RESERVOIR

SEDIMENTATION
Technical Guideline for
Bureau of Reclamation

U. S. Department of the Interior

Bureau of Reclamation

As the Nation's principal conservation agency, the Department of the

Interior has responsibility for most of our nationally owned public
lands and natural resources. This includes fostering the wisest use of
our land and water resources, protecting our fish and wildlife, preserving the environmental and cultural values of our national parks and
historical places, and providing for the enjoyment of life through outdoor recreation. The Department assesses our energy and mineral
resources and works to assure that their development is in the best
interests of all our people. The Department also has a major responsibility for American Indian reservation communities and for people
who live in Island Territories under U.S. Administration.

The information contained in this report regarding commercial products

or firms may not be used for advertising or promotional purposes and is
not to be construed as an endorsement of any product or firm by the
Bureau of Reclamation.
The information contained in this report was developed for the Bureau
of Reclamation; no warranty as to the accuracy, usefulness, or completeness is expressed or implied.

RESERVOIR
SED IMENTATION
by
Robert I. Strand
Ernest L. Penibertori

TECHNICAL GUIDELINE FOR

BUREAU OF RECLAMATION

SEDIMENTATION AND RIVER HYDRAULICS SECTION

HYDROLOGY BRANCH
DIVISION OF PLANNING TECHNICAL SERVICES
ENGINEERING AND RESEARCH CENTER
DENVER, COLORADO

OCTOBER 1982

CONTENTS
P age
Reservoir sedimentation

..............................

Methods of determining sediment inflow ..............

a. Sediment yield rate factors .................
b. Reservoir resurvey data ...................
c. Sediment sampling data ....................
d. Unmeasured sediment load ...................
e. Adjustment to damsite ....................

2
2
4
7
10
15

Reservoir sediment deposition ...................

a. Trap efficiency .......................
b. Density of deposited sediment ................
c. Sediment distribution within a reservoir ...........
d. Delta deposits ........................

15
16
18
21
30

Downstream channel effects .....................

a. General degradation ......................
b. Armoring method .......................
c. Stable slope method .....................

35
35
36
41

Conclusions

............................

45

References

............................

46

General

TABLES
Tab 1 e
1
2
3
4
5
6
7
8
9

Rating chart of factors affecting sediment yield .......

Sediment load computations of Rio Toa Vaca
near Villalba, Puerto Rico .................
Bedload correction ......................
Modified Einstein procedure computation ...........
Design type curve selection .................
Reservoir area and capacity data. Theodore Roosevelt Lake ..................
Determination of elevation of sediment at
Theodore Roosevelt Dam ...................
Elevation of sediment at Theodore Roosevelt Dam .......
Theodore Roosevelt Lake - Type II, Reservoir
sediment deposition study .................

3
12
13
13
23
25
28
30
31

FIGURES
Figure
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

Page
Area and capacity curve for Lake Mead ..........
Sediment distribution from reservoir surveys ......
Lake Mead sediment deposition profile ..........
Average annual sediment yield rate versus drainage
area size .......................
Suspended sediment sampler, DH-48 ............
Suspended sediment rating curve for Rio Toa Vaca near
Villalba, Puerto Rico .................
Flow duration curve for Rio ba Vaca near Villalba,
Puerto Rico ......................
Schematic diagram, reservoir allocations,
sediment deposition ..................
Trap efficiency curves .................
Comparison of densities on Lake Mead at location 5 ...
Sediment deposited in flood control pool ........
Sediment distribution design curves ...........
Sediment distribution for Theodore Roosevelt Lake ....
Area and capacity curves for Theodore Roosevelt Lake
Depth versus capacity for Theodore Roosevelt Lake ....
Curves to determine the depth of sediment at the dam
Typical sediment deposition profile ...........
Topset slope versus original stream slope from
existing reservoirs ..................
Armoring definition sketch ...............
Tractive force versus transportable sediment size ....
Degraded channel by the three-slope method .......

11

..
..
..

4
6
6

..
..

7
8

..

..

11

..
..
..
..
..
..

..

16
17
19
21
23
26
26
29
29
33

..
..
..
..

34
37
40
42

..

RESERVOIR SEDIMENTATION
General. - All reservoirs formed by dams on natural water courses are subject
to some degree of sediment inflow and deposition. The problem confronting
the project planner is to estimate the rate of deposition and the period of
time before the sediment will interfere with the useful function of the
reservoir. Provisions should be made for sufficient sediment storage in the
reservoir at the time of design so as not to impair the reservoir functions
during the useful life of the project or during the period of economic
analysis. The replacement cost of storage lost to sediment accumulation in
American reservoirs amounts to millions of dollars annually (Chow, 1964).
There are a series of basic steps to follow in studying the sedimentation
processes in reservoirs. First, sediment transported by the upstream river
system into a reservoir is deposited and/or transported at a reduced rate
further into the reservoir, the distance being dependent on the decreased
water velocities. As sediment accumulates in the reservoir, storage capacity
is reduced. The continued deposition develops distribution patterns within
the reservoir which are greatly influenced by both operations of the reservoir and timing of large flood inflows. Deposition of the coarser sediments
occurs in the upper or delta reaches while finer sediments may reach the dam
and influence the design of the outlet works. A major secondary effect is
the downsteam degradation of the river channel caused by the releases of
clearer water.
These guidelines cover the essential sedimentation characteristics to be
considered in the design of a dam and reservoir. The sediment related
features requiring study are the sediment inflow, deposition, and degradation
processes. Sedimentation processes in a reservoir are quite complex because
of the wide variation in the many influencing factors. The most important
being, (1) hydrological fluctuations in water and sediment inflow, (2) sediment particle size variation, (3) reservoir operation fluctuations, and
(4) physical controls or size and shape of the reservoir. Other factors that
for some reservoirs may be quite important are: vegetative growth in upper
reaches, turbulence and/or density currents, erosion of deposited sediments
and/or shoreline deposits, and operation for sluicing of sediment through the
dam. The procedures described represent a combination of state-of-the-art
together with methods that are practical, technically sound, and sufficiently
varied to fit the complexity of the problem. It is because of this complexity that empirical relationships developed from surveys of existing reservoirs are being used to define sediment depositional patterns. Many mathematical models are being developed to simulate the physical processes of
sediment transport and deposition in reservoirs. The models, to date, are
not easily adapted to solve problems of reservoir sedimentation without some
simplifications in defining the four most important factors previously
described. With more research and additional reservoir survey data for
verification of the mathematical models, they may become a useful method for
predicting sediment deposition. Changes in these guidelines can also be
expected in many of the empirical relationships with the continuing surveys
of existing reservoirs. Further support to update these guidelines will
occur as loss of storage capacity become more severe along with the economic
and social changes affecting future reservoir uses.

In recent years, critical sediment problems have occurred in some of the

reservoirs in all climatic regions of the world where complete loss of
dependable storage resulted because of sediment deposition. In these situations, sediment control methods are being planned and, in many cases,
construction completed on upstream sediment traps, bypass channels, special
outlets for sluicing sediment, and mechanical dredging techniques. In many
stuations, sediment yields are high and conservation or erosion control
measures in the drainage area are important for a reduction in the long-term
sediment production. In the United States, these measures are usually
carried out under the direction of the Soil Conservation Service, USDA, in
cooperation with landowners and are encouraged by the Bureau of Reclamation.
Methods of determining sediment inflow. - Sediment is the end product of
erosion or wearing away of the land surface by the action of water, wind,
ice, and gravity. Water resource development projects are most affected by
sediment which is transported by water. The total amount of onsite sheet and
gully erosion in a watershed is known as the gross erosion. However, all the
eroded material does not enter the stream system; some of the material is
deposited at natural or manmade barriers within the watershed and some may be
deposited within the channels and their flood plains. The portion of eroded
material which does travel through the drainage network to a downstream
measuring or control point is referred to as the sediment yield. The sediment yield per unit of drainage area is the sediment yield rate.
Most methods for predicting sediment yields are either directly or indirectly
based on the results of measurements. Direct measurements of sediment yields
are considered the most reliable method for determination of sediment yields.
This is accomplished by either surveying of reservoirs or sampling the sediment load of a river, and both methods are described in subsequent sections
of these guidelines. Other methods for predicting sediment yields depend on
measurements to derive empirical relationships or utilize empirically checked
procedures such as the sediment yield rate weighting factors or the Universal
Soil-loss equation (Wischmeier and Smith, 1965).
a. Sediment yield rate factors. - The factors which determine the sediment yield of a watershed can be summarized as follows:
1.
2.
3.
4.
5.
6.
7.
8.
9.

Rainfall amount and intensity

Soil type and geologic formation
Ground cover
Land use
Topography
Upland erosion (nature of drainage network-density, slope, shape,
size, and alinement of channels)
Runoff
Sediment characteristics - grain size, mineralogy, etc.
Channel hydraulic characteristics

Some researchers have deemed it necessary to include some additional

factors; however, even the nine above are interrelated. As an example, a
heavy vegetative cover is dependent upon at least a moderate amount of
rainfall; however, the ground cover conditions could be upset by tillage
practices, overgrazing, or fire. Sediment transported from the drainage
2

basin to a reservoir is controlled by the sediment transport characteristics of the river which is influenced by the first six factors but
reflects a more direct combination of items 7, 8, and 9.
Systems of weighting the individual sediment influencing factors have been
devised (Pacific Southwest Interagency Committee, 1968) to arrive at a
sediment yield rate for an individual drainage basin. This type of
analysis is best applied to preliminary planning studies and has its
greatest reliability when the yield rates can be correlated with a measured sediment yield from an adjacent basin or subbasin.
An example of the techniques for weighting of the nine factors which is
not identical but similar to those used in the report (Pacific Southwest
Interagency Committee, 1968) is given in table 1. The weighted values
would apply to the Pacific Southwest area, but because they are relative
to each other, could be changed for other parts of the United. States.
Table 1. - Rating chart of factors affecting sediment yield

Factors
1.
2.
3.
4.
5.
6.
7.
8.
9.

Rainfall amount and intensity

Soil type and geological
information
Ground cover
Land use
Topography
Upland erosion
Runoff
Sediment characteristics)
Channel hydraulics

Sediment yield level

High
M oderate
Low
10

20
10
10
20
25
10

10
0
0
10
10
5

0
-10
-10
0
0
0

25

10

In computing the sediment yield of a drainage area above a dam or reservoir, a field inspection by a trained sedimentation specialist is needed
to evaluate the factors in table 1 for weighting the significance of the
nine factors affecting sediment yield. Upon completion of an inspection
by the specialist, recommended procedures will be given on (1) available
data and methods for analyzing data, (2) techniques available for predicting sediment yields in gaged as well as ungaged drainage basins, and
(3) additional measurements required to compute sediment yields.
A well-known method for determination of sediment yields from small
drainage areas is the empirical relationship developed by Wischmeieir and
Smith (1965), most commonly referred to as the Universal Soil-loss equation. It should be recognized that gross erosion determined by this
empirical method is, at best, an approximation and considered a rough
estimate. It is normally applied to areas of less than about 4 mi
(10 km2) and even then may have to be corrected by a sediment delivery

ratio when converting gross erosion to sediment reaching a main river

channel
b. Reservoir resurvey data. - Measurement of the sediment accumulation in
a reservoir is considered by many engineers as the best method for determining the sediment yield. Surveys of existing reservoirs for determining
loss of storage space and distribution of sediment deposits within the
reservoir provide data on sediment yield rates as well as for operations
purposes. It is important that when construction is completed on a dam, a
plan be established for surveying or monitoring of the sediment accumulation. Even before construction of the dam is completed, a decision is
needed on the basic method selected for future surveys and technique for
analyzing sediment accumulation (Blanton, 1982).
The main purpose of a reservoir survey is to determine the storage capacity at the time of the survey which when compared to an earlier survey
(usually the original survey) gives the sediment accumulation. The
storage volume computations are made from an area-capacity computer
program involving computation of capacities corresponding to each elevation in the area-elevation data set and fitting the capacity-elevation
relationship using either cubic spline or least square set of equations
(Bureau of Reclamation ACAP Program). The end product of the area-capacity
computations is the plot of the areas and capacities for the original and
new surveys. An example of this plot is shown in figure 1.
- 0 ACRES
00
I

ISO
I
700

400

600

- 0m2
400

500

50

0
I

300

200

100

-Maximum i.vler surface 3744m (1229ft)

1250
200
'ISO

/963 - 64

U0O

w
I-

50

Copoc,fy

00I

Area

950

>
w
-J
w

900
850

i19
II

750
700

d98650ft
20
25
CAPACITY - I0'm3-

800

AREA AND
LAKE MEAD

63 6480tf0m2/9m(720ft)

I
5

I_L___J_I

I__j_I_I

5
CAPACITY -

30
I

20
ACRE-FEET-----

II
25

35
I

40

650

I
30

35

Figure 1. - Area and capacity curve for Lake Mead.

A comparison of capacities between the two surveys as shown in figure 1
gives the measured volume of sediment accumulation. It is important in
this sediment volume computation that the method selected to compute
capacities from contour areas be the same for both of the surveys being
4

compared. That is, if the ACAP method is used for computing the resurveyed reservoir capacity, it should also be used for the original capacity
computations. This would help eliminate any differences in technique
having undue influence on the sediment volume computations. All information from the survey should be documented in the Reservoir Sedimentation
Data Summary sheet which is provided to the Subcommittee on Sedimentation,
Interagency Advisory Committee on Water Data for use in the periodically
published summary on reservoir surveys (U.S. Department of Agriculture,
1978).
Other worthwhile analyses of data from reservoir sedimentation surveys are
to make a plot of percent reservoir depth versus percent sediment deposit
or to plot a sediment deposition profile throughout the length of the
reservoir. The plot of percent depth versus percent sediment (fig. 2)
has been used in developing design curves in predicting the distribution
of sediment deposits in planning studies. The deposition profile provides
valuable information for defining the delta, foreset slopes for possible
density currents, and depth of sediment depositions at the dam. An
example of a dimensionless plot of a sediment deposition profile for
Lake Mead is shown in figure 3.
At the time of the reservoir survey, data are also needed on some of the
characteristics of the sediments both as deposited and moving through the
reservoir. Samples of deposited sediments should be spaced throughout the
reservoir area to be representative of deposits in the topset and foreset
slopes of the delta as well as at the bottomset slopes in the deeper parts
of the reservoir. Analysis of the samples collected consists of density,
particle-size distribution and mineralogic composition. These data on
deposited sediments are used for a better understanding as to the source
of incoming sediments, for use in study of density currents or study of
sluicing capabilities through outlet works, for verification of models
being developed on movement of sediment through reservoirs, and for
development of empirical relationships to be used in the planning and
design of other reservoirs. In addition to the above uses, data on
sediment characteristics when combined with survey data on depths of
sediment near the dam can be used to identify future problems of sediment
deposition associated with inflow to powerplant intakes or plugging of
outlet works. A unique sediment deposition problem to be evaluated in
reporting the results of the survey data is the effects of bank sloughing,
landslides, and valley wall erosion by wave action or unstable slopes.
Reservoir survey data (U.S. Department of Agriculture, 1978) provide an
excellent source for determining sediment yield rates for any part of the
United States. Adjustments in the sediment yield rate will usually be
necessary to account for variation in drainage area characteristics. One
of the most important variations is the size of the drainage basin. Some
investigators have found that the sediment yield varies with the 0.8 power
of the drainage area size (Chow, 1964) (equivalent to sediment yield rate
varying with -0.2 power of the drainage area). Figure 4 is a plot of
sediment yield rate versus drainage area which was developed from selected
reservoir resurvey data in the semiarid climate of southwestern United
States. In using the drainage area versus sediment yield relationship as
shown on figure 4, it is best to make a calibration with a known sediment
5

00 -

F/ep/cnt Butte
(i9/5-/969)
Lcke Meoct

80

(/935-1964)

/
/

I
/KS

uJ

()4Q__

//

w
a.

Mcrfth
(1943-1972)

I,

//
20

/ "
,
/

Horry Sf runk Loke

(/949-19621

Guernsey
(/927 /957)
60>'i
/,/

I,

,"-'

'-

40
20
60
80
PERCENT SEDIMENT DEPOSITED

00

Figure 2. - Sediment distribution from reservoir surveys.

I00

50% Pool El..'.

(/935-64)

17"'

80

'I

/964-,.

LU

/935 (Or/gino/I

I,

I,,
/

w 40

rotci Depth H: 58/ feet (/77,,,)

To tot Distance 0: /2/ mi/es (/95km)

//

LU

, -I

20

20

40
60
80
PERCENT DISTANCE (0)
Dom to Top Operation Pool

100

Figure 3. - Lake Mead sediment deposition profile.

A
tvuuv

DRAINAGE AREA IN Mi 2

tO

0
-

00

1111

liii I

1000

0000

1111

Sediment yields from survey of 28

reservoirs in ssmi-orid climate
of USA.

00

-.

E
- 1000
E

U-

Os 1.844 -0i4
Os lO98A0"4)

z
-

-J
-

I-

I"I

01
0

In

ii
00

000
A r DRAINAGE AREA IN Km2

0000

00000

Figure 4. - Average annual sediment yield rate versus drainage area size.
yield and evaluate the nine sediment contribution factors. This calibration, along with an identification of similar sediment contributing
characteristics, will permit drawing a parallel line similar to that shown
on figure 4 through any measured data point.
c. Sediment sampling data. - Sampling is the surest method of obtaining
an accurate determination of the suspended sediment load being carried by
a stream at a particular location. Suspended sediment sampling in combination with total load computations is the preferred method used for planning
studies in determining the sediment inflow to a proposed reservoir. The
objective of a sediment sampling program on a river is to collect sufficient
samples of sediment carried both as suspended load and as bedload to define
the total sediment being transported. For suspended sediment sampling it
is essential to measure the water discharge, Qw in ft3/ (m3/s) which
is combined with suspended sediment concentration, C, in mg/L to give the
suspended sediment load Qs in tons/day by the equation:
Qs = 0.0027 C Qw (inch-pound units)
or Qs = 0.0864 C Qw (metric units)

(1)

Suspended sediment sampling equipment and techniques for collecting can

vary considerably depending on program objectives and field conditions.
Suspended sediment sampling devices are designed to collect a representative sample of the water-sediment mixture. A thorough discussion of
sediment samplers and techniques for sampling is given in either the
series of reports prepared by U.S. Interagency Sedimentation Project (1940
to 1981) or in the U.S. Government Handbook (1978). An example of the
U.S. Interagency Sedimentation Project designed sampler is shown in
figure 5.

Figure 5. - Suspended sediment sampler DH-48.

In the collection of suspended sediment samples, it is important that
samples represent an integration with width across the channel as well as
with depth from the water surface to the streambed. Although other
methods for sampling are described in the U.S. Government Handbook (1978),
the EWI (equal-width-increment) method provides the most representative
sample of the total suspended sediment load. It is accomplished by
sampling at equally spaced widths or increments across the cross section
and maintaining a constant travel rate in each of the verticals sampled.
In this method, a composite sample is made of all verticals sampled for
only one laboratory analysis of sediment concentration in mg/L and
particle-size distribution.
The sediment sampling program will vary from one river to another, depending on temporal variations in the sediment load and particle-size distribution of the suspended and bed material sediments. The frequency of
sampling suspended sediments will usually vary from daily samples to once
or twice a month but should always include samples during the flood
events. In many situations, the collection and analysis of suspended
sediment samples is an expensive process, and daily sampling yields a good
deal of duplication through a base flow period. For these reasons, the
once or twice a month or miscellaneous sampling which includes sample of
flood flows is more common and economical.
The objective of any suspended sediment sampling program is to develop a
correlation between water discharge and sediment load commonly called a
suspended sediment rating curve. This rating curve is normally a plot on
logarithmic paper of water discharge Qw in ft3/s (m3/s), versus sediment
load, Qs in tons/day from equation 1. These curves can best be computed
by least squares analysis with water discharge as the independent variable
usually defined by one to three such relationships. When two or three
equations are computed from the plotted points, the extrapolation beyond

the observed data, especially at high flows, is considered more reliable

because the skewing effect of the data points at the other extreme has
been eliminated. It is important in this extrapolation that maximum
concentrations of sediment be considered to avoid the potential hazard of
extrapolating beyond either an observed high value for the stream being
sampled or no greater than about 50 to 60 percent concentration by weight.
The one to three equation procedure can also be adjusted so that a second,
parallel set of curves will produce the sediment load equal to the sum of
the observed data points. The result of this procedure for computing
suspended sediment rating curves is shown in figure 6 where the equation
for any segment is in the form:
Q5=aQ

(2)

in which Q5 = suspended transport tons/day

= discharge, ft3/s (m3/s)
a = coefficient
b = exponent
RIO TOA VACA NEAR VILLALBA. PUERTO RICO
S(DJM(NT DISCNARG( 91 TONS /DAY

IN ETRIC TLP4/

StOT1T

Figure 6. - Suspended sediment rating curve for Rio Toa Vaca

near Villalba, Puerto Rico.
An approximate 5-year sampling period may be needed to adequately cover
the full range in water discharges and to avoid extreme curve extrapolation. However, a shorter period may be possible if the range in flows
is adequately covered. The upper portion of the rating curve is most
critical; it significantly affects the rate of sediment transport because

of the extreme large sediment loads carried during flood periods. Another
variation in rating curves is described by Miller (1951) when the source of
runoff can be a combination of either snowmelt or rainstorms. It may be
necessary to develop individual sediment-rating curves for each of the
seasons. Runoff from thunderstorms will usually transport sediment at
higher concentrations than runoff from snowmelt taking place in the higher
elevations.
Suspended sediment rating curves can be combined with available water
discharge records to determine the long-term average sediment yield. The
longer the period of discharge records, the more reliable the results. One
technique for gaging station records that cover a long period is to construct a flow-duration curve from the daily water discharges. This curve
is really a cumulative frequency plot that shows the percent of time that
specific discharges are equaled or exceeded for the period of record. For
some streams, where only short-term discharge records are available, a
lc.g-term flow-duration curve can be computed from a correlation of shortterm to long-term records at a gaging station either on the same stream or
nearby stream. If the flow-duration curve is representative of the longterm flow of the stream, it may be considered a probability curve and used
to represent future conditions. With this assumption, it is combined with
the suspended sediment-rating curve as described by Miller (1951) to
determine the long-term average suspended sediment yield for any projected
period such as 100 years. An example of the flow-duration curve for the
same station used to develop the sediment rating curve on figure 6 is
illustrated in figure 7. Reclamation's (Bureau of Reclamation) computer
facility has linked up with that of the U.S. Geological Survey in Reston
for obtaining flow-duration data for any desired period of flow record.
Table 2 shows the computation of suspended sediment load at the gage
based on combining the sediment rating curve with the flow-duration
curve.
d. Unmeasured sediment load. - To analyze the unmeasured portion of the
total sediment load requires a knowledge of the following terms:
Bed material. - The sediment mixture of which the streambed is composed.
Bedload. - Sediment that moves by rolling or sliding on or near the
streambed.
Bed material load. - That part of the sediment load which consists of
grain sizes represented in the bed.
Wash load. - That part of the sediment load which consists of grain
sizes finer than those of the bed.
Suspended load. - Particles moving outside the bed layer.
Unsampled zone. - The 3 or 4 inches (7.62 to 10.2 cm) from the streambed up to the lowest point of the sampling vertical. Most suspended
sediment samplers cannot sample within this zone.

10

PERCENT OE TIME INDICATED DISCHARGE IS EQUALED OR EXCEEDED

0
SO
40
*0
20

40
30

000

20
a

JO
a
a
4
100
9
S

z2
tD

R a Too Voco
,,eor V,//o/bo, Puerto R,co-,
(1938-1965)

4
I

1
l0
I
6

a
9.9
o .4
2
.9 .3

100

2
.9

.2

.0*
.0*

.04
.03

10

.03

.01
0.01

0.1

0.912

510 20 40 *0 *09095 99 99.9 99,99

PERCENT DE TIME DISCHARGE 5 LESS 11.149 INDICATED 46401,151

Figure 7. - Flow duration curve for Rio Tao Vaca near Villalba,
Puerto Rico.
The suspended sediment load as computed in table 2 represents only a
portion of the total sediment load. The unmeasured load consists of
bedload plus suspended sediments in the unsampled zone between the sampler
nozzle and the streambed. At the time the sediment sampling program is
established, a preliminary appraisal is made on the percentage that the
unmeasured load is of the total load. A useful guide for evaluating the
unmeasured load is the bedload correction shown in table 3. Five conditions are given for defining bedload dependent upon suspended sediment
concentration and size analysis of streambed and suspended materials.
As shown in table 3, either condition 1 or 2 may result in significant
bedload which would require a special sampling program for computing the
unmeasured sediment load. Conditions 3, 4, and 5 usually indicate a 2 to
15 percent correction factor which would not require any special bedload
sampling program.

11

Table 2. - Sediment load computations of Rio Toa Vaca near

Villalba, Puerto Rico
Project

Pe,'tp RicO

Section _____________ Dote

Strearoflow (930-1865

Period at record

o,o
u02-0.i
0.1 - 0.5
0.0- .5
__________
5 5

'miii

25
35
45
35
4555
59' 65
45 - 75
75
85

__t1.L.7L

95-965
985-995
99.599.9
99.9 -99.98
99.98'iOO

0
NiE liv

Checked by

ooo
o,on

' T

..J2.....

_______
O/

o.oi
.0

.i

' 26
' "
30
40
54
60
70
80

1jj._...
...22i_

1.25),,

"559J

01

..!.2....
06.15
99,0
99.7
99.94
99,99

________0 ,

tnn/d

30.161

(29.4
7.5

'"WT'

]L

j J3Jj,
3),
,(

39.925
8.038
1.657
426.6
59,7
0.58.s
0.053
(I 00

16.57

0,114
0.125
0.059
(0.040)

14.93
S.97
0.058
0.005
0.001
11.0002

UiL,

2.080
1.410
1 191.13

J,QJI30)

7 0113

0.1.411
0.460
0.350

5.42
________
1 0.005)

LflJ3D1
.
.L
(... 0..0QJ..

_________

_________
________ I

_________

_________

_,__.

________

Tn

____________________

Annual discharge

10.071)

4.039

_________

(0.034)
(0.030)
(0.0301

(14.46
(29.17

2.61

_________

______
_________

15.94
32.15

_________

_______

(0.0481

1.1
1, 3

0.024
0. 7

4.410

(, fl3

2.47

_________

(fl flee)

,,_

.....J8Z
i4z)
0531
,Q4Q9)

________

08.6)

6.83

0.830

..L5Q4

7
Cal 2, Cnt.5
$-

(0.018
(0.013
(0.010)
- 0.240..,,,
(0.0071
(0.0021
0J360_.
- .DlL.,... ....L1LDDQ4
.
. .
(0.00011
- .Qj)QS__.
0.001
________

in

i'o

0.282

(1.9QU

(1 007

3_

.flsLJ
(30,4

0 (810.0 .(fl4

6.4
4.6
3.5 4
_1

_____________________
_CoI.2 CoL4

5
_______
l, /t I

L4J0)

1,037.8
617.8
250.6
115.4
44.1
20.8
(4.1

0,06

,,,,,__7.3

10
0
0
0
0
0
3,5
I .0
0.4
0.08
0,02

3
U DOLE
L ORDINATE

0.4
.0

1969-1067

Sediment

Computed by
________
LU TO

Toe taco

R,seroOir

Stream Rio Toe vaca

________

21.67

(0.6161

_________

________
_______

_________

________

________

______

______
_________

92.45

(83.90)

. Total
365x 1.9835 .1..21).acretr
- (Total (5,,ft,.8J,5,n 365 086.4 0 1
_______

Annual sediment toad

Total Qs..924Ln 365

(Total Qs.8,3JLx 365

..3D_,ton/y9') (metric)

Average concentration, C

92.45

__________

0.0027 21.67

I________
0.0664
w

0.0027

83.90
0.615 o 0.0864 J.S1.alg/1)

A special sampling program to be undertaken under conditions 1 and 2 in

table 3 is usually established for total sediment transport computations
by use of the Modified Einstein procedure (Colby and Hembree, 1955; Bureau
of Reclamation, 1955 and 1966). Modified Einstein computations require
the collection of the following data for at least 5 to 10 discharges
covering the range of flows with as many measurements at higher discharges
as possible:
Discharge measurements:
Sediment samples:

Cross-section area, channel width, depth,

mean channel velocity, and streamfiow

Suspended sediment samples analyzed for concentration and size distribution, bed material
samples analyzed for size distribution, and water
temperature

12

Table 3.
Suspended
sediment
concentration
mg/L

Condition

1/ 1

Bedload correction

Texture of
suspended
material

<1000

Sand

1000 to 7500

Sand

>7500

Sand

Any
concentration

Compacted clay
gravel, cobbles, or
boulders
Clay and silt

20 to 50
percent sand
20 to 50
percent sand
20 to 50
percent sand
Small amount up
to 25 percent
sand

1/ 2
-

3
2/ 4
-

Any
con centr aton

Percent bedload
in terms of
suspended load

Streambed
material

25 to 150
10 to 35
5
5 to 15

No sand

<2

1/ Special sampling program for Modified Einstein computations required

under these conditions.
2/ A bedload sampler such as the Helley-Smith bedload sampler may be used
or computations made by use of two or more of the bedload equations when
bed material is gravel or cobble size.
Table 4. - Modified Einstein procedure computation
0111 PUT
DETERMINATION OF TOTAL SEDIMENT LOAD IN A STREAM

dOBIDENT

NIOBRARA RIVER- RIVER RANGE 5

METHOD OF COMPUTATION
DATE OF SAMPLE

IUSCI4ARGE

065 -

3060

MILLIMETERS

538
FSO.
(50. U')

VF0CI1Y

158 FT/SEC
482 U/SI

035

512E FRACTION
IN MILLIMETERS

.2360

TOP WIDTH

705.0 FT.
(215. M

EQ*JIV, WIDTH

0.0 FT.

DISTANCE BETWEFN SAMPLER AHO RED (135086)

PERCENT OF MATERIAL
SUSPEM7EO
BED

DATE OF COMPUTATION

TEMPERATURE

CONCENTRATION IN PPM

850. CF5
(24.1 M"/S)

AREA

MODIFIED EINSTEIN

TIME OF SAMPLE

06/13/79

.30 FT.
(.0914 Ml

1808
1/0

73.0 F
(22 2 C)

03/26/82

SLOPE OF ENERGY GRADIENT

SAMPLED SEDIMENT

296.

679.
(616.

.00130 FT/FT
(.00130 H/U)

TONS/DAY
TONS/DAY)

MILLIMETERS
EOUIV.

DEPTH

0.01) FT.

AVERAGE BOTTOM OFPTII

EQ1JIV.

.76 FT.
(.232 Ml

0.00000 FT/FT

SLOPE

HYDR, RADIUS

AVERAGE DEPTH FROM SAMPLE VERTICALS )O5URS(

QPRIME
2 - V A L U E 5 COMPUTATIONAL FACTORS
F(I (Al
F)J )
SIJBS(T/D)
COMPUTED FITTED

.76 FT.
(.232 MI

.76 FT.
(.232 U)

COMPUTED
TOTAL LOAD
I/O

0150
0625
1251)
2500
5000
I 00013
20(530
4 (P300
(11300
IOTA(S

13625
12513
25130
5000
I 0(100
2.0(530
413000
60000
IA 13000

16.50
15.20
34.00
30,80
3.130
0.00
0.00
0.00
0.00

.26
1,84
39.50
90.34
6,11
.99
.74
.17
.05

100,00

100.00

.01
.19
11.28
40.67
7.34
.07
.00
0.00
0.00

74,0
66.5
148.8
135.2
13.1
0.0
0.0
0.0
0.0
437,7

0,00
0.00
.57
.74
.83
0.00
0.00
0.00
0.00

.23
.42
.58
.72
84
.94
105
1.17
1 29

0.00
0.00
0.00
0,00
0.00
0.00
0.00
0,00
0.00

1149.56
162.95
48.77
18,47
11.87
7.64
5.53
4,18
3.21

T/D

114.8
(104,2)
103.3
(93.71
550.3
(499.21
751.3 (681.6)
87 I
(79.0)
.9
(0.5)
.0
0.0
0.0
I6073 (1498 21

13

The Modified Einstein procedure is quite different from the original

Einstein (1950) method. Unlike many formulas for computing sediment transport, it is not a method for predicting sediment transport under future
flow conditions. The unique requirement for a discharge measurement and
collection of depth-integrated, suspended sediment samples as a base in
the computations makes the Modified Einstein procedure serve two main
purposes: (1) it gives the unmeasured load to be added to the suspended
load, and (2) it provides a check or verification on the most reliable
predictive formula. An example of the Modified Einstein computation
results is shown in table 4, a printout from the computer program developed
by Reclamation. The computer program developed by Reclamation follows the
same procedure given in the Bureau of Reclamation (1955) report except for
the suspended load exponent or computation of "zfl which is described in
Bureau of Reclamation (1966) publication.
There are situations where other methods for computing the unmeasured load
are needed to either supplement or to replace the Modified Einstein
procedure. This usually happens at the higher water discharges when
sampling is difficult or with bimodal transport (usually under condition 4 or 5 in table 3) where streambed material is unlike the suspended
materi al
Several methods or formulas for computing the bedload or total bed material
load have been advanced by various investigators over the years. Most of
these formulas are based on the principle that the capacity of the stream
to transport bed materials varies directly with the differences between
the shear stress acting on the bed particles and the critical shear stress
required for initiation of particle motion (Herbertson, 1969). One of the
better known formulas is that of Einstein (1950), which applied a stochastic
approach to sediment transport. Statistical and probability theories are
used as a basis for formulas and experimental results are used to establish
values for various constants and indexes. Of the various refinements of
Einstein's original work, Reclamation has experienced the most success in
predicting sediment transport in streams having graded bed material size
by use of the Velocity-Xi Adjustment to the Einstein formula as described
by Pemberton (1972). Other formulas that are often used to compare with
the Modified Einstein method are:
Meyer-Peter, Muller (1948) and by Sheppard (1960)
Schoklitsch, by Shulits (1935)
Ackers and White (1973)
Engelund and Hansen (1967)
Yang (1973)
A description of the theory and development of the above formulas are much
beyond the scope of this narrative, and the reader is directed to the
listed references for this information.
The recommended approach for extending the range of total sediment loads
is to compute total sediment load using the Modified Einstein procedure
for as wide a range of discharge as possible and then compare these

14

results to those of the predictive formulas. The one giving results most
comparable to the Modified Einstein computations is then used to extend
the range to higher discharges. When data are not available for Modified
Einstein computations, selection of a predictive formula should be of one
which has given good comparative results for streams having similar
hydraulic properties and bed material size distributions.
If the bed material is predominately coarse sand greater than about 0.5 m,
gravel-, or cobble-size material, a special sampling program may be used
either independently or as a check on the bedload formula. This involves
measuring the bedload by a direct measuring sampler such as the HelleySmith bedload sampler described by Emett (1980). The sampling procedure
can be quite extensive, depending on dunes and irregular streambed patterns.
Several samples at 10 to 20 equally spaced verticals in the cross section
are necessary to adequately describe the spatial and temporal variations in
transport rate.
Once the rate of unmeasured sediment movement has been determined from
either the Modified Einstein computations or bedload formulas, an unmeasured load rating curve is drawn. A log-log plot of water discharge
versus unmeasured load for these special samples can be analyzed by least
squares analysis. A computation of unmeasured load from the correlation
of water discharge to unmeasured load is similar to the suspended load
computations shown in table 2. Total load is obtained by combining the
results of the suspended load and unmeasured load computations.
e. Adjustment to damsite. - Any direct measurement of sediment yield
either from reservoir surveys or sediment sampling requires an adjustment
in the yield rate from a specific location to that at the damsite. In
many cases the sediment yields in acre-feet or tons per square mile (cubic
meters or tons per square kilometer) derived from the reservoir survey or
at the gaging station can be applied directly to the drainage area above
the damsite. If the yield rates are not directly applicable to the
drainage area above a damsite, the nine factors shown in table 1 can be
used in a calibration technique for adjustment to the damsite.
Reservoir sediment deposition. - Once the estimated sediment inflow to a
reservoir has been established, attention must be given to the effect the
deposition of this sediment will have upon the life and daily operation of
the reservoir. The mean annual sediment inflow, the trap efficiency of the
reservoir, the ultimate density of the deposited sediment, and the distribution of the sediment within the reservoir, all must be considered in the
design of the dam.
Usually to prevent premature loss of usable storage capacity, an additional
volume of storage equal to the anticipated sediment deposition during the
economic life of the reservoir is included in the original design. The
Bureau of Reclamation requires that provisions be made for sediment storage
space whenever the anticipated sediment accumulation during the period of
project economic analysis exceeds 5 percent of the total reservoir capacity.
A 100-year period of economic analysis and sediment accumulation is typically

15

used for a reservoir; however, less than 100 years of sediment accumulation
may be used if the economic analysis would justify a lesser allocation. The
allocated sediment space is provided to prevent encroachment on the required
conservation storage space for the useful life of the project.
A schematic diagram of anticipated sediment deposition (fig. 8) shows the
effect of sediment on storage. A distribution study with 100-year area and
capacity curves similar to that shown on the left side of figure 8 is needed
whenever the 100-year sediment accumulation is more than 5 percent of the
total reservoir capacity. In operational studies of a reservoir for determining the available water supply to satisfy projected water demands over an
economic life, an average (50 years for a 100-year economic analysis) can be
used of the sediment accumulation during the economic life period. However,
the total sediment deposition is used for design purposes to set the sediment
elevation at the dam to determine loss of storage due to sediment in any
assigned storage space and to be used in determining total storage requirements.
----i

MAXIMUM WS. ELEVATION-----.

--

TOP OF FLOOD CONTROLTOP OF ACTIVE

CONSERVATION STORAGE--- --

SURCHARGE

FLOOD CONTROL POOL

/
TOP OF INACTIVE STORAGE
NACIES:SE
----TOP OF DEAD STORAGE
----STREAMBED
CAPACITY

DISTANCE

SCHEMATIC DIAGRAM
RESERVOIR ALLOCATIONS
SEDIMENT DEPOSITION

Figure 8. - Schematic diagram, reservoir allocations, sediment

deposition.
a. Trap efficiency. - The trap efficiency of a reservoir is defined as
the ratio of the quantity of deposited sediment to the total sediment
inflow and is dependent primarily upon the sediment particle fall velocity
and the rate of flow through the reservoir. Particle fall velocity may be
influenced by size and shape of the particle, viscosity of the water, and
chemical composition of the water. The rate of flow through the reservoir
is determined by the volume of inflow with respect to available storage
and the rate of outflow.
Methods for estimating reservoir trap efficiency are empirically based
upon measured sediment deposits in a large number of reservoirs. Gunnar
Brune (1953) has presented a set of envelope curves for use with normal

16

ponded reservoirs using the capacity-inflow relationship of the reservoirs. The Brune medium curve is reproduced in figure 9.
K

80

70

(SEQ/P.N1ATION INSEX BY C/IURCIIILLI,

/
2

RESERVOIR TRAP
EFFICIENCY

Bruise Medium Curve

Chiit-cM/ rrop Efficiency

60

X-,00-(1600K2-12)

______

40
30

01

ReseroorBr,mls Dots Po,sts

5)0961
-

20

U-I

MO/Dt/IQ (/na)

962-1972

-S

Hit6ud (/96/0)

/957 - /973

GoodS, Sopor (/r,do) /562 -/972

962 ii
Sflo*-a 1n010
I

U-3

0-2

Period
Reserior
Symbol
L)tu oddbo to Colochil re/stos/hol by uO8R

Period

-4

-3

/
0-2
0-3
0

/0

c
0001

IH HJt

I IIIHI

0 0/

Des/son (Lobo Te,iog)

Lake Ccpo5 Christ,
Pcvt SIPPly ReSdr9/97
6/hi/siC)
CRRC/O/ 9tod
Speed Stojy
Spers.sl Study
Jo/hi Mart,o

HIIIL

0
0 I
RATIO OF RESERVO/R CAPAC/TY ro AVE RAGE ANSI/AL
/MFLOW --ACRE - FE El/ACRE - FEET In,3/s3IhY BR/./ NE

Figure 9.

962 - /994
/942-948
/928- /947
/928-1947
Jo/y IT-IS. /960
4ig 3-/k. /960
juy 57-3/196/
/942-1997

/0

/c

Trap efficiency curves.

Using data from Tennessee Valley Authority reservoirs, M. A. Churchill

(1948) developed a relationship between the percent of incoming sediment
passing through a reservoir and the sedimentation index of the reservoir.
The sedimentation index is defined as the ratio of the period of retention
to the mean velocity through the reservoir. The Churchill curve has been
converted to a truly dimensionless expression by multiplying the sedimentation index by g, acceleration due to gravity.
The following description of terms will be helpful in using the Churchill
curve:
Capacity. - Capacity of the reservoir in the mean operating pool for
the period to be analyzed in cubic feet (cubic meters).
Inflow. - Average daily inflow rate during the study period in cubic
feet per second (cubic meters per second).
Period of retention. - Capacity divided by inflow rate.
Length. - Reservoir length in feet (meters) at mean operating pooi
1evl.
Velocity. - Mean velocity in feet per second (meters per second), which
is arrived at by dividing the inflow by the average cross-sectional
area in square feet (square meters). The average cross-sectional area
can be determined from the capacity divided by the length.
Sedimentation index. - Period of retention divided by velocity.
17

Fiyure 9 provides a good comparison of the Brune and Churchill methods for
computing trap efficiencies using techniques developed by Murthy (1980).
A general guideline is to use the Brune method for large storage or normal
ponded reservoirs and the Churchill curve for settling basins, small
reservoirs, flood retarding structures, semidry reservoirs or reservoirs
that are continuously sluiced.
When the anticipated sediment accumulation is larger than one-fourth of the
reservoir capacity, it is necessary that the trap efficiency be analyzed
for incremental periods of the reservoir life. Theoretically, the reservoir trap efficiency will decrease continuously once storage is begun;
however, for most reservoirs it is not practical to analyze the trap
efficiency in intervals of less than 10 years. The variability of the
annual sediment inflow is sufficient reason not to use shorter periods of
analysis.
b. Density of deposited sediment. - Samples of deposited sediments in
reservoirs have provided useful information on the density of deposits.
The density of deposited material in terms of dry mass per unit volume is
used to convert total sediment inflow to a reservoir from a mass to a
volume. The conversion is necessary when total sediment inflow is computed
from a measured suspended and bed material sediment sampling program.
Basic factors influencing density of sediment deposits in a reservoir are
(1) the manner in which the reservoir is operated, (2) the texture and
size of deposited sediment particles, and (3) the compaction or consolidation rate of deposited sediments.
The reservoir operation is probably the most influential of these factors.
Sediments that have deposited in reservoirs subjected to considerable
drawdown are exposed for long periods and undergo a greater amount of
consolidation. Reservoirs operating with a fairly stable pool do not
allow the sediment deposits to dry out and consolidate to the same degree.
The size of the incoming sediment particles has a significant effect upon
density. Sediment deposits composed of silt and sand will have higher
densities than those in which clay predominates. The classification of
sediment according to size as proposed by the American Geophysical Union
is as follows:
Sediment type

Size range in millimeters

Clay
Silt
Sand

Less than 0.004

0.004 to 0.062
0.062 to 2.0

The accumulation of new sediment deposits, on top of previously deposited

sediments, changes the density of earlier deposits. This consolidation
affects the average density over the estimated life of the reservoir such
as for a 100-year period. A good example on consolidation of deposited
sediments is shown in figure 10 taken from the report by Lara and Sanders
(1970) for unit weights (densities) in Lake Mead at a sampling location
with all clay-size material.

18

DENSITY N K/m3
80 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950
0

-0

-2

UI
UI
U

-3w
IUI

I
- 1

-5

Li

20

--

PISTON CORE SAMPLER

GMMO PROBE

24

28

__

____L I I I

10

'4

0
UI

'3

Hb

50

-8
54

58

62

DRY UNIT WEIGHT IN LBS/FT3

Figure 10. - Comparison of densities on Lake Mead at location 5.

The method that takes into account all three factors in determining the
density of deposited sediment is demonstrated in these guidelines. The
influence of reservoir operation is most si9nificant because of the amount
of consolidation or drying out that can occur in the clay fraction of the
deposited material when a reservoir is subjected to considerable drawdown.
The size of sediment particles entering the reservoir will also have an
effect on density as shown by the variation in initial masses. Some 1,300
samples were statistically analyzed by Lara and Pemberton (1965) for determining mathematical equations of variation of the density of the deposits
(sometimes termed unit weight or specific weight) with the type of reservoir operation. Additional data on density of deposited material from
reservoir resurveys have supported the Lara and Pemberton (1965) equations
(equation 3) which are slightly different than the Lane and Koelzer (194)
equations.
Reservoir operations were classified according to operation as follows:
Operation

Reservoir operation

1
2
3
4

Sediment always submerged or nearly submerged

Normally moderate to considerable reservoir drawdown
Reservoir normally empty
Riverbed sediments

Selection of the proper reservoir operation number usually can be made

from the operation study prepared for the reservoir.
Once the reservoir operation number has been selected, the density of the
sediment deposits can be estimated using the following equation:

wp

Wmpm

Wsps

19

(3)

where
W = unit weight in pounds per cubic foot (density in kilograms
per cubic meter)
c' Pm' Ps = percentages of clay, silt, and sand, respectively, of
the incoming sediment
W, Wm, W5 = coefficients of clay, silt, and sand, respectively,
which can be obtained from the following tabulation:
Initial weight (initial mass) in lb/ft3 (Kg/rn3)
Wm
Wc

Operation

26
35
40
60

1
2
3
4

(416)
(561)
(641)
(961)

70
71
72
73

97
97
97
97

(1120)
(1140)
(1150)
(1170)

(1550)
(1550)
(1550)
(1550)

As an example, the following data are known for a proposed reservoir:

Reservoir operation: 1
Size ana'ysis: 23 percent clay, 40 percent silt, and 37 percent sand
then:
W

26(0.23) + 70(0.40)
(1120 kg/rn3)

97(0.37)

6.0

28.0

35.9

70 lb/ft3

In determining the density of sediment deposits in reservoirs after a

period of reservoir operation it is recognized that part of the sediment
will deposit in the reservoir in each of the 1 years of operation, and
each year's deposits will have a different compaction time. Miller (1953)
developed an approximation of the integral for determining the average
density of all sediment deposited in 1 years of operation as follows:
WT

W1

0.4343K

(logT)

i]

(4)

where
WI = average density after 1 years of reservoir operation
= initial unit weight (density) as derived from equation 3
K = constant based on type of reservoir operation and sediment
size analysis as obtained from the following table:
Reservoir operation
1
2
3

K for inch- pound units (metric units)

Clay
Silt
Sand
5.7 (91)
1.8 (29)
0(0)

0
0
0

20

16 (256)
8.4 (135)
0( 0)

Using the same example as was used for the initial unit weight (density)
computation, the 100-year average values to include compaction are computed as follows:
K

16 (0.23)
=

70

57 (0.40)

0 (0.37)

0.4343 (5.96)[

3.68

(4.61)

2.28

1]

70

5.96

2.59 (3.66)

79 lb/ft3

(1270 kg/rn3)
This value may then be used to convert the initial weights (initial
masses) of incoming sediment to the volume it will occupy in the reservoir
after 100 years.
c. Sediment distribution within a reservoir. - The data obtained from
surveys of existing reservoirs (U.S. Department of Agriculture, 1978) as
described in section (b) Reservoir Resurvey Data have been extensively used
to develop empirical relationships for predicting sediment distribution
patterns in reservoirs. The two most common distribution techniques are
illustrated in figures 2 and 3, where sediment is distributed by depth and
by longitudinal profile distance, respectively. Both methods clearly show
that sediment deposition is not necessarily confined to the lower storage
increments of the reservoir.
Sediment accumulations in a reservoir are usually distributed below the top
of the conservation pool or normal water surface. However, if the reservoir
has a flood control pool and it is anticipated that the water surface will
be held within this pool for significant periods of time, a portion of the
sediment accumulation may be deposited within this pool. Figure 11 is a
plot of data from 11 Great Plains reservoirs in the United States which may
be used as a guide in estimating the portion of the total sediment accumulation which will deposit above the normal water surface. This plot should
lOG 80
w.-.
3

60 -

6.0 -

-J

J40

2.0
U.

LC
0.8

III

1111

RESERVOIRS

I
2.
3
4
5.
&

HARRY STRUNK
HARLAN
UN1Y
0 I
KANOPOLIS
/s
LAKE TEXOMA
LA VON
FORT SUPPLY__
___________ _____________

7 BOYSEN

8. GREAT SALT PLAINS

9. ALTUS
0. CONCHAS
I. CANTON

______

0
______

"
/2

______

0.4

rx

0.2
-

[I!
II
II
0.2 0.406 .0 2.0 4060 0 20 40 60100
PERCENT OF SEDIMENT TRAPPED IN FL000 CONTROL POOL

Figure 11. - Sediment deposited in flood control pool.

21

be regarded as a rough guide only, and the estimate obtained from it should
be tempered with some judgment based upon the proposed reservoir operation
and the nature of the incoming sediment. This curve is based on a limited
amount of data and may be revised as more information becomes available.
The term flood pool index is computed as the ratio of
pool depth to the depth below the pool, multiplied by
the reservoir water surface will be within the flood
information for a proposed reservoir must be obtained
operation study.

the flood control

the percent of time
control pool. This
from the reservoir

Once the quantity of sediment which will deposit below the normal water
surface has been established, the Empirical Area-Reduction Method may be
used to estimate the distribution. This method was first developed from
data gathered in the resurvey of 30 reservoirs and is described by Borland
and Miller (1960) with revisions by Lara (1962). The method recognizes
that distribution of sediment is dependent upon (1) the manner in which the
reservoir is to be operated, (2) the texture and size of deposited sediment particle, (3) shape of the reservoir, and (4) volume of sediment
deposited in the reservoir. However, the shape factor was adopted as the
major criteria for development of empirically derived design curves for
use in distributing sediment. The shape of the reservoir is defined by
the depth to capacity relationship where em" is the reciprocal of the
slope of the depth versus capacity plot on a logarithmic paper. The
classification of reservoirs on this basis is as follows:
Reservoir type

Classification

I
II
III
Iv

Lake
Flood plain-foothill
Hill
Normally empty

3.5 to 4.5
2.5 to 3.5
1.5 to 2.5

The procedure now used by Reclamation for distribution with depth is that
of using design curves shown in figure 12. With equal weight applied to
reservoir operation and shape, a type distribution is selected from table 5.
In those cases where a choice of two types are given, then a judicious
decision can be made on whether the reservoir operation or shape of reservoir is more influential. The texture and size of deposited sediments
could be considered in this judgment analysis from the following guidelines:
Predominant size

Type

Sand or coarser
Silt
Clay

I
II
III

The size of sediments in most river systems is a mixture of clay, silt,

and sand and has been found to be least important in selecting the Design
Type Curve from figure 12. Only for those cases with two possible type
distributions should size of sediment be considered in selecting the
Design Type Curve.

22

ID

I
_

I
_

1_

//

I_

//

rip. I
r

/ // 3

S
S

_-

ePIpty

-K
__

__

__ __

PERCENTAGE

__

__

__

SEDIMENT DEPOSITED

Figure 12. - Sediment distribution design curves.

Table 5.

Design type curve selection

Shape

Reservoir operation
Class
Type
Sediment submerged

Type

Class
Lake
Flood plain
foothill
Hill and gorge
Lake
foothill
Flood plain
Hill and gorge
Lake
Flood plain
foothill
Hill and gorge
All shapes
-

Moderate drawdown

II

Considerable drawdown

III

Normally empty

IV

23

I
II
III
I
II
III
I
II
III

Weight ed
type
I
I or II
II
I or II
II
II or III
II
II or III
III
IV

The Lara publication (1962) provides the detail on distributing sediment in

a reservoir by the Empirical-Area Reduction Method. The appropriate design
type curve is selected using the weighting procedure shown in table 5. A
computer program written by Hudspeth and Trietsch (1978) can be also used
for distributing sediment by either the Empirical Area-Reduction Method or
the Area-Increment Method. The Area-Increment Method is based on the
assumption that the area of sediment deposition remains constant throughout
the reservoir depth. It is almost identical to the type II design curve
and is often used to estimate the new zero capacity elevation at the dam.
An example of a sediment distribution study is given for Theodore Roosevelt
Dam located on the Salt River in Arizona. Construction of the dam was
completed in 1909 and a complete survey of the reservoir made in 1981.
The reservoir had an original total capacity of 1 530 500 acre-feet
(188 8006 m3) at elevation 2136 feet (651.0 m), the top of the active conservation pool. The purpose of this example is to (1) compare the actual
survey of 1981 with the distribution procedures, (2) show all of the steps
involved in a distribution study, and (3) provide changes in capacity and
projected sediment depths at the dam for 100, 200, and 300 years.
Table 6 gives the pertinent area-capacity data necessary to evaluate the
actual 1981 survey and for use as a base in thedistribution study. The
total sediment accumulation in Theodore Roosevelt Lake as determined from
the 1981 survey was 193 765 acre-feet (239 x 106 m3). In the 72.4 years
from closure of the dam in May 1909, until the survey in September 198,
the average annual sediment deposited was 2676 acre-feet (3301 x 106 m)
per year. The survey data from table 6 were used to draw the sediment
distribution design curve on figure 13. To check the most appropriate
design curve by the Empirical Area-Reduction Method, the volume of sediment accumulated in Theodore Roosevelt Lake from 1909 to 1981 was distributed by both a type II and III distribution, as shown in figure 13.
This comparison indicates that type II more closely resembles the actual
survey. A plot of the area and capacity data from table 6 is shown on
figure 14.
The first step in the distribution study for the 100-, 200-, and 300-year
period is a determination of the rate of sediment accumulation. In the
case of Theodore Roosevelt Lake, the rate determined from the 1981 survey
used for future projections with the assumption that the compaction or
density of deposits will not change is:
Years
72.4 (1981)
100
200
300

Sediment volume
Acre-feet
(103 m3)
193
267
535
802

765
600
200
800

24

239
330
660
990

009
100
200
300

Table 6. - Reservoir area and capacity data

Theodore Roosevelt Lake

Elevation
Feet
Meters

c-TI

2136
2130
2120
2110
2100
2090
2080
2070
2060
2050
2040
2030
2020
2010
2000
1990
1980
1970
1960
1950
1940
1930
1920
1910
1902

651.0
649.2
646.2
643.1
640.1
637.0
634.0
630.9
627.9
624.8
621.8
618.7
615.7
612.6
609.6
606.6
603.5
600.5
597.4
594.4
591.3
588.3
585.2
582.2
579.7

Orig inal (1909)

Capacity
Area
106 m3
Hectare
Pcres
i03 acre-ft
17
17
16
15
14
13
11
10
9
8
7
6
5
4
3
2
1
1
1

785
203
177
095
104
247
939
638
482
262
106
216
286
264
544
744
985
428
020
677
419
227
117
52
0

7
6
6
6
5
5
4
4
3
3
2
2
2
1
1
1

198
962
547
109
708
361
832
305
837
344
876
516
139
726
434
110
803
578
413
274
170
91.9
47.3
21.0
0

1
1
1
1

53O.5
425.5
258.5
102.2
956.5
819.3
693.3
580.6
479.9
391.2
314.6
248.0
190.3
142.9
103.8
72.3
48.9
31.9
19.7
11.3
5.9
2.7
1.1
0.2
0

1 888
1 758
1 552
1 360
1 180
1 011
855
716
592
483
388
306
235
176
128
89.2
60.3
39.4
24.4
14.0
7.3
3.4
1.3
0.3
0

Actual surve y (1981)

Capacity
Area
106 m3
acre-ft
Hectare
Acres
17
16
15
14
13
12
11
9
8
6
5
4
4
3
2
1

337
670
617
441
555
746
331
842
230
781
569
847
212
387
036
304
903
382
i/O

1/ Sediment elevation at dam for 1981 survey is 1966 feet (599.2 m).

7
6
6
5
5
5
4
3
3
2
2
1
1
1

016
783
320
844
486
158
586
983
331
744
254
962
705
371
824
528
365
155
0

1 336.7
1 234.3
1 072.4
922.3
782.6
650.5
530.0
424.0
333.8
258.9
197.6
145.6
100.3
61.6
35.0
18.7
7.6
0.8
0
i'

1
1
1
1

649
523
323
138
965
802
654
523
412
319
244
180
124
76.0
43.2
23.0
9.4
1.0
0
i'

I0
I-

1991(72.4yr&)

Typ.
10

Type N distr,buticii
40

dIfrib
,

20-

20

100

80
40
60
SEDIMENT DEPOSITED IN PERCENT

Figure 13. - Sediment distribution for Theodore Roosevelt Lake.

73

70

85

50

35

Il

I
Top of Cm,s.rrof(on
5120

2100

ARt* (104 5011081 1T(ft3)

40
35
30

45

II

'__J_.'

__eoft(15l1I)

I._-I

2040

I?

'

,/

,,

//

540

Cocsty

20)I.Qrs

R.4sctloo

300y84r1J

Typ.

'.__

$30

825

820

815$
9

,,,
Actuol S.ar y (191!)
-ActuoI Surv.y (1981)

O,msI C20OCIIY
(1809)

,;''

'S

i//I
bt
-

5
4

50918)

/_V

20
S

2080

25

-' 5'

2090
-'S

v/I/I

-805

'ISO-

\
'S

'.-

800

590

590

$40

1540

520

AREA AND CAPACITY CURVES

THE000R( 8009EVELT ias

- 585
_-O-g201 Sfr,omtJd 902 ft (5791177)

I900Q

cuPuCl1I )4 8CRE01T)
I
I

9
C.SPSCIIV 110

II

IS

I
2
4

I
5

5
5

II

IS

I
4

-580
1

IS

CIJSIC 841(431

Figure 14. - Area and capacity curves for Theodore Roosevelt Lake.
26

There were no data on trap efficiency to apply to the above projections.

The use of the rate from the 1981 survey results assumes that the trap
efficiency for the first 72.4 years will remain the same through 300 years.
In cases where sediment accumulation is determined from the total sediment
load at a gaging station, then trap efficiency by use of figure 9 and
densities from equations 3 and 4 are needed for computing the volume of
sediment accumulation.
To complete this example a logarithmic plot of the depth-capacity relationship for the original (1909) survey (fig. 15) for Theodore Roosevelt Lake,
provided the shape factor for type classification. Although the lower
portion of the reservoir falls slightly in the type III, the upper portion
and overall slope indicates a type II classification. When assigning a
type classification for either an existing reservoir or in distributing
sediment on top of previous sediment deposits that the stage-capacity
relationship only be plotted for the original survey. Studies have shown
that a reservoir does not change type with continued sediment depositions.
Once a reservoir has been assigned a type by shape, this classification
will not change. However, it is possible that a change in reservoir
operation could produce a new weighted type, see table 5.
The next step in the distribution study is computation of the elevation of
sediment deposited at the dam. A set of computations for determining the
depth of sediment at the dam is shown in table 7. The relative depth and a
dimensionless function from the original area and capacity curves for
Theodore Roosevelt Lake are computed as shown in table 7 with the function:
F

S-Vh

(5)

where
F = dimensionless function of total sediment deposition, capacity,
depth, and area
S = total sediment deposition
Vh = reservoir capacity at a given elevation h
H = original depth of reservoir
Ah = reservoir area at a given elevation h
A plot of the data points from table 7 is superimposed on figure 16
and the p value (relative depth) at which the line for any year crosses;
the appropriate type curve will give the relative depth Po equal to the
new zero elevation at the dam. Figure 16 contains plotted curves of the
full range of F values for all four reservoir types and the Area-Increment
Method as developed from the capacity and area design curves. For Theodore
Roosevelt Dam, the intersect points for type II as well as for the AreaIncrement Method curves gave sediment depths shown in table 8. The AreaIncrement Method is often selected because it will always intersect the F
curve and, in many cases, gives a good check on the new zero capacity
elevation at the dam. In the case of Theodore Roosevelt Dam, the 1981

27

Table 7. - Determination of elevation of sediment at Theodore Roosevelt Ov

1981 survey

Elevation
ft 10 is

relative
depth

Original survey (1909)

Ah
area
capacity
106 m
1O is2
acre-ft
acres

cr
2080 6340
2070 6309
2060 6279
2050 6248
2040 6218
20306187
2020 6157
2010 6126
2000 6096
1990 6066
1980 6035
1910 6005
1960 5974
1950 5944

0.761
0.718
0.675
0.632
0.590
0.547
0.504
0.462
0.419
0376
0.333
0.291
0.248
0.205

693 315
580 590
479 928
391 207
314 623
248 009
190 334
142 903
103 787
72 347
48 867
31 935
19 743
11 328

855
716
592
483
388
306
235
175
128
89.2
60.3
39.4
24.4
14.0

11 939
10 638
9482
8 262
7106
6216
5 286
4 264
3 544
2 744
1 985
1 428
1 020
677

48.3
43.1
38.4
33.4
28.8
25.2
21.4
17.3
14.3
11.1
8.03
5.78
4.13
2.74

Year

Total sediment deposition

72.4
100
200
300

193 765 acre-ft (239 106 is3)

267 600 acre-ft (330 106 is3)
535 200 acre-ft (660 106 ml)
802 800 acre-ft (990 106 is3)

2.79
2.49
2.22
1.93
1.66
1.45
1.24
0.998
0.829
0.642
0.464
0.334
0.239
0.158

is3

3 440
3 070
2700
2 380
2 050
1800
1 530
1 230
1 020
791
573
412
294
195

S-Vh
acre-ft

50 862
89 978
121 418
144 898
161 830
174 022
182 437

16

is3

62.7
111
149.8
178.7
199.6
214.6
225

234 feet (71.3 is)

100 years

72.4 years
H'Ah 106
106
acre-ft

Original depth of reservoir

H'Ah

0.0510
0.109
0.189
0.312
0.485
0.730
1.155

S-Yb
acre-ft
106 m3

19591
77 266
124 697
163 813
195 253
218 733
235 665
247 857
256 272

24,2
95.3
154
202
241
270
291
306
315

300 years

200 years
H'Ah

0.0135
0.0623
0.125
0.198
0.304
0.471
0.706
1.037
1.622

F
S-Vs
106 m3
acre-ft

55272
143 993
220 577
287 191
344 866
392 297
431 413
462 853
486 333
503 265
515 457
523 872

68.2
178
272
354
425
484
532
571
600
621
636
646

H'Ah

0.0249
0.0746
0.133
0.198
0.278
0.393
0.520
0.721
1.048
1.507
2.157
3.316

S-Yb
106
acre-ft
109 485
222 210
322 872
411 593
488 177
554 791
612 466
659 897
699 013
730 453
753 933
770 865
783 057
791 472

135
274
398
508
602
584
755
814
862
901
930
951
966
976

is3

H'Ah
0.0392
0.0892
0.145
0.213
0.294
0.383
0.494
0.661
0.843
1.138
1.625
2.308
3.276
5.009

CAPACITY
10
4

0'

4$
I

003-

4
I

CU8IC NTNS

l0
4
2
NIl
________________

10

2
I

2
________________

S
I

l0
&
'C
I0_

________

m-3.i$Typefl

_____

0000

0001

IiHHHI!

HHHII H1HIJHIJ

001
DI
CAPACITY IN NULI.ION ACNE - FEET

00

Figure 15. - Depth versus capacity for Theodore Roosevelt Lake.

':

i-4-

600

TABLE OF F VALUES
________

________

p
0
oo

000

I
I
-t

200

poor

yr-S
5 49
4,53
6 971
4 45
2766

013

00
80

25
3

60

35
'I,

9076
7267

so

.s.oo
.4732

20

6
065
7

II

75
4
85
06

\
\
\
\

0
6

'It

"\

-----203

so..

2253
495
1169
09706
8299

3026
2359
777
262
0.08011
.03*30
01494
007411

99

ssoo
s,s,

038,03

98

_______

--

'.900
I 405

45
8

40

z
0

______

02
05

TYPE

''

7212

6523
0 5565
4900
.4303
375$
3253
0.27*0
2333
.1907
500
1107

007276
.03590
01423
.007109
0.0

___2_____
02023
2330
2716
2911

2057
101$
0.8*21
5180
.417*
.3496
.2968
0.2555
.2212
1917

02932
.2870
2781
.2656
2515
0.2355
2187

2010

1657
.1422
01207
00*

.1*26
.1637
01443
245

.09204
06429
.0473,
005101
01527
006097
003020
20

044
009397
06330
0.04239
.02128
008534
.002470
00

0
2

tI00men T

_______

06

--

02

-z

02

0,,,,

THEOWEE

EVFUL

--

_,

I
'-)

_____ ____- -_____

1J
LID
lID
RELATIVE DEPTH (o)

I
Or

ON

I
09

Ii
7O

Figure 16. - Curves to determine the depth of sediment at the dam.

29

survey had an observed elevation at the dam of 1966 feet (599.2 m) which
was in better agreement with the Area-Increment Method value than any of
the type curves. Data from table 8 can be used to predict useful life of a
reservoir or projection beyond the 300 years.
Table 8. - Elevation of sediment at Theodore Roosevelt Dam
H = 234 feet (71.3 m)

Years
p0
72.4 (1981)
100
200
300

0.23
0.284
0.418
0.553

Type II
Elevation
feet
meters
p0H
54
66
98
129

1956
1968
2000
2031

596.2
599.8
609.6
619.0

p0
0.247
0.290
0.4
0.506

Area increment
Elevation
feet
meters
p0H
58
68
94
118

1960
1970
1996
2020

597.4
600.5
608.4
615.7

The final step in the distribution study is to distribute a specified

volume of sediment which for the example selected involved the 72.4-,
100-, 200-, and 300-year volume in Theodore Roosevelt Lake by the type II
design curve. The results of this distribution using procedures described
by Lara (1962) or the computer program by Hudspeth and Trietsch (1978) are
shown in figure 14. An example of the computer results for the 100-year
distribution by use of the Empirical Area-Reduction Method and type II
design curves is shown in table 9. Although the example given is for
type II, the equations for the relative sediment area, a, for each type
follows:
Type

Equation

a = 5.074 p1.85 (1-p)35

(6)

II

a = 2.487 p0.57 (1-p)41

(7)

III

a = 16.967 p1.15 (1-p)232

(8)

IV

a = 1.486 p025 (1-p)134

(9)

where
a = relative sediment area
p = relative depth of reservoir measured from the bottom
p0 = relative depth at zero capacity
d. Delta deposits. - Another phenomenon of reservoir sediment deposition
is the distribution of sediment longitudinally as illustrated in figure 3
for Lake Mead. The extreme upstream portion of the deposition profile

30

Table 9. - Theodore Roosevelt Lake - Type II reservoir sediment deposition study empirical area reduction method-sediment inflow, 267 600 acre-feet (inch-pound units)

Elev.
(ft)
2136.0
2130.0
2120.0
2110.0
2100.0
2090.0
2080.0
2070.0
2060.0
2050.0
2040.0
2030.0
2020.0
2010.0
2000.0
1990.0
1980.0
1970.0
1968.6
1960.0
1950.0
1940.0
1930.0
1920.0
1910.0
1902.0

Ori ginal
Capacity
Area
(acre-ft)
(acre)
17
17
16
15
14
13
11
10
9
8
7
6
5
4
3
2
1
1
1
1

785.0
203.0
177.0
095.0
104.0
247.0
939.0
638.0
422.0
262.0
106.0
216.0
286.0
264.0
544.0
744.0
985.0
428.0
369.7
020.0
677.0
419.0
227.0
117.0
52.0
0.0

1
1
1
1

530
425
258
102
956
819
693
580
479
391
314
248
190
142
103
72
48
31
29
19
11
5
2
1

499
512
547
215
455
272
315
590
928
207
623
009
334
903
787
347
867
935
983
743
328
893
735
059
211
0

Relative
Depth
Area

1.000
0.974
0.932
0.889
0.846
0.803
0.761
0.718
0.675
0.632
0.590
0.547
0.504
0.462
0.419
0.376
0.333
0.291
0.284
0.248
0.205
0.162
0.120
0.077
0.034
0.000

0.000
0.546
0.795
0.945
1.050
1.127
1.184
1.225
1.254
1.271
1.277
1.274
1.263
1.242
1.212
1.174
1.126
1.068
1.059
0.999
0.918
0.821
0.704
0.558
0.358
0.000

Sediment
Area
Volume
(acre) (acre-ft)
0.0
699.1
1018.8
1210.3
1344.8
1443.6
1516.9
1570.0
1606.3
1628.0
1636.5
1632.8
1617.6
1591.0
1553.1
1503.8
1443.0
1381.5
1369.7
1020.0
677.0
419.0
227.0
117.0
52.0
0.0

267
265
256
245
232
219
204
188
172
156
140
124
107
91
76
60
46
31
29
19
11
5
2
1

600
503
914
768
993
051
248
814
293
761
438
092
840
797
076
792
057
935
983
743
328
893
735
059
211
0

Revised
Capaci ty
Area
(ft)
(acre)
17
16
15
13
12
11
10
9
7
6
5
4
3
2
1
1

785.0
503.9
158.2
884.7
759.2
803.4
422.1
068.0
875.7
634.0
469.5
583.2
668.4
673.0
990.9
240.2
542.0
46.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0

1 262 899
1 160 009
1 001 633
856 447
723 462
600 221
489 067
391 776
306 996
234 446
174 185
123 917
82 494
51 106
27 711
11 555
2 810
33
0
0
0
0
0
0
0
0

Table 9. - Theodore Roosevelt Lake - Type II reservoir sediment deposition study empirical area reduction method-sediment inflow, 330 085 dam3 (metric units) - continued

N)

Original
Capacity
(dam3)

Elev.
(m)

Area
(ha)

651.05
649.22
646.18
643.13
640.08
637.03
633.98
630.94
627.89
624.84
621.79
618.74
615.70
612.65
609.60
606.55
603.50
600.46
600.92
597.41
594.36
591.31
588.26
585.22
582.17
579.73

7197.6
6962.1
6546.8
6108.9
5707.9
5361.1
4831.7
4305.2
3837.4
3343.6
2875.8
2515.6
2139.2
1725.6
1434.3
1110.5
803.3
577.9
554.1
412.8
274.0
169.6
91.9
47.3
21.0
0.0

1
1
1
1
1
1

887 871
758 369
552 418
359 582
179 787
010 572
855 204
716 158
591 991
482 554
388 087
305 919
234 777
176 271
128 021
89 249
60 277
39 392
36 967
24 353
13 973
7 269
3 374
1 306
260
0

Relative
Area
Depth

1.000
0.974
0.932
0.889
0.846
0.803
0.761
0.718
0.675
0.632
0.590
0.547
0.504
0.462
0.419
0.376
0.333
0.291
0.284
0.248
0.205
0.162
0.120
0.077
0.034
0.000

0.000
0.546
0.795
0.945
1.059
1.127
1.184
1.225
1.254
1.271
1.277
1.274
1.263
1.242
1.212
1.174
1.126
1.068
1.059
0.999
0.918
0.821
0.704
0.558
0.358
0.000

Sediment
Volume
Area
(dam3)
(ha)
0.0
282.9
412.3
489.8
544.2
534.2
613.9
635.4
650.1
658.8
662.3
660.8
654.6
643.9
628.5
608.6
584.0
559.0
554.1
412.8
274.0
169.6
91.9
47.3
21.0
0.0

330 085
327 498
316 903
303 155
287 397
270 199
251 940
232 901
213 311
193 364
173 230
153 066
133 019
113 230
93 839
74 985
56 810
39 392
36 967
24 353
13 973
7 269
3 374
1 306
260
0

Area
(ha)

Revised
Capacity
(dam3)

7197.6
6679.1
6134.5
5619.1
5163.7
4776.8
4217.8
3669.8
3187.3
2684.8
2213.5
1854.8
1484.6
1081.8
805.7
501.9
219.4
19.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0

1
1
1
1

557
430
235
056
892
740
603
483
378
289
214
152
101
63
34
14
3

786
871
515
427
391
373
264
256
680
190
858
853
758
041
183
255
468
42
0
0
0
0
0
0
0
0

is the formation of delta deposits. The major consequence of these delta

deposits is the raising of the backwater elevations in the channel upstream
from a reservoir. Therefore, the delta may cause a flood potential that
would not be anticipated from preproject channel conditions and proposed
reservoir operating water surfaces. Predicting the delta development
within a reservoir is a complex problem because of the variables such as
operation of the reservoir, sizes of sediment, and hydraulics (in particular, the width of the upper reaches of the reservoir). Sediments deposited
in the delta are continually being reworked into the downstream storage
area at times of low reservoir stage and during extreme flood discharges.
A delta study is needed for situations involving the construction of
railroad or highway bridges in the delta area, defining inundated property
such as urban or farmland, and design of protective structures to control
inundation of property. The two phases of the delta study are to physically
locate the delta and then with the delta in place to run backwater computations through the upstream channel for defining lands that would be
inundated due to a downstream reservoir and delta. The 100-year flood
peak discharge is often used for inundation comparison in the flood plain,
with a 50-year delta to represent average conditions for the 100-year
event. If structures such as bridges or levees to protect homes are being
designed in the headwater area, then the delta should represent 100 years
of sediment deposits to sustain no damage for at least a 100-year period.
The prediction of delta formation is as yet an empirical procedure based
upon observed delta deposits in existing reservoirs. A typical delta
profile is shown on figure 17. It is defined by a topset slope, foreset
slope, and a pivot point between the two slopes at the median or 50 percent reservoir operating level. The quantity of material to be placed
in the delta is assumed to be equal to the volume of sand-size material or
coarser (>0.062 mm) entering the reservoir for the 50- or 100-year
period. A trial and error method, utilizing survey or topographic data
and volume computations by average end-area method, is used to arrive at a
final delta location.

Figure 17. - Typical sediment deposition profile.

33

The topset slope of the delta is computed by one or more of several methods:
(1) a statistical analysis of existing delta slopes which supports a value
equal to one-half of the existing channel slope (fig. 18), (2) topset slope
from a comparable existing reservoir, or (3) zero bedload transport slope
from a bedload equations such as the Meyer-Peter, Muller (1948) and by
Sheppard (1960) or Schoklitsch by Shulits (1935). An example of the topset
slope computed by the Meyer-Peter, Muller beginning transport equation for
zero bedload transport is given by:
(

\312

116)
(10)

S-K B 90
-

where
S = topset slope
K = coefficient equal to 0.19 (inch-pound units) or
0.058 (metric units)
= ratio of total flow in ft3/s (m3/s) to flow over bed of stream
in ft3/s (m3Is). Discharge is referred to as dominant
discharge and is usually determined by either channel bank
full flow or as the mean annual flood peak.
D = diameter of bed material on topset slope usually determined
as weighted mean diameter in millimeters
diameter of bed material for 90 percent finer than
D90
in millimeters
d = maximum channel depth at dominant discharge in feet (m)
n5 = Mannings roughness coefficient for the bed of channel sometimes
computed as Dg0/6/26.
10-I
8

II HI

H H

NOTES

4 -

-CURVE PERCENT 0 ORIGINAL SLOPE

-

E
10
6
z

0 022>
220 t (jt'

20

CURVE

,'^'
/
/39/

IO-

______
(

Il'

_________________

Of 00001(0
09..

(O1 909

.9'

//

90

/i/t
2

_____
9
00

,.9

C9.

Oz.o

C-no

0900O

ZtJ

U10*
00009

90.00 (C

00009

IO
8
6
4

90.00

k-CURVE 3
____

,02
8
6

/o,

/0,

'<

I0'
iO-

10r0TI0

'

CURVE 2
-

'

a.

0
-J
"

00

5(,)

2
3

,
/

TOPSET BED SLOPE IN

'0-I
8
6

/1 I/lI

H I

_____________

4
6 8 tO
4
6 8 IO
2
2
6 8 IO
ORIGINAL STREAM SLOPE IN FT/FT (m/m3

6 80

Figure 18. - Topset slope versus original stream slope from

existing reservoirs.
34

IO

The Meyer-Peter, Muller equation or any other equation selected for zero
transport will yield a slope at which the bed material will no longer be
transported, which must necessarily be true for the delta to form.
The location of the pivot point between the topset and foreset slopes
depends primarily on the operation of the reservoir and the existing
channel slope in the delta area. If the reservoir is operated near the
top of the conservation pool a large portion of the time, the elevation of
the top of the conservation pool will be the pivot point elevation.
Conversely, if the reservoir water surface has frequent fluctuations and a
deeply entrenched inflow channel, a mean operating pool elevation should
be used to establish the pivot point. In the extreme situation when a
reservoir is emptied every year during the floodpeak flows for sluicing
sediment, there will be no pivot point.
The upstream end of the delta is set at the intersection of the maximum
water surface and the original streambed, and the topset slope is projected from that point to the anticipated pivot point elevation to begin
the first trial computations of delta volume.
The average of foreset slopes observed in Bureau of Reclamation reservoir
resurveys is 6.5 times the topset slope. However, some reservoirs exhibit
a foreset slope considerably greater than this; i.e., Lake Mead foreset
slope is 100 times the topset. By adopting a foreset slope of 6.5 times
the topset, the first trial delta fit can be completed.
The volume of sediment computed from the channel cross sections with the
delta imposed on them should agree with the volume of sand size or larger
material anticipated to come from the delta stream. The quantity of
sediment in the delta above normal water surface elevation should also
agree with that estimated to deposit above the normal operating level as
shown in figure 11. If the adjustment necessary to attain agreement is
minor, it can usually be accomplished by a small change in the foreset
slope. If a significant change in delta size is needed, the pivot point
can be moved forward or backward in the reservoir while maintaining the
previously determined elevation of the point. The topset slope is then
projected backward from the new pivot point location and the delta volume
is again computed. The intersection of the delta topset and the original
streambed may fall above the maximum water surface elevation, a condition
that has been observed in small reservoirs.
Downstream channel effects.

a. General degradation. - The trapping of sediment in a reservoir accompanied with clear water releases from the dam upsets the regime or state
of quasi-equilibrium of the downstream river channel. A natural flowing
stream transporting sediment is usually in equilibrium or regime (Lane,
1955) with no long-term trend toward aggradation or degradation. The
release of clear water either through the outlets, powerplant, or spillway
will upset this natural stable condition with degradation of the channel
bed and banks. The degradation process moves progressively downstream
until it reaches a point where the sediment being transported results in a
stable channel or equilibrium. Some reservoirs that have lower trap
35

efficiencies may release water with colloidal clay material <0.004 mm, but
then releases will usually have a minor influence on retarding the downstream degradation. The one exception to a clear water release would be a
reservoir that has planned sluicing with low-level outlets that have
capacity equal to the high river discharges for moving large amounts of
sediment into the downstream channel. Any sediment sluiced through a dam,
especially of sand-size material >0.062 tin, would reduce the expected
downstream channel degradation.
The techniques for computing degradation below a dam can vary considerably
depending on size of sediments in the bed and banks, release discharges at
the dam, and sophistication desired in results. Sophisticated mathematical
modeling solutions (Corps of Engineers, 1977) for computing degradation by
computer application are becoming available such as the model being developed for Reclamation which is scheduled for completion in 1983. These
models simulate the behavior of an alluvial channel by combining a steadystate backwater computation for defining channel hydraulics with a sediment
transport model. Through the use of the electronic computer, flows can be
simulated over any selected time frame to reflect continual changes in both
water surface and the corresponding bed surface profiles to span a 50- to
100-year period. The models, still undergoing development, are being used
on many river channels but are considered more applicable to some of the
large more uniform width-depth-type river channels in the United States
such as the Missouri, Sacramento, or Mississippi Rivers.
Until the mathematical models prove adaptable to
tions, Reclamation's approach to degradation bel
a stable slope or an armoring analyses. Both of
approaches for estimating the depth or amount of
occur downstream from a dam or similar structure
of material forming the bed of the river channel
and Lara (1982).

meet all river condi)w dams is to apply either

these two distinct
degradation that will
are dependent on the type
described by Pemberton

In cases where the streambed is composed of transportable material and

the material extends to depths greater than that to which the channel can
be expected to degrade, the approach most useful is that of computing the
stable channel slope or limiting slope, estimating the volume of expected
degradation, and then determining a three-slope channel profile which fits
these values. However, if large size or coarse material which cannot be
transported by normal river discharges exists in sufficient quantities, an
armor layer will develop as the finer material is sorted out and transported downstream. Vertical degradation will proceed at a progressively
slower rate until the armor is of sufficient depth to inhibit further
degradation.
b. Armoring method. - A less detailed procedure, which should be tested
first for computing degradation below a dam, is the armoring control
method. This is especially applicable if large size or coarse material
exists in the channel bottom that cannot be transported by normal river
discharge and is in sufficient quantity to provide an armor layer as
described by Pemberton (1976) below Glen Canyon Dam on the Colorado River.
Under the armoring process, the finer transportable material is sorted
out, and vertical degradation proceeds at a progressively slower rate
until armor is of sufficient depth to control further degradation. An
36

armoring layer can usually be anticipated if there is approximately 10

percent or more of the bed material of armoring size or larger. The
armoring computations assume that an armoring layer will form as shown in
figure 19 as follows:
y

(11)

where
ya = thickness of armoring layer
y = depth from original streambed to bottom of the armoring
layer
= depth from the original streambed to top of armoring
layer or the depth of degradation and by definition
=

(L\P)y

(12)

where
Ap = decimal percentage of material larger than the armoring size
The two equations are combined to:
,1
-

'a

1)

(13)

The depth, y, to armor will vary with size of particle needed but is
usually assumed to vary by three armoring particle diameters or 0.5 foot
(0.15 meter), whichever is smaller for use in design. Although armoring
has been observed to occur with less than three particle diameters,
variability of channel bed material and occurrence of peak design discharges
dictate the use of a thicker armor layer.
FLOW

II
Original bed
n,oter,oI-__

000
.

__j

0c

yd

,,-Origino/ streombed
p

y
y-Degroded streombed

yo-!
Y -

Depth to bottom of the ormoring foyer

- Depth of degradation
- Armoring layer
DC - Diameter of armor material
zip - Decimol percentage of original bed material
larger than 13

Figure 19. - Armoring definition sketch.

The sediment particle sizes required for armoring can be computed by
several methods with each regarded as a check on the other. Each method
37

will indicate a different armoring size which requires experience and

judgment in selecting the most appropriate. The basic data required to
make these computations necessitate: (1) samples of the streambed material
through the reach involved and at a depth through the anticipated scour
zone, (2) selection of a dominant discharge usually approximately a 2-year
frequency peak discharge, and (3) average channel hydraulic properties for
the selected dominant discharge obtained from steady flow backwater computations through the selected reach of river. The methods used to compute
a nontransportable size are usually based on some form of a sediment
transport equation or relationship that will form the armoring layer as
described in reports of Pemberton and Lara (1982), ASCE (1975), or Yang
(1973) such as those of:
Method
1
2
3
4
5

Meyer-Peter, Muller (Sheppard, 1960) 1948

Competent bottom velocity (Mavis and Laushey, 1948)
Tractive force (Bureau of Reclamation, 1952)
Shields diagram (Pemberton and Lara, 1982; ASCE, 1975)
Yang incipient motion (Yang, 1973)

An example of a degradation computation limited by armoring using above

methods is given below. The following data are known for the example
computations for a channel downstream of a storage dam:
Dominant discharge
Q = 500 ft3/s (14.2 m3/s)
Channel width = B = 60 ft (18.3 m)
Mean channel depth
d = 4 ft (1.22 m)
= V = 3.4 ft/s (1.04 m/s)
Mean channel veloci
Stream gradient = S = 0.0021
Armoring size = D = diameter in millimeter
(1)

Meyer-Peter, Muller
Sd
In

D
-

(14)

\3/2
1/6)

where
K = 0.19 inch-pound units (0.058 metric units)
n5 = 0.03 (assumed for this example)
D90 = 34 inn (assumed for this example)
D

0.0021 (4.0)

\
0.19 '0.03
34 1/6)
=

3/2

0.0021 (1.22'
[0.058 (0.O0215S

38

0.0048
0.000409

0.00256
0.000125

20

20 rwn

(2)

Competent bottom velocity

Vb = 0.51 CD)1"2 ft/s. (0.155 (D)'2 mIs)

(15)

where
Vb = competent bottom velocity
Vm = mean velocity ft/s (m/s)
D = diameter in millimeters
0 = 3.84

0.7 (Vrn)

inch-pound units

= (41.6 V) metric units

O

3.84 (0.7 x 3.4)2 = 22 mm

[41.6 (0.7 x 1.04)2 = 22 mm]

=
=

(3)

Critical tractive force

t.f.

TwdS

(16)

where
t.f. =
=
d =
S =

tractive force, in lb/ft2 (g/m2)

unit weight (mass) of water 62.4 lb/ft3 (1.0 t/m3)
mean water depth, ft (m)
stream gradient

t.f.

62.4 x 4 0 x 0.0021 = 0.524 lb/ft2

[106 g/m x 1.22 x 0.0021 = 2560 g/m2]

=
=

From figure 20, D = 31 mm

(4)

Shields diagram for material >1.0 mm and Reynold's number R* >500.

Tc
(

T) -D

0.06

(17)

where
=
=
=
=
S =
D =
0

critical shear stress = TdS, lb/ft2 (t/m2)

unit weight (mass) of the particle = 165 lb/ft3 (2 65 t/m3)
unit weight (mass) of water = 62.4 lb/ft3 (1.0 t/m)
mean water depth, ft Cm)
slope, ft/ft (rn/rn)
Diameter of particle, ft (rn)
62.4 (4.0) (0.0021) =
0.0848 ft = 26 mm
0.06 (165
62.4)
1.0 (1.22) (0.0021) = 0.0259 m = 26 mm
0.06 (1.65)
-

39

r1

IL L I
__

___
...itisq
fl5 r
Iracilve Nrc.

I
-

- -

,.shoas of It.chy foc.. IW/ 0.5 ds$s, is

j c05 4 (ep rosins.tsly)
fw.

V _____

R.cD.n n,,ndsd nOteS tar ConalS t..th N i5 cont000 of tots ssd ,msnt
V

p-s
E

3 5.

oats,-

- - ________
I Ift -Con.Is lth 55%

c.ti.o
it, ...
-

cOIIW$0ldSI

V__

______

-.
__________
SchoUi.taclt-

4_

________

$0

fl

*
p

VVVVV

--

'

________

________

2
S

fo' '-anal.
fo
.,tn to. cantons oq f,n.
s.d,,nant at 3355 OUter
-

voiu.s for canes. 'a

SO' COndIS fl sOnd---..

0
U-

Cunoin *4th 0 4,
COl lo.41 fl OUts,

-4) S SR

Fort,. , Sco0(,-f-ftsc )nttsnhSd lar c no a a tots fond arto

Wa ter Co oI ln c oUoefo

_,___ 4,___

I4-,

___
-4
'5
(-4

Ro,rnas.sd.d nOI ,*% to, canals , th c lea r t,03 5,

VV -SIr. 0
V ---

-V::F

U S 5R

olosS

of c,t,cal tOCt,tS torc i

V Canals . 55

4.450'

at5

VV

ffl or 5 5c00S5 - RscotwnaOid

f _: 4

to, Canals n io,t sa nd

and cisar *0 35,

Lii:

----U

fi

IV

NIAN DsANr(A-mm

IV

(0

50

30 30 IV 0 UOWSAJ

Figure 20. - Tractive force versus transportable sediment size.

40

(5) Yang incipient motion criteria for shear velocity Reynold's

number R*>70:
(18)

2.05

where
Vcr = critical average water velocity at incipient motion,
ft/s (m/s)
w = terminal velocity, ft/s (mis),
with Rubey (1933) settling velocity for materials larger than about
2 mm in diameter, the fall velocity can be approximated by:
w = 6.01 D 1/2 inch-pound units
(w = 3.32 D 1/2) metric units

(19)

Equations 18 and 19 can be combined to give:

D
0.00659 Vcr2 inch-pound units
(D = 0.216 Vcr2) metric units

(20)

in the example problem

D
[0

=
=

0.00659 (342 = 0.762 ft

0.216 (1.04) = 0.0233 m]

=
=

23 mm
23 mm

Mean of the above five methods for computing armoring size is 24 mm,
which was adopted as a representative armoring size. By use of equation 13, an assumed three layers of nontransportable material to form
an armor, and assumed 17 percent of bed material >24 mm (from size
analysis of streambed material), the depth of degradation is:
Ya

3D

3(24)

72mm= 0.236 ft (0.072m)

0.236 (0.7

i)

(0.072

(o.7

i)

1.15 ft

0.35 m)

c. Stable slope method. - The method of computing a stable slope to

define degradation below a dam is used when there is not enough coarse
material to develop an armoring layer. The method is used when the
primary purpose is to compute a depth of scour immediately below the dam
for design of the dam and downstream protection against vertical scour of
the streambed. It is also used in early planning stages with a limited
amount of field data and when costs for the more detailed study are
prohibitive. The more detailed electronic computer solutions (Corps of
Engineers, 1977) or their predecessor, the desk calculator method by Lane
(1948), are used when data are available to verify the mathematic model,
41

channel hydraulics ca be easily synthesized, and degradation results

influence the channel morphology for several miles (kilometers) below the
dam.
The stable slope method is illustrated by the sketch in figure 21. The
stable slope is defined as the slope of the stream at which the bed
material will no longer be transported. As shown by figure 21, the
method is also identified as the three-slope method because of the variation expected in slope between the stable slope and the existing slope
further downstream. The computations of stable slope can be made by
application of several methods such as: (1) Schoklitsch (Shulits, 1935)
bedload equation for conditions of zero bedload transport, (2) Meyer-Peter,
Muller (1948) and by Sheppard (1960) bedload equation for beginning transport, (3) Shields diagram (Pemberton and Lara, 1982; ASCE, 1975) for no
motion, and (4) Lane (1952) relationship for critical tractive force
assuming clear water flow in canals. The discharge to be used in any of
the above methods is the dominant discharge and is usually determined by
the channel bankfull flow or 2-year flood peak discharge. With regulation
of the streamfiow by an upstream dam, the problem becomes more complex
because detailed data on future releases are usually not available. If the
releases from the reservoir are fairly uniform, and flood discharges are a
relatively rare occurrence, the average daily discharge may be used as the
dominant discharge. However, if the releases are subject to considerable
fluctuation due to floods, the peak discharge which is equaled or exceeded
on the average of once every 2 years would be considered the dominant
discharge.

Notes
d9

(rn/rnof) the do"

Depth f degrodofon

dSgS,SLnFt/Ft
0,

205

3d

309

Naturol

"S

455

Figure 21. - Degraded channel by the three-slope method.

The next step in degradation computations by the stable slope method is
determination of the average channel hydraulic properties for the dominant
discharge. These data can usually be obtained from the tailwater study
that has been prepared for the dam. The properties of all the tailwater
cross sections when carrying the dominant discharge are averaged to arrive
at a generalized cross section which will be representative of the degradation reach. The water surface slope may be assumed equal to the
hydraulic gradient.
42

The volume of material to be removed b

he stable slope method can be
determined in several ways. From figure 21 it may be expressed by:
Vg = ag Bd

(21)

where
Vg = volume of material to be degraded in ft3 Lm3)
ag = longitudinal area of degradation in ft2 (m')
8d = degraded channel width in feet (m)
If there are no downstream controls or no limit to the length (Lg) for
degradation, the two ways to compute the volume are: (1) assume the river
will pick up a load of coarse sediment (>0.062 mm) equal to that portion
of the historic sediment load >0.062 mm, or (2) compute the outflow from
the degraded reach by a sediment-rating curve, flow-duration curve method.
In the second case, the sediment-rating curve would be defined by use of
one or more of the bedload equations and the flow-duration curve of
anticipated reservoir releases.
By rearranging equation 21 the longitudinal area may be found:
a

=a

(22)

Bd

Once a value has been found for a , the depth of degradation may be
computed using the following equaion:
64 a

\s Y)\

1/2
(23)

dg =(
where
LtSg = the difference between the existing slope and the stable
slope, and the length of the degraded reach can be computed
by:
13 d
L=
8ASg
g

(24)

If it is anticipated that lateral degradation will be a significant factor,

additional study will be necessary to determine the degraded channel width.
Because part of the material will be coming from the streambanks, the
extent of vertical degradation will ordinarily not be as great. Lateral
movernent should always be suspect when the banks are composed of the same
material as the bed and there is not a great deal of vegetation to hold
them. Procedures described by Pemberton and Lara (1982), Bureau of Reclamation (1951) and (1952) are recommended as guides when these conditions
prevail.

43

If a permanent control exists at some point within the degradation reach,

equation 24 may be used to solve for the depth of degradation directly.
The three-slope or stable slope method for computing the depth of degradation at the dam and the degraded channel profile are based on satisfying
the following assumptions:
1. The degradation reach is sufficiently uniform to permit the use of
average cross sections and slope throughout the reach.
2. The bed and bank material throughout the reach is similar enough
that an average composition can be used and that there are no existing
nonerodible barriers in the bed or banks to prevent the stream from
attaining the average section at the stable slope.
3. The degradation will be such that the vertical component will
predominate and horizontal movement will be limited to bank sloughing
resulting from vertical degradation.
The Meyer-Peter, Muller equation for beginning transport has been selected
in an example problem for degradation computations by the stable slope
method. Computations by use of the other more commonly used methods for
computing a stable slope are described by Pemberton and Lara (1982).
The following data are known about a river channel below a diversion
dam:
Dominant discharge = Q = 780 ft3/ (22.1 m3/s)
Channel width = B = 350 ft (107 m)
Mean channel depth = d = 1.05 ft (0.32 m)
Existing stream gradient = S = 0.0014
0.3 rTTITI
Bed material Dm = D50
Dg = 0.96 nm
Manning's "n" for bed of stream, n5 = 0.027
Preliminary studies show that 2160 acre-feet (2.66 x 106 3) of sand would
deposit behind the diversion dam during the 100-year economic life of
the structure. Investigations support an equal volume of sand could be
eroded from the downstream channel.
The stable slope computation by the Meyer-Peter, Muller equation for
beginning transport is:
fn5
SL

)3/2

K% DD1/6
B
\90
d

where
SL = limiting slope
K = 0.19 inch-pound units (0.058 metric units)

44

(25)

QB

= Ratio of total flow in ft3/s (m3/s) to flow over bed of

stream in ft3/s (m3/s). Usually defined at dominant discharge where
= 1 for wide channels.
/0.027 '\ 3/2
SL = 0.19 (0.3) O.96l/6)
= 0.00024
1.05
=(o.058 (0.3)
(0.00448)) = 0.00024
0.32

The difference between the existing and degraded slope, ASg. is 0.00116.
The longitudinal degradation area by equation 22 is:
a9= 43 560 (2160) = 269 000 ft2
350
2.66 x i06
= 24 900 m2
107

The depth of degradation at the dam by equation 23 is:

d

= (64 (269 000) (0.00116)

39

1/2
= 22.6 ft

[= (64 (24 900) (0.00116))h/2 = 6.88 m]

39
and the length by equation 24 of the degradation reach is:
13 (22.6)
Lg = 8 (0.00116) = 31 700 ft
-

13 (6.88)
8 (0.00116) = 9640 m)

Conclusions. - The procedures described in these guidelines for presenting

sediment inflow will depend on sediment yields and available sediment data.
In those cases requiring collection of suspended sediment samples, a 5-year
program of either daily or less frequent sampling on an intermittent basis
can be used to define a sediment-rating curve. This curve, combined with a
flow-duration curve, represents the best method for determining the 100-year
sediment inflow. Some estimate is necessary for the bedload which can be
made by correction factor or one or more of the equations cited in this
report.
Reservoir sediment distribution techniques are described for allocation and
location of deposited sediments. The methods given for estimating density of

45

deposited sediments and trap efficiencies will vary from one reservoir to
another. Every situation is unique, and the methods selected for estimating
total sediment inflow, sediments deposited, and distribution will vary.
Prediction of the degradation in the river channel below a dam may involve
the application of a sophisticated mathematical model. These models are
still undergoing development and change besides being costly and subject to
limitations because of basic data. Certain situations may still warrant the
use of the less detailed models like the limiting slope method. If armoring
is anticipated, the armoring analysis is judged satisfactory.
References.

Ackers, P., and W. R. White, "Sediment Transport: New Approach and Analysis,"
Journal, Hydraulics Division, Proceedings, American Society of Civil
Engineers, , vol. 99, No. HY 11, November 1973.
ASCE, "Sedimentation Engineering," American Society of Civil Engineering
Manual No. 54, V. A. Vanoni, editor, 1975.
Blanton, J. 0. III, "Procedures for Monitoring Reservoir Sedimentation,"
Bureau of Reclamation, October 1982.
Borland, W. M., and C. R. Miller, "Distribution of Sediment in Large
Reservoirs," Transactions, American Society of Civil Engineers, vol. 125,
1960.
Brune, Gunnar M., "Trap Efficiency of Reservoirs," Transactions of American
Geophysical Union, vol . 34, No. 3, June 1953.
Bureau of Reclamation, "Computation of 'Z's' for Use in the Modified
Einstein Procedure," Denver, Colorado, June 1966.
Bureau of Reclamation, "Progress Report on Results of Studies on Design
of Stable Channels," Hydraulic Laboratory Report Hyd 352, Denver,
Colorado, June 1952.
Bureau of Reclamation, "Stable Channel Profiles," Hydraulic Laboratory
Report No. Hyd. 325, Denver, Colorado, September 1951.
Bureau of Reclamation, "Step Method for Computing Total Sediment Load by
the Modified Einstein Procedure," Denver, Colorado, July 1955.
Chow, yen Te, editor-in-chief, Handbook of Applied Hydrology, McGraw-Hill,
New York, N.Y., 1964.
Churchill, M. A., Discussion of "Analysis and Use of Reservoir Sedimentation Data," by L. C. Gottschalk, pp. 139-140, Proceedings, Federal
Interagency Sedimentation Conference, Denver, Colorado, January 1948.
Colby, B. R., and C. H. Hembree, "Computations of Total Sediment Discharge,
Niobrara River near Cody, Nebraska," U.S. Geological Survey Water Supply
Paper No. 1357, 1955.
46

Corps of Engineers, "Scour and Deposition on Rivers and Reservoirs HEC-6," User's Manual for a Generalized Computer Program, Exhibit 3,
1977.
Einstein, Hans Albert, "The Bed-load Function for Sediment Transportation
in Open Channel Flows," U.S. Soil Conservation Service Technical
Bulletin No. 1026, September 1950.
Emmett, W. W., "A Field Calibration of the Sediment-Trapping Characteristics
of the Heltey-Smith Bed-load Sampler," U.S. Geological Survey Professional
Paper No. 1139, 1980.
Engelund, F., and E. Hansen, "A Monograph on Sediment Transport in Alluvial
Streams," Teknisk Forlag, Denmark, 1967.
Herbertson, John G., "A Critical Review of Conventional Bed-load Formulae,"
Journal of Hydrology, vol. VIII, No. 1, May 1969.
Hudspeth, C. L., and J. W. Trietsch, "User's Manual - SWE1, Computer
Program for Sediment Distribution," unpublished report by Bureau of
Reclamation, SW Region, rnarillo, Texas, January 1978.
Lane, E. W., "An Estimate of the Magnitude of the Degradation Which Will
Result in the Middle Rio Grande Channel from the Construction of the
Proposed Sediment Storage Basins and Contraction Works," Bureau of
Reclamation, Hydraulic Laboratory Report Hyd 290, 1948.
Lane, E. W., "The Importance of Fluvial Morphology in Hydraulic Engineering," Journal, Hydraulic Division, Proceedings, merican Society of
Civil engineers, vol. 81, July 1955.
Lane, E. W., and V. A. Koelzer, "Density of Sediments Deposited in Reservoirs," In: A Study of Methods Used in Measurement and
Analysis of
Sedimenttoads in Streams, Report No. 9, Interagency Committee on Water
Resources, 1943.
Lara, J. M., "Revision of the Procedure to Compute Sediment Distribution in
Large Reservoirs," Bureau of Reclamation, May 1962.
Lara, J. M., and E. L. Pemberton, "Initial Unit Weight of Deposited
Sediments," Proceedings, Federal Interagency Sedimentation Conference,
1963, Miscellaneous Publication No. 970, pp. 818-845, U.S. Agriculture
Research Service, June 1965.
Lara, J. M., and H. I. Sanders, "The 1963-64 Lake Mead Survey," Bureau of
Reclamation, 1970.
Mavis, F. 1., and L. M. Laushey, "A Reappraisal of the Beginning of BedMovement-Competent Velocity," International Association for Hydraulic
Research, Second meeting, Stockholm, June 1948.
Meyer-Peter, E., and R. Muller, "Formulas for Bed-load Transport,"
Meeting of the International Association for Hydraulic Structure
Research, Second meeting, Stockhom, June 1948.
47

Miller, C. R., "Analysis of Flow-Duration, Sediment-Rating Curve Method of

Computing Sediment Yield," Bureau of Reclamation, Denver, Colorado,
1951.
Miller, Carl R., "Determination of the Unit Weight of Sediment for Use in
Sediment Volume Computations," Bureau of Reclamation, Denver, Colorado,
1953.
Murthy, B. N., "Life of Reservoir," Technical Report No. 19, Central
Board of Irrigation and Power, New Delhi, reprinted September 1980.
Pacific Southwest Interagency Committee, "Factors Affecting Sediment Yield
and Measures for the Reduction of Erosion and Sediment Yield," October
1968.
Pemberton, E. L., "Channel Changes in the Colorado River Below Glen
Canyon Dam," Proceedings, Third Federal Interagency Sedimentation
Conference, 1976.
Pemberton, E. L., "Einstein's Bed-load Function Applied to Channel Design
and Degradation," Chapter 16, In: Sedimentation, Symposium to honor
Prof. H. A. Einstein, H. W. Shii, editor, 1972.
Pemberton, E. L. and J. M. Lara, "Guide for Computing Degradation and
Local Scour," Bureau of Reclamation, Denver, Colorado, October 1982.
Rubey, W. W., "Settling Velocities of Gravel, Sand, and Silt Particles,"
Mierican Journal, Science, 1933.
Sheppard, John R., "Investigation of Meyer-Peter, Muller Bed-load Formulas,"
Bureau of Reclamation, June1960.
Shulits, Samuel, "The Schoklitsch Bed-load Formula," Engineering, vol. 139,
1935.
U.S. Department of Agriculture, "Sediment Deposition in U.S. Reservoirs,
Summary of Data Reported Through 1975," Miscellaneous Publication No.
1362, Agriculture Research Service, February 1978.
U.S. Interagency Sedimentation Project, "A Study of Methods Used in
Measurement and Analysis of Sediment Loads in Streams," Reports No. 1
through 14 and A through W, Subcommittee on Sedimentation, 1940 to
1981.
U.S. Government Handbook, "Chapter 3 - Sediment, National Handbook of
Recommended Methods for Water-Data Acqusisiton," 1978.
Wischmeier, W. H., and D. D. Smith, "Predicting Rainfall Erosion Losses
from Cropland East of the Rocky Mountains," Agric. Handbook 282,
U.S. agriculture Research Service, 1965.
Yang, C. 1., "Incipient Motion and Sediment Transport," ASCE, vol. 99, Hy
No. 10, October 1973.
48

GPO 838-290

Mission of the Bureau of Reclamation

The Bureau of Reclamation of the US. Department of the Interior is
responsible for the development and conservation of the Nation's
water resources in the Western United States.
The Bureau's original purpose "to provide for the reclamation of arid
and semiarid lands In the West" today covers a wide range of interreIa ted functions. These include providing municipal and industrial water
supplies; hydroelectric power generation; irrigation water for agriculture; water quality improvement; flood control,- river navigation;
river regulation and control,- fish and wildlife enhancement; outdoor
recreation; and research on water-related design, construction, materials, atmospheric management, and wind and solar power.
Bureau programs most frequently are the result of close cooperation
with the U.S. Congress, other Federal agencies, States, local governments, academic institutions, water-user organizations, and other
concerned groups.

A free pamphlet is available from the Bureau entitled, 'Publications

for Sale". It describes some of the technical publications currently
available, their cost, and how to order them. The pamphlet can be
obtained upon request from the Bureau of Reclamation, Attn D-922,
P 0 Box 25007, Denver Federal Center, Denver CO 80225-0007.