Biological Conservation xxx (2010) xxx–xxx

Contents lists available at ScienceDirect

Biological Conservation
journal homepage: www.elsevier.com/locate/biocon

Plant community functional shifts in response to livestock grazing in intermountain
depressional wetlands in British Columbia, Canada
W. Marc Jones a,⇑, Lauchlan H. Fraser b, P. Jefferson Curtis a
a
b

Department of Chemistry, Earth and Environmental Sciences, University of British Columbia Okanagan, Kelowna, British Columbia, Canada V1V 1V7
Department of Natural Resource Science, Thompson Rivers University, Kamloops, British Columbia, Canada V2C 5N3

a r t i c l e

i n f o

Article history:
Received 9 November 2009
Received in revised form 2 October 2010
Accepted 2 October 2010
Available online xxxx
Keywords:
Depressional wetlands
Disturbance gradient
Functional groups
Livestock grazing
Marsh
Wet meadow

a b s t r a c t
Wetlands are ecologically and economically important ecosystems with high conservation value.
Although wetland vegetation is strongly determined by abiotic factors, grazing disturbance may also
be an important influence on this community. We evaluated the effects of livestock grazing on wetland
vegetation in marsh and wet meadow zones in intermountain depressional wetlands in the southern
interior of British Columbia, Canada. We sampled marshes and wet meadows in 36 wetlands along a
grazing intensity gradient ranging from fully fenced and ungrazed wetlands to unfenced wetlands heavily
grazed by livestock. The amount of bare ground was used as a surrogate measure of the intensity of livestock grazing. Vegetation community structure and composition was strongly associated with grazing
intensity. Increased livestock grazing favored shorter-lived and smaller plants; conversely, the frequency
of tall and rhizomatous species, which constitute the dominant plant species in these systems, declined
with higher livestock use. The effects of grazing were more pronounced in the marsh than in the wet
meadow. Associations between species richness and grazing differed between zones: native and exotic
species richness showed a unimodal response in the marsh while in the wet meadow exotic richness
increased and native richness showed no response. The relationship between exotic frequency and
grazing was also inconsistent between zones, with a negative association in the marsh and a positive
but weak association in the wet meadow. Grazing-related changes in vegetation along the grazing intensity gradient were substantial and may affect the habitat value of these wetlands for dependent wildlife
such as breeding waterfowl.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction
Half of the world’s land area is rangeland (Havstad, 2008), and
grazing by domestic livestock is a substantial and pervasive land
use throughout the world (Milchunas and Lauenroth, 1993; Diaz
et al., 2007). The effects of livestock grazing on rangelands and
riparian areas have been well studied. Livestock selectively remove
biomass, trample vegetation, disturb and compact soil, and add
nutrients, and grazing can significantly alter the structure, productivity, diversity, and plant competitive interactions of grazed
ecosystems (Martin and Chambers, 2001; Kauffman et al., 2004;
Frank, 2005; Manier and Hobbs, 2007).
Freshwater depressional wetlands often occur within a rangeland
matrix, such as in the intermountain west and prairie pothole regions of North America. Freshwater wetlands provide important
ecological and economic services, such as wildlife habitat, regulation of water regimes, filtration of polluted water, and production
of forage (Mitsch and Gosselink, 2000) and are critically important
⇑ Corresponding author. Tel.: +1 250 852 6283; fax: +1 250 828 5450.
E-mail address: wmjones547@hotmail.com (W.M. Jones).

for breeding waterfowl and amphibians (Johnson et al., 2005; Piha
et al., 2007). However, our understanding of the effects of livestock
grazing on freshwater wetlands is limited, especially for the study
area (Austin et al., 2007). A better understanding of how grazing
affects vegetation (and thus wetland habitat) is an important step
in determining tradeoffs between the ecological and economic
services of wetlands and maintaining their biodiversity and
productivity.
Several models have been proposed that predict vegetation
community response to grazing with regard to simple functional
attributes, including the generalized model (Milchunas et al.,
1988; Milchunas and Lauenroth, 1993), competitor-stress tolerator-ruderal (CSR) model (Grime, 1977, 2001), and the range succession model (Dyksterhuis, 1949). These models predict that grazing
will favor plant species with an annual life history, short canopy
height, and (in the cases of the generalized and range succession
models) a rhizomatous or stoloniferous root architecture. Empirical
studies have largely validated these predictions in a variety of
rangeland types (McIntyre et al., 1995; Diaz et al., 2001, 2007).
Grazers preferentially forage tall, fast-growing, competitively
dominant species (Fraser and Grime, 1999). Reduction of the

0006-3207/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biocon.2010.10.005

Please cite this article in press as: Jones, W.M., et al. Plant community functional shifts in response to livestock grazing in intermountain depressional wetlands in British Columbia, Canada. Biol. Conserv. (2010), doi:10.1016/j.biocon.2010.10.005

2

W.M. Jones et al. / Biological Conservation xxx (2010) xxx–xxx

dominant species through grazing can reduce the competitive
advantage of these species and benefit shorter-statured, slowgrowing species as well as increase species richness and community diversity (Gough and Grace, 1998; Fraser and Grime, 1999).
Grazing can also increase gaps in vegetation and reduce litter accumulation, and moderate grazing can thereby provide greater establishment opportunities for species with life histories, architectures,
or dispersal strategies that allow them to exploit these gaps (van
der Valk, 1986; Humphrey and Patterson, 2000; Bullock et al.,
2001; Dorrough et al., 2004).
We examined the effects of livestock grazing on intermountain
depressional wetlands to see if predictions of grazing models
developed for rangelands could be extended to highly productive
wetland systems that are subject to flooding stress not encountered in upland systems. We examined differences in the wetland
vegetation community along a grazing intensity gradient and between wetland zones subject to different flooding regimes. We
evaluated vegetation biomass and litter, species composition and
richness, and frequency of functional groups.
We classified species into groups based on readily observable
functional traits such as life history, architecture, height, growth
form, and origin. Study wetlands are currently subject to pervasive
grazing pressure but have a short evolutionary history of grazing
(Daubenmire, 1970; Mack and Thompson, 1982) and are therefore
likely to show a strong functional response to grazing (Milchunas
and Lauenroth, 1993; Diaz et al., 2007). We expected annuals, short
and rhizomatous species, and forbs to increase with increasing
grazing intensity, perennials, medium, tall, and non-rhizomatous
species to decrease with increasing grazing intensity, and the response of graminoids to be neutral.
Following Kantrud et al. (1989), we classified wetland vegetation into two zones: marsh (typically inundated for much of the
growing season and characterized by dense stands of tall emergent
vegetation) and wet meadow (temporarily flooded during the
growing season and dominated by smaller graminoids). Given
the abundance of emergent vegetation, competition for light may
be more intense in the marsh, and we expected the effects of grazing to be more pronounced there, which would be consistent with
predictions of the CSR and generalized models. The marsh zone is
subject to greater flooding stress but is less hydrologically variable
(Kantrud and Stewart, 1977) and less species rich (Keddy, 2000)
than the wet meadow.

2. Methods
2.1. Study Sites
We sampled 36 depressional wetlands from the southern interior of British Columbia, Canada, near the city of Kamloops.
Twenty-five wetlands were sampled in June and July of 2006 and
the remaining eleven were sampled in July and early August of
2007. We selected sites from wetlands that are part of a long-term
(20 years) waterfowl breeding pair survey conducted by Ducks
Unlimited Canada and the Canadian Wildlife Service. Sample
wetlands were either semipermanently or permanently flooded
(except for two sites that were seasonally flooded) and ranged
from oligo- to mesosaline. Sites encompassed a grazing disturbance gradient from fully fenced and ungrazed to unfenced and
heavily grazed by livestock (pasture-level stocking densities at
grazed sites ranged from 0.4 to 2 cattle/ha). Fenced sites included
one fully fenced and three partially fenced wetlands that have been
managed as cattle exclosures for approximately 15 years. Common
wet meadow species include Carex praegracilis (field sedge),
Hordeum jubatum (foxtail barley), and Juncus balticus (Baltic rush);
typical marsh species include Bolboschoenus maritimus (seacoast

bulrush), Schoenoplectus acutus and Schoenoplectus tabernaemontani
(hard- and soft-stemmed bulrush), and Puccinellia nuttalliana
(Nuttall’s alkaligrass).
2.2. Wetland vegetation
We sampled vegetation along six transects systematically
placed around the wetland equidistant from one another. We
placed transects parallel to the hydrological gradient and ran from
the aquatic-emergent vegetation boundary in the marsh zone to
the wetland-terrestrial vegetation boundary in the wet meadow
zone. Transect lengths varied depending on the width of these
zones, from approximately 10 m to 100 m. We delimited wetland
zones based on the dominant vegetation of each wetland using
species-hydrology associations summarized in Kantrud et al.
(1989). We measured plant species frequency of occurrence in 60
0.5-m2 quadrats placed equidistant along transects (30 in each
zone). Species nomenclature followed Douglas et al. (1998–2002).
We separately sampled aboveground biomass and litter from
the middle of the wet meadow and marsh zones at each vegetation
transect (six samples per zone per site) using 0.25-m2 quadrats.
We clipped aboveground biomass to a height of 1 cm. Biomass
and litter samples were oven-dried at 65 °C for at least 72 h to a
constant dry mass.
2.3. Plant traits
We evaluated nine non-mutually exclusive functionally defined
groups. These were based on species’ life history (annual or biennial, perennial), root architecture (rhizomatous, non-rhizomatous),
canopy height (short, medium, tall), and growth form (forb,
graminoid). We also evaluated species of exotic origin. The rhizomatous category included species with rhizomes and stolons as
well as creeping species that develop nodal roots. Breaks for the
height category were <40 cm (short), between 40 and 80 cm (medium), and >80 cm (tall) (Diaz et al., 2001; Pykälä, 2004). All species
classifications were based on Douglas et al. (1998-2002). To measure functional group similarity, we calculated the proportion of
species shared between each pair of groups using the Jaccard coefficient of community.
2.4. Grazing intensity
We measured bare ground and soil bulk density at each wetland. For each wetland zone, bare ground was calculated as the
mean number of vegetation quadrat corners that intersected bare
ground (n = 30/zone), while soil bulk density was calculated for
each wetland (n = 6/site). Where quadrats were inundated, a rod
projected vertically from the quadrat corner to the substrate was
used. The site-level measure of bare ground was highly correlated
with soil bulk density (r = 0.83), and we used bare ground as
our indicator of grazing intensity within each wetland zone
(Greenwood and McKenzie, 2001; Hendricks et al., 2005; Manier
and Hobbs, 2007). Although bare ground is used as a surrogate
measure of grazing intensity, it can also be indicative of mechanical disturbance through hoof shear and trailing (M. Jones, personal
observation). To improve the interpretability of this measure, it
was rescaled so that it ranged from 0 to 100.
2.5. Statistical analysis
We used a Wilcoxon rank sum test to compare whether bare
ground was different between the marsh and wet meadow zones
and whether aboveground biomass was different between the
zones at wetlands with no measurable grazing pressure (bare
ground = 0). We used linear regressions to model the relationships

Please cite this article in press as: Jones, W.M., et al. Plant community functional shifts in response to livestock grazing in intermountain depressional wetlands in British Columbia, Canada. Biol. Conserv. (2010), doi:10.1016/j.biocon.2010.10.005

W.M. Jones et al. / Biological Conservation xxx (2010) xxx–xxx

of marsh- and wet meadow-level mean biomass and litter to grazing
intensity (n = 35 for marsh, n = 36 for wet meadow). Assumptions
of linear regression (normal distribution of errors, constant variance) and the effect of influential data were checked by examining
plots of residuals against fitted values, residuals against each datum’s leverage, and quantile–quantile plots of residuals against
quantiles.
We used an indirect ordination technique, nonmetric multidimensional scaling (NMDS) to examine associations between plant
community composition and grazing intensity. NMDS was calculated based on a Bray-Curtis dissimilarity matrix of vegetation frequency data aggregated to the level of wetland zone using the
metaMDS routine in the vegan package for the R statistical environment (Oksanen et al., 2008; R Development Core Team, 2008).
This routine chooses multiple random starts with Procrustes rotations to find a global (vs. local) NMDS solution and uses principal
components analysis to rotate the final solution so that the first
axis explains the greatest variation in the data. NMDS was run
for k = 2–4 dimensions. The optimal number of dimensions was selected by examining a scree plot of stress vs. dimensionality and
choosing the dimension number at the main inflection point (i.e.,
where the reduction in stress accompanying increased dimensionality was minimal).
Associations between plant functional groups, richness of native and exotic species, and grazing intensity were evaluated with
generalized linear models in R. We modeled the probability of
occurrence of each functional group in each wetland zone as well
as overall species richness in each wetland zone conditional to
the amount of bare ground (n = 35 for marsh, n = 36 for wet meadow). We used quasi-binomial models with a logit link function
to evaluate functional group response and quasi-Poisson models
with a log link function to evaluate the response of species richness. Diagnostic plots of the models similar to the ones described
previously were examined to verify model assumptions of distribution and constant variance. We included a quadratic term for
bare ground in models to account for a unimodal response to grazing (Hobbs et al., 1996; Fensham et al., 1999; Grime, 2001). We
evaluated the importance of the quadratic term using F-ratio tests
and retained it in the models only where it was significant at
a = 0.05.

3. Results
Bare ground was not different between the marsh and wet meadow (W = 608, p = 0.97). Mean values of the bare ground index
were 34.2 ± 34.4 (SD) in the marsh and 31.4 ± 28.8 (SD) in the
wet meadow. Aboveground vegetation biomass and litter were significantly negatively associated with bare ground in both the
marsh (biomass: F1,33 = 44.7, R2 = 0.58, p < 0.0001, litter: F1,33 =
16.0, R2 = 0.33, p = 0.0003) and wet meadow zone (biomass:
F1,34 = 22.9, R2 = 0.40, p < 0.0001, litter: F1,34 = 15.5, R2 = 0.31, p =
0.0004). In the marsh zone, a 10 unit increase in the bare ground
index was associated with a 42.2 ± 12.9 g/m2 (95% CI, n = 35) decrease in aboveground biomass and a 21.9 ± 11.2 g/m2 (95% CI,
n = 35) decrease in litter. In the wet meadow zone, a 10 unit increase in bare ground was associated with a 30.8 ± 13.1 g/m2
(95% CI, n = 36) decrease in biomass and a 23.2 ± 12.0 g/m2 (95%
CI, n = 36) decrease in litter. At sites with no measured livestock
grazing (n = 7), mean aboveground biomass was higher in the
marsh (507 ± 185 g/m2 (SD)) than in the wet meadow (348 ±
68 g/m2 (SD)), although this difference was not statistically significant (W = 37, p = 0.13).
Ordination results for wetland vegetation showed a high level
of separation between wetland zones as well as clustering along
the grazing intensity gradient (Fig. 1). The greatest variation

3

Fig. 1. Unconstrained ordination results for wetland vegetation (stress-based
R2 = 0.96). Marshes and wet meadows are represented by s’s and D’s, respectively.
Symbol size is scaled by grazing intensity with larger symbols representing greater
amounts of bare ground. Vector label refers to bare ground (bare); vector length
corresponds to the strength of correlation.

among samples was explained by the first axis (NMDS 1) and
appeared to reflect wetland hydrology given the strong separation
between marsh and wet meadow zones along this axis. The second
axis (NMDS 2) was associated with grazing intensity, which was
strongly correlated with bare ground (r = 0.73). Although grazing
appeared to be an important factor in structuring the vegetation
community in both the marsh and wet meadow, this pattern is
weaker for the marsh community, which was less clustered in ordination space, than for the wet meadow.
Although functional groups were not mutually exclusive in all
cases, the proportion of shared species exceeded 50% for only
two group comparisons. Perennial and rhizomatous groups shared
69% and 60% of species in the marsh and wet meadow zones,
respectively, and annual and non-rhizomatous groups shared 61%
and 51% of species.
The frequency of occurrence of most functional trait-based
groups evaluated was strongly associated with the amount of bare
ground, although the deviance explained (analogous to R2 for linear models) for some functional groups was not high (Table 1).
Many groups had a consistent response to bare ground in both
the marsh and wet meadow zones. The predicted probability of
encountering a non-rhizomatous or short species in a quadrat were
positively associated with bare ground in both zones, rhizomatous
and tall species were negatively associated with bare ground in
both zones, and graminoids and perennial species had no statistically significant association with bare ground in either zone
(Fig. 2). Forbs and annual species showed a unimodal response to
bare ground in the wet meadow, with frequencies of occurrence
highest at intermediate levels of bare ground, but were positively
associated with bare ground in the marsh (Fig. 2). In contrast, plant
species of medium height and exotic species were positively associated with bare ground in the marsh zone but negatively associated in the wet meadow zone (Fig. 2). The richness of both
native and exotic species exhibited a unimodal response in the
marsh zone, with the greatest numbers of species occurring at
intermediate levels of bare ground. Conversely, in the wet meadow
zone, the richness of exotic species was positively associated with
bare ground but there was no statistically significant association
between native richness and bare ground (Fig. 2). It should be
noted that for both forbs and exotic species in the wet meadow,
the site with the highest measure of bare ground exerted a strong

Please cite this article in press as: Jones, W.M., et al. Plant community functional shifts in response to livestock grazing in intermountain depressional wetlands in British Columbia, Canada. Biol. Conserv. (2010), doi:10.1016/j.biocon.2010.10.005

4

W.M. Jones et al. / Biological Conservation xxx (2010) xxx–xxx
Table 1
Deviance explained, df, and significance of bare ground for generalized linear models describing the association of plant functional group
frequency and species richness with bare ground in the marsh and wet meadow zones (n = 35 for marsh, n = 36 for wet meadow).
Functional group/richness

Zonea

Deviance

Forb

M
W

0.31
0.20

1.34
2.34

3.72
2.24

<0.001
0.032

Graminoid

M
W

0.12
0.24

1.34
1.35

1.68
1.88

0.100
0.069

Annual

M
W

0.49
0.33

1.34
2.,33

5.19
3.84

<0.001
<0.001

Perennial

M
W

0.12
0.24

1.34
1.35

1.50
1.50

0.140
0.142

Rhizomatous

M
W

0.39
0.59

1.34
1.35

3.74
5.10

<0.001
<0.001

Non-rhizomatous

M
W

0.53
0.52

1.34
1.35

4.93
4.99

<0.001
<0.001

Short

M
W

0.49
0.44

1.34
1.35

5.03
4.22

<0.001
<0.001

Medium

M
W

0.32
0.57

1.34
1.35

3.73
5.17

<0.001
<0.001

Tall

M
W

0.50
0.21

1.34
1.35

4.68
3.07

<0.001
0.004

Exotic

M
W

0.12
0.12

1.34
1.35

2.19
2.08

0.036
0.045

Exotic richness

M
W

0.22
0.17

2.33
1.35

3.08
2.63

Native richness

M
W

0.24
0.06

2.33
1.35

2.73
1.43

df

explained

a
b

Bare groundb

Bare ground
t

P

t

P
2.57

0.015

3.18

0.003

0.004
0.013

2.73

0.010

0.010
0.160

2.22

0.034

M = marsh, W = wet meadow (bare ground).
Is the quadratic term for bare ground.

influence on the model fit. When this site was removed from the
analysis, the relationships between the functional groups and bare
ground were no longer statistically significant. Given that we have
no biological basis for the removal of this site and that it represents
the most extreme end of the grazing intensity gradient, we have
retained it in the analysis.

4. Discussion
We found that livestock grazing strongly affects the richness,
structure, and composition of wetland vegetation communities in
intermountain depressional wetlands. Community structure was
more variable in the marsh zone compared to the wet meadow
regardless of grazing intensity. Our results were largely consistent
with model predictions and previous studies (Dyksterhuis, 1949;
Grime, 1977; Milchunas et al., 1988; Diaz et al., 2007). The frequency of annuals, forbs, and short species increased with grazing
intensity, although the frequency of annuals and forbs in the wet
meadow zone decreased at high levels of bare ground. Tall and
medium height species decreased with grazing intensity (the latter
only in the wet meadow), and graminoid frequency was not associated with grazing intensity. Contrary to our expectations and
previously developed models, perennial species showed no response to grazing intensity, rhizomatous species were negatively
associated with grazing, and non-rhizomatous were strongly positively associated with grazing.
Grazing-vegetation models were largely developed for grasslands, and although the generalized and CSR models (Milchunas
et al., 1988; Milchunas and Lauenroth, 1993; Grime, 2001) as well
as a global review of previous research (Diaz et al., 2007) have
incorporated productivity into model predictions, these models

have not been applied to such highly productive systems as wetlands. Our results show that many model predictions are applicable to wetlands, which tend to be more productive than
grasslands as well as subject to flooding stress. In general, the response of functional groups that were expected to be positively
associated with grazing intensity was more consistent with model
predictions than were functional groups that were expected to be
negatively associated with grazing.
There are several mechanisms whereby ungulate grazing can affect plant communities. Grazers can remove biomass, physically
damage plants by trampling, and create gaps, as well as alter plant
reproductive output, competitive interactions, and nutrient dynamics. In the wetlands studied, mechanisms that affect plant competition and colonization may be of primary importance. Vegetation of
ungrazed depressional wetlands is typically relatively species poor
and is characterized by dense stands of tall, clonal, competitively
dominant species. Competition for light is likely to be an important
constraint in such a system (Twolan-Strutt and Keddy, 1996; Geho
et al., 2007), as is the ability to successfully colonize gaps (Carlyle
and Fraser, 2006). This is attested by the near-ubiquity of tall and rhizomatous species at ungrazed sites in the study.
Grazing-mediated changes in competitive interactions and
shifts in dominance have been documented in wetlands (Ausden
et al., 2005), and clipping has been shown to significantly affect
biomass, reproductive output, and juvenile growth of wetland
plants (Crosslé and Brock, 2002; Hayball and Pearce, 2004). Grazing-associated reductions in aboveground biomass and litter observed in this study as well as increases in bare ground from
trampling are likely to alter competitive outcomes and provide
greater establishment opportunities. Because the frequency of
occurrence of both tall and rhizomatous species declined with
increasing grazing intensity, it is likely that the reduction in

Please cite this article in press as: Jones, W.M., et al. Plant community functional shifts in response to livestock grazing in intermountain depressional wetlands in British Columbia, Canada. Biol. Conserv. (2010), doi:10.1016/j.biocon.2010.10.005

W.M. Jones et al. / Biological Conservation xxx (2010) xxx–xxx

5

Fig. 2. Associations between bare ground and the frequency of occurrence of plant functional groups and native and exotic species richness. Data from the marsh and wet
meadow zones are shown by d’s and D’s, respectively. Lines show significant relationships at a = 0.05, with solid lines representing the marsh zone and dashed lines
representing the wet meadow zone.

frequency of these groups, which contain many of the competitively dominant species, would provide opportunities for less competitive species, especially if they were unpalatable or had
attributes that allowed them to either avoid or tolerate grazing
pressure. This may explain why the non-rhizomatous functional
group, which shared many species with the annual group, was positively associated with grazing intensity.
Perennials and graminoids showed no response to grazing
intensity. Both these groups are ubiquitous in the study wetlands

and encompass species with very different responses to grazing.
Common members of both groups were P. nuttalliana and H. jubatum, which were positively associated with grazing, and S. acutus
and J. balticus, which were negatively associated with grazing.
For the study wetlands, these functional groups may be too general
to make consistent predictions of their response to grazing intensity. This was expected for graminoids but not for perennials,
which other studies have consistently found to be negatively associated with grazing (Diaz et al., 2007).

Please cite this article in press as: Jones, W.M., et al. Plant community functional shifts in response to livestock grazing in intermountain depressional wetlands in British Columbia, Canada. Biol. Conserv. (2010), doi:10.1016/j.biocon.2010.10.005

6

W.M. Jones et al. / Biological Conservation xxx (2010) xxx–xxx

The relationship between species richness and livestock grazing
was not consistent between zones. In the marsh zone, the relationship was hump-shaped for both native and exotic species, while in
the wet meadow it was positive for exotic species but there was no
effect on native species. It is a well established hypothesis that species richness will be maximized at intermediate levels of disturbance (Grime, 1973; Connell, 1978; Pacala and Crawley, 1992);
however, both productivity and species pools can influence this
pattern. Species richness and disturbance have been shown to be
positively associated in highly productive communities and negatively associated in communities with low productivity (Grime,
1973; Olff and Ritchie, 1998; Kondoh, 2001; Frank, 2005). The relative size of a community’s regional species pool (i.e., the number
of species capable of living within that community) can also
strongly influence species richness (Zobel, 1997; Grace, 2001). Productivity, as measured by aboveground biomass, was not significantly different between the marsh and wet meadow zones at
ungrazed sites. The number of species sampled from the marsh
(n = 82) was substantially less than the number sampled from
the wet meadow (n = 127), implying that the regional species pool
of marsh species is more restricted. It is possible that the decline in
species richness associated with very high levels of bare ground
may reflect the more limited pool of available species in this
community.
Similarly, the relationship between the frequency of exotic species and grazing was not consistent between the marsh and wet
meadow zones. Although exotic species are often positively
associated with grazing (McIntyre and Lavorel, 1994; Rusch and
Oesterheld, 1997; Diaz et al., 2007), this is not always the case
(Marty, 2005), and whether grazing facilitates or inhibits exotic
species may be contingent on specific community and species
dynamics. In the wet meadow, exotic species occurred at very high
frequencies at undisturbed wetlands, and the three most abundant
species, Poa pratensis (Kentucky bluegrass), Sonchus arvensis (field
sowthistle), and Agrostis stolonifera (redtop), were tall and competitively dominant. In contrast, in the marsh, exotic species included
higher numbers of forbs, annuals, and short species.
Marsh zone vegetation in this study was both more species poor
and more dissimilar among wetlands than wet meadow vegetation. The response of many functional groups was largely consistent with grazing models and was strongly directional in
response to grazing intensity in both zones; however, change in
community structure was more variable in the marsh than in the
wet meadow. Thus grazing exerted a strong filter on functional
composition in both zones but had a weaker influence on overall
species composition in the marsh than in the wet meadow.
Differences in richness and functional group frequency between
the marsh and wet meadow zones may have to do with the interaction of grazing with hydrology or with differential hydrological
constraints between the two zones. Vegetation zonation in these
wetlands is determined by water depth through species-specific
effects on seed germination, seedling survival, and adult mortality
(van der Valk and Welling, 1988). Water depth acts as a very strong
filter on wetland vegetation regeneration (van der Valk, 1981;
Weiher and Keddy, 1995), and the marsh vegetation species pool
may be limited due to constraints caused by flooding. It is possible
that the greater hydrologic stability and more limited species pool
of the marsh provide for fewer regeneration opportunities and a
stronger founder effect which leads to greater community
dissimilarity.
4.1. Management implications
Compositionally and functionally, wetland vegetation changed
substantially along the observed livestock grazing gradient. These
changes can have broader implications for the habitat and biodi-

versity values of these wetlands. Decreases in biomass and the frequency of tall species, which comprise the dominant species in
these wetlands, can adversely affect the habitat value of these wetlands for wetland-associated wildlife. Both breeding waterfowl and
amphibians are sensitive to livestock grazing in wetlands (Kirsch,
1969; Klett et al., 1988; Belanger and Picard, 1999; Schmutzer
et al., 2008). For the study wetlands, the intensity of livestock grazing was negatively associated with waterfowl breeding use, especially of dabbling ducks, which are more likely to use emergent
vegetation as escape cover (M. Jones, unpublished data). Compared
to ungrazed wetlands, grazing effects on wetland vegetation appeared to occur at even low levels of grazing intensity. However,
low intensity grazing is much more likely to be compatible with
the management of these wetlands for conservation values. Further work is needed to link the observed grazing intensity gradient
with stocking rates and seasonality of livestock use. These linkages
may better direct wetland management.
Role of funding sources
Funding was provided by Ducks Unlimited Canada Western
Region office and the Institute for Wetlands and Waterfowl
Research as well as the British Columbia Ministry of Forests and
Range Forest Science Program. In addition, M. Jones was supported
by the Natural Sciences and Engineering Research Council of
Canada Industrial Postgraduate Fellowship. These sponsors had
no involvement in the study design, data collection or analysis,
writing or submission of this manuscript.
Acknowledgments
We thank Bruce Harrison with Ducks Unlimited Canada for his
collaboration on this project. For field assistance we thank
Montana Burgess, Eleanor Bassett, Jessica Gosling, Anna Maria Pellett, and Amy Bitz. We also thank three anonymous reviewers for
constructive comments that greatly helped to improve the manuscript. Funding was provided by grants from Ducks Unlimited Canada’s Institute for Wetland and Waterfowl Research and the British
Columbia Ministry of Forests and Range Forest Science Program. M.
Jones was also supported by an NSERC Industrial Postgraduate
Scholarship.
References
Ausden, M., Hall, M., Pearson, P., Strudwick, T., 2005. The effects of cattle grazing on
tall-herb fen vegetation and molluscs. Biological Conservation 122, 317–326.
Austin, J.E., Keough, J.R., Pyle, W.H., 2007. Effects of habitat management treatments
on plant community composition and biomass in a montane wetland. Wetlands
27, 570–587.
Belanger, L., Picard, M., 1999. Cattle grazing and avian communities of the St.
Lawrence River Islands. Journal of Range Management 52, 332–338.
Bullock, J.M., Franklin, J., Stevenson, M.J., Silvertown, J., Coulson, S.J., Gregory, S.J.,
Tofts, R., 2001. A plant trait analysis of responses to grazing in a long-term
experiment. Journal of Applied Ecology 38, 253–267.
Carlyle, C.N., Fraser, L.H., 2006. A test of three juvenile plant competitive response
strategies. Journal of Vegetation Science 17, 11–18.
Connell, J.H., 1978. Diversity in tropical rain forests and coral reefs. Science 199,
1302–1310.
Crosslé, K., Brock, M.A., 2002. How do water regime and clipping influence wetland
plant establishment from seed banks and subsequent reproduction? Aquatic
Botany 74, 43–56.
Daubenmire, R., 1970. Steppe Vegetation of Washington. Washington Agricultural
Experiment Station Technical Bulletin 62. Washington State University,
Pullman.
Diaz, S., Lavorel, S., McIntyre, S., Falczuk, V., Casanoves, F., Milchunas, D.G., Skarpe,
C., Rusch, G., Sternberg, M., Noy-Meir, I., Landsberg, J., Zhang, W., Clark, H.,
Campbell, B.D., 2007. Plant trait responses to grazing – a global synthesis.
Global Change Biology 13, 313–341.
Diaz, S., Noy-Meir, I., Cabido, M., 2001. Can grazing response of herbaceous plants be
predicted from simple vegetative traits? Journal of Applied Ecology 38, 497–
508.

Please cite this article in press as: Jones, W.M., et al. Plant community functional shifts in response to livestock grazing in intermountain depressional wetlands in British Columbia, Canada. Biol. Conserv. (2010), doi:10.1016/j.biocon.2010.10.005

W.M. Jones et al. / Biological Conservation xxx (2010) xxx–xxx
Dorrough, J., Ash, J., McIntyre, S., 2004. Plant responses to livestock grazing
frequency in an Australian temperate grassland. Ecography 27, 798–810.
Douglas, G.W., Straley, G.B., Meidinger, D.V., Pojar, J., 1998–2002. Illustrated Flora of
British Columbia. BC Ministry of Sustainable Resource Management and BC
Ministry of Forests, Victoria, BC.
Dyksterhuis, E.J., 1949. Condition and management of range land based on
quantitative ecology. Journal of Range Management 2, 104–115.
Fensham, R.J., Holman, J.E., Cox, M.J., 1999. Plant species responses along a grazing
disturbance gradient in Australian grassland. Journal of Vegetation Science 10,
77–86.
Frank, D., 2005. The interactive effects of grazing ungulates and aboveground
production on grassland diversity. Oecologia 143, 629–634.
Fraser, L.H., Grime, J.P., 1999. Interacting effects of herbivory and fertility on a
synthesized plant community. Journal of Ecology 87, 514–525.
Geho, E.M., Campbell, D., Keddy, P.A., 2007. Quantifying ecological filters: the
relative impact of herbivory, neighbours, and sediment on an oligohaline
marsh. Oikos 116, 1006–1016.
Gough, L., Grace, J.B., 1998. Herbivore effects on plant species density at varying
productivity levels. Ecology 79, 1586–1594.
Grace, J.B., 2001. The roles of community biomass and species pools in the
regulation of plant diversity. Oikos 92, 193–207.
Greenwood, K.L., McKenzie, B.M., 2001. Grazing effects on soil physical properties
and the consequences for pastures: a review. Australian Journal of Experimental
Agriculture 41, 1231–1250.
Grime, J.P., 1973. Control of species density in herbaceous vegetation. Journal of
Environmental Management 1, 151–167.
Grime, J.P., 1977. Evidence for existence of three primary strategies in plants and its
relevance to ecological and evolutionary theory. American Naturalist 111,
1169–1194.
Grime, J.P., 2001. Plant Strategies, Vegetation Processes, and Ecosystem Properties.
John Wiley & Sons, Inc., Chichester.
Havstad, K.M., 2008. Mongolia’s rangelands: is livestock production the key to the
future? Frontiers in Ecology and the Environment 7, 386–387.
Hayball, N., Pearce, M., 2004. Influences of simulated grazing and water-depth on
the growth of juvenile Bolboschoenus caldwellii, Phragmites australis and
Schoenoplectus validus plants. Aquatic Botany 78, 233–242.
Hendricks, H.H., Bond, W.J., Midgley, J., Novellie, P.A., 2005. Plant species richness
and composition a long livestock grazing intensity gradients in a Namaqualand
(South Africa) protected area. Plant Ecology 176, 19–33.
Hobbs, N.T., Baker, D.L., Bear, G.D., Bowden, D.C., 1996. Ungulate grazing in
sagebrush grassland: mechanisms of resource competition. Ecological
Applications 6, 200–217.
Humphrey, J., Patterson, G., 2000. Effects of late summer cattle grazing on the
diversity of riparian pasture vegetation in an upland conifer forest. Journal of
Applied Ecology 37, 986–996.
Johnson, W.C., Millett, B.V., Gilmanov, T., Voldseth, R.A., Gunterspergen, G.R.,
Naugle, D.E., 2005. Vulnerability of northern prairie wetlands to climate change.
Bioscience 55, 863–872.
Kantrud, H.A., Hillar, J.B., van der Valk, A.G., 1989. Vegetation of wetlands in the
Prairie Potholes Region. In: van der Valk, A.G. (Ed.), Northern Prairie Wetlands.
Iowa State University Press, Ames, Iowa, pp. 132–187.
Kantrud, H.A., Stewart, R.E., 1977. Use of natural basin wetlands by breeding
waterfowl in North Dakota. Journal of Wildlife Management 41, 243–253.
Kauffman, J.B., Thorpe, A.S., Brookshire, E.N.J., 2004. Livestock exclusion and
belowground ecosystem responses in riparian meadows of Eastern Oregon.
Ecological Applications 14, 1671–1679.
Keddy, P.A., 2000. Wetland Ecology: Principals and Conservation. Cambridge
University Press, Cambridge.
Kirsch, L.M., 1969. Waterfowl production in relation to grazing. Journal of Wildlife
Management 33, 821–828.
Klett, A.T., Shaffer, T.L., Johnson, D.H., 1988. Duck nest success in the Prairie Pothole
Region. The Journal of Wildlife Management 52, 431–440.

7

Kondoh, M., 2001. Unifying the relationships of species richness to productivity and
disturbance. Proceedings of the Royal Society B 268, 269–271.
Mack, R.N., Thompson, J.N., 1982. Evolution in steppe with few large, hooved
mammals. American Naturalist 119, 757–773.
Manier, D.J., Hobbs, N.T., 2007. Large herbivores in sagebrush steppe ecosystems:
livestock and wild ungulates influence structure and function. Oecologia 152,
739–750.
Martin, D.W., Chambers, J.C., 2001. Effects of water table, clipping, and species
interactions on Carex nebrascensis and Poa pratensis in riparian meadows.
Wetlands 21, 422–430.
Marty, J.T., 2005. Effects of cattle grazing on diversity in ephemeral wetlands.
Conservation Biology 19, 1626–1632.
McIntyre, S., Lavorel, S., 1994. Predicting richness of native, rare, and exotic plants in
response to habitat and disturbance variables across a variegated landscape.
Conservation Biology 8, 521–531.
McIntyre, S., Lavorel, S., Tremont, R.M., 1995. Plant life-history attributes - their
relationship to disturbance responses in herbaceous vegetation. Journal of
Ecology 83, 31–44.
Milchunas, D.G., Lauenroth, W.K., 1993. Quantitative effects of grazing on
vegetation and soils over a global range of environments. Ecological
Monographs 63, 327–366.
Milchunas, D.G., Sala, O.E., Lauenroth, W.K., 1988. A generalized-model of the effects
of grazing by large herbivores on grassland community structure. American
Naturalist 132, 87–106.
Mitsch, W.J., Gosselink, J.G., 2000. Wetlands, third ed. John Wiley & Sons, Inc., New
York.
Oksanen, J., Kindt, R., Legendre, P., O’Hara, B., Simpson, G., Solymos, P., Stevens, M.,
Wagner, H., 2008. Vegan: Community Ecology Package for the R Statistical
Environment. Version 1.15. <http://vegan.r-forge.r-project.org>.
Olff, H., Ritchie, M.E., 1998. Effects of herbivores on grassland plant diversity. Trends
in Ecology and Evolution 13, 261–265.
Pacala, S.W., Crawley, M.J., 1992. Herbivores and plant diversity. American
Naturalist 140, 243–260.
Piha, H., Luoto, M., Merila, J., 2007. Amphibian occurrence is influenced by current
and historic landscape characteristics. Ecological Applications 17, 2298–2309.
Pykälä, J., 2004. Cattle grazing increases plant species richness of most species trait
groups in mesic semi-natural grasslands. Plant Ecology 175, 217–226.
R Development Core Team, 2008. R: and Language and Environment for Statistical
Computing, Version 2.8.1. R Foundation for Statistical Computing, Vienna.
<http://www.R-project.org>.
Rusch, G., Oesterheld, M., 1997. Relationship between productivity, and species and
functional group diversity in grazed and non-grazed Pampas grassland. Oikos
78, 519–526.
Schmutzer, A.C., Gray, M.J., Burton, E.C., Miller, D.L., 2008. Impacts of cattle on
amphibian larvae and the aquatic environment. Freshwater Biology 53, 2613–
2625.
Twolan-Strutt, L., Keddy, P.A., 1996. Above- and belowground competition intensity
in two contrasting wetland plant communities. Ecology 77, 259–270.
van der Valk, A.G., 1981. Succession in wetlands: a Gleasonian approach. Ecology
62, 688–696.
van der Valk, A.G., 1986. The impact of litter and annual plants on recruitment from
the seed bank of a lacustrine wetland. Aquatic Botany 24, 13–26.
van der Valk, A.G., Welling, C.H., 1988. The development of zonation in freshwater
wetlands: an experimental approach. In: During, H., Werger, M., Willems, J.
(Eds.), Diversity and Pattern in Plant Communities. Academic Publishing, The
Hague, pp. 145–158.
Weiher, E., Keddy, P.A., 1995. The assembly of experimental wetland plant
communities. Oikos 73, 323–335.
Zobel, M., 1997. The relative role of species pools in determining plant species
richness: an alternative explanation of species coexistence? Trends in Ecology
and Evolution 12, 266–269.

Please cite this article in press as: Jones, W.M., et al. Plant community functional shifts in response to livestock grazing in intermountain depressional wetlands in British Columbia, Canada. Biol. Conserv. (2010), doi:10.1016/j.biocon.2010.10.005