Species richness is the core of biodiversity-ecosystem functioning (BEF) research. Nevertheless, it is difficult to accurately predict changes in plant species richness under different climate scenarios, especially in alpine biomes. In this study, we surveyed plant species richness from 2009 to 2017 in 75 alpine meadows (AM), 199 alpine steppes (AS), and 71 desert steppes (DS) in the Tibetan Autonomous Region, China. Along with 20 environmental factors relevant to species settlement, development, and survival, we first simulated the spatial pattern of plant species richness under current climate conditions using random forest modelling. Our results showed that simulated species richness matched well with observed values in the field, showing an evident decrease from meadows to steppes and then to deserts. Summer precipitation, which ranked first among the 20 environmental factors, was further confirmed to be the most critical driver of species richness distribution. Next, we simulated and compared species richness patterns under four different precipitation scenarios, increasing and decreasing summer precipitation by 20% and 10%, relative to the current species richness pattern. Our findings showed that species richness in response to altered precipitation was grassland-type specific, with meadows being sensitive to decreasing precipitation, steppes being sensitive to increasing precipitation, and deserts remaining resistant. In addition, species richness at low elevations was more sensitive to decreasing precipitation than to increasing precipitation, implying that droughts might have stronger influences than wetting on species composition. In contrast, species richness at high elevations (also in deserts) changed slightly under different precipitation scenarios, likely due to harsh physical conditions and small species pools for plant recruitment and survival. Finally, we suggest that policymakers and herdsmen pay more attention to alpine grasslands in central Tibet and at low elevations where species richness is sensitive to precipitation changes.
