pp.
P. W. J., ed.), Proc., Open Pit Mining Symposium, Johan- 9.68 Martin, D. C., and Piteau, D. R. Application of Waviness
nesburg, South African Institute of Mining and Metallurgy, of Structural Discontinuities to Rock Slope Design. Proc.,
Balkema, Amsterdam, 1971, pp. 87-1 02. 29th Canadian Geotechnical Conference on Slope Design,
9.48 John, K. W. Graphical Stability Analysis of Slopes in Vancouver, B.C., Canadian Geotechnical Society, Montreal,
Jointed Rock. Journal of Soil Mechanics and Foundations 1976, pp. VIIl.1-VIII.17.
Division, American Society of Civil Engineers, New York, 9.69 McAnuff, A. L. Blasting Is an Art as Well as a Science.
Vol. 94, No. SM2, 1968, pp. 497-526. Canadian Consulting Engineer, Vol. 15, No. 6, 1973, pp.
9.49 Kaden, R. A. Steel Fibrous Shotcrete. Western Construc- 24-26.
tion, Vol. 49, No. 4, April 1974. 9.70 McCauley, M. L. Engineering Geology Related to High-
9.50 Kilburn, J. Discussion of Papers. In Planning Open Pit ways and Freeways. In Engineering Geology in Southern
Mines (van Rensburg, P. W. J., ed), Proc., Open Pit Mining California, Association of Engineering Geologists, Special
Symposium, Johannesburg, South African Institute of Publ., 1966, pp. 117-121.
Mining and Metallurgy, Balkema, Amsterdam, 1971, pp. 9.71 Mearns, R. Solving a Rockfall Problem in Nevada County,
270-271. California. Highway Research News, No. 49, Autumn
9.51 Kley, R. J., and Lutton, R. J. A Study of Selected Rock 1972, pp. 14-17.
Excavations as Related to Large Nuclear Craters. U.S. 9.72 Mehra, S. R., and Natarajan, T. K. Landslide Analysis and
Army Corps of Engineers, Plowshare, PNE-5010, 1967, Correction. Central Road Research Institute, New Delhi,
157 pp. India, 1966.
9.52 Knill, J. L. Collecting and Processing of Geological Data 9.73 Merrill, R. H., and Wisecarver, D. W. The Stress in Rock
for Purposes of Rock Engineering. Univ. of Alberta, Ed- Around Surface Openings. In Failure and Breakage in
monton, 1971, 35 pp. Rock (Fairhurst, C., ed.), Proc., 8th Symposium on Rock
9.53 Kraus, J., and Tyc, P. Sanace Svahu Zeleznicnich Nasypu Mechanics, American Institute of Mining, Metallurgical
a Zarezu/Stabilization of Railway Fills and Cuts. and Petroleum Engineers, New York, 1967, pp. 337-350.
Nakladatelstvi Dopravi a spoiu, Prague, 1965, 367 pp. 9.74 Mollard, J. D. Photo Analysis in Interpretation in Engi-
9.54 Krsmanovic, D. Initial and Residual Shear Strength in neering Geology Investigations: A Review. In Reviews
Hard Rocks. Geotechnique, Vol. 17, No. 2, 1967, pp. in Engineering Geology, Geological Society of America,
145-16 0. Vol. 1, 1962.
9.55 Lambooy, P., and Espley-Jones, R. C. Practical Considera- 9.75 Morgenstern, N. R., and Eisenstein, Z. Methods of Esti-
tions of Blasting in Open Pit Cast Mines. In Planning Open mating Lateral Loads and Deformations. Proc., Specialty
Pit Mines (van Rensburg, P. W. J., ed.), Proc., Open Pit Conference, Cornell Univ., Ithaca, N.Y., Soil Mechanics
Mining Symposium, Johannesburg, South African Institute and Foundations Division, American Society of Civil En-
of Mining and Metallurgy, Balkema, Amsterdam, 1971, pp. gineers, New York, 1970, pp. 5 1-102.
22 7-234. 9.76 MUller, L. Application of Rock Mechanics in the Design
9.56 Lancaster-Jones, P. F. F. Methods of Improving the Prop- of Rock Slopes. In State of Stress in the Earth's Crust
erties of Rock Masses. In Rock Mechanics in Engineering (Judd, W., ed.), Elsevier, New York, 1964, pp. 575-598.
Practice (Zienkiewicz, (5. C., and Stagg, D., eds.), Wiley, 9.77 MUller, L. The Stability of Rock Bank Slopes and the Ef-
New York, 1968, pp. 385-429. fect of Rock Water on Same. International Journal of
9.57 Lang, L. C., and Favreau, R. F. A Modern Approach to Rock Mechanics and Mining Sciences, Vol. 1, No. 4,
Open-Pit Blast Design and Analysis. Bulletin, Canadian 1964, pp. 475-504
Institute-of Mining, June 1972, pp. 3745. 9.78 Murrell, S. A, and Misra, A. K. Time-Dependent Strain
9.58 Lang, T. A. Rock Reinforcement. Bulletin, Association on Creep in Rocks and Similar Non-Metallic Materials.
of Engineering Geologists, Vol. 9, No. 3, 1972, pp. 215- Trans., Institution of Mining and Metallurgy, London,
239. Vol. 71, 1962, pp. 353-378.
9.59 Langefors, U., and Kihlstrom, B. The Modern Technique 9.79 National Academy of Sciences. Better Contracting for
of Rock Blasting. Wiley, New York, 1963. Underground Construction. U.S. National Committee on
226
Tunneling Technology, National Academy of Sciences, 9.100 Root, A. W. Prevention of Landslides. In Landslides and
Washington, D.C.,.1974, 143 pp. Engineering Practice (Eckel, E. B., ed.), Highway Research
9.80 Nesbitt, M. D. The Stabilization of Rock Slopes at Hell's Board, Special Rept. 29, 1958, pp. 113-149.
Gate, British Columbia. Proc., Canadian Good Roads As- 9.101 Ross-Brown, D. M., and Atkinson, K. B. Terrestrial Photo-
sociation, Sept. 1967, pp. 295-309. grammetry in Open Pits: Description and Use of the
9.81 Norman, J. W. The Photogeological Detection of Unstable Phototheodolite in Mine Surveying. Trans., Institution
Ground. Journal of Institution of Highway Engineers of Mining and Metallurgy, London, Vol. 81, pp. 205-213.
(London), Vol. 18, No. 2, 1970, pp. 19-22. 9.102 Seegmiller, B. L. Artificial Stabilization in Open Pit Mines:
9.82 Northwood, T. D., and Crawford, R. Blasting and Build- A New Concept to Achieve Slope Stability. Paper pre-
ing Damage. Canada National Research Council, Ottawa, sented at Intermountain Minerals Conference, Vail, Cob.,
CBD 63, 1965,4 pp. American Institute of Mining, Metallurgical and Petrol-
9.83 Peck, R. B., Hanson, W. E., and Thornburn, T. H. Founda- eum Engineers, New York, 1974.
tion Engineering. Wiley, New York, 2nd Ed., 1973, 514 9.103 Serafim, J. L. Influence of Interstitial Water on the Behav-
pp. ior of Rock Masses. In Rock Mechanics in Engineering
9.84 Peckover, F. L. Treatment of Rock Falls on Railway Lines. Practice (Zienkiewicz, 0. C., and Stagg, D., eds.), Wiley,
American Railway Engineering Association, Bulletin 653, New York, 1968, pp. 55-97.
Chicago, 1975, pp. 471-503.. 9.104 Shuk, T. E. A Simple Method to Minimize Cost of Rock
9.85 Peckover, F. L., and Kerr, J. W. G. Treatment and Main- Slopes. In Estimation of Upper Bounds to Rock Slopes
tenance of Rock Falls on Transportation Routes. Cana- by Analysis of Existing Slope Data, Mines Branch, Canada
dian Geotechnical Journal, Vol. 14, No. 4, 1977, pp. Department of Energy, Mines and Resources, Ottawa,
487-50 7. Rept. 76-14, 1976, pp. 18-34.
9.86 Phillips, F. C. The Use of Stereographic Projection in 9.105 Stacey, T. R. Stability of Rock Slopes in Mining and Civil
Structural Geology. Arnold, London, 1971. Engineering Situations. National Mechanical Engineering
9.87 Piteau, D. R. Geological Factors Significant to the Stabil- Research Institute, Council for Scientific and Industrial
ity of Slopes Cut in Rock. In Planning Open Pit Mines Research, Pretoria, MEG/4415, 1973, 193 pp.
(van Rensburg, P. W. J., ed.), Proc., Open Pit Mining Sym- 9.106 Steffen, 0. K. H., and Jennings, J. E. Definition of Design
posium, Johannesburg, South African Institute of Mining Joints for Two-Dimensional Rock Slope Analyses. Proc.,
and Metallurgy, Balkema, Amsterdam, 1971, pp. 33-53. 3rd Congress, International Society of Rock Mechanics,
9.88 Piteau, D. R. Characterizing and Extrapolating Rock Joint Denver, Vol. 2, Part 13, 1974, pp. 827-832.
Properties in Engineering Practice. Rock Mechanics, Supp. 9.107 Stimpson, B. S. Physical Properties of Rock. Univ. of
2, Springer-Verlag, Vienna, 1973,31 pp. California, Berkeley, March 1976, 70 pp.
9.89 Piteau, D. R. Plan Geometry and Other Factors Relating 9.108 Terzaghi, K. Stability of Steep Slopes in Hard Unweathered
to Natural Rock Slope Stability Applied to Design of Rock. Geotechnique, Vol. 12, No. 4, 1962, pp. 25 1-270.
DeBeers Mine. Mines Branch, Canada Department of En- 9.109 Terzaghi, R. Sources of Error in Joint Surveys. Geotech-
ergy, Mines and Resources, Rept. TB 190, 1974, 120 pp. nique, Vol. 15, No. 3, 1965, pp. 287-304.
9.90 Piteau, D. R. Regional Slope Stability Controls and Re- 9.110 U.S. Army Corps of Engineers. Rock Reinforcement in
lated Engineering Geology of the Fraser Canyon, British Civil Engineering Works. U.S. Department of the Army,
Columbia. In Landslides (Coates, D. R., ed.), Geological EM 1110-1-2907, Feb. 1975.
Society of America, 1977. 9.111 Yu, Y., and Coates, D. F. Analysis of Rock Slopes Using
9.91 Piteau, D. R., and Clayton, R. Discussion of paper, Com- the Finite Element Method. Mines Branch, Canada De-
puterized Design of Rock Slopes Using Interactive Graphics partment of Energy, Mines and Resources, Ottawa, Re-
for the Input and Output of Geometrical Data, by P. A. search Rept. R229, 1970.
Cundall, M. D. Voegele, and C. Fairhurst. In Design 9.112 Záruba, Q., and Mend, V. Landslides and Their Control.
Methods in Rock Mechanics (Fairhurst, C., and Crouch, Elsevier, New York, and Academia, Prague, 1969, 205 pp.
S. L., eds.), Proc., 16th Symposium on Rock Mechanics,
Univ. of Minnesota, Minneapolis, American Society of
Civil Engineers, New York, 1977, pp. 62-63.
9.92 Piteau, D. R., and Jennings, J. E. The Effects of Plan PHOTOGRAPH CREDITS
Geometry on the Stability of Natural Slopes in Rock in
the Kimberley Area. Proc., 2nd Congress, International
Society of Rock Mechanics, Belgrade, Vol. 3, 1970, pp. Figure 9.1 D. R. Piteau, Piteau and Associates
289-295. Figure 9.2 D. R. Piteau, Piteau and Associates
9.93 Piteau, D. R., McLeod, B. C., Parkes, D. R., and Lou, J. K. Figure 9.11 D. Johnson; courtesy of British Columbia Department
Overturning Rock Slope Failure at Hell's Gate, British of Highways
Columbia. In Geology and Mechanics of Landslides and Figure 9.12 G. Miller, Miller Engineering Surveys, Ltd.
Avalanches (Voight, B., ed.), Elsevier, New York, 1977. Figure 9.13 D. R. Piteau, Piteau and Associates (photograph);
9.94 Pit Slope Manual. Canada Department of Energy, Mines British Columbia Department of Highways (sketch)
and Resources, Ottawa, 1977. Figure 9.14 Courtesy of A. Rutka, Ontario Ministry of Transporta-
9.95 Rawlings, G. E. Stabilization of Potential Rock Slides in tionand Communications
Folded Quartzite in Northwestern Tasmania. Engineering Figure 9.16 Courtesy of G. E. Crippen and Associates, Ltd.
Geology, Vol. 12, No. 5, 1968, pp. 283-292. Figure 9.18 J. Rupert, British Columbia Department of Highways
9.96 Redlinger, J. F., and Dodson, E. L. Rock Anchor Design. Figure 9.19 D. Johnson, courtesy of British Columbia Department
Proc., 1st Congress, International Society of Rock Me- of Highways
chanics, Lisbon, Vol. 3, 1966, pp. 173-1 75. Figure 9.21 D. R. Piteau, Piteau and Associates
9.97 Reiche, P. A Survey of Weathering Processes and Products. Figure 9.22 D. R. Piteau, Piteau and Associates
Univ. of New Mexico, Albuquerque, Geology PubI. 3, Figure 9.23 J. W. Stewart; courtesy of Dolmage, Mason and Stewart,
1950,95 pp. Ltd.
9.98 Ritchie, A. M. Evaluation of Rockfall and Its Control. Figure 9.25 D. R. Piteau, Piteau and Associates
Highway Research Board, Highway Research Record 17, Figure 9.26 D. R. Piteau, Piteau and Associates
1963, pp. 13-28. Figure 9.27 F. L. Peckover
9.99 Robertson, A. M. The Interpretation of Geological Fac- Figure 9.28 Courtesy of Dofasco
tors for Use in Slope Theory. In Planning Open Pit Mines Figure 9.29 D. R. Piteau, Piteau and Associates
(van Rensburg, P. W. J., ed.), Proc., Open Pit Mining Sym- Figure 9.30 B. McLeod; courtesy of British Columbia Department
posium, Johannesburg, South African Institute of Mining of Highways
and Metallurgy, Balkema, Amsterdam, 1971, pp. 55-71. Figure 9.31 F. L. Peckover
227
Figure 9.32 D. Johnson; courtesy of British Columbia Department Figure 9.48 D. R. Piteau, Piteau and Associates
of Highways (top); D. R. Piteau, Piteau and Associates Figure 9.50 T. Kirkbride; courtesy of British Columbia Department
(bottom) of Highways
Figure 9.35 K. Barron and others Figure 9.51 D. R. Piteau, Piteau and Associates
Figure 9.36 F. L. Peckover Figure 9.52 B. Peirone; courtesy of Maccaferri Gabions of Canada,
Figure 9.38 D. R. Piteau, Piteau and Associates Ltd.
Figure 9.39 F. L. Peckover Figure 9.53 F. L. Peckover
Figure 9.41 D. R. Piteau, Piteau and Associates Figure 9.54 F. L. Peckover
Figure 9.42 D. R. Piteau, Piteau and Associates Figure 9.55 F. L. Peckover
Figure 9.43 Courtesy of B. Peirone, Maccaferri Gabions of Canada, Figure 9.56 George Allen Aerial Photos, Ltd.; courtesy of Canadian
Ltd. National Railway
Figure 9.44 Courtesy of American Hoist and Derrick Company Figure 9.57 D. C. Martin, Piteau and Associates
Figure 9.46 F. L. Peckover Figure 9.58 D. R. Piteau, Piteau and Associates
228
Index
229
net 221 Crimmins, R. S. 201 surface 176, 209 Feld,J. 28
waIl 221-223 cross section of slide 98, 99, drainage tunnel 178 Fellenius method of analysis
cause of slope movement 110 drained test 142 162
4, 26-28, 38, 39, 177, Crossman, N. F. 198 driving-force reduction Ferguson, H. F. 28
196 crown of slide Fig. 2.1t 17 5-182 Fernandez Concha, J. 5, 21,
Cedergren, H. R. 165, 176, Crozier, M. J. 25,28 Drnevich, V. P. 151 27
177, 182, 209 Cruden, D. M. 21, 167 Drury, P. 27 field investigation 81-110
CedrCin,G. 12 Dubois, J. 159, 169 area 84
Chagnon, J. Y. 6, 23,66,68,70 Da Silveira, A. F. 198 Dullage, C. 169 period 84
chaining survey method 114 Daido,A. 18 Duncan, J. M. 5, 143, 168, scope 81-85
Chan, H. T. 165 Dale,W.J. 58 169 stages 84, 85
Chandler, R. J. 146, 147, 148 Dallaire, E. E. 117 Duncan, N. 193 field test 105-110
Charles,J. A. 117 D'Appolonia, D. J. 150, 169 Dunlop, P. 169 life, P. K. 9
Chassie, R. G. 4, 39 D'Appolonia, E. 150 Dunnicliff, C. J. 113, 116, fill
chemical injection 189 data acquisition 129-132 117, 119, 121, 123, counterweight 183-185
Chieruzzi, R. 36 data correlation 110 130 lightweight 182, 183
Christian, J. 169 Davies, W. E. 4, 6, 27, 36, 37 Dupree, H. K. 27 finite element method of
circular slip surface 16 1-164 Davis, E. H. 158 Durr,D.L. 112 analysis 168, 169., 197
clay soil 144-148 de Freitas, M. H. 12, Fig. 2.1 Dutch cone 106 lisher, C. P. 38,56,59
Clayton, R. 218 Deal, C. S. 38, 56, 59 Dutro, H. B. 125 Fisher, R. V. 18
Cleaves, A. B. 49, 55, 81, debris 24 Duvall, W. 1. 128 Fiske, R. S. 27
92,94 avalanche 18, 22, 37, 67 Dylik,J. 19 flank of slide Fig. 2.lt
Cleveland, G. B. 28 Fig. 2.1q3 Fleming, R. W. 26
climate 35, 195, 196 fall 12, Fig. 2.1b earth 24 Fletcher, G. F. A 106
Close, U. 20 flow 18, 20, 74, Fig. 2.lql block slide Fig. 2.11 flow 17, 23, 74, 77, Fig. 2.1
Clough,G.W. 144 flow and rock fall 21, 22 fall 12, Fig. 2.1c debris 18, 20, 74,
Clough,R.W. 169 Fig. 2.lsl now 19, 20, 64, 65, 74, Fig. 2.lql
Cluff, L. S. 5,21,22 slide 17, 18,60,61,72 Fig. 2.1r2 debris and rock fall 21, 22,
coastal plain 6 1-63 Fig. 2.1k lateral spread Fig. 2.10 Fig. 2.lsl
Coates, D. F. 193, 196, 197, slump Fig. 2.1h slump Fig. 2.1i earth 19, 20, 64, 65, 74,
215, 216 topple Fig. 2.le topple Fig. 2.1f Fig. 2.1r2
Cobbold, P. R. 27 Deere, D. U. 12, 92, 93, 150, earthquake 27, 151, 152 earth and slump 15,
cohesionless soil 144 165, 166, 193, 194 Eckel, E. B. 2 Fig. 2.ls5
cohesive soil 144-148 deflectometer 114 Eckert,O. 217 bess 20, Fig. 2.1r5
Coldwell, K. L. 146 deformation analysis 168, 169 Eden,W.J. 19,28 mud 18, 19, 23, 25
Cole,K.W. 169 Denness, B. 28 effective Stress 139, 149 retrogressive 19
Coleman, T. A. 182 Dennis, T. H. 177 analysis 157, 160-165 rock fragment 21
collapsible soil 149 Denny, C. S. 58 Ehrlich, M. E. 208 sand 21, 63, 74, Fig. 2.lrl,
Collins, K. 149 Denton, P. E. 113, 205 Eigenbrod, K. D. 164, 165 Fig. 2.1r4
colluvium 149, 150 depletion zone of slide Eisenstein, Z. 169, 213 silt 21, 74, Fig. 2.1rl
Colton, R. B. 36, 37 Fig. 2.lt electronic distance soil 18
common slices method of Desai, C. S. 168, 169 measurement 114, 117 spontaneous liquefaction
analysis 162 design rock joint 198 electromagnetic spectrum 19
complex slope movement Dibiagio, E. 127 48,49 subaerial 19
21, 23, Fig. 2.1 Dickinson, R. 0. 125 electroosmosis 189 subaqueous 19
composite slope movement DiMillio, A. F. 184 Elm (Switzerland) slide 21, Foggin, G. T., 111 23
23 direct shear test 108, 143 Fig. 2.lel foliation 93
Compton, R. R. 92 discontinuity in rock mass eluvium 150 Fookcs, P. G. 28, 201
concave slope 196 193,194 embankment-area drainage foot of slide Fig. 2.1
conc test 141 Dishaw, H. E. 49 177-180 Foott, R. 140, 141
Conlon,R.J. 146 disrupted slide 14 Embleton, C. 19 Förstcr, W. 196
consolidated-drained test ditch Embody, D. R. 3, 19, 20, 21, Forsyth, R. A. 134, 188, 189
142 diversion 176 25, 26, 27 Fort Benton (Montana) slide
con solida ted-u n drained test for falling rock 2 19-221 engineering soil 24 135,136
142, 143 D/L ratio 25 Ericksen,G. E. 5,21,27 Foster, R. A. 19
construction 104 Dobr,J. Fig. 2.1 erosion 54 fracture 40
contour of slide 90, 91 Dodson, E. L. 212 Espley-Jones, R. C. 207 fracture-zone slide 25
convex slope 196 Donald, I. B. 149 Esrig, M. 1. 151, 152 Francis, P. W. 27
Conway, B. W. 28 Donn,W.L. 198 Esu, F. 146, 157 Frank (Alberta, Canada) slide
Cooling, L. F. 127 Douglas, G. R. 18 excavation 205-208 6,21, 167
Corbett, E. S. 28 Dow, D. H. 58 excavation-area drainage Franklin, J. A. 113, 205
Cording, E. J. 113-116, 119, dowel 214 180-181 lreeze, R. A. 165
124, 132, 150, 166, drain 176-181 extensometer 115, 117, 124, Fröhlich, 0. K. 23
167, 193, 194 horizontal 176, 179-181 125, 132, 133 frost action 35, 196
coring 96, 97 interceptor 176 Eyles, R. J. 26 Fuchirnoto, M. 48
Cornforth, D. H. 121 siphon 180
Corte, A. E. 19 trench 177, 178, 180 Fahnestock, R. K. 21 Gagnon, H. 66
cost of slope movement 3,4 vertical 176, 177, 179, fall 12, 73, 74, Fig. 2.1 Gano,D. 112
Coulter, H. W. 19 180 debris 12, Fig. 2.1b Gartner, J. F. 64
counterweight fill 183-185 drain hole 209,210 earth 12, Fig. Z.lc Gay, T. E., Jr., 4
Coward, M. 27 drainage 54, 93, 102, 176-182, bess 12, Fig. 2.1c gelifluction 18
cracks 7 1-73, 78, 87-89, 118 189, 209, 210 rock 12,13, 73, 76, 218 geologic map 44, 98, 198
Crandell, D. R. 21 embankment area 177-180 Falls, C. P. 69 geologic survey 91-94, 198
Crawford, R. 208 excavation area 180-181 fatalities 5, 6, 20, 21 geology 82, 83
creep in soil Fig. 2.1q4 subsurface 176-18 1, 189, fault 92, 93 geomorphic process 35
crib wall 187 209,210 Favreau, R. F. 207 geomorphic stages 36
230
geomorphology 35 Hill, R. 143 Jones, R. A. 113-116, 119, Legget, R. F. 26, 201
geophysical study 95, 98, 109, Hill, R. A. 178, 190 124, 132 Leighton, F. 128
110 Hilts, D. E. 135, 136 Jordan, R. H. 19 Leighton, F. B. 5, 23, 24, 201
Gibbs,H.J. 144 Hirao,K. 28 Jrstad, F. A. 28, 195, 217 Lemke, R. W. 19
Gibson, R. E. 158, 159 Höeg,K. 169 juxtaposed block slide 23, 24 Lewin,P.I. 145
glacial deposit 59, 62 Hoek, E. 12, 162, 166, 167, Lewis, H. 7, 208
Goguel,J. 28 199, 209, 212 Kaden,R.A. 211 Liebling, R. S. 28
Golder, H. Q. 146 Hofmann,H. 12 Kankare, E. 146, 157 Leith, C. J. 38 1 56,59
Goldstein, M. N. 23, 24, 25 Holcomb, R. T. 27 Kawasaki, T. 48 liability 6-9
Goodman, R. E. 128, 150, Hollingworth, S. E. 22 Kealy,C.D. 165 engineers and geologists
167, 169, 193, 199 Holtz, W. G. 144 Kendorski, F. S. 198 8,9
Goughnour, R. D. 4, 39 Hoover, T. 128 Kennard, M. F. 126, 127, 165 private agencies 8
Gould,J.P. 113, 116, 117, horizontal component of Kenney, T. C. 27, 28, 147, public agencies 6-8
123,130 slide 25 148, 149, 157, 165 Liang, T. 48
graben 77 horizontal drain 176, 179-181 Kent, P. E. 28 lightweight fill 182, 183
granite 59, 60 horizontal-movement Kern and Company, Ltd. 117 limestone 57
granular soil 144 measurement instrument Kerr,J.W.G. 218,224 limit equilibrium method of
gravity 98 114,115 Kerr, P. F. 28 analysis 155-167, 212
Gray, D. H. 27, 28 hot spring-volcanic slide 25 Kesseli, J. E. 13, 17 Lindberg, D. A. 125, 127
Green, G. E. 120, 121, 123, Houston, W. N. 148 Kiersch, G. A. 5 liquefaction 15, 152
144 Hoyer, B. E. 52 Kihlstrom, .B. 207 Little, A. L. 161
grid for slope-movement Hsu,K.J. 22 Kilburn,J. 211 Lo, K. Y. 23, 145, 159
measurement 88, 89, 91 human activity 4, 27, 28, 54, Kimata, K. 48 load test 108, 151
ground survey 86-89 103, 104 King,C.A.M. 19 loading of soil 151, 152
groundwater 83, 100, 101, hummocky ground 40 Kittitas County (Washington) Lobeck, A. K. 36
103, 125-128, 194 Hunter, J. H. 158, 159 slide 39 bess 12, 20, 63, 65, Fig. 2.1
measurement 103, 125-128 Hutchinson, J. N. 12, 18, 22, Kjaernsli, B. 146, 164 fall 12, Fig. 2.1c
observation well 103, 125 23, 24, 25, 26, 27, 141, Kjellman, W. 23 flow 20, Fig. 2.lr5
table 100, 103 146, 147, 148, 157, Klengel, K. J. 25, 26, 201 I4ken,T. 19
Gubin, I. Y. 20, 27, 28 161, 164, 166 Kley, R. J. 197 Londe,P. 212
gunite 210 Hvorslev, M. J. 97, 106, 108, Knill, J. L. 198 Long, A. E. 196
Gupton, C. P. 38,56,59 125 Koca,J.F. 69 Lorente de No, C. 196
Guther, H. 166 Hwang, C. T. 169 Kojan,E. 23 Lou,J.K. 12,196
Guy, H. P. 6,18,28,35 Kolb,C.R. 26,27 Loughney, R. W. 189
Gyenge,M. 197,216 igneous rock 57 Komarnitskii, N. I. 28 Louis, C. 166
inactive slope 26 Kondner, R. L. 169 Love, C. L. 6, 7, 8
Hadley, J. B. 6, 27 infinite slope method of Koppejan, A. W. 27 Lovell, C. W., Jr. 23
Halstead, P. N. 198 analysis 160-161 Koyanagi, R. Y. 27 Lucchita, 1. 36, 37
Hamel, J. V. 12, 199 infrared reflectance 67 Krahn,J. 21, 167 Lueder, D. R. 53
Handy, R. L. 189 infrared imagery 67-71, 85, 86 Kraus, J. 212 Lunib, P. 150
Hanna,T. H. 113, 121, 125, inclinometer 114, 118-125, Krishnayya, A. V. 169 Lundstrom, R. 16
127 132 Krohn,J.P. 3,4,6 Lutton, R. J. 197
Hansbo,S. 141 casing 120, 122-124 Krinitsky, E. L. 26, 27
Hajisen, J. B. 156, 159 data evaluation 132 Krivoy, H. L. 27 Madison Canyon (Montana)
Hansen, W. R. 27, Fig. 1.2 data reduction 131 Krsmanovic, D. 194 slide 5
Hansmire, W. H. 113-116, 119, maintenance 131 Kulhawy, F. H. 169 Mamulca, M. A. 28
124, 132 readings 130 Kurdin,R.D. 19 MacPhcrson, H. H. 113-116,
Hanson,W.E. 211 sensor 120-122 Kwann,D. 146 119, 124, 132
Hardin, B. 0. 151 insequent slide 25 Mahar,J.W. 113-116,119,
Harr, M. E. 23, 165 instrumentation 112-137 La Rochelle, P. 6, 23, 146 124, 132
Hast,N. 197 instrumented slide areas laboratory test 141-143 Mahtab, M. A. 198
Haugen,J.J. 184 133-13 7 L.add,C.C. 140, 141, 146 Maini, Y. N. T. 166
Hauscarán Mountain (Peru) interceptor drain 176 Ladd, G. E. 26 Malone, A. W. 128
slide 5, 21 International Society of Rock Lahee, F. H. 92 map
Hayley,D.W. 23 Mechanics 198 lake bed 51,61-63 geologic 44, 98, 198
Hays, W. V. 8 Ireland, H. 0. 145, 146, 173 Lake Roosevelt (Washington) quadrangle 44, 45
Hecht, M. L. 8, 9 slide 20, 21 soil 47
Hedley, D. G. F. 205 Jackson, E. D. 27 Lambe, T. W. 96, 97,141, 169 topographic 43, 44, 87, 89,
Heiberg, S. 19 Jacobs, D. M. 7, 208 Lambooy, P. 207 91
Heim, A. 21, Fig. 2.1 Jaeger, J. C. 150, 194 Lancaster-Jones, P. F. F. 213 Markell, A. R. 16, 19, 28
Heinz, R. A. 125, 130 Jahn, A. Fig. 2.1 landform 35, 53, 93 Marshall, H. C. 175, 180,
Hendron, A. J., Jr. 113-116, James, E. L. 121 landslide definition 1 201, 209, 220
119, 124, 132, 150, Janbu,N. 164 Landva, A. 143 Marsland, A. 145
166, 167, 193, 194, Japan Ministry of Construction Lane, K. S. 5, 27 Martin,D.C. 194
199 5 Lang, L. C. 207 Maruyasu, T. 48
Henkel, D. J. 28, 141, 142, Jennings, J. E. 28, 193, 194, Lang,T.A. 213 Matcker, 13. J., Jr. 208
143, 145, 147, 152 196, 198, 199 Langefors, U. 207 Matheson, D. S. 28
Hennequin, M. 212 John, K. W. 167, 198 Larocque, G. 207 Mathur, B. S. 64
Herbert, R. 165 Johnson, A. M. 18,23 laser instruments 117 Matich, M. A. J. 189
Herbold, K. D. 181 Johnson, N. M. 22 lateral spread 14-16, 77, maturity (landform) 36
Herget,G. 198 Johnson, S. J. 180 Fig. 2.1 McAnuff, A. L. 208
Herlinger, E. W. 180 joint of rock 92, 93, 150, 193, Laverdière, C. 25 McCaul, D. 121
Heuze, F. E. 167, 199 198 Lawson, A. C. 27 McCauley, M. L. 128, 202
highway damage 5 Jones, D. E. 3 LeeK. L. 5 McConnell, R. G. 6,21
Highway Focus 205, 206, 207 Jones, F. 0. 3, 18, 19, 20, 21, Lcfebvre, G. 168 McCormick, E. 20
Highway Research Board 209 25, 26, 27, 28 legal aspects of landslides 69 McDaniel, A. H. 7, 208
231
McDowell, B. 5 Nemöok, A. 12, 17, 28, piezometrie level 101 mass 150
McGown, A. 149 Fig. 2.1 Pichler, E. 146 metamorphic 59, 61
McLeod, B. C. 12, 196 New York Avenue (District of piles 185, 188 physiealproperties 83
McNutt, R. B. 46,47 Columbia) slide 188 Pilot,G. 158 scaling 207
McRoberts, E. C. 19 Newmark,N.M. 27 Pinole (California) slide 179 sedimentary 56, 57
Mearns, R. 128, 189, 221 Nilsen, T. H. 4 pit test 95, 97, 108, 109 shed 222-224
Meese, R. H. 112, 119, 121, Nishio, M. 48 Piteau, D. R. 12, 27, 28, 128, slump Fig. 2.lg
126, 127 noncircular slip surface 164, 193, 194, 196, 197, 198, stress-strain 197
Mehra,S.R. 201 165, 167, 168 218 topple 166, 167, Fig. 2.1d
Melzer, K. J. 106 Noorishad,J. 166 planar slip surface 160, 161, trimming 207
Mend, V. 12, 26, 27, 28, Nordin,J.G. 7 166, 167 rock fall 12, 13, 73, 76, 218
35, 92, 201, 209, Norell, W. F. 39, 58 plane-strain 140, 143 and debris flow 21, 22,
Fig. 2.1 normal water table 101 condition 143 Fig. 2.lsl
Mendes, F. 198 normally consolidated clay soil test 143 and rock slide Fig. 2.1s3
Merriam, R. 3, 113 144, 145 Pollock,J.P. 7 avalanche 21
Merrill,R. H. 128, 196, 197 Norman, J. W. 204 Pomeroy,J.S. 4 characteristics 218
metamorphic rock 59,61 Nossin, J. J. 26 Popov, I. V. 26 model 218
Mexico Ministry of Public Northwood, T. D. 208 pore pressure 125-128, 139, rock fragment flow 21
Works 120 161, 165 rock joint 92, 93, 150, 193,
microwave radiometry 68 O'Colman, E. 188 distribution 165 194, 198
Migliaccio, R. R. 19 Odenstad, S. 141 measurement 125-128 rock-mass model 198-200
Mikkelsen, P. E. 119, 120, 128 Okubo, S. 28 ratio 161 rock-noise detection equipment
Miller, W. J. 13, 16 old age (land form) 36 Portuguese Bend (California) 128, 129
Minneapolis-St. Paul (Minnesota) optical leveling survey method slide 7 rock slide 73, 74, 76, 166,
slide 133, 134, 187 114 Poulos, S. 180 167, Fig. 2.1j3
Mintzer, 0. W. 42,48 ordinary method of analysis precipitation 35, 195 and debris flow 21
Misra,A.K. 196 162 pressuremeter 107 and rock fall Fig. 2.1s3
Mitchell, J. K. 148 O'Rourk, T. D. 113-116, 119, Prevost,J-H. 169 rock slope
Mitchell, R. J. 16, 19, 28, 149 124, 132 Price, V. E. 161, 164, 173 analysis 197-201
model of slide 98, 99 orthophotomap 43 Prior, D. B. 18, 23 design factors 193-197
Mohr-Coulomb criterion 139 Ostaficzuk, S. Fig. 2.1 profile of slide 90,91,98 engineering 192-225
Mollard,J. D. 204 outcrop 92 progressive failure 23, 196 failure mode 199-201
Monahan, B. P. 201 overconsolidated clay soil property damage 5, 20, 21 geometry 196
Montaro Valley (Peru) slide 5 145-147 Prysock, R. H. 180 monitoring 205
moraine 59,61 Ozarki, Y. 48 protection methods 201,
Moreau,M. 158 quadrangle map 44, 45 218-222
Morgenstern, N. R. 19,143, Pachoud, A: 28 quick test 143 remedial measures 201-225
147, 158, 161, 162, Pain, C. F. 27 stabilization methods 201,
164, 165, 166, 169, Palladino, D. J. 134, 185 Radbruch-Hall, D. H. 12, 13, 205-2 17
173,213 Parkes, D. R. 12, 196 17, 24, 36, 37 Rockwood (Tennessee) slide
Moore,J. F. A. 117, 118,121, partially saturated soil 149 Ragel, R. H. 22 185, 186
124 Paiek,J. 12, 17, 25, 26, Rahan,P.H. 18 Rodrigues, A. F. 198
Moore,J.G. 27 Fig. 2.1 Rapp, A. 12, 18,26,28 RodrIquez, M. A. 12
Morton, D. M. 7, 27 Patton, F. D. 12, 27, 92, 93, Ray, R. C. 56 Romani, F. 23
mud 150, 165, 166, 193, 194 Rawlings, G. E. 215 Roobol, M. J. 27
flow 18, 19, 23, 25 Patton, J. H., Jr. 8 Redlinger, J. F. 212 Root, A. W. 175, 177-1 80,
slide 23 Paul,M.J. 28 Reiche, P. 25, 196 187, 201
Muktabhant, C. 146 Pautre, A. 21 reinforced earth wall 185, 186 Ross-Brown, D. M. 204
Muller, L 17, 27, 167, 194, Peacock, W. H. 151 reinforcement systems 212-216 rotational slide 13, 73, 74,
196,197 Peck, R. B. 23, 28, 72, 95, 106, relict structure 93 Fig. 2.1
multiple slope movement 23 112, 134, 157, 162, 173, remote sensing 48-69, 85, 86 Royster, D. L. 181, 185, 186
multisensor method of analysis 185,211 removal of unstable material Rushton, K. R. 165
48 Peckover, F. L. 195, 197, 215, 173, 174 Rybâ, J. 12, 14, 16, 17,
multispectral imagery 68 217, 218, 221, 223, 224 residual angle of friction 147 Fig. 2.1
multistoried slide 24, 25 penetration test 102 residual soil 56-59, 93, 149,
MuToz,A.,Jr. 112 Penman, A. D. M. 117, 125, 150 Sabarly, F. 21
Murray, D. W. 169 126,127 resisting-force increase 183-190 safety factor 156, 172, 192
Murrell, S. A. 196 perched water table 101 resistivity 95, 98, 109 Sage, R. 215, 216
Myers, L. D. 66 perfobolt 214 retaining wall 187, 21 1-213, Saint-Jean-Vianney (Quebec,
Mylrea, F. H. 128 permeability 102 215 Canada) slide 6,68
Peters, R. B. 7, 208 retrogressive failure 23 St.John,C.M. 117
Nakano,T. 48 Peterson, W. L. 3, 19, 20, 21, retrogressive flow 19 St.John,B.J. 28
Nakase, A. 158, 159 25, 26, 27 Rib, H. T. 66, 67,69 sampling 94-97, 105
Natarajan, T. K. 201 Petley, D. J. 28, 145, 147 Rice, R. M. 23, 28 Sams, C. E. 112
National Academy of Sciences Pflafker, G. 5, 21, 27 Richmond, C. M. Fig. 1.2 Samuels, R. 201
202 Philbrick, S. S. 81, 92, 94 Rippere, K. H. 198 Samuels, S. G. 145
National Aeronautics and Phillips, F. C. 198 Rissler, P. 166 sand flow 21, 63, 74, Fig. 2.lrl,
Space Administration Phillips, S. H. E. 121 Ritchie, A. M. 219, 220 Fig. 2.1r4
66 photogrammetry 49, 85, 86, Roads and Streets 178 sandstone 57, 58
National Research Council, 114,117 Robertson, A. M. 28, 198, Sanglerat, G. 96, 97
Canada and United physical properties of soil and 199 Sangrey, D. A. 28, 145, 148,
States 19 rock 83 Rocha, M. 97 149, 151, 152
Nesbitt, M. D. 213 physiographic regions 36 rock Sanjines, A. G. 5
Nelson County (Virginia) slide Piercy, N. H. 206 alteration 195 satellite imagery 66, 67, 69
37 piezometer 101-103, 115, bolt 213, 214, 217-223 Savarensky, F. P. 25
Nelson, D. S. 182, 183 125-128, 133 igneous 57 scaling rock 207
232
scarp Fig. 2.lt debris 17, 18, 60, 61, 72, So, C. L. 28 successive slope movement 24
Scharon, H. L. 208 Fig. 2.1k soil Suklje, L. 28
Schmertmann, J. H. 97, 106 depletion zone Fig. 2.lt clay 144-148 superposed block slide 23, 24
Schmidt, B. 113, 119 disrupted 14 cohesioriless 144 surface drainage 176, 209
Schneider, B. 21 flank Fig. 2.lt cohesive 144-148 surface survey 113-125
Schofield, A. 152 foot Fig. 2.1t collapsible 149 surface water 100
Schuster, R. L. 158, 159 fracture-zone 25 creep Fig. 2.1q4 survey
Schultze, E. 106 horizontal component 25 engineering 24 geologic 91-94, 198
Schweizer, R. J. 187 hot spring-volcanic 25 flow 18 ground 86-89
Scott, K. M. 18 insequent 25 granular 144 soil 46,47
Scott, R. C., Jr. 18 investigation by aerial loading 151, 152 surface 113-125
Scully,J. 48,57 photographs 66 map 47 topography 85-9 1
Seattle (Washington) slide model 98, 99 normally consolidated survey method
134, 135, 185, 188 mud 23 clay 144, 145 baseline offset 114
sedimentary rock 56, 57 multistoried 24, 25 overconsolidated clay chaining 114
Seed, H. B. 16, 23, 27, 143, profile 90, 91, 98 145-147 optical leveling 114
151, 152, 169 rock 73, 74, 76, 166, 167 partially saturated 149 Sutter, J. H. 8,9
Seegmiller, B. L. 216 Fig. 2.1j3 physical properties 83 Swanston, D. N. 28
seeps 39, 83, 88, 102, 165 rotational 13, 73, 74, residual 56-59, 93, 149, Sweeney, M. 201
seismic measurement 98, 109 Fig. 2.1 150
Semprich, S. 166 slab 25 sensitive 19, 148, 149 Tabor, R. W. 17
sensitive soil 19, 148, 149 subsidence zone 87 strength properties Taechathummarak, A. 146
Serafim,J. L. 194 tertiary 25 144-150 Takada, Y. 25
serpentinite areas 59 toe Fig. 2.lt stress state 140 Takeda, Y. 48
Sevaldson, R. A. 146, 162 translational 13, 16, 17, stress-strain 140, 151, 152 talus 58, 175
shale 57,58 73, 74, Fig. 2.1 survey 46, 47 Tanguay, M. G. 66, 68, 70
Shannon and Wilson, Inc. 187, vertical component 25 tone 54 Tanner, R. G. 146
188 warning system 129, 201, soil-slope Taranik, J. V. 52
Shannon,W. L. 112, 119, 121, 223-225 design 172-190 Taueher,G.J. 27
126, 127 slide-susceptible landforms stability analysis 157-165 Tavenas, F. 6, 23
Sharp,J.C. 166 36-3 8, 54-56 stress analysis 157-165 Taylor, D. W. 157, 161
Sharp, R. P. 22 slide-vulnerable locations Sokolov, N. 1. 25 Taylor, F. A. 3
Sharpe, C. F. S. 12, 19, 23, 25, 37, 55, 56 solifluction 18, Fig. 2.1q2 Taylor, J. H. 22
26, Fig. 2.1,37 slip mechanism 156 Solonenko, V. P. 27 Taylor, R. L. 165
shear strength 26, 27, 139-143, slip surface Sowers, G. B. 26, 95, 102, Tchalenko, J. S. 143, 147
150 circular 161-164 105, 106, 108 Temple, P. H. 18, 28
measurement 141-143 noncircular 164, 165, 167, Sowers, G. F. 7, 12, 26, 28, tendon anchor 215
of rock joint 150 168 85, 95, 102, 105, 106, terrace 62
reduction 26, 27 planar 160, 161, 166, 167 108 terrain
undrained 140 slope Spencer, E. 163 evaluation 34, 35
shear stress increase 26, 27 active 26 Spencer, G. S. 26 investigation 42,43
shearing resistance 156, 166 inactive 26 Spencer, J. M., Jr. 69 Ter-Stepanian, G. 17, 23, 24,
Shelton, J. S. 23, 25 concave 196 split-tube sampler 105, 106 25, 28
Shimer,J.A. 198 convex 196 spontaneous liquefaction now tertiary slide 25
shoterete 210, 211 flattening 181 19 Terzaghi, K. 23, 26, 27, 72,
Shreve, R. L. 22, 25, 28 stability analysis 155-169 Springs 83, 88, 102 95, 103, 106, 116, 157,
Shroder, J. F. 24, 25, 28 weight reduction 181-183 Spirek, M. 28 162, 194, 195
Shuk,T. E. 197 slope movement Squier, R. 119 Terzaghi, R. 198
silt flow 21, 74, Fig. 2.lrl complex 21, 23, Fig. 2.1 Stacey, T. R. 193 test
Silver, M. L. 151 composite 23 Stafford, G. 180 cone 141
Simonett, D. S. 27 cost 3,4 Stanculescu, 1. 1. 190 consolidated-drain 142
Simons, N. 146 cause 4, 26-28, 38, 39, Stagg, K. G. 169 consolidated-undrained
simple shear test 143 177,196 Stallard, A. H. 47, 48, 66 142,143
simplified Bishop method of evidence 71, 72, 74, 75 Stateham, R. M. 128 direct shear 108, 143
analysis 161, 163, 164 identification 72-78 Steffen, 0. K. H. 198 drained 142
Singh, R. 145 material 24 Stephens, N. 18, 23 field 105-110
siphon drain 180 multiple 23 Stimpson, B. S. 194 laboratory 141-143
Skempton, A. W. 12, 23, 24,. process 11-28 Strain meter 114, 124, 125 load 108,151
25, 26, 28, 141, 142, rate 24, Fig. 2.lu Streitz, R. 7 penetration 102
145, 146, 147, 148, recognition 34-78 strength pit 95, 97, 108, 109
157, 161, 166 successive 24 measurement 139-152 plane strain 143
Skipp, B. A. 36,37 type 12-24, 72-76, Fig. 2.1 properties 139-152 quick 143
slab slide 25 Slosson, J. E. 3,4,5,6 stress simple shear 143
slices method of analysis slow test 142 effective 139, 149, 157, slow 142
159, 160 slump 14, 15, 61, 74, 76-78 160-165 triaxial 141, 142
slickenside 15, 147 and earth flow 15, total 139, 140, 157-160 uneonfined compression
slide 12, 73, 74, Fig. 2.1 Fig. 2.1s5 stress-strain 141
accumulation Fig. 2.1t and topple Fig. 2.1s2 in rock 197 undrained 143
asequent 25 debris Fig. 2.1 h in soil 140, 151, 152 vane shear 107, 141
avoidance 173-175 earth Fig. 2.1i Struble, R. A. 42 thermal infrared imagery 68,
block 14, 16, 23, 24, 40, rock Fig. 2.1g structural domain 198 70, 71, 85, 86
73, 74, Fig. 2.1j2, Smalley, 1. J. 27, 28 subaqueous flow 19 thermal treatment 189, 190
Fig. 2.11 Smith, R. 3,6 subaerial flow 19 Thomas,T. L. 117
contour 90, 91 Smith, T. W. 134, 177, 180, subsurface drainage 176-181, Thompson, G. A. 60
cross section 98, 99, 110 182, 188, 189 189, 209, 210 Thomson, S. 23, 28
crown Fig. 2.1t Snopko, L. 25 subsurface exploration 91-99 .Thornburn,T. H. 211
233
Thornbury, W. D. 35, 36 Uddin, S. 157 Vésic, A. S. 144 Whitman, R. V. 161
Tice, J. A. 112 unconfined compression test Voight,B. 28 Wiggins, J. H. 4
tied-back wall 188-189, 211, 141 volcanic flow 57 Williams, G. P. 6, 18, 28, 35
212 undrained test 143 Williams, W. W. 189
Tilling, R. 1. 27 undrained shear strength 140 Wade, N. H. 143 Wilson, D. D. 28
tiltmeter 118, 119 U.S. Army Corps of Engineers Walbancke, H. J. 113 Wilson, E. L. 168
toe of slide Fig. 2.1t 212 Walker, G. P. L. 27 Wilson, S.D. 16, 112, 119,
Toms,A.H. 121 U.S. Bureau of Public Roads wall 120, 121, 125, 126,
topographic map 43, 44, 87, 207 cable 217 127, 133, 134, 135,
89, 91 U.S. Bureau of Reclamation catch 221-223 141,188
topography 53, 82 102 crib 187 wind deposits 63, 64
survey 85-91 method of analysis 162 reinforced earth 185, 186 wire mesh 219-221
topple 12, 73, 74, 166, 167, U.S. Committee on Large retaining 187 Wisecarver, D. W. 128, 196,
Fig. 2.1 Dams 149 tied-back 188-189, 211, 197
and slump Fig. 2.1s2 U.S. Geological Survey 6 212 Witherspoon, P. A. 165, 166
debris Fig. 2.1e "unloading" slide 181 Ward, W. H. 27, 145 Witkin, I. J. 6
earth Fig. 2.1f Warren, C. R. 58 Wittke,W. 166
rock 166, 167, Fig. 2.1d Vaiont reservoir (Italy) slide Washburn, A. L. 19 Wong, K. K. 149
Torrance, J. K. 28 5,167 water 83, 99-103 Wood, A. M. 27
total stress 139, 140 valley bulging 22, 23, water content of slope Woodward, R. J. 169
analysis 157-160 Fig. 2.1s4 movement 24 Wroth, P. 152
Toth,J. 165 Van Wamelon, B. M. 19, 27 water-laid deposits 61 Wright, S.G. 187
traffic-induced vibration 197 vane shear test 107, 141 water-level probe 103
translational slide 13, 16, 17, Vargas, M. 146 Yemel'ianova, Ye. P. 24
water pressure 100, 101, 166,
73, 74, Fig. 2.1 Varnes,D.J. 11,12,21,36, Yoder,E.J. 188
199
Transportation Research 37 water table Yorke, C. A. 38, 56, 59
Board 112,119 Varnes, H. D. 13, 15 Youd,T.L. 16,28
normal 101
traverse line 88, 89, 91 Vaughan, P. R. 113, 125, 126, youth (landforni) 36
perched 101
trench drain 177, 178, 180 127, 128 Yu,Y. 196
Watters, R. J. 12, Fig. 2.1
triangulation 86, 87, 114 vector of slide movement 91 wave velocity 109 Yudhbir 147
triaxial test 141, 142 Veder, C. 28, 189 Way, D. S. 53 Zruba, Q. 12, 26, 27, 28,
Trigo Ramirez, J. 188 vegetation 54, 83, 84, 104, weather 103, 199 35, 92, 201, 209,
trimming rock 207 105 weathering 28, 195 Fig. 2.1
Trollope, D. H. 28 vertical component of slide Weaver, C. E. 28 Zelasko,J.S. 169
tunnel 222-224 25 Webb, D. L. 145 Zienkiewicz, 0. C. 169
Turner, B. L. 4 vertical drain 176, 177, 179, Weinberg,L. J. H. 19, 27 Zischinsky, U. 12, 17, Fig. 2.1
Tyc,P. 212 180 White, D. E. 60 Zolotarev, G. S. 26
234
TRANSPORTATION RESEARCH BOARD 1978
Officers
KARL S. BOWERS, Federal Highway Administrator, U.S. Department of Transportation (ex officio)
HARVEY BROOKS, Chairman, Commission on Sociotechnical Systems, National Research Council (ex officio)
HOWARD L. GAUTHIER, Department of Geography, Ohio State University (ex officio, MTRB)
ROBERT N. HUNTER, Chief Engineer, Missouri State Highway Department (ex officio, Past Chairman, 1977)
HAROLD L. MICHAEL, Head, Civil Engineering Department, School of Civil Engineering, Purdue University (ex officio, Past Chairman, 1976)
RICHARD S. PAGE, Urban Mass Transportation Administrator, U.S. Department of Transportation (ex officio)
HENRIK E. STAFSETH, Executive Director, American Association of State Highway and Transportation Officials (ex officio)
JOHN McGRATH SULLIVAN, Federal Railroad Administrator, U.S. Department of Transportation (ex officio)
The Transportation Research Board is an agency of the National Re- The National Academy of Sciences was established by a congres-
search Council, which serves the National Academy of Sciences and sional act of incorporation signed by President Abraham Lincoln on
the National Academy of Engineering. The Board's purpose is to March 3, 1863, to further science and its use for the general welfare
stimulate research concerning the nature and performance of trans- by bringing together the most qualified individuals to deal with sci-
portation systems, to disseminate information that the research pro- entific and technological problems of broad significance. It is a pri-
duces, and to encourage the application of appropriate research find- vate, honarary organization of more than 1000 scientists elected on
ings. The Board's program is carried out by more than 260 commit- the basis of outstanding contributions to knowledge and is supported
tees and task forces composed of more than 3100 administrators, by private and public funds. Under the terms of its congressional
engineers, social scientists, and educators who serve without compen- charter, the Academy is called upon to act as an official—yet inde-
sation. The program is supported by state transportation and high- pendent—advisor to the federal government in any matter of science
way departments, the major administrations of the U.S. Department and technology, although it is not a government agency and its activ-
of Transportation, the Association of American Railroads, and other ities are not limited to those on behalf of the government.
organizations interested in the development of transportation. To share in the task of furthering science and engineering and of
The Transportation Research Board operates within the Commis- advising the federal government, the National Academy of Engineer-
sion on Sociotechnical Systems of the National Research Council. ing was established on December 5, 1964, under the authority of the
The Council was organized in 1916 at the request of President Wood- act of incorporation of the National Academy of Sciences. Its ad-
row Wilson as an agency of the National Academy of Sciences to en- visory activities are closely coordinated with those of the National
able the broad community of scientists and engineers to associate Academy of Sciences, but it is independent and autonomous in its
their efforts with those of the Academy membership. Members of organization and election of members.
the Council are appointed by the president of the Academy and are
drawn from academic, industrial, and governmental organizations
throughout the United States.
TYPE OF MATERIAL
(DOMINANT) (BEFORE MOVEMENT)'
TYPE OF MOVEMENT
BEDROCK
I. FALLS
Mass in motion travels most of the distance through the air. Includes free fall,
(2.1a)
a b c
movement by leaps and bounds, and rolling of fragments of bedrock or soil.
II . TOPPLES
Movement due to forc es that cause an overturning moment abou t a pivot point
t2.1d1) ROCK TOPPLE (de Freitas and Watters, 2.371.
d (2.1o) DEBRIS TOPPLE
e f
below the center of gravity of the unit. If unchecked, will result in a fall or
slide.
EARTH TOPPLE
B. TRANSLATIONAL
Movement predominantly along more or less planar or
(2.1j1) (Nomeok. 2115. af1or Bock. 251 (2.1j2) ROCK BLOCK SLIDE
J
l2.1k) DEBRIS SLIDE . very slow to rapid
k
gently undulatory surfaces.
Movement frequently is structurall y controlled by surfaces
of W'eakness, such as faults, joints, bedding planes, and (2.1j3) ROCK SLIDE
~~,.-i~HiQn~ Jn ~J'l~~r ~!n~ng!h b11t:m~n l~y~r~ 9f !;;ledded
deposi ts, or by the contact between firm bedrock and
overlying detritus.
500 m
11640 tt1
~~::;:::·::.~~~~~~~~~~~~~~~
flow o r subjacent material. 328ft
silt <11nd and liih'8fS
Soft clavwilh
Firm clayey
500m
1640 ft
V. FLOWS
A. IN BEDROCK
A'2.1p1) (2.1p3) (Zil<hinsky, 2.194) (2.1p4) {Rybai and Dobr, 2 .1371 p B (2.1r11 WETSANDORSILTFLOW ,
rapid to very rapid
MUD FLOW r
Includes spatially continuous deformation and surficial as well as deep
creep. Involves extremely slow and generally nonaccelerating dif-
ferential movements among rela tive ly in tac t units. Movements may
~~~~~~~{~:
1. Be along many shear surfaces that are apparently not connected;
2 . Result in folding, bending, or bulging; or
--.. '
~
3. Roughly simulate those of viscous fluids in distribution of
velocities.
B. IN SOIL
Movement within displaced mass such that the form taken by moving
material or the apparent distribution o f velocities and displacements
'-~
resemble those of viscous fluids. Slip surfaces within moving material
are usually not visible or are short-lived. Boundary between moving
~
mass and material in place may be a sharp surface of differential move- Riviere Blanche,
t
ment or a zone of distributed shear. Movement ranges from extremely Quebec
rapid to extremely slow. {2.1q3) DEBRIS AVALANCHE, very rapid to -1 2.1q4) SOIL CREEP , extremely slow
extreme ly rapid l2.1r31 EARTH FLOW, "ery slow to rapid
Gravity downslope movement of (Z8ruba and Mencl , 2.1931
clayey rocks and coal on the margin of
a sedimentary basin
tf. . ·. ·
w~thered bedroc... .
·~'"'·
..,,.,. T
Note: Sackung or
gravitational sagging (2.1r5) LOESS FLOW (dry, caused by
earthquakes), extremely rapid
is illustrated by (2.1q51 BLOCK STREAM Also moist or wet
p1, p2, and p3.
DRY
VI. COMPLEX
Movement is by a combination of one or more of the five principal types of
EXAMPLES
(2.1s1) ROCK FALL-DEBRIS FLOW IROCK·FALL AVALANCHE), (2.1s41 CAMBERING AND VALLEY BULGING 12.1s5) SLUMP-EARTH FLOW s
movement described above. Many landslides are complex, although one Extremely rapid (After Heim , 2.58, Elm , Switzerland, 1881} Normal ~ub-horll'Ot1tll