Abstract
In theory, changes in the amount of rainfall can change plant biomass allocation and subsequently influence coupled plant-soil microbial processes. However, testing patterns of combined responses of plants and soils remains a knowledge gap for terrestrial ecosystems. We carried out a comprehensive review of the available literature and conducted a meta-analysis to explore combined plant and soil microbial responses in grasslands exposed to experimental precipitation changes. We measured the effects of experimental precipitation changes on plant biomass, biomass allocation, and soil microbial biomass and tested for trade-offs between plant and soil responses to altered precipitation. We found that aboveground and belowground plant biomass responded asynchronically to precipitation changes, thereby leading to shifts in plant biomass allocation. Belowground plant biomass did not change under precipitation changes, but aboveground plant biomass decreased in precipitation reduction and increased in precipitation addition. There was a trade-off between responses of aboveground plant biomass and belowground plant biomass to precipitation reduction, but correlation wasn't found for precipitation addition. Microbial biomass carbon (C) did not change under the treatments of precipitation reduction. Increased root allocation may buffer drought stress for soil microbes through root exudations and neutralize microbial responses to precipitation reduction. However, precipitation addition increased microbial biomass C, potentially reflecting the removal of water limitation for soil microbial growth. We found that there were positive correlations between responses of aboveground plant biomass and microbial biomass C to precipitation addition, indicating that increased shoot growth probably promoted microbial responses via litter inputs. In sum, our study suggested that aboveground, belowground plant biomass and soil microbial biomass can respond asynchronically to precipitation changes, and emphasizes that testing the plant-soil system as a whole is necessary for forecasting the effects of precipitation changes on grassland systems.