A woody plant community and tree-cacti associations change with distance to a water source in a dry Chaco forest of Argentina

Abstract

In semiarid regions, livestock is concentrated around water sources generating a piosphere pattern (gradients of woody vegetation degradation with increasing proximity to water). Close to the water source, livestock may affect the composition, structure and regeneration strategies of woody vegetation. We used the proximity from a water source as a proxy of grazing pressure. Our objectives were (1) to compare woody vegetation attributes (richness, diversity, species composition, density and basal area) and ground cover between sites at two distances to a water source: near (higher grazing pressure) and far from the water source (lower grazing pressure), and (2) to quantify and compare cases of spatial association among the columnar cacti Stetsonia coryne (Salm-Dyck) Britton and Rose (Cactaceae), and the dominant tree Bulnesia sarmientoi Lorentz ex Griseb. (Zygophyllaceae). We used a paired design with eight pairs of rectangular plots distributed along a large and representative natural water source. We found lower total species richness, plant density and soil cover near than far from water source, and more cases of spatial associations between the two species studied. Our results show evidence of increased livestock impacts around water sources. However, we found no difference in terms of species composition or basal area at near versus far sites. We conclude that grazing pressure might be changing some attributes of the woody plant community, and that the association of young trees with thorny plants (grazing refuge) could be a regeneration mechanism in this semiarid forest with high grazing pressure.

Full text

A woody plant community and tree-cacti associations
change with distance to a water source in a dry Chaco
forest of Argentina
Carolina B. Trigo
A,B,E
, Andrés Tálamo
A,B
, Mauricio M. Núñez-Regueiro
C
,
Enrique J. Derlindati
A
, Gustavo A. Marás
A
, Alicia H. Barchuk
D
and Antonio Palavecino
A
A
Facultad de Ciencias Naturales, Universidad Nacional de Salta, Avda. Bolivia 5150, Salta, Argentina.
B
Instituto de Bio y Geociencias del NOA (IBIGEO), Universidad Nacional de Salta, Consejo Nacional de
Investigaciones Científicas y Técnicas, Mendoza 2, Salta, Argentina.
C
School of Natural Resources and the Environment and Center for Latin American Studies,
University of Florida, Gainesville, FL 32601, USA.
D
Facultad de Ciencias Agropecuarias, Universidad Nacional de Córdoba, Ing Agr. Felix Aldo Marrone
746 –Ciudad Universitaria, Córdoba, Argentina.
E
Corresponding author. Email: carolinatrigo88@gmail.com
Abstract. In semiarid regions, livestock is concentrated around water sources generating a piosphere pattern (gradients
of woody vegetation degradation with increasing proximity to water). Close to the water source, livestock may affect the
composition, structure and regeneration strategies of woody vegetation. We used the proximity from a water source as a
proxy of grazing pressure. Our objectives were (1) to compare woody vegetation attributes (richness, diversity, species
composition, density and basal area) and ground cover between sites at two distances to a water source: near (higher grazing
pressure) and far from the water source (lower grazing pressure), and (2) to quantify and compare cases of spatial association
among the columnar cacti Stetsonia coryne (Salm-Dyck) Britton and Rose (Cactaceae), and the dominant tree Bulnesia
sarmientoi Lorentz ex Griseb. (Zygophyllaceae). We used a paired design with eight pairs of rectangular plots distributed
along a large and representative natural water source. We found lower total species richness, plant density and soil cover near
than far from water source, and more cases of spatial associations between the two species studied. Our results show evidence
of increased livestock impacts around water sources. However, we found no difference in terms of species composition or
basal area at near versus far sites. We conclude that grazing pressure might be changing some attributes of the woody plant
community, and that the association of young trees with thorny plants (grazing refuge) could be a regeneration mechanism
in this semiarid forest with high grazing pressure.
Additional keywords: grazing pressure, regeneration, water resource.
Received 12 February 2016, accepted 21 November 2016, published online 5 January 2017
Introduction
In many arid and semiarid areas around the world, domestic
animal husbandry is the main subsistence economic activity
among residents of natural ecosystems (Asner et al.2004;
Bedunah and Angerer 2012). In dry forests, such as the dry Chaco
in Argentina, one of the main livestock systems used is ‘extensive
ranching’, where animals move freely in search of forage
and water (Quiroga et al.2009). As a result of trampling
and grazing, livestock grazing may have multiple effects on
vegetation, which need to be recognised and quantified in order
to evaluate the environmental sustainability of this important
economic activity. Understanding the relationship between
livestock and native woody vegetation is critical for future
management schemes that enhance long-term conservation of
native forests.
Grazing pressure is one of the factors affecting vegetation.
Increased grazing pressure on a site may cause higher herbivory
rates and mechanical trampling over seeds and seedlings. As a
result, the vegetation composition may be modified directly by
changes in abundance of plants consumed or trampled over
(Milchunas and Lauenroth 1993; Bisigato et al.2005). Grazing
can also catalyse indirect changes via soil compaction, which
might reduce seed germination and the establishment of
seedlings and saplings (Winkel and Roundy 1991; Mohseni
Saravi et al.2015). However, despite the very clear possible
negative consequences, not all livestock effects are detrimental.
Journal compilation Australian Rangeland Society 2017 www.publish.csiro.au/journals/trj
CSIRO PUBLISHING
The Rangeland Journal
http://dx.doi.org/10.1071/RJ16014

For some species, seed germination is dependent on livestock
dispersal (Winkel and Roundy 1991; Carmona et al.2013), and
high rates of grazing and moderate trampling may lead to an
increase in primary productivity and greater root or buried stem
regrowth rates (Belsky 1986; Harris et al.2016).
In arid and semiarid ecosystems, livestock interacts with
water sources creating a unique pattern, which shapes plant
communities. Here, livestock usually concentrates around water
sources creating areas with bare soil and a gradient of grazing
pressure that decreases with increasing distance to water; this is
known as a piosphere pattern (Lange 1969; Andrew 1988; Todd
2006). Consequently, vegetation attributes such as richness,
diversity and species composition, density, basal area and ground
cover might change (Noy-Meir et al.1989; Fleischner 1994; Pettit
et al.1995; Hernández Vargas et al.2000; Guevara et al.2006;
Todd 2006; Plieninger et al.2011; Macchi and Grau 2012).
However, much of what we know about the piosphere pattern
comes from studies focused on herbaceous communities in desert
or semi-desert environments. Very few studies have considered
the piosphere pattern in woody ecosystems such as subtropical
dry forests. This is a true shortcoming, given that dry forests are
one of the most rapidly disappearing ecosystems of the planet
(Hansen et al.2013). Therefore, studies that help to understand
how woody plants are potentially impacted around water sources
in subtropical dry forests will provide key information in the face
of continued habitat degradation.
In environments with high grazing pressure, woody plants
spatially associate with thorny plants (e.g. cactus), which provide
effective protection against herbivores and facilitate regeneration
in woody plants (i.e. indirect positive interaction; Rebollo et al.
2002,2005). This grazing refuge provided by a nurse plant is
a type of indirect facilitation mechanism (Callaway 1995). As a
result of the protection against herbivores, germination rates may
increase, as well as growth and survival of protected seedlings
(McAuliffe 1986; Bertness and Callaway 1994; Callaway 1995;
Zamora et al.2001). The positive effects of the refuge may vary
with the pressure of grazing, and this association should be
stronger and more common in areas with high grazing pressure
(Rebollo et al.2005; Tálamo et al.2015a). Positive interactions
can increase functional diversity in grazed ecosystems. These
interactions can provide protection to grazing-sensitive species
(Milchunas and Noy-Meir 2002; Rebollo et al.2002; Boughton
et al.2011), which is very important for the restoration of native
vegetation in grazed and browsed environments. Although many
studies exist on this topic, the role of positive interactions or
facilitation as a regeneration mechanism in the context of the
piosphere pattern remains largely unexplored.
In Latin America, extensive livestock in the remaining
forest patches is considered a major cause of environmental
degradation of dry Chaco forest (Morello and Adámoli 1974;
The Nature Conservancy et al.2005; Torres et al.2014).
However, very few empirical studies examine possible effects of
livestock on native woody vegetation in the semiarid Chaco. In
this important ecoregion, water tends to accumulate around
artificial dams, causing a clear piosphere pattern on the vegetation
and the bird community (Macchi and Grau 2012). However, in the
alluvial Chaco of Salta province, in the NW of Argentina, water
is usually found in natural water bodies in areas of ancient
riverbeds, locally known as ‘madrejones’(Adámoli et al.1972).
In this sub-region, different vegetation units exist, characterised
by their dominant tree species, such as ‘Palo Santo’forests
(Bulnesia sarmientoi), locally known as ‘palosantales’(Adámoli
et al.1972). This species is of outmost importance from an
economic and cultural point of view. For hundreds of years, local
indigenous and creole communities have used its wood for several
purposes such as firewood and fiver (Arenas and Suárez 2006;
Suárez 2014). A second dominant species of this unit is the
arboreal cactus known as ‘cardón’(Stetsonia coryne). This is a
species that might deter domestic livestock due to its typical stem
and thorny branches, thus protecting other species associated to it.
This system offers the possibility of conducting an observational
study on the relationship between livestock grazing and woody
vegetation in the context of the piosphere pattern.
Here, our objectives were (1) to compare woody vegetation
attributes and ground cover at two distances to a water source
(close and far away), and (2) to quantify and compare tree-cacti
spatial association between both distances. Furthermore, we test
the hypotheses that (1) livestock pressure alters the attributes of
native woody vegetation, in the context of the piosphere pattern,
and (2) spatial associations among woody and thorny plants under
high livestock pressure can influence the regeneration of native
woody species. We predict that woody vegetation attributes
(richness, diversity, density and basal area) and ground cover
will be higher, and species composition will differ in areas far
away from water source (where livestock pressure is lower) in
comparison to areas close to water (where livestock pressure
is high). We also expect that the proportion of individuals of
Bulnesia sarmientoi spatially associated with Stetsonia coryne
will be higher close to the water source in comparison to sites far
from the water source.
Methods
Study area
The study was conducted on the northern edge of a 5000-ha farm
‘Finca el Paraíso’(23.8138S, 62.7948W), which is located in
the department of San Martín, Salta Province, Argentina. The
environment corresponds to the Dry Chaco ecoregion, which
is composed of a xeric and semi-deciduous forest, with the
overstory dominated by Schinopsis lorentzii (Griseb.) Engl.,
Aspidosperma quebracho-blanco Schltdl., and Bulnesia
sarmientoi Lorentz ex Griseb. (Adámoli et al.1972). This forest is
composed of other shorter trees, like Ziziphus mistol Griseb.,
Tabebuia nodosa (Griseb.) Griseb., Ruprechtia triflora Griseb.,
different species of genus Prosopis and Acacia. Cacti, such as
Opuntia quimilo K. Schum., Stetsonia coryne (Salm-Dyck)
Britton and Rose, and Cereus haenkeanus F.A.C. Weber ex
K.Schum. (Torella and Adámoli 2006; Derlindati et al.2012), are
also present.
The subtropical climate is characterised by strong seasonality,
with a hot and humid season, between the months of November
and April (510 mm) and warm and dry season, between
the months of May and October (65 mm) (Bianchi and Yáñez
1992). Numerous water sources are used by animals in the study
area; the majority of these run out of water during the dry season.
For this reason, we chose one of the most prominent, permanent
and representative water source to perform the study. This water
BThe Rangeland Journal C. B. Trigo et al.

source is a large pond where the water level varies with rainfall (in
summer it expands and in winter it contracts).
Sampling design
To test our hypotheses, we utilised a paired design, where the
design factor (or treatment factor) evaluated was ‘distance from
water source’, with two discrete levels: areas near the water
source (15 m from it, where livestock-use intensity is higher) and
areas far from the water source (200 m from the previous level,
where livestock pressure is lower). The separation distance used
between areas could not be greater than 200–300 m, because
at that distance the vegetation unit changes by edaphic
characteristics, regardless of the existence of livestock (i.e. a
‘quebrachal’, dominated by A. quebracho-blanco, was
immediately beyond a ‘palosantal’, dominated by B. sarmientoi,
see Fig. 1). Thus, by restricting our sampling to the ‘palosantal’
unit we are controlling the possible effect of an undesirable
confounding factor. For the same reason, it was not possible to
work with a gradient by adding an intermediate distance.
Similarly, the first 15 m were not sampled, because the vegetation
(species composition, identity of the species and productivity)
surrounding the water bodies was very different from that of the
‘palosantal’vegetation unit. Camera-trapping confirmed almost
6 times higher frequency of livestock near water sources in
comparison to sites far away from water sources (Poisson
regression parameter b= 5.87; 95% C.I. = 4.92, 6.04). For this,
five camera traps were placed <150 m from the water point and
eight cameras were set in areas >1 km from the water source.
Camera traps were active for 16 days. We built a generalised linear
model using log as the link function and the Poisson distribution
as the assumed probability distribution function of the response
variable (number of livestock individuals in 12 h). We used the
categorical classification of near (<150 m from the water point)
and far (>1 km from the water point) as predictor variables.
The design consisted in eight blocks with two plots each (one
near the water source and another far from it), located parallel
to the longest axis of the water source (Fig. 1). The
representativeness of the sampling was increased spreading the
plot pairs spatially; therefore they were separated by at least
200 m.
Each plot consisted of a rectangle, with two different sizes
depending on the size categories of plants. The size categories
were: (1) ‘saplings’(individuals with height <1 m), (2) ‘adult
shrubs and young trees’(individuals with height >1 m and
diameter at breast height <10 cm) and (3) ‘adult trees’(individuals
Water sources
Plots near
Plots far
“Quebrachal”
“Palosantal”
0400 m
Fig. 1. Map showing the study area, the natural water source and a sketch with the sampling design.
Plots are located within the vegetation unit called ‘Palosantal’. The other vegetation unit
(‘Quebrachal’) was excluded of the sampling because it has a different soil type in comparison to
‘Palosantal’.
Piosphere pattern in Chaco forest The Rangeland Journal C

with height >1 m and diameter at breast height >10 cm). To
sample categories 1 and 2, we used plots of 2 m 50 m. To
sample adult trees (size category 3) we used plots of 10 m 50 m
because of their lower density. In plots of 2 m 50 m we recorded
species richness (number of species of all woody plants), plant
density (size categories 1 and 2), and ground cover (with a rod
of 1 m, we counted the number of touches of the following
categories every 1 m: bryophytes, litter, bare soil, wood, dicots
and Bromeliaceae). In plots of 10 m 50 m we recorded tree
density, diameter at breast height, and number of cases of spatial
associations between S. coryne and B. sarmientoi.Wedefined a
spatial association when a B. sarmientoi individual was found
within 20 cm from the trunk of an individual of S. coryne.
The response variables analysed to meet the first objective
(to compare woody vegetation attributes and ground cover
between two distances to a water source) were: (1) Species
richness, (2) Diversity, (3) Species composition, (4) Density,
(5) Basal area, and (6) Ground cover in different categories. For the
second objective (to quantify and compare tree-cacti association
between both distances) the response variable was the percentage
of spatial associations counted as number of cases of tree-cacti
association in relation to the total number of individuals per plot.
Statistical analyses
Mean species richness, density, basal area, and percentage of
tree-cacti associations was analysed using a paired-sampled t-test
(Zar 1999). All component of diversity were visually compared
by range-abundance curves (Feinsinger 2004). To assess whether
both distances differ in their species composition, we used a
multivariate technique known as blocks multiple response
permutation procedure. This technique allows assessing if the
differences in species composition found between groups are
greater than what might be expected by chance (McCune and
Grace 2002). This analysis was performed using the PC-ORD
software (McCune and Mefford 1999). To determine whether the
different ground cover categories are distributed equally between
both distances, we used the X
2
homogeneity test (Zar 1999).
Results
We found a total of 21 woody plant species, 14 in plots near the
water source (Table 1) and 17 in plots far away from the water
source (Table 1). Both near and far sectors differed in the mean
number of species per plot (|

d| = 3.63 sps; s.d.
(dif)
= 2.33 sps;
T= 4.41; P= 0.0031; Fig. 2). With regard to species diversity,
saplings and adult shrubs and young trees assemblages had grater
evenness in areas far from the water source in comparison to areas
close to the water source (Fig. 3). The species composition was
not different in both distances (multiple response blocked
permutation (MRBP): T=–0.25; P= 0.30). We recorded a total of
243 woody plants (for the three size categories) in plots near the
water source, and a total of 943 woody plants in plots far from the
water source. We found the highest density of saplings in plots
far from water source (|

d| = 7413 saplings/ha; s.d.
(dif)
= 6028
saplings/ha; T=–3.48; P= 0.0103; Fig. 4a). Both distances had a
similar density of adult shrubs and young trees (|

d| = 375 adult
shrubs and young trees/ha; s.d.
(dif)
= 913 adult shrubs and young
trees/ha; T=–1.16; P= 0.2834; Fig. 4a), but more adult trees were
found in plots far from the water source than in plots close to water
Table 1. List of woody plants species presents (‘saplings’,‘adult shrubs and young trees’and ‘adult trees’) in areas near and far from
a water source in the northern edge of the Finca el Paraíso, Salta, Argentina
Scientific name Common name Sapling Far Adult Sapling Near Adult
Adult trees Adult trees
shrubs
and
young
trees
shrubs
and
young
trees
Acacia praecox Garabato X
Aspidosperma sp. Asp sp. X
Aspidosperma quebracho-blanco Quebracho blanco X
Aspidosperma triternatum Quebrachillo X X X
Bromelia sp. Chaguar X
Bulnesia sarmientoi Palo santo X X X X X X
Capparis retusa Sacha poroto X X X X X
Capparis salicifolia Sacha sandía X X X X X
Capparis speciosa Bola verde X X
Castela coccinea Meloncillo X X X
Cercidium praecox Brea X
Mymozyganthus carinatus Iscayante X X X
Opuntia quimilo Quimil X
Prosopis elata Algarrobillo X X X
Prosopis ruscifolia Vinal X X X
Prosopis sericantha Barba de tigre X X
Ruprechtia triflora Duraznillo X X
Schinopsis lorentzii Quebracho colorado X
Stetsonia coryne Cardón X X X X X
Tabebuia nodosa Palo cruz X X X X X
Zizyphus mistol Mistol X X
DThe Rangeland Journal C. B. Trigo et al.

source (|

d| = 167.5 adult trees/ha; s.d.
(dif)
= 173.3 adult trees/ha;
T=–2.73; P= 0.0292; Fig. 4c). On the contrary, both near and far
sectors were similar in terms of basal area (|

d| = 1.44 m
2
/ha;
s.d.
(dif)
=9.79m
2
/ha; T= 0.42; P= 0.6892; Fig. 4d). The proportions
of ground cover categories were different between both distances
(X
2
= 42.97; d.f. = 5; P<0.0001). Near the water sources bare soil
was higher and cover of dicots and Bromeliaceae was lower than
areas far away from water source (Fig. 5). Finally, sectors near
the water source had a higher percentage of spatial associations
(|

d| = 20.56%; s.d.
(dif)
= 17.47%; T= 3.33; P= 0.0126; Fig. 6).
Discussion
The results of this study suggest a potential effect of livestock
pressure on native woody vegetation, confirming both
hypotheses. In comparison to areas far from the water source,
sectors near the water source showed lower total and mean species
richness, lower woody plant density, higher bare soil and a greater
degree of tree-cacti spatial association. Species composition and
basal area were similar in sites far and near the water source.
Our study contributes to fill the gap in relation to the effect
of livestock –that use water sources –on dry forest vegetation.
Also, we provide baseline information on a possible positive
plant–plant interaction, which could be used to design strategies
for forest restoration. Here, we highlight novel results regarding
an unexplored field (relationships between livestock and woody
plants, while simultaneously exploring positive interactions
between plants) for the Chaco forest, which is being lost and
degraded as a consequence of increased farming in this region
(Volante et al.2012; Torres et al.2014; Vallejos et al.2014;
Núñez-Regueiro et al.2015).
This work is similar to other studies on the piosphere pattern in
several arid and semiarid areas of the world. In relation to species
richness, we found a lower number of species in the areas near the
water source, which coincides with similar results reported in
Australian grasslands (Landsberg et al.2003) and in dry Chaco
forest (Macchi and Grau 2012). A similar pattern was found in the
community of shrubs in Nama-Karoo, where species richness
decreased with increasing proximity to the water source, because
of the loss of local palatable species for livestock (Todd 2006).
In contrast, another study in the grasslands of southern Ethiopia
found that the richness of woody vegetation was higher in
intermediate distances to water sources (Tefera et al.2007).
In relation to woody species composition, our results present
a unique pattern, which differs from what was found in
other studies. In general, livestock changes the community
composition of plants, especially affecting palatable species
(Brits et al.2002; Landsberg et al.2003; Todd 2006). Contrary
to these trends, our results indicate that grazing and trampling
apparently did not affect species composition between near
and far areas from the water source. The Chaco forest has a
long history experiencing livestock pressures (Bucher 1987).
Consumption of palatable species over a long period of time could
result in a community composed of grazing-tolerant species
(less palatable) regardless of the distance from the water source.
This could explain why we found a similar community
composition in near and far areas.
0
5
12345678 Tot a l
Near Far
Block
10
Species richness
(no. sps./plot)
15
Fig. 2. Species richness per plot (2m 50 m) in relation to both distances
to the water source (near and far) in different blocks (1–8), and mean values
per plot 1s.e. (Total) in the northern edge of the Finca el Paraíso, Salta,
Argentina.
Near Far
Relative density
0.8 (a)
(b)
(c)
0.6
0.4
Bul sar
Ste cor
Cap ret
Bul sar
Cap ret y Cap sal
Bul sar
Ste cor
Cap ret
Bul sar
Bul sar
Ste cor
Tab nod
Rank
0.2
0
0.8
0.6
0.4
0.2
0
0.8
0.6
0.4
0.2
0
Fig. 3. Rank-abundance curves for (a) saplings, (b) adult shrubs and young
trees and (c) adult trees in relation to both distances to the water source (near
and far) in the northern edge of the Finca el Paraíso, Salta, Argentina. Bul sar:
Bulnesia sarmientoi; Ste cor: Stetsonia coryne; Cap ret: Capparis retusa;
Cap sal: Capparis saliscifolia; Tab nod: Tabebuia nodosa.
Piosphere pattern in Chaco forest The Rangeland Journal E

Livestock could change the abundance of woody plants
at different distance from the water point, even when the
composition of species remains the same. With regard to woody
plants density, our results agree with Landsberg et al.(2003),
who also showed lower densities of woody plants near the water
source. It is noteworthy that our results show far lower adult tree
densities near than far despite the short distance between these
areas. In contrast, Tefera et al.(2007) indicated that the density of
woody plants was similar at different distances from water source,
because the strength of other disturbances was similar along
the sampled gradient. When grazing is the dominant source of
disturbance, higher livestock pressure near a water source may be
driving differences in plant density (Lange 1969). In relation to
basal area (m
2
/ha), we have not found differences between areas
near and far from water source. These results agree with Gandiwa
et al.(2012).They too were not able to find significant differences
in basal area, except after at 2000 m from water source. Our
results may be related to the density and the diameter of
individuals, where although areas near water source had a lower
density, individuals had a higher stem diameter. One explanation
could be related to resources availability and grazing tolerance
(Gao et al.2008). In sites near the water source, with less plant
density, plants may have more resource availability (space, light)
and probably experience lower competition pressures. As a result,
fewer established plants can grow faster than plants located far
away from the water source (higher plant density, less resources
availability, and more competition). In agreement with our results
Macchi and Grau (2012) measured diameter at breast height
and found decreased diameter at breast height at greater distance
from the water source.
12345678 To t a l
25 000 (a)
(b)
(c)
(d)
Sapling density (no./ha)
20 000
15 000
10000
5000
0
25 000
Adult shrub and
young tree density (no./ha)
20 000
15 000
10000
5000
0
800
Adult tree density (no./ha)
600
400
200
0
30
Basal area (m2/ha)
20
10
0
Block
Near Far
Fig. 4. (a) Density of saplings, (b) adult shrubs and young trees, (c) adult
trees, and (d) basal area in relation to both distances to the water source (near
and far) in different blocks (1–8), and mean values per plot 1 s.e. (Total) in
the northern edge of the Finca el Paraíso, Salta, Argentina.
Ground cover (%)
Bromeliaceae
Dicots
Wood
Bare soil
Trash
Bryophytes
100
80
60
40
20
0
Near Far
Fig. 5. Percentage of ground cover in relation to both distances to the water
source (near and far), on the northern edge of the Finca el Paraíso, Salta,
Argentina.
12
60
40
20
0
345678 Total
Block
% of associations
Near Far
Fig. 6. Percentage of spatial associations Stetsonia coryne and Bulnesia
sarmientoi for both distances from water source (near and far) in different
blocks (1–8), and mean values per plot 1 s.e. (Total), on the northern edge of
the Finca el Paraíso, Salta, Argentina.
FThe Rangeland Journal C. B. Trigo et al.

Forest regeneration is a critical aspect in grazing ecology, and
we found higher sapling density in areas far from the water source.
In sectors with high livestock pressure (close to the water source),
regeneration of trees was lower than in areas with low stocking
rates (far from the water source). In the dry Chaco, the use of
a traditional livestock managing system with high stocking
rates may be the main factor that inhibits installation of tree
saplings, because livestock can graze directly and trample over
newly germinated seedlings (Saravia Toledo 1988). In forests,
ungulates tend to consume young plants and therefore may affect
community dynamics, changing the species composition of the
future canopy (Augustine and McNaughton 1998). In addition,
the effect of higher trampling in areas near the water source
can have a negative impact on the survival of saplings of woody
species, favouring the growth of herbaceous vegetation (Brits
et al.2002).
Ground cover also changed, as seen by the proportion of bare
soil that was higher near the water source in comparison to areas
far away from the water source. This is consistent with results
reported by Landsberg et al.(2003) and Guevara et al.(2006),
where the proportion of bare soil tended to increase with the
proximity to the water source. In our study we found differences
in Bromeliaceae (one category of ground cover) between near
and far areas. This differs with the results obtained by Landsberg
et al.(2003) where the shrub cover did not vary significantly with
distance from the water source. In contrast, other studies
demonstrated the dominance of grass cover in highly disturbed
areas near from water source (Todd 2006). Usually, presence of
species resistant to grazing and trampling by livestock tend to
increase near water sources, whereas the amount of more sensitive
species tend to decrease (Guevara et al.2006), which is an aspect
that could be addressed in future studies.
Finally, in this arid environment with domestic livestock
grazing, tree saplings may be protected against the effects of
grazing by associating with thorny plants, which could be a
regeneration strategy for some tree species like B. sarmientoi.
Additionally, this positive effect was more marked in sectors with
higher herbivory pressure. Surprisingly, despite a large literature
on grazing refuges, we have not found studies that analysed
plant–plant interaction in the context of piosphere. However,
compared with studies of grazing refuges, our results agree with
patterns found by others. In grazed ecosystems of Swiss Jura
Mountains (France and Switzerland) unpalatable plant species act
as nurse plants, protecting young trees from livestock grazing
(Smit et al.2006,2007). Olff et al.(1999) also described how the
thorny bushes protected tree seedlings of large herbivores in the
floodplain forests in Western Europe. Rebollo et al.(2002,2005)
observed that the cactus Opuntia polycantha provides refuge
for some plant functional groups, maintaining production of
inflorescences, seeds and sustaining abundance of some species
sensitive to grazing. In the mountainous Chaco forests of
central Córdoba, the regeneration of different tree species in the
presence of exotic livestock can depend on the shelter provided
by shrubs (Torres and Renison 2015,2016). Similarly, in the
semiarid Chaco forests (dominated by Schinopsis lorentzii and
Aspidosperma quebracho-blanco)the regeneration of key tree
species in the presence of exotic livestock might be related to the
presence of shelter provided by thorny shrubs (Tálamo et al.
2015a,2015b). Our study supports the hypothesis that in arid and
semiarid environments with high herbivore pressure, indirect
facilitation would be a common interaction between some plant
species, and its effect would be more important with increasing
grazing pressure (i.e. near the water source).
Here we present results from an observational study (not
manipulative) and therefore the trends we report could be
explained, at least in part, by factors other than livestock pressure.
For example, the humidity gradient, soil differences, or different
disturbance frequencies (e.g. fire) could also vary at different
distance from the water source. Observational studies are
abundant in the piosphere literature and have limited ability to
untangle the effects of the factors mentioned above. However,
such factors will likely have a weaker effect than the potential
effect of livestock. It is unlikely that areas near the water source
have a higher frequency of wildfires because they lack fine
woody debris that could fuel a wildfire (A. Tálamo, pers. obs.).
Furthermore, informal interviews with local people who live in
the vicinity of the water source report that this area never
experienced an intentional or wildfire. The soil and environmental
humidity gradient could be an important factor, albeit only at
short distances from the water source. The satellite image from
our site (Fig. 1) shows a clear vegetation difference on the first
15 m from the water source in comparison with areas far from
the water source. For this reason, our sampling did not consider
sites located <15 m from the water source. Finally, we did not
consider soil variables because the soil type remains constant
throughout all our sites (A. Tálamo, pers. obs.). However, it might
be interesting for future studies to explore if other soil variables,
besides soil type, vary at different distances from the water source.
Thus, we firmly believe that the stronger factor potentially
impacting the observed trends is livestock pressure.
Although our study was conducted in only one water source,
we believe that the same trends might be taking place in other
similar water sources (large and permanent water) of this region.
Satellite imaging can reveal the same piosphere pattern in
many other water sources in the area. Thus, we strongly suggest
conducting similar studies in the area, covering more water
sources with the aim of getting a better understanding of
the relationship between livestock and the vegetation of the
region, considering other response variables such as the
browsing rate.
Implications for management and conservation
The study of the piosphere provides a framework for basic
ecological research, as well as for rangeland management
(Andrew 1988). The dry Chaco forest provides food and fuel
resources to indigenous and peasant families that rely on the
many sources grouped surrounding water where cattle focus,
exerting a piosphere pattern (Macchi and Grau 2012).
Understanding how woody plant communities respond in
forested lands that are managed for livestock will help preserve
these natural habitats and ensure continued provision of vital
ecosystem services. For example, according to our study a
possible management recommendation could be to reduce the
livestock load or to promote the exclusion of sectors near the
water source in order to allow the natural regeneration of some
species. We plan to experimentally test the effectiveness of this
recommendation in future studies.
Piosphere pattern in Chaco forest The Rangeland Journal G

In general, understanding the drivers of spatial associations in
plants at the community level is essential to design restoration and
management strategies for natural environments. For example
on intensively grazed pastures, some managers have increased
survival and growth of seedlings by transplanting seedlings of
trees under natural bush nurses (Smit et al.2006). Moreover,
Rebollo et al.(2002) argued that the ecological role of cactus
O. polycantha should be considered for management practices,
acting as a nurse plant for seedlings by protecting them from
herbivores. Similarly, the cactus S. coryne in our study site could
be used as a possible tool for reforestation and management
in grazed and degraded areas in this dry forest. This can be
particularly important for some woody plant species that are
heavily used, such as B. sarmientoi, which is extensively used
throughout the region (Arenas and Suárez 2006; Suárez 2014).
In cases where these techniques are expensive, we recommend
protecting shrubs and cacti as potential nurses to promote
spontaneous regenerations.
Acknowledgements
This study was supported by the Catholic University of Salta under the project
‘Development of sustainable management plan in Finca el Paraíso.
Experimental Field of the Catholic University of Salta’and by the Research
Council of the National University of Salta (Project 2101/0). N. Cruz helps us
with the data collection, and C. Trucco y Luna N., help us with the
characterisation of woody plants species according to their palatability.
Cecilia Moraga and Joshua Hiller helped us translating the manuscript
into English. We thank all those who assisted in drafting the manuscript
anonymously.
References
Adámoli, J., Neumann, R., Ratier de Colina, A. D., and Morello, J. (1972).
El chaco aluvional salteño. Revista de Investigación Agropecuaria,
Instituto Nacional de Tecnologia Agropecuaria 9, 165–237.
Andrew, M. H. (1988). Grazing impact in relation to livestock watering points.
Trends in Ecology & Evolution 3, 336–339. doi:10.1016/0169-5347(88)
90090-0
Arenas, P., and Suárez, M. (2006). Woods employed by Gran Chaco Indians
to make fire drills. Candollea 62,27–40.
Asner, G., Elmore, A., Olander, L., Martin, R. E., and Harris, A. T. (2004).
Grazing systems, ecosystem responses, and global change. Annual
Review of Environment and Resources 29, 261–299. doi:10.1146/
annurev.energy.29.062403.102142
Augustine, D. J., and McNaughton, S. J. (1998). Ungulate effects on the
functional species composition of plant communities: Herbivore
selectivity and plant tolerance. The Journal of Wildlife Management 62,
1165–1183. doi:10.2307/3801981
Bedunah, D. J., and Angerer, J. P. (2012). Rangeland degradation, poverty,
and conflict: how can rangeland scientists contribute to effective responses
and solutions? Rangeland Ecology and Management 65, 606–612.
doi:10.2111/REM-D-11-00155.1
Belsky, A. (1986). Does herbivory benefit plants? A review of the evidence.
American Naturalist 127, 870–892. doi:10.1086/284531
Bertness, M. D., and Callaway, R. M. (1994). Positive interactions in
communities. Trends in Ecology & Evolution 9, 191–193. doi:10.1016/
0169-5347(94)90088-4
Bianchi, A. R., and Yáñez, C. E. (1992). ‘Las Precipitaciones en el Noroeste
Argentino.’Segunda Edición. pp. 383. (Instituto Nacional de Tecnología
Agropecuaria: Salta, Argentina.)
Bisigato, A. J., Bertiller, M. B., Ares, J. O., and Pazos, G. E. (2005). Effect
of grazing on plant patterns in arid ecosystems of Patagonian Monte.
Ecography 28, 561–572. doi:10.1111/j.2005.0906-7590.04170.x
Boughton, H. E., Quintana-Ascencio, P. F., and Bohlen, P. J. (2011). Refuge
effects of Juncus effusus in grazed, subtropical wetland plant
communities. Plant Ecology 212, 451–460. doi:10.1007/s11258-010-
9836-4
Brits, J., Van Rooyen, M. J., and Van Rooyen, N. (2002). Ecological impact of
large herbivores on the woody vegetation at selected watering points
on the eastern basaltic soils in the Kruger National Park. African Journal
of Ecology 40,53–60. doi:10.1046/j.0141-6707.2001.00344.x
Bucher, E. H. (1987). Herbivory in arid and semi-arid regions of Argentina.
Revista Chilena de Historia Natural 60, 265–273.
Callaway, R. M. (1995). Positive interactions among plants. Botanical Review
61, 306–349. doi:10.1007/BF02912621
Carmona, C. P., Azcarate, F. M., and Peco, B. (2013). Does cattle dung cause
differences between grazing increaser and decreaser germination
response? Acta Oecologica 47,1–7. doi:10.1016/j.actao.2012.11.001
Derlindati, E., Núñez-Regueiro, M., Tálamo, A., Vázquez, D., Marás, G.,
Palavecino, A., Trigo, C., López Morales, R., and Cruz, N. (2012).
Inventario de biodiversidad. Línea de base para la propiedad Finca El
Paraíso, de la Universidad Católica de Salta. In:‘Formulación del Plan
de Manejo Sostenible de Finca El Paraíso, Campo Experimental de la
Universidad Católica de Salta’. (Ed. Universidad Católica de Salta.)
pp. 7–47. (Universidad Católica de Salta: Salta, Argentina.)
Feinsinger, P. (2004). ‘El diseño de estudios de campo para la conservación
de la biodiversidad.’pp. 242. (Editorial FAN: Santa Cruz de la Sierra,
Bolivia.)
Fleischner, T. L. (1994). Ecological costs of livestock grazing in western
North America. Conservation Biology 8, 629–644. doi:10.1046/j.1523-
1739.1994.08030629.x
Gandiwa, E., Tupulu, N., Zisadza-Gandiwa, P., and Muvengwi, J. (2012).
Structure and composition of woody vegetation around permanent-
artificial and ephemeral-natural water points in northern Gonarezhou
National Park, Zimbabwe. Tropical Ecololgy 53, 169–175.
Gao, Y., Wang, D., Ba, L., Bai, Y., and Liu, B. (2008). Interactions
between herbivory and resource availability on grazing tolerance of
Leymus chinensis. Environmental and Experimental Botany 63,
113–122. doi:10.1016/j.envexpbot.2007.10.030
Guevara, J. C., Estevez, O. R., and Stasi, C. R. (2006). Respuesta de la
vegetación en un gradiente de intensidad de pastoreo en Mendoza,
Argentina. Multequina 15,27–36.
Hansen, M. C., Potapov, P. V., Moore, R., Hancher, M., Turubanova, S. A.,
Tyukavina, A., Thau, D., Stehman, S. V., Goetz, S. J., Loveland, T. R.,
Kommareddy, A., Egorov, A., Chini, L., Justice, C. O., and Townshend,
J. R. G. (2013). High-resolution global maps of 21st-century forest cover
change. Science 342, 850–853. doi:10.1126/science.1244693
Harris, R. B., Samberg, L. H., and Yeh, E. T. (2016). Rangeland responses
to pastoralists’grazing management on a Tibetan steppe grassland,
Qinghai Province, China. The Rangeland Journal 38,1–15. doi:10.1071/
RJ15040
Hernández Vargas, G., Velásquez Sánchez, L. R., Carmona Valdovidos, T. F.,
Pineda López, M. R., and Cuevas Guzmán, R. (2000). Efecto de la
ganadería extensiva sobre la regeneración arbórea de los bosques de la
Sierra de Manantlán. Madera y Bosques 6,13–28.
Landsberg, J., James, C. D., Morton, S. R., Muller, W. J., and Stol, J. (2003).
Abundance and composition of plant species along grazing gradients
in Australian rangelands. Journal of Applied Ecology 40, 1008–1024.
doi:10.1111/j.1365-2664.2003.00862.x
Lange, R. T. (1969). The piosphere: sheep track and dung patterns. Journal
of Range Management 22, 396–400. doi:10.2307/3895849
Macchi, L., and Grau, H. R. (2012). Piospheres in the dry Chaco. Contrasting
effects of livestock puestos on forest vegetation and bird communities.
Journal of Arid Environments 87, 176–187. doi:10.1016/j.jaridenv.2012.
06.003
McAuliffe, J. R. (1986). Herbivore-limited establishment of a Sonoran
desert tree Cercidium microphyllum. Ecology 67, 276–280. doi:10.2307/
1938533
HThe Rangeland Journal C. B. Trigo et al.

McCune, B., and Grace, J. B. (2002). ‘Analysis of Ecological Communities.’
MjM Software Design. (Gleneden Beach, OR, USA). Available at: www.
researchgate.net/profile/James_Grace/publication/216769990_Analysis_
of_ecological_communities/links/0a85e5318e69b2ae7f000000.pdf
(accessed 25 October 2014).
McCune, B., and Mefford, M. (1999). ‘PC-ORD. Multivariate Analysis
of Ecological Data, Version 4 for Windows.’(MjM Software Design:
Gleneden Beach, OR, USA.) Available at: www.pcord.com/ (accessed
10 October 2014)
Milchunas, D. G., and Lauenroth, W. K. (1993). Quantitative effects of
grazing on vegetation and soils over a global range of environments.
Ecological Monographs 63, 327–366. doi:10.2307/2937150
Milchunas, D. G., and Noy-Meir, I. (2002). Grazing refuges, external
avoidance of herbivory and plant diversity. Oikos 99, 113–130.
doi:10.1034/j.1600-0706.2002.990112.x
Mohseni Saravi, M., Chaichi, M. R., and Attaeian, B. (2015). Effects of soil
compaction by animal trampling on growth characteristics of Agropyrum
repens (Case Study: Lar Rangeland, Iran). International Journal of
Agriculture & Biology 7, 909–914. doi:doi:1560–8530/2005/07–6–
909–914
Morello, J., and Adámoli, J. (1974). Las grandes unidades de vegetación y
ambiente del Chaco Argentino. Segunda Parte: Vegetación y ambiente de
la Provincia del Chaco. Secretaría de Estado de Agricultura y Ganadería de
la Nación. INTA, Serie Fitogeográfica No. 13, 130 pp. (Buenos Aires,
Argentina.) Available at: www.sidalc.net/cgi-bin/wxis.exe/?IsisScript=
LIBROS.xis&method=post&formato=2&cantidad=1&expresion=mfn=
003833 (accessed 20 January 2015)
Noy-Meir, I., Gutman, M., and Kaplan, Y. (1989). Responses of
Mediterranean grassland plants to grazing and protection. Journal of
Ecology 77, 290–310. doi:10.2307/2260930
Núñez-Regueiro, M. M., Branch, L., Fletcher, R. J. Jr, Marás, G. A.,
Derlindati, E., and Tálamo, A. (2015). Spatial patterns of mammal
occurrence in forest strips surrounded by agricultural crops of the Chaco
region, Argentina. Biological Conservation 187,19–26. doi:10.1016/
j.biocon.2015.04.001
Olff, H., Vera, F. W. M., Bokdam, J., Bakker, E. S., Gleichman, J. M., de
Maeyer, K., and Smit, R. (1999). Shifting mosaics in grazed woodlands
driven by the alternation of plant facilitation and competition. Plant
Biology 1, 127–137. doi:10.1111/j.1438-8677.1999.tb00236.x
Pettit, N. E., Froend, R. H., and Ladd, P. J. (1995). Grazing in remnant
woodland vegetation: changes in species composition and life form
groups. Journal of Vegetation Science 6, 121–130. doi:10.2307/3236263
Plieninger, T., Schaich, H., and Kizos, T. (2011). Land-use legacies in
the forest structure of silvopastoral oak woodlands in the Eastern
Mediterranean. Regional Environmental Change 11, 603–615.
doi:10.1007/s10113-010-0192-7
Quiroga, E. R., Blanco, L. J., and Ferrando, C. A. (2009). A case study
evaluating economic implications of two grazing strategies for cattle
ranches in Northwest Argentina. Rangeland Ecology and Management
62, 435–444. doi:10.2111/08-044.1
Rebollo, S., Milchunas, D. G., Noy-Meir, I., and Chapman, P. L. (2002). The
role of a spiny plant refuge in structuring grazed shortgrass steppe plant
communities. Oikos 98,53–64. doi:10.1034/j.1600-0706.2002.980106.x
Rebollo, S., Milchunas, D. G., and Noy-Meir, I. (2005). Refuge effects of a
cactus in grazed short-grass steppe. Journal of Vegetation Science 16,
85–92. doi:10.1111/j.1654-1103.2005.tb02341.x
Saravia Toledo, C. (1988). Compatibilizacion de manejo de pastizales,bosque
y fauna en los sistema agrosilvopastoriles del Chaco Semiárido. In:
‘Forrajeras y Cultivos Adecuados para la Región Chaqueña Semiárida’.
pp. 99–105. Ed. Oficina Regional de la FAO para América Latina y el
Caribe (Santiago, Chile)
Smit, C., Ouden, J. D., and Muller-Scharer, H. (2006). Unpalatable plants
facilitate tree sapling survival in wooded pastures. Journal of Applied
Ecology 43, 305–312. doi:10.1111/j.1365-2664.2006.01147.x
Smit, C., Vandenberghe, C., Ouden, J. D., and Muller-Scharer, H. (2007).
Nurse plants, tree saplings and grazing pressure: changes in facilitation
along a biotic environmental gradient. Oecologia 152, 265–273.
doi:10.1007/s00442-006-0650-6
Suárez, M. (2014). ‘Etnobotánica wichí del bosque xerófito en el Chaco
semiárido salteño.’(Autores de Argentina: Buenos Aires.)
Tálamo, A., Barchuk, A., Cardozo, S., Trucco, C., Maras, G., and Trigo, C.
(2015a). Direct vs. indirect facilitation (herbivore-mediated) among
woody plants in a semiarid Chaco forest: a spatial association approach.
Austral Ecology 40, 573–580. doi:10.1111/aec.12224
Tálamo, A., Barchuk, A. H., Garibaldi, L. A., Trucco, C. E., Cardozo, S., and
Mohr, F. (2015b). Disentangling the effects of shrubs and herbivores
on tree regeneration in a dry Chaco forest (Argentina). Oecologia 178,
847–854. doi:10.1007/s00442-015-3269-7
Tefera, S., Snyman, H. A., and Smit, G. N. (2007). Rangeland dynamics of
southern Ethiopia: (2). Assessment of woody vegetation structure in
relation to land use and distance from water in semi-arid Borana
rangenlands. Journal of Environmental Management 85, 443–452.
doi:10.1016/j.jenvman.2006.10.008
The Nature Conservancy, Fundación Vida Silvestre Argentina, Fundación
Para El Desarrollo Sustentable Del Chaco, and Wildife Conservation
Society Bolivia (2005). ‘Evaluación Ecorregional del Gran Chaco
Americano/Gran Chaco Americano Ecoregional Assessment.’
(Fundación Vida Silvestre Argentina: Buenos Aires, Argentina.)
Available at: www.nature.org/media/aboutus/dossier_eval_ecorregional_
del_gran_chaco_americano.pdf (accessed 23 December 2014).
Todd, S. W. (2006). Gradients in vegetation cover, structure and species
richness of Nama-Karoo shrublands in relation to distance from watering
points. Journal of Applied Ecology 43, 293–304. doi:10.1111/j.1365-
2664.2006.01154.x
Torella, S. A., and Adámoli, J. (2006). Situación ambiental de la ecoregión del
Chaco seco. In:‘Brown, A., Martínez Ortiz, U., Acerbi, M., and Corcuera,
J. (2005). La situación ambiental Argentina. pp. 75–82. Ed. Fundación
Vida Silvestre Argentina, (Buenos Aires, Argentina) Available at: http://
oab.org.ar/capitulos/cap01.pdf (accessed 18 December 2014).
Torres, R. C., and Renison, D. (2015). Effects of vegetation and herbivores on
regeneration of two tree species in a seasonally dry forest. Journal of Arid
Environments 121,59–66. doi:10.1016/j.jaridenv.2015.05.002
Torres, R. C., and Renison, D. (2016). Indirect facilitation becomes stronger
with seedling age in a degraded seasonally dry forest. Acta Oecologica 70,
138–143. doi:10.1016/j.actao.2015.12.006
Torres, R., Gasparri, N. I., Blendinger, P. G., and Grau, H. R. (2014). Land
use and land-cover effects on regional biodiversity distribution in a
subtropical dry forest: a hierarchical integrative multi-taxa study.
Regional Environmental Change 14, 1549–1561. doi:10.1007/s10113-
014-0604-1
Vallejos, M., Volante, J. M., Mosciaro, M. J., Vale, L. M., Bustamante, M. L.,
and Paruelo, J. M. (2014). Transformation dynamics of the natural cover
in the Dry Chaco ecoregión: A plot level geo-database from 1976 to 2012.
Journal of Arid Environments 123,1–9. doi:10.1016/j.jaridenv.2014.
11.009
Volante, J. N., Alcaraz-Segura, D., Mosciaro, M. J., Viglizzo, E. F., and
Paruelo, J. M. (2012). Ecosystem functional changes associated with
land clearing in NW Argentina. Agriculture, Ecosystems & Environment
154,12–22. doi:10.1016/j.agee.2011.08.012
Winkel, V., and Roundy, B. (1991). Effects of cattle trampling and mechanical
seedbed preparation on grass seedling emergence. Journal of Range
Management 44, 176–180. doi:10.2307/4002318
Zamora, R., Castro, J., Gómez, J. M., García, D., Hodar, J. A., Gómez, L., and
Baraza, E. (2001). El papel de los matorrales en la regeneración forestal.
Quercus 187,41–47.
Zar, J. H. (1999). ‘Biostatical Analysis.’4th edn. pp. 663. (Prentice-Hall Inc.:
Upper Saddle River, NJ.)
Piosphere pattern in Chaco forest The Rangeland Journal I
www.publish.csiro.au/journals/trj