Ecohydrological characteristics of biocrusts explain divergent responses of biocrusts and grasses to rainfall reduction

Drylands are mainly structured by both vascular plants and biological soil crusts (biocrusts), the latter often regarded as a self‐organizing component of these ecosystems. Yet how the covers of these two components respond jointly to rainfall variability remains unclear—a critical gap for predicting future dryland resilience and carbon–water dynamics. To that end, we conducted a 4‐year rainfall reduction experiment (−75%, −50%, −25% and control) in a semi‐arid grassland with a large number of replicates, allowing a robust assessment of ecosystem responses under natural conditions. We found that grass cover responded mainly to rainfall reduction treatments, showing consistent declines due to their dependence on water input. Biocrust cover was more resilient to rainfall reduction but strongly promoted by years with more frequent rainfall events, reflecting their poikilohydric nature. Ecohydrological characteristics of biocrusts—specifically their rainfall thresholds for minimum carbon gain (carbon‐balancing rainfall size) and maximum water absorption (maximum usable rainfall size)—explained these divergent responses by mediating water availability. Synthesis . Our findings show that biocrusts mediate grass water supply by intercepting a significant fraction of rainfall; failing to account for this hydrological interaction leads to a biased overestimate of plant‐available water. This underscores the need to integrate biocrust‐plant interactions into experimental and modelling frameworks to improve predictions of dryland ecosystem sustainability under climate change. Read the free Plain Language Summary for this article on the Journal blog.