Chemical heterogeneity and mixed pollution are associated with microbial functional potential across soils in a coal mining subsidence landscape.

Coal mining subsidence occurs when underground coal extraction removes structural support for overlying strata, causing ground deformation and localized surface collapse. However, it remains unclear whether soils across a subsidence landscape respond in a similar way, or whether different habitats develop distinct nutrient conditions and mixtures of heavy metals and PAHs that are associated with different microbial communities and carbon, nitrogen, and sulfur cycling potentials. Local ecological risks would be overlooked if the entire subsidence zone is generalized as one disturbed ecosystem. Here, we sampled surface soils (0-20 cm) in September 2024 from four ordered habitats along a local subsidence associated topographic habitat in Jining, China: subsidence soil (SS), ditch soil adjacent to the subsidence area (SL), ditch soil near farmland (SR), and farmland soil (SF). We quantified soil physicochemical properties, heavy metals, polycyclic aromatic hydrocarbons (PAHs), and enzyme activities, and characterized bacterial and fungal communities and functional genes using metagenomic sequencing and downstream multivariate analyses. Soil chemical conditions differed consistently across habitats, with clear separation along nutrient related gradients and distinct contaminant profiles among habitats. Microbial α-diversity and β-diversity showed specific habitat patterns, and community composition differed among habitats for both bacteria and fungi. Functional profiles related to carbon, nitrogen, and sulfur cycling also varied across habitats, indicating that chemical heterogeneity and mixed pollution coincided with reconfiguration of microbial metabolic potential at the pathway and gene levels. Ordination analyses further showed that microbial community structure and functional potential were strongly associated with soil physicochemical gradients and pollutant variables, while enzyme activities covaried with key soil properties and contaminants. Our results indicate that coal mining subsidence landscapes contain multiple habitat-specific chemical and contaminant filters rather than a single uniform "subsidence effect", supporting chemically informed, habitat-stratified assessment of soil condition and microbial functional potential in mining-affected ecosystems.