Electronic Scientific Paper Archive

Clustered animal burrows yield higher spatial heterogeneity

Y. Yoshihara, T. Okuro, B. Buuveibaatar, J. Undarmaa K. Takeuchi

Abstract

    An understanding of the relationshipsbetween spatial heterogeneity and disturbance regime is important for establishing the mechanisms necessary to maintain biodiversity. Our objective was to examine how the configuration of disturbance by burrowing rodents (Siberian marmot) affected the
spatial heterogeneity of vegetation and soil nutrient properties. We established three 2500-m2 (50 m 9 50 m) isolated-burrows plots and three2500-m2 clustered-burrows plots in a Mongolian grassland. Each plot was subdivided into 4-m2quadrats, and the plant species richness, percent coverage, and soil nutrient properties in the quadrats were surveyed. Spatial heterogeneity was calculated for vegetation using the mean dissimilarity of species composition among sample quadrats, and geostatistical analysis was used to calculate soil properties.Heterogeneous patches of plants such as Achnatherum splendens and higher nutrient concentrations were found only near the clustered burrows. As a result,
spatial heterogeneities of vegetation and soil nutrient properties were higher in the clustered colony than those in the isolated colony. The configuration of disturbance patches affected the spatial heterogeneity at the landscape level through the spatial pattern of disturbance frequency.

Keyword: Patch configuration,Disturbance regime,Mongolia,Siberian marmot,Soil nutrition,Plant composition

Introduction

   Understanding the patterns, causes, and consequences of spatial heterogeneity is a research frontier in the landscape ecology, population ecology, and ecosystem ecology (Hanski and Simberloff 1997; Turner et al. 2001; Wu 2004). Spatial heterogeneity has been recognized as a key focus in biodiversity from the individual to the species, community, and ecosystem levels (Huston 1994; Tilman 1999; Tews et al. 2004; Dufour et al. 2006; North and Ovaskainen 2007) through its effects on survival rate (Corlett 2000), competition (Fransen et al. 2001; Day et al. 2003), dispersal success (Rees et al. 2000), lateral flux of matter or energy (Reiners and Driese 2001), and ecosystem functions (Lovett et al. 2005). Disturbance is the one of main factors generating spatial heterogeneity (Levin 1992; White and Jensch 2001). Species respond differently to disturbance according to their various abilities to tolerate or thrive under disturbed conditions. These various responses are related to species’ life-history traits; for instance,
generally annual plants do better in disturbed areas than do perennials; plants propagated clonally regenerate better in disturbed areas than do those propagated by seed (Halpern 1988; 1989; Turner et al. 1997; White and Harrod 1997; Dorrough et al. 2004). If we consider nonspatial components of disturbance at the patch scale, different types of patches (succession stage) will form according to the disturbance magnitude gradient. Recent studies have examined the importance of the spatial and temporal components of individual disturbances (e.g., Halpern 1989; Moloney and Levin 1996; Platt and Connell 2003). In the context of communities at the landscape scale, because the disturbance regime (e.g., size, frequency, and proportion of disturbed area; temporal variation) controls the proportion of the landscape at each succession stage (Wu and Levin 1994; Moloney and Levin 1996; White and Harrod 1997; Platt and Connell 2003), it is essential to consider the disturbance regime when examining the dynamics of spatial heterogeneity. Romme et al. (1998) suggested that some unusually frequent or intense disturbances produce qualitatively different responses as compared with similar disturbances of lesser magnitude; such events cause dramatic changes in impact at some point along the disturbance intensity or frequency gradient (threshold responses). Similarly, compounded disturbances within the normative recovery time of a community appear to cause that community to enter new domains or produce alternative stable states that are ecologically
surprising (Paine et al. 1998). For example, in Jamaican coral reefs, three sequential disturbances failed to allow coral recovery, which would have developed, had the disturbance events occurred individually (Hughes 1994; Andres and Witman 1995). These findings indicate that the effects of
compounded disturbances on a community are not simply additive but multiplicative. If such emergent patches are generated within a spatial extent, spatial heterogeneity at the landscape scale will be affected. In a tallgrass prairie ecosystem, frequent fires modified the species composition of vegetation to a greater extent than did a single fire (Collins 2000) and, consequently, affected spatial heterogeneity (Knapp et al. 1999, Collins and Smith 2006). The configuration of disturbance patches should affect spatial heterogeneity by changing the spatial pattern of the disturbance frequency. Assuming that each disturbance patch is spatially concentrated, these patches may overlap in some areas. Different communities will be created in these frequently disturbed overlapping patches, resulting in an increase of spatial heterogeneity at the landscape scale. Simulation models have evaluated the effects of disturbance configuration (spatial autocorrelation among individual disturbances) on the distribution and abundance of species at the landscape scale (Green 1989; Moloney and Levin 1996); however, empirical work designed to characterize the effects at the landscape scale has not been conducted. Herbivores affect the spatial patterns of vegetation and soil nutrition through defoliation, trampling,
excavating, and excretion. For instance, prairie dogs, rabbits, and voles in grasslands increase the spatial heterogeneity of biomass at the landscape scale (Bangert and Slobodchikoff 2000; Questad and Foster 2007; Olofsson et al. 2008). Urination by cattle creates nitrogen-enriched patches on the order of 80 cm in diameter (Stillwell et al. 1983), resulting in an increased late-succession species such as C4 plants
and increased b diversity (Steinauer and Collins 1995). These influences are mitigated by grazing, however. Since such high nutrient patches attract grazers, urine appears to impact spatial heterogeneity of vegetation to a larger scale via defoliation by grazers than do by urine alone (Steinauer and Collins 2001). In Mongolian grasslands, Siberian marmots (Marmota sibirica; ‘‘tarbagan’’ in Mongolian) modify plant communities and soil physical and chemical properties through activities such as burrowing, herbivore, and urinating (Adiya 2000; Van Staalduinen and Werger 2007). In order to minimize their predation risk (Holmes 1984), marmots return frequently to their burrows. Because their burrows serve as focal points for their activities (Behrends et al. 1985; Branch 1993), the relative frequency of marmot activity decreases with increasing distance from their shelters (Karels et al. 2004). Therefore, the compositional heterogeneity of vegetation increases in areas closer to marmot burrows (English and
Bowers 1994; Yoshihara et al. 2009). The isolated-burrows plot includes two patch types that differ in disturbance frequency: an undisturbed patch and frequently disturbed patches (A and B in Fig. 1). The clustered-burrows plot includes three types of patches: an undisturbed patch, frequently disturbed patches, and an extremely frequently disturbed patch near the center of the cluster (C in Fig. 1).

In this grassland, we predict that the disturbance- free patches (A) would be dominated by perennial graminoids, the disturbed patches (B) would be dominated by annual forbs, and unexpected communities would be established in the extremely frequently disturbed patch (C), which should contribute to a greater degree of landscape level heterogeneity in the clustered-burrows plot. The aim of this study was to investigate the effects of the spatial configuration of disturbance patches on spatial heterogeneity at the landscape level by comparing areas with isolated marmot burrows versus those with
clustered burrows. At the colony scale (contain multiple burrows), we expected that the spatial heterogeneities of vegetation and soil nutrients in the clustered-burrows plot would be higher than that in the isolated-burrows plot (Fig. 1).

Material and Methods

Study areas
   We selected study sites in Hustai National Park (HNP), which is located 100 km west of Ulaanbaatar, Mongolia (47500N, 106000E, elevation 1100–1840 m). HNP occupies *600 km2 of the forest steppe region of Mongolia and receives an average annual precipitation of 232 mm. The annual average temperature is 0.2C and average monthly temperatures vary between –23C (in January) and?20C (in July). About 88% of the area of HNP is covered by grassland and shrubland steppe and about 5% is covered by birch-dominated forest (Wallis de Vries et al. 1996). The types of vegetation that grow on the mounds created by Siberian marmots (M. sibirica) can differ greatly from those on the surrounding, comparatively homogeneous, grasslands, which are dominated by a few graminoids (Van Staalduinen and Werger 2007). When this study was conducted, livestock had been excluded from the HNP’s core areas for 15 years. HNP is thus a suitable site for detecting the effect of marmots on landscape heterogeneity in grasslands.

Sampling design
Because dung density is considered a good indicator of marmot disturbance (Karels et al. 2004), we located and noted the position of marmot burrows and dung within a 25-ha area (500 m 9 500 m) using a global positioning system receiver (Garmin Inc., Taiwan). Based on these data, we identified an area with clustered burrows and an area with isolated burrows. Our field surveys were conducted in July and August in 2006 and 2007. According to Addicott et al. (1987), the spatial extent that encompasses most of an organism’s activity relevant to some process (and over
a relevant amount of time) is the appropriate scale for investigating this process. Thus, the effect of grazing by marmots at the colony scale was investigated by comparing the 2500-m2 plots. We established three 2500-m2 (50 m 9 50 m) plots including some isolated marmot burrows (isolated colony, ISO) and three 2500-m2 plots including the same number of clustered burrows (clustered colony, CLU). Each plot was subdivided into 625 adjacent 4-m2 (2 m 9 2 m) quadrats (total = 3750 quadrats), because the 4-m2 quadrat size is a suitable grain size to reflect marmot disturbances (Yoshihara et al. 2009). The plots were separated by *300 m within the same landscape (valley foot slope); all plots had the same soil type (Haplic Kastanozems) according to the soil-type layer in the GIS database produced by the HNP and were dominated by Stipa krylovii. In 2006, we recorded the absence/presence of each plant species in each quadrat of plots ISO1 and CLU1. In many studies on heterogeneity, the comparative
analysis of multiple types of measurements most likely yields accurate outcome (Tews et al. 2004). We therefore recorded the area of ground covered by each species in each quadrat of ISO2 and CLU2 in 2007, and ISO3 and CLU3 in 2008, using a modified Daubenmire percent cover scale (Daubenmire 1959; Collins and Smith 2006): 0, absent; 1,1%; 2, 2–5%; 3, 6–25%; 4, 26–50%; 5, 51–75%; 6, 76–95%; and 7,[95%. In order to investigate soil properties, we extracted soil core samples (10-cm diameter 9 15-cm depth) from 40 m 9 40 m areas situated at the centers of
plots ISO2 and CLU2, with 5-m spacing between points (64 points). At each point, we extracted three replications (three-point method) and the samples were homogenized. In addition, we extracted an extra eight random samples at the center of the 64 points, at a 2.5-m spacing. For comparison, we performed the same soil sampling in an area without marmot burrows (off-colony plot). This nested design was well suited for quantifying heterogeneity across as broad a range of spatial scales as possible (Gross et al. 1995; Augustine and Frank 2001). In the laboratory of the Department of Ecosystem Studies of the Mongolian State University of Agriculture, soil samples were dried in an oven at 70C
for 48 h and weighed. Dry combustion with decalcified soil samples (Matejovic 1993) was used to measure total nitrogen (TN), digestion with a mixture of salicylic acid was used to determine nitrate nitrogen (NO3–N), and Colwell methods were used to measure potassium (K).

Data analysis
On a map of the plots generated in ArcMAP 9.2 (ESRI 2007), we radiated out 11.6-m buffers from each burrow, where the 11.6 m equals the normal activity range (X) based on a study of hoary marmots (Marmota caligata) by Karels et al. (2004). With the aid of spatial analysis, we divided the 25-ha area into three patch types: outside the buffers (analogous to A in Fig. 1), within single buffer (B in Fig. 1), and within overlapped buffers (C in Fig. 1). Dung density was calculated in each patch type. We performed the same methods with changing buffer size (i.e., activity range), radiated out 5.8-m (1/2X) and 23.2-m (2X) buffers from each burrow. In order to compare plant species composition between the isolated-burrows plots and the clusteredburrows plots, we calculated the species richness for each quadrat. Spatial heterogeneity for each plot was calculated as the mean dissimilarity in community species composition among sample points in space (Collins and Smith 2006). We used a 1-Sorensen’s quotient of similarity (1-Sorensen’s QS) index for the absence/presence measurements of vegetation heterogeneity and a Euclidian distance index for quantitative measurement of vegetation heterogeneity: a greater value of each index indicates higher spatial heterogeneity. We calculated the heterogeneity over all possible pairs of quadrats within a plot. We used Euclidian distance because it is less affected by species richness and therefore more accurately measures community heterogeneity (Collins 2000). In order to statistically compare the vegetation heterogeneity between in the clustered-burrows plots and in the isolated-burrows plots, we used the three values of 1-Sorensen’s QS index in each plots with a t test. If the P value was smaller than 0.05, we concluded that the dissimilarity index between the plots was significantly different. We input all the soil data (ISO2 and CLU2) and used kriging in ArcGIS to predict values of unsampled locations. In a first step, to predict an unknown value for a specific location, the kriging technique requires the values of the measured sample points around the prediction location and the spatial data configuration. We input all soil data and related locations using the geostatistical analysis extension of ArcGIS. Then, we examined changes in the amount of variance among sampling points separated by a given distance (h, spatial lag) at increasing distances to quantify the spatial structure of the data (experimental variogram). Second, we fit several models for each variogram to construct a theoretical variogram. Then we compared
the root mean squared errors (RMSEs) of practical models: spherical, exponential, Gaussian, and J-Bessel (Ella et al. 2001; Sarangi et al. 2006). We then selected the model with the lowest RMSE as the best-fit model (Pastor et al. 1998; Augustine and Frank 2001). Parameters derived from the best-fit model were used to quantify the spatial pattern of patchiness in a variate’s distribution. Parameter C0, the y intercept of the best-fit model in a variogram, is called the ‘‘nugget’’ and indicates local random effects or measurement error at a short distance. Parameter (C ? C0), where the fitted model reaches an asymptote, is called the ‘‘sill’’ or semivariance and indicates the intensity of the spatial pattern (Robertson
and Gross 1994). The proportion of sample variance explained by patchiness was calculated as 1 - [C0/(C ? C0)]. The ‘‘spatial range’’ was measured as the distance separating sampling points at which semivariance reaches an asymptote (Robertson and Gross 1994), which indicates the maximum distance at which the variable is spatially autocorrelated. Finally, maps of soil TN, NO3–N, and K in each plot (off-colony plot, ISO2, CLU2) were constructed for the 40 m 9 40 m sample areas using the best-fit models. Based on these maps, soil properties were compared among the off-colony, isolated-burrows, and clustered-burrows plots.

Result

Marmot disturbances

  Marmot dung density in the clustered-burrows plots was *2.3, 1.8, and 3.2 times greater than in the isolated-burrows plots when 5.8-, 11.6-, and 23.2-m buffers were radiated out from each burrow, respectively (Fig. 2). This indicated that the clusteredburrows plot was extremely frequently disturbed by marmots. Plant communities and spatial heterogeneity In plots ISO1, ISO2, and ISO3, a total of 52, 40, and 49 plant species were recorded, and in plots CLU1, CLU2, and CLU3, 48, 49, and 51 species were recorded. Spatial heterogeneity was significantly higher in the clustered burrows than in the isolated-burrows plots (Fig. 3, t = -3.59, df = 4, P�.05). The mean Euclidian distances were 3.47 in the isolatedburrows plots and 3.71 in the clustered-burrows plots. At the clustered-burrows plots, the positions of the highly heterogeneous patches roughly overlapped those of the marmot burrows (Figs. 4, 5, 6). Large, highly heterogeneous patches were apparent in the clustered-burrows plots. Achnatherum splendens (Compositae) was found among the clustered burrows but not in the isolated-burrows plots (Figs. 4, 5, 6). Artemisia adamsii also showed the same tendency, although these plants were distributed more widely than A. splendens (Fig. 5). Potentilla spp. were found on isolated burrows or a little distance away from the clustered burrows (Figs. 4, 5, 6).


Spatial heterogeneity of soil properties
Total nitrogen (F = 50.12, df = 1, P�.001), NO3–N (F = 3.93, df = 1, P�.001), and K (F = 21.62, df = 1, P�.001) exhibited significant differences between ISO2 and CLU2 (Table 1).
     

 
The coefficients of variance showed a greater degree of variance for TN and NO3–N among plots. For all soil nutrients, the mean and coefficients of variance were largest in CLU2.
Overall, semivariance at every lag size and sill values for semivariogram models were higher in CLU2 than in the off-colony or ISO2 (Fig. 7). The off-colony plot showed neither a spatial pattern nor patchiness. For all soil nutrients, the nugget values were largest in CLU2 (Table 2). For soil TN, exponential models best fit the data in all plots (Table 2). For soil NO3–N, ISO2 showed no spatial pattern, while CLU2 showed an oscillating spatial pattern (Fig. 7).
In addition, the proportion of variance explained by small-scale patchiness was greater and the range of autocorrelation was lower in CLU2 (Table 2), indicating that CLU2 had greatly increased fine-grain variability and that patches of NO3–N show a repetitive pattern across the plot. For soil K, ISO2 showed oscillation in the spatial pattern, although the magnitude of semivariance was low (Fig. 7).

Based on the kriged map, each soil property exhibited highly heterogeneous patches in the CLU2, which corresponded to the positions of the clustered burrows (Fig. 8). In contrast, we could find no relationship between the heterogeneous patches and burrow positions in the ISO2.


 

Discussion

Researchers have long debated whether the effects of disturbance on spatial heterogeneity depend on the size or intensity of the disturbance (Moloney and Levin 1996; Su et al. 2006). Our results demonstrated that the configuration of disturbance patches affected spatial heterogeneity by modifying the spatial pattern of disturbance frequency and had stronger effects on heterogeneity than the presence or absence of disturbance (Figs. 3, 8). The highly heterogeneous landscape in the clustered-burrows plots was attributed to the emergence of communities that do not occur until multiple disturbance patches overlap within the landscape (Figs. 4, 5, 6). In a simulated landscape, Moloney and Levin (1996) noted that the configuration of disturbance patches had a minor effect on the number of species present. This conclusion most likely arose because they ignored the emergent property when individual disturbance patches are overlapped; that is, they only considered the presence or absence of disturbance. In accordance with our prediction, clustered burrows increased the spatial heterogeneity of vegetation by permitting the coexistence of several communities with various disturbance-tolerance levels. Various disturbance regimes divided areas into several floristic communities, corresponding with the disturbance magnitude (Figs. 4, 5, 6). For example, areas that experienced infrequent disturbance (A in Fig. 1) were dominated by S. krylovii, considered to be the Mongolian grassland species with the weakest tolerance of disturbance (Van Staalduinen and Werger 2007; Van Staalduinen et al. 2007). Those areas that experienced frequent disturbance (B in Fig. 1) were covered by Potentilla acaulis and Potentilla bifurca
(Rosaceae). Previous studies indicated that Potentilla coverage increases with moderate grazing due to the plants’ prostrate growth form in semiarid grasslands of eastern Asia (Hilbig 1995; Liu et al. 2007). Finally, the most frequently disturbed areas (C in Fig. 1) were covered by A. splendens or Urtica angustifolia, which contributed to greater spatial heterogeneity in plots with clustered marmot burrows. An A. splendens community was emergent among the clustered burrows but not on the isolated burrows, indicating that this species can adapt to frequent disturbance (Figs. 4, 5, 6). Unlike other species, A. splendens could adapt to fluctuating ground water levels under the clustered burrows. Clustered burrows can capture run-off from the upper mountain areas, resulting in temporal floods below the burrows, as tunnels connect the burrows underground. In arid environments, dense burrows dug by small mammals lead more water to deeper positions of the soil profile (Laundre 1993; Boeken et al. 1998). Achnatherum
splendens tussocks grow on terraces in valleys and on the sides of banks and ridges along highelevation streams and rivers, all of which experience such fluctuating ground water levels (Van Staalduinen 2005). Another reason that A. splendens was able to become established on the clustered burrows is due to the species’ high resistance to grazing. Because of its hard leaves, A. splendens is unpalatable to grazers most of the year, except to horses, and thus would be expected to increase in relative cover on heavily grazed sites (Fernandez-Gimenez and Allen-Diaz 2001). Further, this grass roots can reach depths of 1.6–2.0 m and can thus access deep water, including phreatic water. Hence, this might help A. splendens avoid the unfavorable condition of local salt accumulation derived from the sediments left after
evaporation of urine in the topsoil (Pasturage and Veterinary Department of Agricultural Ministry of
China 1996). Urtica angustifolia was also found only in the frequently disturbed area among the clustered
burrows (Fig. 4). In a study of Urtica dioica, Teckelmann (1987) noted that its growth, development,
and distribution are mainly dependent on high nutrient availability, such as is found in floodplains.
This species also prefers elevated places on depressions where flooded conditions commonly prevail but
no long-term flooding occurs (Srutek 1993, 1997). Although there were no differences in soil nutrient
properties between the off-colony and isolated-burrows plots, a marked difference was found between
the isolated-burrows and clustered-burrows plots (Table 1, Fig. 8). Thus, isolated marmot burrows do
not provide a degree of disturbance significant enough to affect soil properties, and the small amount of
nutrients left in soils may have been absorbed by plants. In contrast, large patches of bare ground
occurred near the clustered burrows due to frequent disturbance by marmots. Therefore, when marmots
excrete dung or urine on these bare patches, high nutrient concentrations would remain in soils without
significant absorption by plants. In other grassland systems, the spatial heterogeneity of soil nutrients is
primarily associated with the presence or absence of bunchgrass (Gibson 1988; Canals et al. 2003).
Figure 7 indicates that there is spatial dependence at smaller spatial scales than what can be detected with
this sampling design and at larger spatial scales than what can be detected with the total plot size used.
In the clustered-burrows plot, the spatial patterns differed among the three soil nutrients measured in
this study (Fig. 8), which can be attributed to differences in uptake rates by plants (Bachand and
Horne 2000; Reilly et al. 2000) or diffusion rates in soil. Highly heterogeneous, large patches of TN and
K likely reflect the places where these substances were excreted by marmots (Van Staalduinen and
Werger 2007). In contrast, the complex mosaic of NO3–N patches likely reflects the spatial distribution
of patches of bare ground and vegetation. Indeed, those patches with lower levels of NO3–N near the
clustered burrows had relatively abundant plant coverage above them. Previous studies indicated that
prairie dog colony sites exhibited greater net mineralization rates than sites away from the colony,
leading to enhanced inorganic nitrogen (Holland and Detling 1990; Fahnestock and Detling 2002). However,
the higher inorganic nitrogen concentration (e.g., NO3–N) in soil is absorbed by plants, resulting
in lower concentrations in soil. Alternatively, soluble NO3–N moves more rapidly than TN in soil (leaching)
because it has no significant attachment to soil solids. The coarser particles and related large macropores
in the mound soil as a result of burrowing may have accelerated vertical motion of NO3–N
(Bundy et al. 1994; Sherrod and Seastedt 2001). Overhunting has caused a sharp decline of Siberian
marmot populations in Mongolia (Adiya 2000; Wingard and Zahler 2006), resulting in greater
isolation of their burrows. Although it was unknown what process causes their burrows to be either
clustered or isolated on the landscape, the clustered burrows were often observed in highly populated
areas. In unprotected areas outside the HNP, the fraction of clustered burrows was lower than inside
the park (38% inside, 17% outside). Numerous studies have shown that, in general, spatial heterogeneity
is positively correlated with biodiversity (Huston 1994; Tews et al. 2004). Thus, through their
direct and indirect effects on spatial heterogeneity of plant communities, marmot colonies are affecting
biodiversity numerically (Yoshihara et al. 2009) or spatially (Fig. 3). Of the novel plant communities,
A. splendens can make a strong contribution to biodiversity, especially due to its complex physical
structure. Therefore, we caution that the isolation of marmot burrows that results from continued overhunting
will deteriorate this animal’s positive ecological roles in the grasslands of Mongolia.

Acknowledgments

We thank S. Takatsuki, University of Asabu, the project members, and the staff at Hustai National
Park for their kindness in helping with the field survey and other aspects of the study. We also thank T. Okayasu,
University of Tokyo, for providing comments on an earlier version of this manuscript. This work was supported financially
by the Global Environmental Research Fund of Japan’s Ministry of the Environment (G-071). The field works
performed complies with the current laws of Mongolia.

Acknowledgement

none

Reference

Addicott JF, Aho JM, Antolin MF et al (1987) Ecological neighborhoods—scaling environmental patterns. Oikos 49:340–346
Adiya Y (2000) Mongolian marmots: biology, ecology, conservation and use (in Mongolian). Mammalian Ecology
Laboratory, Institute of Biological Sciences, Mongolian Academy of Sciences, Mongolia
Andres NG, Witman JD (1995) Trends in community structure on a Jamaican reef. Mar Ecol Prog Ser 118:305–310
Augustine DJ, Frank DA (2001) Effects of migratory grazers on spatial heterogeneity of soil nitrogen properties in a
grassland ecosystem. Ecology 82:3149–3162
Bachand PAM, Horne AJ (2000) Denitrification in constructed free-water surface wetlands. I. Very high nitrate removal
rates in a macrocosm study. Ecol Eng 14:9–15
Bangert RK, Slobodchikoff CN (2000) The Gunnison’s prairie dog structures a high desert grassland landscape as a
keystone engineer. J Arid Environ 46:357–369
Behrends P, Daly M, Wilson M (1985) Range use patterns and spatial relationships of Merriams’s kangaroo rats (Dipodomys
merriami). Behavior 96:187–209
Boeken B, Lipchin C, Gutterman Y et al (1998) Annual plant community responses to density of small-scale soil disturbances
in the Negev Desert of Israel. Oecologia 114:106–117
Branch LC (1993) Intergroup and intragroup spacing in the plains vizcacha, Lagostomus maximus. J Mammal 74:890–900
Bundy LG, Knobeloch L, Webendorfer B et al (1994) Nitrate in Wisconsin groundwater: sources and concerns. Cooperative
Extension Service, Madison
Canals RM, Herman DJ, Firestone MK (2003) How disturbance by fossorial mammals alters N cycling in a California
annual grassland. Ecology 84:875–881
Collins SL (2000) Disturbance frequency and community stability in native tallgrass prairie. Am Nat 155:311– 325
Collins SL, Smith MD (2006) Scale-dependent interaction of fire and grazing on community heterogeneity in tallgrass
prairie. Ecology 87:2058–2067
Corlett RT (2000) Environmental heterogeneity and species survival in degraded tropical landscapes. In: Hutchings
MJ, John EA, Stewart A (eds) The ecological consequences of environmental heterogeneity. Blackwell Science, Oxford, pp 333–355
Daubenmire R (1959) A canopy-coverage method of vegetational analysis. Northwest Sci 33:43–66
Day KJ, John EA, Hutchings MJ (2003) The effects of spatially heterogeneous nutrient supply on yield, intensity of
competition and root placement patterns in Briza media and Festuca ovina. Funct Ecol 17:454–463
Dorrough J, Ash J, McIntyre S (2004) Plant responses to livestock grazing frequency in an Australian temperate
grassland. Ecography 27:798–810
Dufour A, Gadallah F, Wagner HH et al (2006) Plant species richness and environmental heterogeneity in a mountain
landscape: effects of variability and spatial configuration. Ecography 29:573–584
Ella VB, Melvin SW, Kanwar RS (2001) Spatial analysis of NO3–N concentration in glacial till. Trans ASAE 44:317– 327
English EI, Bowers MA (1994) Vegetational gradients and proximity to woodchuck (Marmota monax) burrows in an
old field. J Mammal 75:775–780
Fahnestock JT, Detling JK (2002) Bison-prairie dog–plant interactions in a North American mixed-grass prairie.
Oecologia 132:86–95
Fernandez-Gimenez M, Allen-Diaz B (2001) Vegetation change along gradients from water sources in three grazed
Mongolian ecosystems. Plant Ecol 157:101–118
Fransen B, de Kroon H, Berendse F (2001) Soil nutrient heterogeneity alters competition between two perennial grass
species. Ecology 82:2534–2546
Gibson DJ (1988) The relationship of sheep grazing and soil heterogeneity to plant spatial patterns in dune grassland. J
Ecol 76:233–252
Green DG (1989) Simulated effects of fire, dispersal and spatial pattern on competition within forest mosaics. Vegetatio 82:139–153
Gross KL, Pregitzer KS, Burton AJ (1995) Spatial variation in nitrogen availability in 3 successional plant communities.
J Ecol 83:357–367
Halpern CB (1988) Early successional pathways and the resistance and resilience of forest communities. Ecology
69:1703–1715
Halpern CB (1989) Early successional patterns of forest species: interactions of life-history traits and disturbance.
Ecology 70:704–720
Hanski I, Simberloff D (1997) The metapopulation approach: its history, conceptual domain, and application to conservation.
In: Hanski IA, Gilpin ME (eds) Metapopulation biology. Academic Press, San Diego, pp 5–25
Hilbig W (1995) The vegetation of Mongolia. SPB Academic Publishing, Amsterdam
Holland EA, Detling JK (1990) Plant-response to herbivory and belowground nitrogen cycling. Ecology 71:1040– 1049
Holmes WG (1984) Predation risk and foraging behavior of the hoary marmot in Alaska. Behav Ecol Sociobiol 15:293– 301
Hughes TP (1994) Catastrophes, phase-shifts, and large-scale degradation of a Caribbean coral-reef. Science 265:1547– 1551
Huston MA (1994) Biological diversity: the coexistence of species on changing. Cambridge University Press, Cambridge
Karels TJ, Koppel L, Hik DS (2004) Fecal pellet counts as a technique for monitoring an alpine-dwelling social rodent,
the hoary marmot (Marmota caligata). Arct Antarct Alp Res 36:490–494
Knapp AK, Blair JM, Briggs JM et al (1999) The keystone role of bison in north American tallgrass prairie: bison
increase habitat heterogeneity and alter a broad array of plant, community, and ecosystem processes. Bioscience 49:39–50
Laundre JW (1993) Effects of small mammal burrows on water infiltration in a cool desert environment. Oecologia
94:43–48
Levin SA (1992) The problem of pattern and scale in ecology. Ecology 73:1943–1967
Liu ZG, Li ZQ, Dong M et al (2007) Small-scale spatial associations between Artemisia frigida and Potentilla
acaulis at different intensities of sheep grazing. Appl Veg Sci 10:139–148
Lovett GM, Jones CG, Turner MG et al (2005) Ecosystem function in heterogeneous landscapes. Springer, New
York
Matejovic I (1993) Determination of carbon, hydrogen, and nitrogen in soils by automated elemental analysis (dry
combustion method). Commun Soil Sci Plant Anal 24:2213–2222
Moloney KA, Levin SA (1996) The effects of disturbance architecture on landscape-level population dynamics.
Ecology 77:375–394
North A, Ovaskainen O (2007) Interactions between dispersal, competition, and landscape heterogeneity. Oikos
116:1106–1119
Olofsson J, de Mazancourt C, Crawley MJ (2008) Spatial heterogeneity and plant species richness at different spatial
scales under rabbit grazing. Oecologia 156:825–834
Paine RT, Tegner MJ, Johnson EA (1998) Compounded perturbations yield ecological surprises. Ecosystems 1:535– 545
Pastor J, Dewey B, Moen R et al (1998) Spatial patterns in the moose-forest-soil ecosystem on Isle Royale, Michigan,
USA. Ecol Appl 8:411–424
Pasturage and Veterinary Department of Agricultural Ministry of China (1996) General Station of National Pastuage and
Veterinary, Meadow Resources of China. Agricultural Science and Technology Press, Beijing
Platt WJ, Connell JH (2003) Natural disturbances and directional replacement of species. Ecol Monogr
73:507–522
Questad EJ, Foster BL (2007) Vole disturbances and plant diversity in a grassland metacommunity. Oecologia
153:341–345
Rees M, Mangel M, Turnbull L (2000) The effects of heterogeneity on dispersal and colonization in plants. In:
Hutchings MJ, John EA, Stewart AJA et al (eds) The ecological consequences of environmental heterogeneity.
Blackwell Science, Oxford, pp 237–265
Reilly JF, Horne AJ, Miller CD (2000) Nitrogen removal from a drinking water supply with a large-scale free-surface
constructed wetlands prior to groundwater recharge. Ecol Eng 14:33–47
Reiners WA, Driese KL (2001) The propagation of ecological influences through heterogeneous environmental space.
Bioscience 51:939–950
Robertson GP, Gross KL (1994) Assessing the heterogeneity of belowground resources: quantifying scale. In: Caldwell
MM, Pearcy RW (eds) Exploitation of environmental heterogeneity by plants. Academic Press, New York, USA
Romme WH, Everham EH, Frelich LE et al (1998) Are large, infrequent disturbances qualitatively different from small,
frequent disturbances? Ecosystems 1:524–534 Sarangi A, Madramootoo CA, Enright P (2006) Comparison of
spatial variability techniques for runoff estimation from a Canadian watershed. Biosyst Eng 95:295–308
Sherrod SK, Seastedt TR (2001) Effects of the northern pocket gopher Thomomys talpoides on alpine soil characteristics,
Niwot Ridge, CO. Biogeochemistry 552:195–218
Srutek M (1993) Distribution of the stands with Urtica dioica L. along the Luznice River floodplain on the border
between Austria and Czechoslovakia and land management. Vegetatio 106:73–87
Srutek M (1997) Growth responses of Urtica dioica L. to different water table depth. Plant Ecol 130:163–169
Steinauer EM, Collins SL (1995) Effects of urine deposition on small-scale patch structure in prairie vegetation. Ecology
76:1195–1205
Steinauer EM, Collins SL (2001) Feedback loops in ecological hierarchies following urine deposition in tallgrass prairie.
Ecology 82:1319–1329
Stillwell MA, Senft R, Rittenhouse LR (1983) Total urine collection from free-grazing heifers. J Range Manag
36:798–799
Su YZ, Li YL, Zhao HL (2006) Soil properties and their spatial pattern in a degraded sandy grassland under post-grazing
restoration, Inner Mongolia, northern China. Biogeochemistry 79:297–314
Teckelmann M (1987) Kohlenstoff-, Wasser- und Stickstoffhaushalt von Urtica dioica L. an natu¨rlichen Standorten.
PhD Thesis, Bayreuth University
Tews J, Brose U, Grimm V et al (2004) Animal species diversity driven by habitat heterogeneity/diversity: the
importance of keystone structures. J Biogeogr 31:79–92
Tilman D (1999) The ecological consequences of changes in biodiversity: a search for general principles. Ecology
80:1455–1474
Turner MG, Romme WH, Gardner RH et al (1997) Effects of fire size and pattern on early succession in Yellowstone
National Park. Ecol Monogr 67:411–433
Turner MG, Gardner RH, O’Neill RV (2001) Landscape ecology in theory and practice: pattern and process.
Springer, New York
Van Staalduinen MA (2005) The impact of herbivores in a Mongolian forest steppe. PhD Thesis, Utrecht University
Van Staalduinen MA, Werger MJA (2007) Marmot disturbances in a Mongolian steppe vegetation. J Arid Environ
69:344–355
Van Staalduinen MA, During H, Werger MJA (2007) Impact of grazing regime on a Mongolian forest steppe. Appl Veg
Sci 10:299–306
Wallis de Vries MF, Manibazar N, Du¨gerlham S (1996) The vegetation of the forest-steppe region of Hustain Nuruu,
Mongolia. Vegetatio 122:111–127
White PS, Harrod J (1997) Disturbance and diversity in a landscape context. In: Bissonette JA (ed) Wildlife and
landscape ecology effects and pattern of scale. Springer, New York, pp 128–159
White PS, Jensch A (2001) The search for generality in studies of disturbance and ecosystem dynamics. Prog Bot 62:399–
449
Wingard JR, Zahler P (2006) Silent steppe: the illegal wildlife trade crisis in Mongolia: Mongolia Discussion Papers,
East Asia and Pacific Environment and Social Development Department. World Bank, USA
Wu JG (2004) Effects of changing scale on landscape pattern analysis: scaling relations. Landsc Ecol 19:125–138
Wu JG, Levin SA (1994) A spatial patch dynamic modeling approach to pattern and process in an annual grassland.
Ecol Monogr 64:447–464
Yoshihara Y, Ohkuro T, Buuveibaatar B et al (2009) Effects of disturbance by Siberian marmots (Marmota sibirica) on
spatial heterogeneity of vegetation at multiple spatial scales. Grassl Sci 55:89–95