pH-dependent overestimation of heavy metals mobility risk in soils: Implications for soil quality assessment

Risk Assessment Code (RAC), which treats the percentage of the Bureau Communautaire de Référence (BCR) acid-extractable fraction (P_F1) as a proxy for heavy metal (HMs) mobility, is widely used to evaluate soil HMs mobility risk. Yet it may lead to pH-dependent bias if P_F1 increases or remains constant with rising pH. However, the P_F1-pH relationship remains unclear. This study synthesizes 5151 observations from 162 studies and use nonparametric statistics, random forests, and SHapley Additive exPlanations to quantify P_F1-pH relationships for eight HMs, and assess modulation by environmental properties. Results show P_F1 increases with pH for Cr, Fe, Mn, and Pb (p<0.05); peaks near neutral pH (6.5-7.0) for Zn; shows a V-shaped trend with an inflection around neutral pH ≈ 7.0 for Cu; exhibits weak dependence for Cd and Ni. Mean annual temperature, mean annual precipitation, clay, and organic carbon modulate P_F1-pH patterns for Cu, Cd, Fe, Pb and Ni, respectively. These findings indicate RAC systematically overestimates risk under alkaline conditions, as HMs mobility declines with pH whereas P_F1 for Cr, Fe, Mn, and Pb rises. Based on these findings, we propose a conceptual fraction-environment coupling framework integrating subfraction weighting and environmental-factor correction to recalibrate RAC for more accurate soil quality assessment.