Divergent effects of litter manipulation and nitrogen enrichment on aboveground-belowground stability are mediated by soil biodiversity in contrasting grassland states

Abstract Grassland degradation poses a severe threat to semi-arid ecosystems. However, the interactive effects of litter management and nitrogen (N) enrichment on ecosystem stability remain unclear, especially when jointly considering above- and belowground processes. We coupled N enrichment with a seven-year asymmetric litter manipulation experiment in Inner Mongolian grasslands. Over three consecutive years, we assessed how these drivers influenced plant and soil biodiversity and the stability of above- and belowground multifunctionality. In the undegraded grassland, both litter removal and N enrichment reduced bacterial diversity and belowground functionality, yet whole-system stability remained largely unchanged. In the degraded grassland, however, N enrichment reduced belowground functional stability by simplifying soil communities and amplifying functional variability, whereas litter addition buffered these destabilizing effects, consistent with improved moisture retention. Notably, soil biodiversity, rather than plant diversity, emerged as the primary biotic regulator of stability, but the dominant stabilizing taxa shifted with grassland state: nematode and bacterial diversity underpinned stability in the undegraded and degraded grassland, respectively. Above- and belowground stability were decoupled, with belowground processes exhibiting greater sensitivity to both litter and N manipulation. Consequently, management for stability in semi-arid grasslands should explicitly prioritize soil biodiversity over plant diversity alone. In undegraded systems, minimizing litter removal better safeguards belowground functioning. In degraded systems, strategic litter addition enhances resistance to N enrichment by alleviating moisture limitation but continued N inputs risk long-term destabilization via soil biotic homogenization. Monitoring soil bioindicators (e.g., nematode trophic diversity and functional bacterial groups) can provide early warnings and guide adaptive management strategies.