J. Range Manage.

48:517-522 November 1995

Ecosystem changes associated with grazing

intensity on the Punta Ninfas rangelands of
Patagonia, Argentina
ANA M. BEESKOW, NESTOR 0. ELISSALDE AND CESAR M. ROSTAGNO

Authors are range ecologist, CENPAT-CONICET, Puerto Madryn, Argentina; range ecologist, IhKA, Trelew, Argentina:
and soil scientist, CENPAT-CONICET, Puerto Madryn, Argentina.

Abstract dition, and are usually assessedfrom measurements of the quanti-

ty and composition of the vegetation (Stoddart et al. 1975).
According to Wilson and Tupper (1982), range condition should
Changes in the vegetation and soil surface were assessed along be based primarily on soil stability as soil degradation is the most
a grazing intensity gradient on rangelands of the Punta Nias serious manifestation of a decline in range condition. Vegetation
area in southern Argentina. Thirty-two transects were sampled changes and soil degradation processes are closely related and
in areas with different grazing intensity. Bray-Curtis polar ordi- may be site specific. Bosch and Kelhter (1991) emphasized the
nation and simple correlation were used to display changes in importance of understanding the process of rangeland degrada-
commuuity composition and measure association between differ- tion before assessingthe range condition of any area.
ent community attributes. The first axis expressed the changes in The development of a conceptual model of ecosystem changes
species composition along a gradient of grazing intensity. The through statistical procedures may contribute to an understanding
extremes of the gradient were represented by shrub and grass of speciesresponsesto grazing, community changes that could be
steppes. Shrub steppes dominated in heavily grazed areas close to expected during degradation, and the recovery potential of land
permanent water points, while grass steppes dominated in lightly degraded to different degrees. Westoby et al. (1989) proposed
grazed areas in the extremes of the paddocks. A significant nega- that rangeland dynamics can be described by a set of discrete
tive relation (r = -0.81, p-zO.05) between grass and shrub cover “states” of vegetation on a site and a set of “transitions” between
suggested that grasses decreased as shrub increased. Flechilla states. They indicated that management criteria should be used in
(Mpa tenuis Phil.) and fiechilla negra [Piptochaefirrm napostaense selecting and defining the states that occur in a given situation.
@peg.) Hackel ap Stuckert.] were the main decreaser grasses Although the central problem of Patagonian rangelands is the
while quilembai (Chuquiruga avellanedue Cav.) was the main
advance of desertification due to overgrazing (Tnstituto National
shrub invading the grass steppes. Uneroded soil surface condi-
tions decreased, and the size and frequency of crusted and desert de Tecnologia Agropecuatia, 1992), few studieshave been under-
pavement areas and mounds increased with shrub cover. Three taken to assessthe deleterious changes in vegetation composition
states or stages of range degradation were identified along the and soil degradation processesassociatedwith grazing by domes-
gradient of grazing intensity. Grass steppe represented the most tic livestock. The objectives of this study were to define the
desirable state in term of livestock production and soil stability, extent of change along a gradient of grazing intensity, and to
while shrub steppe represented the most degraded and least pro- identify indicators of community deterioration and the different
ductive state. degradation stagesin the Punta Ninfas area of Patagonia.

Key Words: range condition, erosion, shrub invasion,

shrub/grass relationships, soil surface conditions. Materials and Methods
Sheep grazing, one of the main economical activities of
Patagonia, is based almost entirely on the natural vegetation. In Study Area
most of the Patagonian rangelands, grazing appearsto have modi- The Punta Ninfas area (l3eeskow et al. 1991) has an area of
fied the vegetation and accelerated soil degradation processes approximately 250 km2 ( and is located in the north-east of the
(Soriano et al. 1983, Beeskow et al. 1987, Ares et al. 1990). Chubut province in Patagonia, 70 km E of Puerto Madryn
These changes are generally referred to as changes in range con- (43”OO’S, 64”3O’W). It is an undulating plateau dissected by
coastal valleys and interrupted by numerous depressions with
The authors thank MR. Mxin and R Mazzanti for computation assistance and playa lakes. Climate is arid and temperate. Mean annual precipi-
M. Cornejo for help with the drawings. The research was supported by the tation from 1955-1992 was 254 mm, and varied from 81 mm to a
Institute National de Tecnologia Agropecuaria, Trelew, and by the Consejo maximum of 519 mm in 1992, when sampling was conducted.
National de Investigaciones Cientificas y Tecnicas, Puerto hladryn.
Manuscript accepted 5 Feb. 1995. Monthly and annual amounts are extremely variable. Most of the

JOURNAL OF RANGE MANAGEMENT 48(6), November 1995 517

rainfall occurs during the cold season from April to September. lowest similarity value with that site were automatically selected
Mean annual temperature is 12.5”C. The mean annual wind as end points of the first axis along which all other samples were
velocity at 10 m above ground level is 4.6 m 8’ (Barros 1983). ordinated. Two sites located near the center of the first axis, but
Dominant soils are Xerollic Haplargids with some Xerollic with low similarity, were selected as end points to define a sec-
Calciorthids and Typic Torriorthents (Soil Survey Staff, 1975). ond axis of ordination. Regression analyses were used to deter-
Haplargids are shallow with a loamy sand A horizon 5-15 cm mine the relatinships of shrub cover to grass and mound cover.
thick, and a sandy loam B2t horizon 20-30 cm thick. A gravelly
sand to sandy clay alluvium 50 to 80 cm thick forms the soil sub-
stratum. This deposit of Holocene age rests either on the Results and Discussion
“Patagonian Gravel” (Rodados Patagonicos) formation of Plio-
Pleistocene age on the plateaus, or directly on the Tertiary sedi- Sample Ordination and Degradation Gradient
ments in the coastal plains and in the depressions. Sample ordinations along the first axis are given in Fig. 1. The
Two main physiognomies characterize the vegetation of the x-axis expressedthe relative position of the samples along a gra-
Puma Ninfas area, herbaceous and shrub steppes. Perennial, cool- dient of community composition between samples 1 and 27.
season bunch grassesand evergreen shrubs are the dominant bio- Sample 1 was located adjacent to a water point representing
forms of these communities; dwarf shrubs and ephemerals are heavily grazed areas, and sample 27 was located in the extreme
secondary components. Foliar cover varies from 35% to 65%, but of a paddock representing lightly grazed areas. This ordination
may increase substantially in rainy periods when annuals con- fitted the preconceived utilization gradient, represented by a dis-
tribute a high proportion of the total cover. tance to permanent water points, which in turn was considered
The area is located in the ecotone between the Patagonian and the main factor controlling range utilization. However, the fact
Monte phytogeographic provinces. Principal species representing that other temporary water points developed in the paddocks (i.e.,
the first province are the shrubs quilembai (Chuquiraga avel- in the playa lakes) could have altered the development of the pat-
Zunedue Cav.), colapiche [Nzssuuviufueguiunu (Speg.) Cabrera], tern of utilization usually found in large paddocks with 1 water
and neneo [Mulinum spinosum (Cav.) Pers.]. The cool-season point. This, other factors controlling range utilization, and the
grasses flechilla (Stipu tenuis Phil.) and flechilla negra variation in plant community due to environmental heterogeneity,
[Piptochuetium nupostuense (Speg. Hackel ap Stuckert.] are the could explain the distance of some stands to the x axis. The ordi-
main speciesof the Monte province. nation along the y axis separated sites differing mainly in cover
Sheep grazing for wool production was introduced in the area of pasto hilo (Pou lunuginosu Poiret). The high cover of pasto
at the begining of this century. Continuous grazing is practiced hilo, a rhizomatous cool-season grass that grows on loose sandy
extensively in paddocks commonly exceeding 2,500 ha in size soils (Rostagno 1989), in samples close to one of the end points
with a single permanent water point common to 3 or 4 paddocks. might indicate that the y-axis was related to the texture and thick-
Mean stocking rate is 1 sheep 4 ha’. nessof the A horizon.

Sampling Procedure
Sample sites were selected on aerial photographs of scale
1:60,000. Distances to permanent water points were taken into 0 Thm A
account in order to select sample sites with different grazing Hortzon
intensities. In October and November of 1992, 32 transects were I
located at these sites and sampled using the point quadrat method
(Goodall 1952). Transects were 50 m long, and 100 points were
recorded on each. A 1 m-long pin was lowered at 50 cm intervals
60
and foliar cover by species and soil surface condition (desert
pavement, uneroded soil and mound) were recorded. Cover was
calculated as the percentage of direct hits per transect.
Standing crop was determined at 6 of the sampling sites (2 per
state or physiognomic classification, as described later). s-1
Herbaceous plants were clipped at a height of 0.5 cm in ten 0.5 m
x 0.5 m quadrats per site, layed at fixed intervals along the tran- S-27

sects. Samples were dried for 24 hours at 60°C and weighed. 20

Data Analysis
Ordination techniques allow major elements in the distribution
patterns of different locations to be compared and related to inde-
0 10 20 30 40 50 60 70 80 90 100
pendent environmental information (Friedel 1991). This approach
does not assume a climax, but produces classes and orders of
locations that define a degradation gradient according to known Fig. 1. Bray-Curtis polar ordination of 32 cover transects. The x axis
site factors, seasonal conditions, and land management. The polar represents a grazing intensity gradient. S-l is a heavily grazed
area located adjacent to a permanent water point and S-27 is a
ordination method (Cottam et al. 1973) with Jaccard’s index of lightly-grazed area in the opposite extreme of the paddock. The y
similarity as modified by Spatz (Mueller-Dombois and Ellenberg axis represents a gradient in the depth of the soil A horizon.
1974) was used in this study. The sampling site with the lowest
mean similarity with all other samples, and the sample with the

518 JOURNAL OF RANGE MANAGEMENT 46(6), November 1995

 Areas considered to be relatively unchanged or nondegraded steppe. The low palatability of these shrubs has likely been an
were characterized by a grass steppe physiognomy and an homo- important factor in the dominance in heavily grazed areas. The
geneous and stable soil surface condition. Changes in composi- existance of mature plants of quilembai in the grass steppe pro-
tion and abundance of forage species were observed along the moted the formation of bare patches with desert pavements. This
main gradient, apparently reflecting vegetation response to graz- change appeared to have a positive effect on quilembai establish-
ing (Pig. 2). Under continuous grazing a decrease of palatable ment since small individuals of quilembai were found in the
speciesand an increase in unpalatable speciesoccurs. The 2 dom- degraded patches.
inant species in the grass steppe, the cool-season bunchgrasses Palatable annual species, mostly introduced, such as verdin
flechilla and flechilla negra, classified as highly palatable and [Schismus barbatus (L.) Thell.], wild barleys (Hordeum sp.), and
palatable, respectively (Elissalde and Miravalle 1983), were con- the winter annual forb stork’s bill or ahilerillo [G-odium cicutari-
urn (L.) L’Her.], significantly contributed to the total forage pro-
duction in moderately degraded areas and were the main forage
cl Llghlly Heawly 1 source in heavily degraded areas. Annuals were especially abun-
grazed
I dant in years with wet falls and/or winters, as occurred in the year
of sampling.

Community Indicators of Ecosystem Change

: The most important change recorded in the Puma Ninfas range-
6 lands was the transformation of the grass steppe into a shrub
, . . . _- . -
E steppe. The cover of the perennial grasses,principally the domi-
v)
nant grass flechilla, decreased while that of shrubs increased.
Quilembai was the most abundant shrub and the main indicator of
range degradation. A highly negative relationship (r = -0.81,
~~0.05) between grass and shrub cover (Pig. 3) substantiated this
observation. The transformation of former grassland into shrub-
land has been documented elsewhere (Buffrngton and Herbel
1965, Tueller and Blackbum 1974, Wickens and Whyte 1979,
Herbel 1981).
Modifications of spatial plant distributions were associated
with changes in botanical composition. As quilembai and other
shrubs invade the grass steppe, the size of bare patches increases,
I
I
BUS flDP OM flStte RChav C]Nafu OPoli q Pola II
leading to an acceleration of the erosion process and the forma-
tion of mounds (Pig. 4). However, the invading shrub neneo did
not appear to promote the pattern of patches of desert pavement
Fig. 2. Relative abundance of the principal species and soil surface
types along a grazing intensity gradient. US: uneroded soil; DP:
desert pavement; M. mounds; Stte: Sfipa tenuis; Chav:
Chuquirugu avellanedae; Nafu: Nassauviafiegiana; Poli: Poa ligu-
la&; Pola: P. languginosa.

sidered decreasersunder light to moderate grazing. These species

accounted for most of the forage production in nondegraded
areas. Plechilla negra seemed to be either less tolerant to utiliza-
tion or to changes in soil surface conditions as it decreased
abruptly in even slightly grazed areas. At distances from water
representative of moderately grazed areas,the peremdal and high-
ly palatable grasses coiron poa (Pea ligularis Nees ap. Steudel)
and flechilla grande (Stipa longiglumis Philippi) were abundant
and frequently found in the shrub interspaces. Near water, where
grazing was considered to be severe, these species were only
found where protected under shrubs.
As one moved toward water points, from lightly grazed to
heavily grazed areas, the abundance of shrubs increased (Pig. 2).
In the shrub-invaded areas, most of the A horizon had been ~0 5 10 15 20 25 30 35 40 45 50
translocated and accumulated beneath shrubs. Quilembai and
colapiche, dominant shrubs in low areas along drainageways and Shrub cover (%)
around playa lakes with fine textured soils, invaded the adjacent
grass steppe and became dominant in heavily grazed areas.
Fig. 3. Relationship between grass and shrub foliar cover in the
Neneo, a shrub present on sandy soils, also invaded the grass Pnuta Niias rangelands of Patagonia.

JOURNAL OF RANGE MANAGEMENT 48(6), November 1995 519

and mounds as did quilembai. The concept of determinant or of the changes in ecosystem function underlying various forms of
“driver” species, originally used in the context of removing a desertification.
species from an ecosystem (Walker 1992), could well be applied How much of the shrub encroachment and accelerated erosion
to quilembai because its invasion directly affects ecosystem in the Punta Ninfas area can be attributed to overgrazing is debat-
processessuch as erosion, soil infihrability and nutient cycling. able. According to Wright and Bailey (1982), fire in combination
Accelerated erosion was followed by the formation of a desert with other factors would have been required to prevent shrub
pavement in the eroded areas and the formation of mounds invasion in some North American desert grasslands. In those
beneath the shrubs. Thus, soil surface conditions were considered ecosystems, grazing would have reduced fine fuel for fires and
allowed shrubs to invade (Chew and Chew 1965). In the Punta
Ninfas rangelands, a role of fire is suggested by the observation
that recent accidental fires on nearby areas have promoted the
reestablishment of a grass steppe. Without fire, quilembai seems
to replace perennial grasses, at least under moderate to heavy
grazing*

States and Transitions

Cover data were readily categorized into 3 groups (Table I),
each representing a plant physiognomy that was easily recog-
nized in the field. These physiognomic classifications may be
viewed as the stable states of the state-and transition (boxes and
arrows) concept of Westoby et al. (1989) (Pig. 5). Transitions
between states can occur following certain climatic events and
management practices. In the Punta Ninfas area, no references to
climatic conditions conducive to a given state (i.e., shrub estab-
lishment) are available. However, heavy continuous grazing, by
reducing perennial herbaceous vigor, appears to be a prerequisite
for shrub establishment in the grass steppe (transition 1). The 2
“0 5 10 15 20 25 30 35 40 extreme states, I and III, showed little internal variation in their
species composition and structure. They were characterized by a
Qutlembac cover (%) grass steppe with little or no shrub cover, or a shrub or dwarf-
shrub steppe with little grass cover. Soil degradation in state I
Fig. 4. Relationship between the percentage of soil surface covered was neglegible although some mounds had developed. On the
by mounds and folk cover of the shrub quilembai in the Punta contrary, soil degradation in state III was generalized with maxi-
Niias rangelandsof Patagonia.
mum development of desert pavement and crust. Considerable
variation in species composition occurs in the intermediate state.
Grass/shrub steppes were more associated with soils with a thick
good indicators of soil degradation. These mounds seemed to be sandy A horizon, with flechilla, pasto hilo, and desert needlegrass
formed mainly by the accumulation of wind-blown material. In (Stipa speciosa Trin. and Rupr.) as dominant grasses, while
contrast, mounds associated with shrub clumps in the Puerto shrub/grass steppes were more associated with soils with thin A
Madryn area 70 km west of the Punta Ninfa area were formed horizons and coiron poa and flechilla grande as dominant grasses.
mainly by water erosion, with the mounds representing a relict of Many quilembai seedlings were found in bare areas between
an ancient land surface (Rostagno and de1 Valle 1988). Mounds mature plants in areas assigned to states II and III. In contrast,
of colic origin associatedwith shrubs, also termed coppice dunes few or no young shrubs were present in adjacent grass-dominated
(Melton 1940), were reported on soils with texture ranging from
sand to sandy loam (Wood et al. 1978. Gile et al. 1981, HeMeSSy Table 1. Mean foliar cover of principal plant species in each of the 3
et al. 1986). Hodgkinson (1983) concluded that the presence of states identified in the Punta Ninfas rangelands of Patagonia.
coppice dunes associatedwith Cutler Mormon-tea (Ephedra cut- Coefficients of variation appear ia parentheses (II = 6). Data for States
I, II, and III represent the left aad right extremes and the center of the
Zeri Peebles) on a loamy fine sand soil in northeastern Arizona
x axis of Fig. 1, respectively.
indicated a decline in range condition. Soils of the Punta Ninfas
area have a loamy sand A horizon that appears to be very suscep-
State I state II State III
tible to wind erosion. As the A horizon was eroded from the Species
shrub interspace, a bare and stable surface, characterized by a - -‘462(35;- - - - - ;gii; - - ---- - - - - - _
crust and a desert pavement, developed on the top of the sandy Stipa tenuis 3.8 (67)
S. speciosa 3:0 (69) 2:2 (95) 1.3 (84)
clay B2t horizon. Plant establishment on this surface appeared to Poa ligularis 3.2 (76) 5.2 (57) 3.1 (56)
be limited by excessivecompaction. P. lanuginosa 9.6 (62) 6.2 (76) cl (95)
Mounds and shrub interspace areas differ greatly in soil proper- Piptochaetium napostaense 3.3 (33) <I (75) cl (130)
ties (Rostagno et al. 1991), with the mound soils representing bet- Chuguiraga avellanedae 10.5 (34) 22.1 (16) 33.3 (13)
Mulinum spinosum cl (124) 2.5 (103) 1.3 (74)
ter conditions in term of fertility and water balance. Schlesinger Nassauvia fueguiana cl (110) 9.1 (87)
et al. (1990) proposed that the increase in heterogeneity of soil
properties caused by shrub invasion could be a general measure

JOURNAL OF RANGE MANAGEMENT 48(6), November 1995

 I .ooo

300

jO0

DOMINANT SPECIES

Transitions State1 State2 State3

Ti: Reduced herbaceousvigor by continuous heavy grazing.
Climatic conditions favoumble for quilembal seedings Fig. 6. Standing herbaceous biomass at 3 sampling points in each of
establishment. the 3 defined states in the Punta Ninfas rangelands of Patagonia.
T2: Fire or other shrub control technique. Vertical bars are the standard error of the mean.
T3: Shrub driven succession.

T4: Chemical or mechanical treatment for shrub control plus improvement

in hydrological conditions (piMing)
doubtful that recovery from shrub to grass steppe will occur fol-
lowing a decrease in the stocking rate or grazing exclusion, as
T5: The same as in T4 but less probable due to changes in soil surface
conddlons. similar examples indicate (Hennessy et al. 1983). Manipulations
such as brush control, reseeding, and other reclamation practices
seem necessary if rapid return to grass dominance is desired, as
Fig. 5. State (S) and transition (T) diagram for the Punta Ninfas proposed by West et al. (1984) for a similar ecosystem that had
rangelands of Patagonia, including dominant plant species. not recovered after 13 years of grazing exclusion.

areas. The degradation of the soil surface could explain the alter-
Conclusions
nating patches of grass steppe and patches of shrub steppe and be
a principal factor in transition 3.
Fire or other shmb control practices might be used to encour- In the Punta Ninfas rangelands of Argentina, the main vegeta-
age transition 2, the change from the grass/shrub state (II) to tion change associated with a gradient of utilization was the
grass steppe (I). However, the transition from shrub or dwarf- transformation of a grass steppe into a shrub steppe. The invasion
shrub steppe (state III) to either state D[ (transition 4) or state I of shrubs with low forage value into the grass steppe has
(transition 5) would be possible only by mechanical shrub control decreased forage production. The invasion of shrubs, mainly
and improvement in hydrological conditions, such as pitting of quilembai, also affected soil stability by promoting the formation
the soil surface, because infiltration rates are very low (Rostagno of crusts, desert pavement, and mounds. Shrub, desert pavement,
1989). and mound cover can be used as quantitative indicators of the
Availability of forage decreases dramatically from the grass extent of ecosystem degradation. Grass steppe represented the
steppe, through grass/shrub steppe, to shrub steppe (Fig. 6), most- most desirable state in terms of forage production and soil stabili-
ly due to replacement by shrubs. The most common shrubs, ty, while shrub steppe represented the least productive state. The
quilembai, colapiche, and neneo, have very low forage value. In change from a shrub-dominated to grass-dominated community
state III most of the herbaceous forage biomass was represented may be difficult to achieve through grazing management, and
by annuals. In the succession model, condition represents the more drastic manipulations may be required if grass dominance is
position of the vegetation along a continuum from heavily- to be restored.
grazed, early successional, poor condition, to ungmzed, climax,
excellent condition. An important implication of this model for
management is that range condition can be modified along this Literature Cited
continuum by adjusting the stocking rate and providing time for
the vegetation to equilibrate with it (Westoby et al. 1989). Ares, J.O., A.M. Beeskow, M.B. Bertiller, CM. Rostagno, M.P.
Irisarri, J. Anchorena, O.E. Deffosse, and C.A. Merino. 1990.
Although the different states (Fig. 5) might well represent seral Structural and dynamic characteristics of the overgrazed grasslands of
stages with the grass steppe being the climaxic condition, it is northern Patagonia, Argentina, p. 149-175. In: A Breymeyer (ed.),
Managed Grasslands: Regional Studies. Elsevier, Amsterdam.

JOURNAL OF RANGE MANAGEMENT 48(6), November 1995 521

Barros, V.R. 1983. Atlas de1 potential eolico de la Patagonia. M. Canevari, P. Canevari, and W.G. Vasina. 1983. Deserts and
Conhibucion no. 69, Centro National Patagonico. Puerto Madryn, semi-deserts of Patagonia, p. 423-460. In: N.E. West (ed.), Ecosystems
Chubut, Argentina. of the world. Temperate deserts and semi-deserts. Elsevier,
Beeskow, AM., H.F. de1 Valk, and CM. Rostagno. 1987. Los sis- Amsterdam.
temas fisiograficos de la region arida y semiarida de la Provincia de1 Stoddart, L.A., A.D. Smith, and T.W. Box. 1975. Range management.
Chubut. Secretaria de Ciencia y Tecnol. Bariloche, Argentina. Third ed. McGraw-Hill, New York, N.Y.
Beeskow, AM., N. Eiissaide, and CM. Rostagno. 1991. Delimitation Tueiier, P.T., and W.H. Biackburn. 1974. Condition and trend of the
y caracterizacion de 10s sitios de pastoreo de1 sector noreste de la big sagebrushlneedleandthread habitat type in Nevada. J. Range
provincia de Chubut. X Reunion National para el Estudio de las Manage. 27:36-40.
Regiones Aridas y Semiaridas. Bahia Blanca, Argentina. Walker, B.H. 1992. Biological and ecological redundancy. Conserv.
Bosch, O&H, and K. Keihrer. 1991. The use of degradation gradient Biol. 6:18-23.
for the ecological interpretation of condition assessments in the west- West, N.E., F.D. Provenza, P.S. Johnson, and M.K. Owens. 1984.
em grassland biome of southern Africa. J. Arid Environ. 21:21-29. Vegetation changes after 13 years of livestock grazing on sagebrush
Buffmgton, L.C., and C.H. Herbel. 1965. Vegetation changes on a semidesert in west central Utah. J. Range Manage. 37:262-264.
semidesert grassland range from 1858 to 1963. Ecol. Monog. Westoby, M., B. Walker, and I. Noy-Meyer. 1989. Gpporhmistic man-
35:139-164. agement for rangelands not at equilibrium. J. Range Manage.
Cottam, G., F.G. Goff, and R.H. Whittaker. 1973. Wisconsin compar- 42~266-274.
ative ordination, p. 195-221. In: R.H. Whitaker (ed.), Ordination and Wickens, G.E., and L.P. White. 1979. Land use in the southern margius
classification of communities. Junk, The Hague. of the Sahara, p. 205-242. In: B.H. Walker (ed.), Management of
Chew, R.M., and A.E. Chew. 1965. The primary productivity of a Semiarid Ecosystems. Elsevier, Amsterdam.
desert-shrub (Larrea tridentata) community. Ecol. Monog. Wiison, AD., and G.J. Tupper. 1982. Concepts and factors applicable
35:355-375. to the measurements of range condition. J. Range Manage.
E&aide, N.O., and H.R. Miravaiie. 1983. Evaluation de 10s campos 35:6&l-689.
de pastoreo de Peninsula Valdes. Contribution no. 70, Central Wood, M.K., W.H. Blackbum, R.E. Eckert, and F.F. Peterson. 1978.
National Patagonico, Puerto Madryn, Chubut, Argentina. Interrelations of the physical properties of coppices dunes in determin-
Friedel, M.H. 1991. Range condition assessment and the concept of ing range trend in northeastern Arizona. J. Range Manage.
thresholds: A viewpoint J. Range Manage. 44:422-426. 313189-192.
Gile, L.H., J.W. Hawley, and R.B. Grossman. 1981. Soils and geomor- Wright, H.A., and A.W. Bailey. 1982. Fire Ecology. John Wiley and
phology in the basin and range area of southern New Mexico. Sons, N.Y.
Guidebook to the Desert Project. New Mexico Bureau of Mineral
Resources. Memoir 39.
Goodall, D.W. 1952. Some considerations in the use of point quadrats
for the analysis of vegetation. Aust. J. Sci. Res., Series B:l-ll.
Hennessey, J.T., R.P. Gibbens, J.M. Tromble, and M. Cardenas.
1983. Vegetation changes from 1935 to 1980 in mesquite dunelands
and former grasslands of southern New Mexico. J. Range Manage.
36:370-374.
Hennessey, J.T., B. Kies, R.P. Gibbens, and J.M. Tremble. 1986. Soil
sorting by forty-five years of wind erosion on a southern New Mexico
range. Soil Sci. Sot. Amer. J. 50:391-394.
Herbel, C.H. 1981. Manipulative range improvements. Principles and
practices, p. 557-579. In: Manassah, J.T., and E.J. Briskey (eds.),
Advances in food-producing systems for arid and semiarid lands.
Academic Press, N.Y.
Hodgkinson, H.S. 1983. Relationship between Cutler Mormon Tea
[Ephedru cutleri] and coppice dunes in determining range trend in
northeastern Arizona. J. Range Manage. 36:375-377.
Institnto National le Tecnologia Agropecuaria. 1992. Proyecto de pre-
vencion y control de la desertificacion en la Patagonia. Secret, de. Agr.
Ganaderia y Pesca (SAGYP), INTA. Buenos Aires.
hlelton, F.A. 1940. A tentative classification of sand dunes: its applica-
tion to dune history in the southern High Plains. J. Geology
48:113-174.
Mueller-Dumbois, D., and H. Eiienberg. 1974. Aims and methods of
vegetation ecology. John Wiley and Sons, N.Y.
Rostagno, C.M. 1989. Infiltration and sediment production as affected
by soil surface conditions in a shrubland of Patagonia, Argentina. J.
Range Manage. 42~382-385.
Rostagno, CM., and H.F. de1 Valle. 1988. Mounds associated with
shrubs in aridic soils of northeastern Patagonia: characteristics and
probable genesis. Catena 15:347-359.
Rostagno, CM., H.F. de1 Vaiie, and L. Videla. 1991. The influence of
shrubs on some chemical and physical properties of an aridic soil in
north-eastern Patagonia, Argentina. J. Arid Environ. 20: 179-I 88.
Schlesinger, W.H., S.F. Reynolds, O.L. Cunningham, L.F. Huenneke,
W.hl. Jarrel, R.A. Virginia, and W.G. Whifford. 1990. Biological
feedbacks in global desertification. Sci. 247:1043-1048.
Soil Survey Staff. 1975. Soil taxonomy: a basic system of soil classifica-
tion for making and interpreting soil surveys. USDA Agr. Handb. 436.
Soriano, A., W. Voikheimer, H. Waiter, E.O. Box, A.A. Marcoiin,
J.A. Vaierini, C.P. Movia, RJ. Leon, J.M. GaBardo, M. Rumboli,

522 JOURNAL OF RANGE MANAGEMENT 48(6), November 1995