The effect of small mammal exclusion on grassland recovery from disturbance in the Chihuahuan Desert

https://doi.org/10.1016/j.jaridenv.2019.04.001Get rights and content

Highlights

  • Small mammal herbivores did not constrain grassland regeneration.

  • Herbivore exclusion may not be effective in improving restoration of Bouteloua eriopoda grasslands, in some time periods.

  • Further research into the demography of B. eriopoda ramets and seedlings is needed.

Abstract

In many arid ecosystems, shrub encroachment is coupled to the loss of perennial grasses and associated ecosystem services. Increased native herbivore abundance associated with shrub encroachment can have negative effects on grass restoration. In the Chihuahuan Desert, native herbivore abundance can be two times greater in shrubland states dominated by Prosopis glandulosa (honey mesquite) than in historical Bouteloua eriopoda (black grama) grasslands. We compared the recovery of B. eriopoda patches following disturbance in plots that were exposed to or protected from native herbivores in shrub-dominated, grass-dominated and ecotone (grass and shrub co-dominated) states. We created a disturbance in the center of B. eriopoda grass patches in each treatment quadrat in 2001. The disturbed areas were measured for recruitment and re-establishment of B. eriopoda in 2002, 2005 and 2008. Although mean rodent abundance was generally greater in shrub-dominated states than other states over the study period, reproductive potential of B. eriopoda was similar in shrub and grass dominated states. Additionally, there was no revegetation by B. eriopoda in any state in response to rodent exclusion. Because increased herbivore abundance in shrub-dominated states did not constrain grass revegetation, other factors are likely to be more important constraints on perennial grass recovery in the Chihuahuan Desert.

Introduction

Woody plant encroachment and associated decline of perennial grasses is a key management and restoration challenge in rangelands worldwide (Parizek et al., 2002; Ratajczak et al., 2012; Roques et al., 2001; Van Auken, 2009). The persistent loss of perennial grasses is usually ascribed to competitive preemption of water and other resources by shrubs and to soil erosion feedbacks due to the absence of ground cover (Archer et al., 2017; Peters et al., 2013; Pierce et al., 2018). Another possible and little understood effect in these systems involves the impact of native mammalian herbivores, specifically rodents and lagomorphs, on perennial grass recovery following shrub encroachment and grass loss (Bestelmeyer et al., 2007).

In the northern Chihuahuan Desert, historical grasslands dominated by the perennial grass Bouteloua eriopoda (black grama) have undergone state transitions to shrublands dominated by Prosopis glandulosa (mesquite) over the last century (Peters et al., 2006). Attempts to restore B. eriopoda in areas where it has been completely lost have met with minimal success (Herrick et al., 2006). Lack of B. eriopoda grassland regeneration and restoration success is attributed to abiotic factors in these cases (Peters, 2000). Where B. eriopoda remnants exist, however, gradual recovery is possible (Bestelmeyer et al., 2013). Native herbivores, however, can limit the ability of grasses to recolonize from remnant plants through consumption of seeds and tillers (Bestelmeyer et al., 2007; Brown and Heske, 1990). This effect may be exacerbated by the presence of shrubs. Shrubs often serve as habitat and a water source for many rodent and lagomorph species (Jaksic, 1986; Jaksic and Soriguer, 1981; Kerley, 1992), and certain rodent and lagomorph species may increase in abundance and/or activity with increasing shrub cover (Daniel et al., 1993; Reynolds, 1950; Schooley et al., 2018; Whitford, 1993). While evidence exists for small herbivore impacts on native grasses, our understanding of herbivore impacts on plant dynamics under a range of biotic and abiotic conditions is lacking (Maron and Crone, 2006). Furthermore, there is scant evidence for the interaction of increases in shrub cover and impacts of herbivores. In one study along a gradient between P. glandulosa and B. eriopoda dominated states, herbivory on seedlings was greatest in P. glandulosa-dominated states, but small mammal abundance measured in that study was not higher in those states—likely indicating increases in foraging effort by small mammals within these shrub-dominated areas (Bestelmeyer et al., 2007).

Bouteloua eriopoda spreads predominantly through asexual reproduction by stolons and ramets (Nelson, 1934; Smith et al., 2004). Although seeds are produced in years with sufficient rainfall (Nelson, 1934), their viability is often low (<50%) as a result of genetic (Hanson, 1972) and insect-related (Watts, 1965) factors. Based on model simulations and field studies, Peters (2000) suggested that the frequency of optimal conditions for viable seed set and seedling establishment of B. eriopoda is low, and successful seedling establishment events are rarely observed in this region (Neilson, 1986). Thus, disturbed areas within B. eriopoda grasslands are likely to be recolonized primarily through the establishment of ramets; herbivory on stolons and ramets would limit grass recovery.

We conducted an experiment to examine the impact of native small herbivores on recovery of disturbed B. eriopoda patches through vegetative spread (stolons, non-rooted ramets) and recruitment (rooted ramets, seedlings). We simulated a disturbance by removing sections of B. eriopoda patches and observed recovery over seven years. We hypothesized that exposure to rodent and lagomorph activity would constrain grass recovery in disturbed patches. We also hypothesized that the magnitude of this constraint would increase along local grassland to shrubland gradients due either to greater rodent numbers or increased small mammal foraging activity in shrublands.

Section snippets

Study area

The study was established on the Jornada Experimental Range (JER) and the Chihuahuan Desert Rangeland Research Center (CDRRC), 37 km north of Las Cruces, New Mexico, USA (32° 35′ N, 106° 51’ W; 1334 m a.s.l.). The study sites are currently or formerly dominated by Bouteloua eriopoda grasslands that have been invaded to varying degrees by Prosopis glandulosa shrubs. This desert grassland ecosystem type historically extended across the U.S. states of Arizona, New Mexico, and Texas, as well as

Analysis

SAS Version 9.4 (SAS Institute, Cary NC) was used for data manipulation and analyses. Rodent abundance per year was modeled with a Poisson generalized linear mixed model (PROC GLIMMIX) with the Laplace method for likelihood approximation. Habitat, year, and their interaction were modeled as fixed effects. A G-side random effect for habitat*wet/dry period was included to account for interannual variability and overdispersion.

Bouteloua eriopoda canopy cover and counts of stolons, non-rooted

Results

The main effects of year and state on rodent abundance were highly significant (p < 0.0001 and p = 0.0001 respectively) but there was no evidence of a year*state interaction (p = 0.9699) (Fig. 3). Across states rodent numbers increased from 2004 to 2005 (p = 0.0386) and 2006 to 2007 (p = 0.0017) before declining between 2007 and 2008 (p = 0.0002) to a level no different than 2004 (p = 0.4678). Compared to Grass states, rodent abundance was 3.87 ± 1.72 rodents/ha higher in the Ecotone state (p

Impact of herbivory on B. eriopoda

We hypothesized that reproductive effort and success of B. eriopoda in rodent-excluded plots would be greater than non-excluded plots for all states. Rodent and lagomorph exclusion in Grass, Ecotone, and Shrub states did not have a significant effect on the recolonization of disturbed B. eriopoda patches over 7 years. Our results are not consistent with research findings that documented an increased impact of herbivores on herbaceous vegetation in shrub-dominated versus grass-dominated states (

Conclusions

Models of vegetation dynamics in desert grassland ecosystems predict accelerating loss of perennial grasses due to altered biogeochemical and erosion processes resulting from shrub encroachment and climate change (Schlesinger et al., 1990). Shrub encroachment could induce grass loss by facilitating increases in rodent and lagomorph populations with subsequent impacts on grasses via herbivory and disturbance to root systems (Bestelmeyer et al., 2007). Multiple studies suggest that either greater

Author contributions

BTB, DPCP conceived of the study and designed the research; BTB, LNS, DKJ conducted the research; BTB, DKJ analyzed the data; LNS, BTB, DKJ wrote the manuscript; LNS, BTB, DKJ, DPCP edited the manuscript. All authors approve the final article.

Acknowledgments

Funding for this research was provided by the National Science Foundation to the Jornada Basin Long Term Ecological Research Program through New Mexico State University [grant number DEB-1235828] and USDA-ARS CRIS Project [grant number 3050-11210-007-00D]..

References (48)

  • J.H. Brown et al.

    Control of a desert-grassland transition by a keystone rodent guild

    Science (Washington, D.C.)

    (1990)
  • J.H. Brown et al.

    Resource utilization and coexistence of seed‐eating desert rodents in sand dune habitats

    Ecology

    (1973)
  • E.M. Christensen et al.

    Long‐term community change through multiple rapid transitions in a desert rodent community

    Ecology

    (2018)
  • A. Daniel et al.

    Jackrabbit densities on fair and good condition Chihuahuan desert range

    J. Range Manag.

    (1993)
  • K.A. DaVanon et al.

    Effects of exurban development on trophic interactions in a desert landscape

    Landsc. Ecol.

    (2016)
  • C. Elzinga et al.

    Measuring and Monitoring Plant Populations

    (1998)
  • A. Hanson

    Grass Varieties in the United States

    (1972)
  • C.H. Herbel et al.

    Drought effects on a semidesert grassland range

    Ecology

    (1972)
  • J.E. Herrick et al.

    Remediation research in the Jornada Basin: past and future

  • E.J. Heske et al.

    Effects of kangaroo rat exclusion on vegetation structure and plant species diversity in the Chihuahuan Desert

    Oecologia

    (1993)
  • F.M. Jaksic

    Predation upon small mammals in shrublands and grasslands of southern South America: ecological correlates and presumable consequences

    Rev. Chil. Hist. Nat.

    (1986)
  • F.M. Jaksic et al.

    Predation upon the European rabbit (Oryctolagus cuniculus) in Mediterranean habitats of Chile and Spain: a comparative analysis

    J. Anim. Ecol.

    (1981)
  • G.I. Kerley

    Ecological correlates of small mammal community structure in the semi‐arid Karoo, South Africa

    J. Zool.

    (1992)
  • G.I. Kerley et al.

    Mechanisms for the keystone status of kangaroo rats: graminivory rather than granivory?

    Oecologia

    (1997)
  • 1

    Present address: School of Veterinary and Life Sciences, Murdoch University; Murdoch Western Australia 6150 Australia; and Biodiversity Conservation Centre; Kattidj Close, Kings Park, Western Australia, 6005, Australia.

    View full text