Clean water irrigation promotes microbial community recovery in acid mine drainage-contaminated paddy soil: a spatiotemporal analysis based on simulated soil column experiments from Dabaoshan mine, China

Microbial communities serve as critical bioindicators and functional drivers of soil restoration processes, particularly in mining-impacted ecosystems undergoing remediation. However, systematic insights into microbial dynamics during clean water restoration of contaminated paddy soils remain limited. This study systematically investigated, by means of column experiments, the temporo-spatial dynamics of microbial community structure and metal speciation in acid mine drainage (AMD)-contaminated paddy soil from the Dabaoshan mining area. The soil was subjected to constant flooding with clean water, including experiments with artificial AMD as a control, for 176 days. The heavy metal fractions present in the soil were determined by sequential extraction. The bacterial community was analyzed at 7 time points and 5 depths using high-throughput 16S rRNA gene amplicon sequencing of the V5-V7 region. Long-term flooding increased the dominance of <i>Firmicutes</i>, <i>Acidobacteria</i>, and <i>Proteobacteria</i>, with limited overlap in significantly enriched taxa during restoration, indicating specialized microbial adaptation or microbial selection. The metal mobility increased as a result of flooding, most strongly in the mobile fractions of Cd at 5 cm depth (FM increased from 62.6% to 68.7%) and Cu at 20 cm depth (FM increased from 16.2% to 21.6%). This was accompanied by a substantial reduction in the residual total reducible-phase Cu (the sum of Fe/Mn oxide-bound fraction F3 and the organic-matter-bound fraction F4) was reduced from 188.4 to 30.8 mg kg^-1. Likewise, residual easily migratable Cd (the sum of exchangeable fraction F1 and carbonate-bound fraction F2) was reduced from 5.8 to 0.3 mg kg^-1. Such increased mobility might present an increased environmental risk. Canonical correspondence analysis identified the pH, Cu/Cd concentrations, and SO₄^2- as primary environmental drivers (cumulative explanation: 72.3%) governing microbial community restructuring. Complementary LEfSe analysis further elucidated potential microbial interaction networks underlying the rehabilitation process. The identified microbial-metal dynamics highlight the importance of integrating biological indicators with geochemical parameters when assessing the rehabilitation efficacy in heavy metal-contaminated agricultural systems.