Litter decomposition is a key ecosystem process that drives carbon (C) and nutrient cycling, which could be affected by shifts in plant community composition caused by plant invasion or expansion. However, how changes in leaf litter composition (e.g., litter-mixing effect) and soil microbial community induced by shift in plant community composition affect decomposition remains elusive.

Here, by deploying 432 litterbags (50-μm mesh screen), we treated leaf litter of bamboo (Phyllostachys edulis­) and tree species (in a mixture or alone) and decomposed them in forest sites expanded or not expanded by bamboo at three regions, to assess the effects of bamboo expansion on decomposition, microbial community structure and function.

Bamboo expansion reduced microbial C to nitrogen (N) ratio and nutritional stress indicator (cy/pre), which was very likely ascribed to elevated N availability as a result of accelerated litter N release. The N loss of bamboo litter (28.1 ± 03%) was more than double of tree species litter (13.2 ± 3.1%), and decomposing litter of bamboo and tree species in a mixture increased litter N loss from 15.3± 3.1% to 25.9± 3.1% (by 69%). However, no consistent effect of litter type and litter treatment on C loss was observed. The non-additive litter-mixing effect is attributable to increased microbial nutrient use efficiency, which is illustrated by the interactive effect between litter treatment and cy/pre on decomposition. Moreover, the forest type interactively affected decomposition with cy/pre as ignoring litter treatment, where higher decay rate and litter N loss at bamboo expanded than non-expanded sites would be expected when both sites having comparable cy/pre, but the reality is that bamboo expansion tends to have lower cy/pre. These results imply a trade-off between introduced litter and nutrient use efficiency of local microbial community to drive decomposition.

Our results demonstrate that litter decomposition affected by bamboo expansion was jointly modulated by litter-mixing and microbial composition, which sheds new insights on predicting changes in soil C and nutrient cycling induced by shifts in plant community composition.