Summary 1.The concept of threshold has become important in ecology, but the nature of potential threshold responses of vegetation to grazing in rangeland ecosystems remains poorly understood. We aimed to identify ecological thresholds in vegetation changes along a grazing gradient and to examine whether threshold changes were expressed similarly at a variety of ecological sites. 2. To accomplish this, we surveyed the vegetation along grazing gradients at 10 ecological sites, each located at different landscape positions in Mongolia's central and southern rangelands. Evidence for a threshold in changes in floristic composition along the grazing gradient was examined by comparing linear models of the data with nonlinear models fitted using an exponential curve, an inverse curve, a piecewise regression and a sigmoid logistic curve. 3. Three nonlinear models (piecewise, exponential and sigmoid) provided a much better fit to the data than the linear models, highlighting the presence of a discontinuity in vegetation changes along the grazing gradient. The shapes of the best-fit models and their fit to the data were generally similar across sites, indicating that the changes in floristic composition were relatively constant below a threshold level of grazing, after which the curve changed sharply. 4. Except for two sites, the best-fit models had relatively narrow bootstrap confidence intervals (95% CI), especially around threshold points or zones where the rate of change accelerated, emphasizing that our results were robust and conclusive. 5. Synthesis . Our study provided strong evidence for the existence of ecological thresholds in vegetation change along a grazing gradient across all ecological sites. This suggests that vegetation responses to grazing in the study areas are essentially nonlinear. The recognition that real threshold changes exist in real grazing gradients will help land managers to prevent the occurrence of undesirable states and promote the occurrence of desirable states, and will therefore permit a major step forward in the sustainable management of rangeland ecosystems.

The concept of functional redundancy is at the core of theory relating changes in ecosystem functioning to species loss. However, few empirical studies have investigated the strength and form of the relationship between species and functional diversity (i.e., the presence of functional redundancy in ecological communities) in this context. In particular, we know little about how local extinctions in real communities might impact functional diversity. Here, we examined the relationship between species and functional diversity in plant communities along a grazing gradient across Mongolian rangeland ecosystems. We applied a recently described measure of functional diversity that incorporates species' dissimilarities defined from plant functional traits and tested several hypothesized forms of the relationship between species and functional diversity using linear and nonlinear modeling techniques. We found a significant sigmoid logistic relationship between species richness and functional diversity in relatively benign environmental conditions. This indicates high functional redundancy at low levels of species richness followed by a rapid increase at intermediate levels, until functional diversity reaches an asymptote at high levels (i.e., two-phase functional redundancy). In contrast, we generally observed a positive linear relationship between these parameters in relatively harsh environmental conditions, indicating low functional redundancy. Observed functional redundancy probably resulted from two factors, intrinsic redundancy in species' functional traits and extrinsic redundancy caused by nonrandom compositional change that is nonrandom with respect to functional traits. Lack of either intrinsic or extrinsic redundancy may result in low functional redundancy. Two-phase functional redundancy suggests that functional traits are abruptly lost from a community below a certain level of species richness, and a community then shifts into a contrasting state that has a few limited functional groups characterized by disturbance-resistant traits, as a consequence of disturbances such as livestock grazing. This study represents a major step forward in predicting the consequences of livestock grazing on the functioning of Mongolian rangeland ecosystems.

The current growing body of evidence for diversity-disturbance relationships suggests that the peaked pattern predicted by the intermediate disturbance hypothesis (IDH) may not be the rule. Even if ecologists could quantify the diversity-disturbance relationship consistent with the IDH, the applicability of the IDH to land management has rarely been addressed. We examined two hypotheses related to the generality and management applicability of the IDH to Mongolian rangeland ecosystems: that the diversity-disturbance relationship varies as a function of landscape condition and that some intermediate scales of grazing can play an important role in terms of sustainable rangeland management through a grazing gradient approach. We quantified the landscape condition of each ecological site using an ordination technique and determined two types of landscape conditions: relatively benign and harsh environmental conditions. At the ecological sites characterized by relatively benign environmental conditions, diversity-disturbance relationships were generally consistent with the IDH, and maximum diversity was observed at some intermediate distance from the source of the grazing gradient. In contrast, the IDH was not supported at most (but not all) sites characterized by relatively harsh environmental conditions. The intermediate levels of grazing were generally located below the ecological threshold representing the points or zones at which disturbance should be limited to prevent drastic changes in ecological conditions, suggesting that there is little "conundrum" with regard to intermediate disturbance in the studied systems in terms of land management. We suggest that the landscape condition is one of the primary factors that cause inconsistencies in diversity-disturbance relationships. The ecological threshold can extend its utility in rangeland management because it also has the compatibility with the maintenance of species diversity. This study thus suggests that some intermediate scales of grazing and ecological thresholds are mutually supportive tools for sustainable management of Mongolian rangelands.

No abstract

We focused on responses to grazing by individual species and functional groups in relation to ecological thresholds in Mongolian rangelands, with repeated measures from the same ecological sites to account for rainfall variability. At all sites, even under rainfall fluctuations, there were robust combinations of indicator species that could be used to forewarn managers to take action to minimize the probability of crossing ecological thresholds. Depending on the landscape condition of each site, the cover of functional groups, which shared traits of perennial life history, grass or forb growth form, linear leaf shape, and alternate leaf attachment, or the cover of functional groups of woody shrubs dramatically decreased before an ecological threshold was crossed.Thus, across all sites, the responses of certain functional groups to grazing appeared to predict the crossing of an ecological threshold. The ecological indicators derived in this study should help to improve land managers' ability to prevent adverse changes in states before ecological thresholds are reached.

We focused on land units as landscape characteristics and selected seven typical land units on a land catena comprising two areas of southern Mongolia. Hierarchical analysis was used to test the hypothesis that a land unit's edaphic factors could explain the differences in vegetation responses to grazing. We established the survey sites at increasing distances from a livestock camp or water point within each land unit, then analysed patterns of change in floristic and functional compositions, vegetation volume and soil properties within each land unit to reveal differences in vegetation responses to grazing. We also examined the variations in floristic and functional compositions across land units to identify the edaphic factors that may underlie these differences. Changes in vegetation and soil properties at increasing distances from a camp or water point within each land unit were into three different patterns. Ordination techniques consistently indicated that land unit groups categorised using edaphic factors corresponded to those categorised using response patterns. Our study revealed that edaphic factors were responsible for the observed landscapescale differences in vegetation responses to grazing in the study areas. In addition, the mechanisms underlying vegetation responses to grazing may have been primarily determined by edaphic factors.

We surveyed the floristic composition and analyzed the patterns of floristic composition under different grazing intensities in the semiarid area of Bulgan, in the South Gobi, Mongolia. We sampled 90 quadrats at six points (15 quadrats each) along a transect from a spring and grouped them into five vegetation types by using TWINSPAN. All or most of the quadrats at each point corresponded to one of the five vegetation types. DCA axis 1 arranged all the quadrats in order of grazing intensity, and the floristic composition changed dramatically in a lopsided stepwise‐pattern along this axis. The composition ratio for five groups clustered by TWINSPAN according to plant functional type (toxicity or palatability, and life‐form) and family varied to a greater or lesser extent. Background environmental factors such as soil moisture or landform did not affect the pattern of the floristic composition at this site. The patterns of floristic composition were determined by grazing intensity (determined as distance from the spring) together with palatability, which was the dominant environmental effect at the site.

With our enhanced understanding of the factors that determine biodiversity and assemblage structure has come increasing acknowledgment that the use of an appropriate disturbance regime to maintain spatial heterogeneity is an effective conservation technique. A herbivore's behavior affects its disturbance regime (size and intensity); this, in turn, may modify the associated spatial heterogeneity of plants and soil properties. We examined whether the pattern of spatial disturbance created by the Siberian marmot (Marmota sibirica) affects the spatial heterogeneity of vegetation and soils at a colony scale on the Mongolian steppe. We expected that the difference in management between two types of area (protection against hunting marmots vs. hunting allowed) would result in different behavioral patterns; therefore, we estimated the patterns of spatial disturbance separately in protected and unprotected areas. We then surveyed plant communities and soil nutrients in these areas to assess their spatial heterogeneity. We found that disturbance of both vegetation and soil was more concentrated near marmot burrows in the unprotected area than in the protected area. In addition, the degrees of spatial heterogeneity of vegetation and soil NO(3)-N were greater in the unprotected area than in the protected area, where disturbance was more widely distributed. These results indicate that the spatial pattern of disturbance by herbivores affects the spatial heterogeneity of vegetation and soil properties through changes in the disturbance regime. Our findings also suggest that the intensity of disturbance is more important than its size in determining community structure in Mongolian grasslands.