Jinling Xu, Shuangfeng Liu, Yan Li, Qianqian Fu, Yuzhen Liu, Juan Xiong, Enfeng Liu
Straw return is widely used to recycle crop residues and improve soil phosphorus (P) availability. However, its effects on microbial biomass P (MBP), phosphatase activity, and P fractions have rarely been integrated across studies, limiting the interpretation of available P (AP) responses in relation to P resupply and fertilizer management. Here, we compiled 177 studies with 4438 paired observations and used meta-analysis and machine-learning approaches to quantify soil P responses to straw return and identify their main predictors. Straw return only modestly increased soil total P (TP) by 5.5%, but significantly increased AP and MBP by 18.1% and 47.4%, respectively. Acid and alkaline phosphatase activities increased by 21.2% and 15.7%, respectively. Within the Hedley-type fractionation dataset, responses decreased in the order labile P > moderately labile P > non-labile P. AP responded more strongly under moderate hydrothermal conditions. AP and alkaline phosphatase showed the greatest increases at AN < 50 mg kg-1. Moreover, AP, alkaline phosphatase, and labile and moderately labile P showed pronounced responses in the 20-40 cm soil layer, indicating that straw-induced P changes extend beyond surface soils. Overall, assessing AP together with MBP, phosphatase activity, and P fractions provides a basis for improving site-specific phosphorus management under residue return systems.