Context-dependent effects of microplastics on soil available phosphorus: An AP-centered multilevel meta-analysis

Microplastics are increasingly recognized as modifiers of soil nutrient availability, but their effects on soil available phosphorus (AP) remain difficult to generalize across soil-plastic contexts. Building on broader P syntheses, this AP-centered multilevel meta-analysis addressed within-study dependence, missing variance information, and AP-specific heterogeneity. The full dataset contained 597 paired observations from 79 studies, with 546 complete-SD effect sizes from 75 studies used for primary inference. Microplastic exposure was associated with an average 11.8% decrease in soil AP (lnRR = -0.126, 95% CI: -0.199 to -0.053), and this negative mean effect remained stable across complementary robustness analyses. However, the wide prediction interval showed that AP depletion was not universal. Reported background pH, particle size, and polymer identity were the clearest heterogeneity axes. Depletion was strongest in acidic soils and under exposure to particles ≤ 100 μm, while polymer identity significantly structured AP responses and the binary biodegradability classification did not. Exposure-duration class and soil sampling location showed limited explanatory power. The evidence base was concentrated in Asia, especially China, and AP extraction methods were poorly reported. Overall, microplastics shifted average AP toward depletion, but this response is best interpreted as context dependent rather than universal.