Zokor Mounds Reshape Alpine Grasslands in Ways That Defy Simple Labels

On the Qinghai-Tibetan Plateau, a subterranean engineer is quietly rewriting the rules of grassland recovery. The plateau zokor, a fossorial rodent in the genus Myospalax, spends most of its life tunneling through the soil of alpine meadows, and in doing so it pushes fresh, mineral-rich material to the surface in the form of mounds. These mounds have long been treated in the literature and in management debates as either ecological wounds or regeneration hotspots, depending on who is asked. A new study published in Plant and Soil suggests that both camps are partly right and partly wrong, because the answer depends overwhelmingly on where you look. The research, led by Wenzhao Hu of Lanzhou University together with colleagues at the University of Münster and partner institutions in Gansu, demonstrates that the ecological signature of a newly formed zokor mound is not a fixed property of the mound itself but a product of the surrounding landscape context.

The team set out to resolve a persistent uncertainty in grassland ecology: whether the contrasts in soil, vegetation and biomass between zokor mounds and adjacent undisturbed grassland follow similar patterns across sites, or whether they vary in magnitude and direction from place to place. This question matters because alpine meadows on the Tibetan Plateau are among the most extensive high-altitude rangelands on Earth, supporting herding communities, carbon storage, water regulation and a distinctive cold-adapted flora. Degradation of these meadows, driven by a combination of overgrazing, climate change and rodent population dynamics, has prompted large-scale restoration programs across China, and rodents such as zokors and pikas have frequently been cast as pests to be controlled. Understanding whether their digging actually harms or helps vegetation recovery is therefore far from an academic exercise.

The experimental design was elegantly simple and unusually patient. In 2021, the researchers marked newly formed zokor mounds at three alpine meadow sites representing a gradient of degradation, from lightly degraded through moderately degraded to heavily degraded grassland. They then waited. Sampling took place in 2023, during the third growing season after the mounds were created, allowing the team to capture an early but not instantaneous stage of secondary succession. At each site, they compared mounds with adjacent grassland across a comprehensive suite of measurements: soil physical and chemical properties, plant diversity metrics including species richness and evenness, vegetation structure and cover, aboveground biomass, and the full composition of the plant community.

The analytical framework combined mixed-effects models, which allowed the researchers to separate mound effects from site effects and random plot-to-plot variation, with ordination techniques that visualize how communities differ in species composition. To test whether those compositional differences were statistically robust, they applied PERMANOVA, a non-parametric multivariate analysis of variance well suited to ecological count data. Finally, indicator species analysis identified which plant species were reliably associated with mounds versus adjacent grassland, a technique that goes beyond simple presence-absence comparisons to quantify the fidelity and specificity of species to particular microhabitats.

The headline finding is a study in nuance. Mounds did differ from adjacent grassland in soil properties, vegetation structure, biomass and community composition, and those differences persisted into the third growing season. But the size and pattern of the contrasts shifted substantially among the three sites. Some patterns were consistent: adjacent grassland generally supported higher species richness, greater plant cover and greater biomass than the mounds, which is unsurprising given that mounds begin as bare, freshly excavated substrate. Yet several soil variables and diversity measures displayed site-specific patterns, differing in direction as well as magnitude depending on the degradation status of the surrounding meadow. In other words, a mound in a lightly degraded meadow is not simply a smaller or larger version of a mound in a heavily degraded one; it is functionally a different microsite.

Community composition told a similar story of context dependence. Across all sites, and within each site individually, the plant assemblages on mounds were significantly distinct from those on adjacent grassland, confirming that zokor digging creates genuine ecological niches rather than mere copies of the surrounding vegetation. However, the indicator species analysis revealed an asymmetry that carries real ecological weight. Species associated with adjacent grassland were relatively numerous and consistent, reflecting the established, mature community. Species associated with mounds, by contrast, were fewer and more site-specific, suggesting that early colonizers of fresh mounds are drawn from a local pool whose identity depends on the seed rain, soil seed bank and dispersal dynamics of each particular site. The mound is a blank canvas, and what gets painted on it depends on what paints are available nearby.

These results resonate with a broader theoretical framework in plant ecology: the distinction between seed limitation and microsite limitation. Classic work on plant recruitment has shown that regeneration can fail either because propagules never arrive or because suitable safe sites for germination and establishment are absent. Zokor mounds are textbook examples of newly created microsites, patches of disturbed soil where competition from established vegetation is temporarily removed and where soil conditions differ from the surrounding sward. Whether such microsites accelerate recovery or entrench degradation depends on whether desirable species reach them and establish. The new study implies that on the Tibetan Plateau, this outcome is not uniform: in some landscape contexts mounds may function as regeneration niches that promote diversity, while in others they may be colonized by less desirable species or remain bare for longer, contributing to erosion and productivity loss.

The findings also complicate the management narrative that has dominated rodent control policy on the plateau. If mound effects were uniformly negative, poisoning or exclusion programs could be justified on straightforward ecological grounds. If they were uniformly positive, zokors could be embraced as restoration allies. The reality documented by Hu and colleagues is more awkward and more interesting: mounds are neither uniformly beneficial nor uniformly detrimental, but context-dependent early-stage microsites. In lightly degraded meadows, where propagule availability and soil conditions are relatively intact, mounds may recover quickly and contribute to the fine-scale heterogeneity that sustains biodiversity. In heavily degraded sites, the same digging may produce persistent bare patches in a landscape already short of vegetation cover, potentially amplifying the very degradation managers are trying to reverse. A single blanket policy for zokor control or tolerance is therefore unlikely to serve all grassland situations equally well.

Methodologically, the study offers a model for how to study bioturbation, the physical reworking of soils by animals, in a rigorous way. By marking mounds at the moment of creation and returning across multiple growing seasons, the researchers avoided the confounding inherent in sampling mounds of unknown age, a weakness that has plagued earlier work. The multi-site degradation gradient allowed them to test for context dependence directly rather than assuming homogeneity, and the combination of mixed-effects modeling, ordination, PERMANOVA and indicator species analysis triangulated the mound effect from complementary angles. The authors have also made their data and R scripts publicly available through Figshare, supporting reproducibility and reuse. Earlier studies on zokor mounds, including four-year vegetation dynamics on subalpine meadow mounds and work linking mound density to plant diversity across rangeland types, had hinted at variability, but this study places that variability at the center of the interpretation rather than treating it as noise.

What remains open is the longer trajectory. Three growing seasons capture the earliest phase of succession, when differences between mounds and grassland are expected to be at their widest. Prior work on related systems, including molehill succession in European meadows and gopher mounds in North American cascades, suggests that community convergence can occur over time, though the pace and endpoint depend on environmental drivers. Whether zokor mounds on the Tibetan Plateau converge toward the adjacent community, persist as distinct patches, or follow divergent trajectories depending on site degradation is a question the authors’ marked-mound design is well placed to answer in future resurveys. For now, the message for ecologists and land managers is clear: the humble zokor mound is not a fixed ecological object but a lens through which the condition of the surrounding landscape is revealed. Reading it correctly requires looking not just at the mound, but at everything around it.

Subject of Research: Context-dependent effects of plateau zokor mounds on plant succession, biomass and soil properties in alpine grasslands

Article Title: Effects of zokor mounds on plant succession, biomass and soil properties in alpine grasslands are strongly context dependent

Article References: Hu, W., Hölzel, N., Zhao, J., Wang, S., Li, Y., Ma, Y., Yu, H., Yang, H., Chen, X., An, H., Liu, Z., Shi, X., Neuenkamp, L., Heim, R. J., & Zhao, Z. (2026). Effects of zokor mounds on plant succession, biomass and soil properties in alpine grasslands are strongly context dependent. Plant and Soil. https://doi.org/10.1007/s11104-026-09054-w

Image Credits: AI Generated

DOI: 10.1007/s11104-026-09054-w

Keywords: plateau zokor, alpine meadow, Qinghai-Tibetan Plateau, bioturbation, plant succession, soil properties, plant community composition, biomass, microsites, grassland degradation, indicator species, ecological restoration