Skip to main content
NCBI home page
RSC Advances logoLink to RSC Advances
. 2026 Sep 25. Online ahead of print. doi: 10.1039/d6ra03694e

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

Yajun Yin a,b,†, Yujiao Ma c,d,†, Ya Wang e, Peike Gao c,d, Chuanxing Jia c,d, Chuanxia Li a,b, Qing Yang f, Fanglong Zhang a,b, Huaizhi Bo a,b, Renjun Wang c,d,✉, Junfeng Chen c,d
PMCID: PMC13613652  PMID: 42798947

Abstract

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.

Coal mining subsidence occurs when underground coal extraction removes structural support for overlying strata, causing ground deformation and localized surface collapse.graphic file with name d6ra03694e-ga.webp

Introduction

Growing energy demand and increasing emphasis on sustainable development have brought the ecological consequences of resource extraction into sharp focus.1–3 Coal mining remains central to energy supply but imposes substantial and long-lasting pressures on terrestrial ecosystems, particularly on soils that underpin ecosystem stability and recovery.4,5 Coal mine subsidence occurs when underground coal extraction destabilizes the overlying strata and results in ground deformation, surface sinking, and, in high groundwater regions, seasonal or permanent inundation.5,6 As a dominant ecological disturbance during coal extraction, mining-induced subsidence triggers a series of progressive soil degradation issues: it begins with surface deformation and structural damage, progresses to altered hydrological conditions and nutrient imbalance, and ultimately leads to changes in contaminant distribution in soils.7–11 As a result, soils in mining subsidence areas often exhibit pronounced physicochemical deterioration, encompassing strong fluctuations in pH, declines in organic matter content, and intensified mineral salt leaching.8,12–15 These physical and chemical stressors generate highly heterogeneous soil environments.14,16 However, how such coupled disturbances regulate soil ecological processes and constrain ecosystem functioning remains insufficiently understood, particularly regarding the contribution of chemical stressors.

Among the multiple interacting stressors induced by coal mining subsidence, chemical contamination represents a critical entry point for understanding the mechanisms underlying soil ecological degradation. Subsidence does not directly produce soil contaminants. However, subsidence-related changes in topography, water movement, and soil conditions may affect the transport, retention, and local accumulation of pollutants from mining and other human activities. Therefore, contaminant patterns across a subsidence landscape may reflect both external pollutant inputs and their redistribution among habitats.5 In particular, heavy metals and polycyclic aromatic hydrocarbons (PAHs) are consistently recognized as important chemical pollutants in mining-affected soils.17,18 These contaminants primarily originate from mining-related material migration and occur as complex mixtures rather than isolated compounds.19,20 The ecological relevance of heavy metals and PAHs stems from their environmental persistence and recalcitrance to degradation, as well as their frequent co-occurrence and potential to interact within contaminated soils.21,22 Long-term accumulation of these pollutants in soils adversely interferes with soil structure, biological activity, and biogeochemical functioning.23,24 Importantly, the combined presence of heavy metals and PAHs can generate non-additive and synergistic chemical stress, creating complex soil environments in which the responses of ecological processes are difficult to predict.25,26 Addressing this mechanistic uncertainty requires an improved understanding of the biological processes through which co-occurring contaminants interact with soil ecosystems.

In mining-affected soils, microbial communities constitute the primary biological interface through which chemical contamination influences soil ecological processes.27,28 As core drivers of soil ecosystem functioning, microorganisms are highly susceptible to anthropogenic and environmental disturbances.29–31 Consequently, shifts in microbial diversity, community structure, and functional metabolism are commonly used to reflect changes in soil health.32,33 Previous studies have shown that chemical pollutants (including heavy metals) and land-use practices can strongly regulate microbial functional pathways, particularly those governing nutrient cycling.34,35 Chemical contamination reshapes microbial community structure by selectively favoring pollution-tolerant microbial taxa while suppressing sensitive groups, thereby altering the structural organization and stability of soil microbial assemblages.26,36 Functionally, microorganisms modulate key biogeochemical processes, encompassing carbon cycling, nitrification, denitrification, nitrogen fixation, and sulfur metabolism.37–39 Disruption of microbial functional metabolism under contaminant stress can diminish soil enzyme activities and impair nutrient cycling, ultimately exacerbating soil ecological degradation.40–42 In this context, microbial biomass and enzyme activities are widely applied as integrative indicators of microbial metabolic intensity and ecosystem stress in contaminated soils.

Soil environmental conditions in mining-disturbed landscapes often exhibit pronounced habitat heterogeneity, arising from variations in mining intensity, land-use practices, and hydrological regimes across subsidence-flooded zones, transition areas, and surrounding farmlands.10,12,43 This spatial heterogeneity, when coupled with mining-induced physicochemical alterations and chemical contamination, further complicates the regulation of soil ecological processes.27,44–46 Current research has predominantly focused on single pollutant types, isolated ecological indicators, or individual microbial attributes. Consequently, the interactive effects of multiple co-occurring environmental drivers in shaping microbial community assembly and mediating biogeochemical functions across heterogeneous mining habitats remain insufficiently clarified. In particular, the synergistic mechanisms through which habitat heterogeneity and chemical pollutants jointly influence microbial community differentiation and carbon, nitrogen, and sulfur metabolic functions remain poorly understood.

Unravelling how habitat heterogeneity and chemical pollutants jointly regulate microbial community structure and associated carbon, nitrogen, and sulfur metabolic functions is essential for explaining soil ecological processes in mining-affected landscapes. In this study, soils were collected from a coal mining subsidence area and adjacent habitats in Jining City, China. By integrating soil physicochemical properties with comprehensive assessments of microbial community diversity, structure, and carbon, nitrogen, and sulfur metabolic functions, this study aims to identify key environmental drivers and evaluate how habitat heterogeneity and chemical contaminants jointly shape microbial community assembly and functional profiles in mining-affected landscapes.

Materials & methods

Study area

The study was conducted in a coal mining subsidence area located in Taiping Town, Jining City, Shandong Province, China (35°39′–35°40′ N, 116°52′–116°53′ E). Long-term underground coal extraction has resulted in land subsidence and the formation of persistently waterlogged soils. The local terrain elevation ranges from +38 to +41 m, while the surface water level is approximately +36.4 m, yielding an average water depth of about 1.5 m in the subsidence zone. The region experiences a temperate monsoon climate, with a mean annual precipitation of 700–800 mm, more than 65% of which occurs between June and September. Winters and springs are relatively dry, leading to pronounced seasonal variability in soil moisture conditions. Prevailing winds are southeasterly in summer and northwesterly in winter. The surrounding landscape is dominated by agricultural land, primarily wheat-maize rotation systems, interspersed with forest patches and built-up areas. Agricultural irrigation mainly relies on nearby rivers and shallow groundwater. None of the sampled habitats had undergone engineering backfilling or reclamation using coal gangue, mine tailings, fly ash, or other mine-derived materials.

Sample collection and pretreatment

Soil sampling was conducted in September 2024 across four habitat types selected to represent the main topographic and hydrological settings within the local subsidence area: subsidence soil (SS), ditch soil adjacent to the subsidence zone (SL), ditch soil near farmland (SR), and farmland soil (SF), which respectively represented the lowest and most strongly subsided position, the lower transitional position, the upper transitional position, and the highest and least subsidence-affected position (Fig. 1). At each habitat, three independent composite samples were collected from the 0–20 cm soil layer. Each composite sample consisted of five subsamples collected using a five-point sampling method within a 10 m × 10 m plot to reduce microscale spatial heterogeneity. All soil samples were placed in sterile polyethylene bags, transported on ice, and delivered to the laboratory within 24 h. Visible stones, roots, and plant residues were removed manually. Each sample was homogenized and divided into two subsamples. One subsample was air-dried, ground, and sieved through a 100-mesh (150 µm) screen for physicochemical and contaminant analyses. The second subsample was stored at −80 °C for soil enzyme activity assays and metagenomic sequencing.

Fig. 1. Sampling areas and sampling point locations. The red dots in figure (a) indicate the location of the study area, while the colored dots in figure (b) indicate the locations of the various sampling points.

Fig. 1

Open in a new tab

Soil physicochemical properties

Soil pH and electrical conductivity (EC) were measured in a soil–water suspension (1 : 5, w/v) using a portable multi-parameter analyzer (WTW, Germany). Soil inorganic nitrogen forms, comprising nitrate nitrogen (NO3−–N), nitrite nitrogen (NO2−–N), and ammonium nitrogen (NH4+–N), as well as total nitrogen (TN), total phosphorus (TP), sulfate (SO42−), and soil organic matter (SOM), were determined following the National Environmental Standards of the People's Republic of China (HJ 634-2012; HJ 717-2014; HJ 632-2011; HJ 635-2012; HJ 615-2011).

Heavy metal concentrations (Cr, Cu, Pb, Ni, Cd, and As) were quantified after microwave-assisted acid digestion using inductively coupled plasma optical emission spectrometry (ICP-OES; iCAP 7400, Thermo Fisher Scientific, USA). Soil enzyme activities, including catalase (CAT), sucrase (SC), alkaline phosphatase (AKP), and urease (UE), were measured in fresh soils using commercial assay kits (Solarbio, Beijing, China).

Polycyclic aromatic hydrocarbons (PAHs) were extracted and quantified within one week after sampling using gas chromatography-mass spectrometry (GC-MS; Thermo Trace 1310 ISQ) following the national standard method HJ 805-2016. Target compounds included naphthalene, fluorene, phenanthrene, fluoranthene, pyrene, benzo[a]anthracene, benzo[b]fluoranthene, indeno[1,2,3-c,d]pyrene, dibenzo[a,h]anthracene, and benzo[g,h,i]perylene, with concentrations expressed as ng g−1 dry soil.

Soil DNA extraction and sequencing

Total soil DNA was extracted from frozen soil using the Omega Soil DNA Kit (Omega Bio-Tek, USA) according to the manufacturer's instructions. DNA quality and purity were assessed using OD260/280 and OD260/230 ratios, and DNA concentrations were quantified prior to library construction. Metagenomic libraries were prepared using the TruSeq DNA Library Preparation Kit (Illumina, USA) and quantified with a Qubit fluorometer. Paired-end sequencing (2 × 150 bp) was performed on the Illumina sequencing platform by Ling'en Biotechnology Co., Ltd (Shanghai, China). Raw sequencing reads were quality-filtered to remove low-quality reads, adaptor sequences, and ambiguous bases. Clean reads were subsequently used for metagenomic assembly, gene prediction, and functional annotation. Functional genes were annotated against the KEGG and SEED databases, and gene abundances were normalized prior to downstream analyses. Detailed metagenomic analytical pipelines are provided in the SI.

Statistical analysis

Differences in soil physicochemical properties, enzyme activities, and contaminant concentrations among habitats were assessed using one-way analysis of variance (ANOVA), followed by least significant difference (LSD) post hoc tests when normality and homoscedasticity assumptions were met. When assumptions were violated, nonparametric alternatives were applied. Statistical significance was defined at p < 0.05.

Microbial alpha diversity indices, covering Richness, Chao1, Shannon, and Simpson, were calculated using the vegan package in R. Differences in microbial alpha diversity indices among habitats were tested using the same statistical procedures as those applied to soil physicochemical properties. Beta diversity was assessed using Bray–Curtis distance matrices, and differences in community composition among habitats were tested using permutational multivariate analysis of variance (PERMANOVA).

Microbial community composition at the phylum and genus levels was analyzed and visualized on the Galaxy platform. Linear discriminant analysis effect size (LEfSe) was also performed on the Galaxy platform to identify taxa with significantly different abundances among habitats (LDA score > 4.0).

Redundancy analysis (RDA) was performed using the vegan package in R to examine relationships between microbial community composition and environmental variables. Pearson correlation analyses were conducted to assess associations between environmental factors and soil enzyme activities, with p-values adjusted using the Benjamini–Hochberg method to control for multiple comparisons. Most figures were generated using the ggplot2 package in R.

Results

Soil physicochemical properties

pH, EC, SOM, and soil nutrient pools can serve as indicators of soil condition and fertility. Soil physicochemical properties differed significantly among the four habitats (SS, SL, SR, and SF) (Fig. 2A and Table S1). The highest pH was observed in SL, whereas SF exhibited the lowest value, and SS and SR showed comparable pH levels. EC varied markedly across habitats and increased sequentially from SS to SF, spanning a range of 256.67 to 1790.00 µS cm−1. Nitrogen-related variables exhibited distinct spatial patterns. NH4+–N concentrations were consistently higher than NO3−–N across all habitats and peaked in SL, whereas the lowest NO3−–N concentration was observed in SS and the highest in SR. TN differed substantially among habitats, with SS showing the highest TN content, followed by SF, while SL and SR exhibited similarly low values. SOM displayed a pattern similar to TN, with SS exhibiting significantly higher SOM than the other habitats. SO42− concentrations decreased progressively from SS to SF. TP ranged from 0.25 to 0.49 g kg−1, with relatively higher values in SS and SF and the lowest concentration in SR. The soils were slightly to moderately alkaline, with pH values ranging from 7.14 to 8.34. This range is comparable to the alkaline conditions reported in other coal-mining subsidence soils in northern China, where soil pH values were generally close to 8.47

Fig. 2. Habitat-specific variation in soil chemical properties, contaminants, and enzyme activities across subsidence-affected landscapes. Values were log 10-transformed for visualization because the measured variables differed widely in their numerical ranges; raw values are provided in Tables S1–S4. (A) Soil physicochemical properties, including pH, electrical conductivity (EC), nutrient pools (TN, NO3−–N, NO2−–N, NH4+–N, TP), sulfate (SO42−), and soil organic matter (SOM). (B) Concentrations of heavy metals (Cr, Cu, Pb, Ni, Cd, and As). (C) Concentrations of polycyclic aromatic hydrocarbons (PAHs). (D) Activities of soil enzymes involved in oxidative stress regulation and nutrient cycling, including catalase (CAT), sucrase (SC), urease (UE), and alkaline phosphatase (AKP). Soils were collected from four habitat types: subsidence soil (SS), ditch soil near subsidence (SL), ditch soil near farmland (SR), and farmland soil (SF). Bars represent means ± standard deviations. Different lowercase letters indicate significant differences among habitats (one-way ANOVA, p < 0.05).

Fig. 2

Open in a new tab

Heavy metal concentrations

Heavy metal concentrations varied significantly among habitats (Fig. 2B and Table S2). Farmland soil (SF) exhibited the highest concentrations of Cr, Cu, Ni, and Cd, whereas SR generally showed the lowest levels. Cr concentrations ranged from 28.88 to 43.82 mg kg−1 and were significantly higher in SF than in the other habitats. Cu showed a similar distribution pattern, with the highest concentration in SF and the lowest in SR. Lead (Pb) concentrations peaked in SL, while SS, SR, and SF displayed relatively comparable levels. Arsenic (As) was detected only in SS and SR, whereas concentrations in SL and SF were below the detection limit. Most metal concentrations were below the applicable screening values; however, Cd in SF exceeded the corresponding screening value of 0.6 mg kg−1, indicating a potential agricultural-soil contamination risk that warrants further assessment (GB 15618-2018).

Polycyclic aromatic hydrocarbons (PAHs)

All 16 priority PAHs were analyzed across the four habitats. Ten compounds were detected above the reporting limit, whereas six remained below the method detection limit (<5.0 ng g−1). PAHs were detected in all soil habitats, although individual compound concentrations varied substantially among habitats (Fig. 2C and Table S3). Naphthalene concentrations ranged from 16.58 to 66.54 ng g−1, with the highest value observed in SL and the lowest in SR. Phenanthrene and benzo[a]anthracene exhibited their highest concentrations in SS. Fluoranthene and pyrene showed moderate variation among habitats, with concentrations ranging from 5.48 to 19.84 ng g−1 and from 9.34 to 25.57 ng g−1, respectively. Dibenzo[a,h]anthracene ranged from 11.76 to 19.42 ng g−1 and reached its highest concentration in SR. Several PAH compounds were detected only in specific habitats. To account for these non-detected compounds, Σ16PAHs was expressed as a lower-to-upper estimate, giving ranges of 435.25–475.25 ng g−1 in SS, 363.52–403.52 ng g−1 in SL, 225.49–275.49 ng g−1 in SR, and 256.19–301.19 ng g−1 in SF. Both estimates produced the same order of SS > SL > SF > SR, indicating that the treatment of non-detected compounds did not change the habitat pattern. Benzo[a]anthracene was the main contributor to total PAHs across all habitats. According to the concentration-based classification proposed by Maliszewska-Kordybach (1996), all four habitats fell within the weakly contaminated range of 200–600 ng g−1. The Σ16PAHs concentrations were within the lower part of the ranges reported for other Chinese coal-mining soils. These comparisons indicate weak overall PAH contamination rather than severe accumulation, although PAH composition differed among habitats.

Soil enzyme activities

Soil enzyme activities differed significantly among habitats (Fig. 2D and Table S4). Catalase (CAT) activity ranged from 0.11 to 0.14 mg (g 20 min)−1 and was highest in SF, whereas SS and SR exhibited lower activities. SC activity showed pronounced habitat-specific variation, with SS exhibiting the highest activity, followed by SL and SF, and the lowest values observed in SR. UE activity peaked in SL and was lowest in SR. AKP activity also differed markedly among habitats, with SS showing substantially higher activity than the other soils, SF exhibiting intermediate levels, and SL and SR displaying similarly low activities.

The degree and changing pattern of differences among habitats varied for different environmental factors, thus forming unique combinations of soil nutrients, pollutants, and extracellular enzyme activities.

Microbial community variability across habitats

Microbial alpha diversity indices revealed significant differences among the four soil habitats (Fig. 3). For bacterial communities, Richness and Chao1 indices indicated that SR harbored the highest species richness, whereas SL consistently exhibited lower values (Fig. 3A and B). In contrast, Shannon and Simpson indices were highest in SS and lowest in SL, reflecting differences in community evenness among habitats (Fig. 3C and D).

Fig. 3. Habitat-specific patterns of bacterial and fungal α- and β-diversity across subsidence-affected soils. (A–D) Bacterial α-diversity indices, including Richness, Chao1, Shannon, and Simpson indices. (E–H) Fungal α-diversity indices calculated using the same metrics. (I–J) Principal coordinate analysis (PCoA) of bacterial (I) and fungal (J) community composition based on Bray–Curtis dissimilarity. Soil samples were collected from four habitat types: subsidence soil (SS), ditch soil near subsidence (SL), ditch soil near farmland (SR), and farmland soil (SF). Boxes represent interquartile ranges with median values, and whiskers indicate minimum and maximum values. Different lowercase letters above boxplots denote significant differences among habitats (one-way ANOVA, p < 0.05). Percentages on PCoA axes indicate the proportion of variance explained.

Fig. 3

Open in a new tab

For fungal communities, Richness and Chao1 indices decreased in the order SR > SF > SS > SL, with SR exhibiting the highest richness and SL the lowest (Fig. 3E and F). Shannon and Simpson indices showed a similar pattern, with consistently reduced fungal diversity observed in SL (Fig. 3G and H).

Beta diversity analysis further demonstrated clear differentiation in microbial community composition among habitats. Principal coordinate analysis (PCoA) revealed that samples from the same habitat clustered closely, whereas distinct separation was observed among different habitats for both bacterial and fungal communities (Fig. 3I and J).

Microbial community composition and differentially abundant taxa

At the phylum level, bacterial communities across all habitats were dominated by Pseudomonadota, accounting for 45.88–62.21% of total sequences (Fig. 4A). Other abundant phyla included Bacteroidota, Acidobacteriota, and Actinomycetota, with their relative abundances varying among habitats. Chloroflexota was more abundant in SF and SS, whereas Cyanobacteriota and Desulfobacterota exhibited higher relative abundances in SF and SL, respectively.

Fig. 4. Habitat-specific variation in bacterial and fungal community composition of subsidence-affected soils inferred from annotated protein-coding genes in the non-redundant gene catalogue. (A) Relative abundance of dominant bacterial phyla across subsidence soil (SS), ditch soil near subsidence (SL), ditch soil near farmland (SR), and farmland soil (SF). (B) Relative abundance of the top 20 bacterial genera across the four soil habitats. (C) Relative abundance of dominant fungal phyla across the four soil habitats. (D) Relative abundance of the top 20 fungal genera across the four soil habitats. Only taxa with mean relative abundance exceeding the threshold for dominance are shown; remaining taxa are grouped as “others”. Relative abundances are expressed as percentages of total sequences within each sample.

Fig. 4

Open in a new tab

At the genus level, Burkholderia and Bradyrhizobium were dominant across all soils (Fig. 4B). SL and SR exhibited similar bacterial community composition patterns, with Algoriphagus showing relatively high abundance in both habitats. In contrast, SF and SS harbored distinct dominant genera, with Enterococcus and Ramlibacter enriched in SF and Caulobacter showing higher relative abundance in SS.

Fungal communities were dominated by Ascomycota, Mucoromycota, and Basidiomycota across habitats (Fig. 4C). In SS, Mucoromycota was the most abundant phylum, whereas SR exhibited relatively balanced contributions from Ascomycota, Basidiomycota, and Mucoromycota. At the genus level, Rhizopus and Aspergillus dominated across habitats, while habitat-specific genera were also detected, including Linnemannia in SF and Gaertneriomyces uniquely in SS (Fig. 4D).

LEfSe analysis further identified differentially abundant microbial taxa among habitats (LDA > 4.0, p < 0.05). Distinct bacterial and fungal biomarkers were detected across the four soil types, with SS and SF characterized by unique bacterial lineages, whereas SR and SL exhibited a greater number of fungal biomarkers (Fig. S1).

Microbial functional profiles related to carbon, nitrogen, and sulfur cycling

Functional gene profiles related to carbon, nitrogen, and sulfur cycling differed markedly among habitats (Fig. 5). For carbon metabolism, genes participating in glucose degradation, pyruvate conversion, and the pentose phosphate pathway exhibited habitat-specific abundance patterns, reflecting variation in functional potential (Fig. 5A and D). Genes associated with pyruvate metabolism and acetyl-CoA production showed pronounced variation among soils, whereas the gluconeogenesis-related gene pycA displayed lower relative abundance in SF.

Fig. 5. Habitat-specific differentiation of microbial functional gene profiles related to carbon, nitrogen, and sulfur cycling across subsidence-affected soils based on gene-level annotations from the CCycDB, NCycDB, and SCycDB databases. (A–C) Relative abundances of functional gene categories implicated in microbial carbon (A), nitrogen (B), and sulfur (C) cycling across subsidence soil (SS), ditch soil near subsidence (SL), ditch soil near farmland (SR), and farmland soil (SF). Values are expressed as the percentage contribution of each functional gene category to the total functional profile within each elemental cycle. Different lowercase letters indicate significant differences among habitats (p < 0.05). (D–F) Schematic representation of major microbial metabolic steps related to carbon (D), nitrogen (E), and sulfur (F) cycling. Heatmaps indicate the relative abundance of key functional genes annotated to each metabolic step across the four soil habitats (SS, SL, SR, SF), with warmer colors representing higher gene abundance. The abbreviations used in the bar charts are defined as follows: CF, carbon fixation; CR, carbon release; CB, organic carbon biosynthesis; CD, organic carbon degradation; CT, organic carbon transformation; CTp, carbon transport; OND, organic nitrogen degradation; NF, nitrogen fixation; Nit, nitrification; Denit, denitrification; DNR, assimilatory nitrate reduction; ANR, dissimilatory nitrate reduction; Si, inorganic sulfur metabolism; So, organic sulfur metabolism; SOx, sulfur oxidation; SRed, sulfur reduction; SA, sulfur assimilation; and SOX, sulfur oxidation system.

Fig. 5

Open in a new tab

Nitrogen-cycling gene profiles also varied among habitats, particularly those associated with potential nitrification and denitrification (Fig. 5B and E). The nitrification gene amoA showed lower relative abundance in SS and higher relative abundance in SL. Genes involved in denitrification, including norB and norC, were relatively concentrated in SS, whereas nir genes associated with nitrogen reduction pathways showed higher relative abundance in SF.

Sulfur metabolism pathways displayed clear habitat-specific patterns (Fig. 5C and F). Genes involved in sulfate activation, including sat, cysD, and cysN, were detected at high relative abundance across all habitats. In contrast, genes associated with sulfite and sulfide oxidation, as well as dimethyl sulfur compound metabolism, exhibited distinct distribution patterns among soil types.

Relationships between soil properties, microbial communities, and enzyme activities

Redundancy analysis (RDA) revealed strong associations between soil environmental factors and microbial community composition across subsidence-affected habitats (Fig. 6). For bacterial communities, soil physicochemical variables explained a large proportion of community variation (adjusted R2 = 0.813; Fig. 6A). EC, NO3−–N, and pH were positively associated with samples from SF and SR, whereas SOM, NH4+–N, SO42−, and TN were strongly correlated with SS. SL occupied an intermediate position along these environmental gradients.

Fig. 6. Relationships between soil environmental factors and microbial community structure in subsidence-affected soils. (A and B) Redundancy analysis (RDA) showing the relationships between soil physicochemical properties and bacterial (A) and fungal (B) community composition across subsidence soil (SS), ditch soil near subsidence (SL), ditch soil near farmland (SR), and farmland soil (SF). Arrows represent environmental variables, and their lengths indicate the strength of correlation with community composition. Percentages on axes indicate the proportion of variance explained by each RDA axis. (C and D) RDA ordinations illustrating the relationships between heavy metals and PAHs and bacterial (C) and fungal (D) community composition. Points represent individual soil samples, and arrows indicate contaminant variables related to microbial community structure. (E) Pearson correlation heatmap showing relationships between soil enzyme activities (CAT, SC, UE, AKP) and soil physicochemical properties, nutrients, heavy metals, and PAHs. Colors indicate the strength and direction of correlations, with significant correlations marked by asterisks (p < 0.05).

Fig. 6

Open in a new tab

For fungal communities, soil physicochemical variables also explained a substantial proportion of variation (adjusted R2 = 0.683; Fig. 6B). Fungal community composition in SS was primarily associated with higher SOM, SO42−, and reduced nitrogen forms, whereas communities in SF were more closely linked to elevated EC, NO3−–N, and pH.

RDA analyses incorporating heavy metals and PAHs further highlighted habitat-specific drivers of microbial community variation (Fig. 6C and D). In bacterial communities, PAHs and As were closely associated with SS, whereas Cr, Ni, Cu, and Pb showed stronger associations with SR and SF (adjusted R2 = 0.809). For fungal communities, PAHs and As were also linked to SS, while Pb and Ni were more strongly correlated with SL and SF (adjusted R2 = 0.577).

Pearson correlation analysis revealed significant relationships between soil enzyme activities and environmental variables (Fig. 6E). SC and AKP activities were strongly positively correlated with SOM, TN, and SO42−. CAT activity showed positive correlations with EC and several heavy metals. UE activity was positively associated with inorganic nitrogen forms, including NO3−–N and NH4+–N.

Discussion

Habitat heterogeneity within the coal mining subsidence landscape corresponded to distinct chemical settings and contaminant mixtures, and these differences were consistently aligned with microbial community turnover, functional potentials for C/N/S cycling, and enzyme activities. The key implication is that “subsidence-affected soils” are not a single ecological unit: different habitats appear to be constrained by different dominant filters, so microbial structure and function reorganize in habitat-specific ways.

A first-order driver is the contrast between subsidence-associated soils and farmland-associated soils in their chemical regimes. Subsidence soils were characterized by a chemical signature consistent with prolonged wet conditions and organic matter accumulation, whereas farmland soils showed patterns consistent with stronger solute enrichment and management influence. Such divergence matters because it changes the primary resource and stress landscape experienced by microbes. In reduced, organic-matter-rich soils, microbes are expected to rely more on anaerobic or low-oxygen metabolic options and to couple carbon processing with alternative electron acceptors.48–50 In contrast, higher ionic strength and nitrate-enriched conditions in farmland-associated habitats can favor taxa and functions adapted to more oxidized and nutrient-input environments. This provides a parsimonious explanation for why community composition separated strongly by habitat and why the ordination axes tracked a small set of coherent environmental gradients rather than many unrelated variables.

Contaminant mixtures represent a second, habitat-specific filter that likely reinforces community differentiation rather than acting as a uniform “pollution effect.” In this system, heavy metals and PAHs did not peak in the same habitat, implying that different locations experience different dominant contaminant suites. Because contaminant effects depend on local chemistry (e.g., pH, organic matter, ionic strength) and mixture context, the same landscape can impose qualitatively different selective pressures across short distances.51–54 Therefore, the strong alignment between contaminant profiles and community structure is best interpreted as evidence that contaminants contribute to habitat-specific selection on top of baseline geochemical constraints, rather than as a claim that any single pollutant controls the system alone.

The divergent responses of bacteria and fungi further support the view that different habitats impose different physiological constraints.55–58 Bacterial diversity patterns suggested that subsidence-associated conditions did not necessarily reduce community complexity, whereas fungal diversity was more consistently constrained near the subsidence zone. This contrast is consistent with the general expectation that many fungi are more limited by low-oxygen environments and hydrological stress, while bacteria more often include facultative or anaerobic lineages that can exploit fine-scale redox heterogeneity.59–61 The particularly low diversity observed in the ditch habitat adjacent to the subsidence zone is consistent with a “compound constraint” scenario in which multiple stressors overlap (e.g., alkalinity combined with contaminants and transitional hydrology), narrowing the set of taxa that can persist. Importantly, this interpretation does not require invoking dramatic ecosystem collapse; it simply frames the ditch habitat as a locally restrictive environment relative to other habitats in the same landscape.

Functional profiles of C/N/S cycling provide an interpretive link between habitat conditions and microbial functional potential, but they should be interpreted as metabolic potential rather than realized rates.62–65 Because metagenomic sequencing detects DNA-based functional potential, differences in the relative abundance of annotated functional genes should not be interpreted as direct evidence of gene expression, enzyme activity, or in situ metabolic fluxes. The most informative pattern is that nitrogen and sulfur functions were reconfigured across habitats in a direction consistent with the dominant chemical regimes: subsidence-associated soils showed functional signatures compatible with more reduced nitrogen transformation options, whereas habitats closer to farmland showed signatures compatible with more oxidized nitrogen pathways. Sulfur functions were broadly present but differentially distributed among habitats, consistent with spatially variable electron acceptor availability and substrate conditions. Independently measured enzyme activities provided complementary biochemical information, revealing that enzymes associated with carbon and phosphorus cycles were correlated with organic matter and nutrient pools, whereas stress-related enzyme responses were linked to ionic strength and metal exposure. Taken together, the convergence among chemistry, community structure, functional potentials, and enzyme activities supports the interpretation that habitat heterogeneity is associated with meaningful differences in microbial community structure, functional potential, and biochemical activity, without implying direct changes in biogeochemical fluxes. These findings have two practical implications for assessment and restoration. First, monitoring in subsidence landscapes should be habitat-stratified; aggregating subsidence, ditches, and farmland into a single category risks masking the dominant constraints operating in each habitat. Second, interventions should prioritize the dominant limiting factors in each habitat (e.g., managing solute stress and metal accumulation near farmland, and addressing combined wetness-contaminant contexts in subsidence-associated habitats) rather than applying a uniform remediation logic across the entire landscape.

This study also has clear boundaries. Sampling was conducted at one time point, so seasonal shifts in hydrology and agricultural inputs could modify chemical conditions and microbial responses. Functional profiles derived from metagenomes capture potential rather than activity, and direct measurements of redox status and process rates would further strengthen causal inference. Future work combining seasonal sampling with redox measurements and targeted process assays (e.g., nitrification/denitrification and sulfur transformation rates) would allow a tighter linkage from genes and enzymes to realized biogeochemical fluxes.

Conclusion

The central contribution of this study is to show that habitat heterogeneity within a coal mining subsidence landscape is linked to coupled chemical and contaminant gradients, microbial community differentiation, and variation in C/N/S functional gene profiles. Recognizing these habitat-specific filters provides a more realistic basis for diagnosing soil constraints and designing habitat-appropriate management in coal mining subsidence landscapes.

Author contributions

Software, Yajun Yin and Yujiao Ma; methodology, Yajun Yin and Yujiao Ma; data curation, Yajun Yin, Yujiao Ma, Ya Wang, Peike Gao and Junfeng Chen; formal analysis, Yajun Yin and Yujiao Ma; validation, Yajun Yin and Yujiao Ma; visualization, Yajun Yin and Yujiao Ma; supervision, Yajun Yin, Yujiao Ma and Renjun Wang; writing – original draft, Yajun Yin and Yujiao Ma; writing – review & editing, Ya Wang, Peike Gao, Chuanxing Jia, Chuanxia Li, Qing Yang, Fanglong Zhang, Huaizhi Bo, Renjun Wang and Junfeng Chen; conceptualization, Renjun Wang; funding acquisition, Renjun Wang.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Supplementary Material

RA-OLF-D6RA03694E-s001

Acknowledgments

The authors are very grateful for the financial support provided by the Foundation of Shandong Provincial Lunan Geology and Exploration Institute (Grant No. LNY202308) and Jining Key Research and Development Program (2023HHCG003).

Data availability

The data that support the findings will be available in the NCBI Sequence Read Archive at https://www.ncbi.nlm.nih.gov/sra/PRJNA1438108 following an embargo from the date of publication to allow for commercialization of research findings.

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d6ra03694e.

References

  1. Pedro J. Nicolas T. Martin W. Bowler D. E. Coscieme L. Golden A. S. et al., The direct drivers of recent global anthropogenic biodiversity loss. Sci. Adv. 2022;8(45):eabm9982. doi: 10.1126/sciadv.abm9982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Schandl H. Martinez R. M. West J. Miatto A. Lutter S. Lieber M. et al., Global material flows and resource productivity: The 2024 update. J. Ind. Ecol. 2024;28(6):2012–2031. doi: 10.1111/jiec.13593. [DOI] [Google Scholar]
  3. Sonter L. J. Dade M. C. Watson J. E. M. Valenta R. K. Renewable energy production will exacerbate mining threats to biodiversity. Nat. Commun. 2020;11(1):4174. doi: 10.1038/s41467-020-17928-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Raza M. A. Karim A. Aman M. M. Khasawneh M. A. A. Faheem M. Global progress towards the Coal: Tracking coal reserves, coal prices, electricity from coal, carbon emissions and coal phase-out. Gondwana Res. 2025;139:43–72. doi: 10.1016/j.gr.2024.11.007. [DOI] [Google Scholar]
  5. Zhang K. Liu S. Bai L. Cao Y. Yan Z. Effects of Underground Mining on Soil–Vegetation System: A Case Study of Different Subsidence Areas. Ecosyst. Health Sustain. 2023;9:0122. doi: 10.34133/ehs.0122. [DOI] [Google Scholar]
  6. Wang J. Wang P. Qin Q. Wang H. The effects of land subsidence and rehabilitation on soil hydraulic properties in a mining area in the Loess Plateau of China. Catena. 2017;159:51–59. doi: 10.1016/j.catena.2017.08.001. [DOI] [Google Scholar]
  7. Hu J. Zheng L. Liu S. Chen Y. Li C. Ni J. et al., Quantifying the impacts of coal mining activities on topsoil using Hg stable isotope: A case study of Guqiao mining area, Huainan City. Environ. Pollut. 2023;335:122378. doi: 10.1016/j.envpol.2023.122378. [DOI] [PubMed] [Google Scholar]
  8. Jing Z. Wang J. Zhu Y. Feng Y. Effects of land subsidence resulted from coal mining on soil nutrient distributions in a loess area of China. J. Cleaner Prod. 2018;177:350–361. doi: 10.1016/j.jclepro.2017.12.191. [DOI] [Google Scholar]
  9. Ouyang Z. Gao L. Yang C. Distribution, sources and influence factors of polycyclic aromatic hydrocarbon at different depths of the soil and sediments of two typical coal mining subsidence areas in Huainan, China. Ecotoxicol. Environ. Saf. 2018;163:255–265. doi: 10.1016/j.ecoenv.2018.07.024. [DOI] [PubMed] [Google Scholar]
  10. Wang J. Lu C. Sun Q. Xiao W. Cao G. Li H. et al., Simulating the hydrologic cycle in coal mining subsidence areas with a distributed hydrologic model. Sci. Rep. 2017;7:39983. doi: 10.1038/srep39983. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Xing C. Chunlu J. Liugen Z. Xianglin D. Yongchun C. Chang L. Identification of nitrate sources and transformations in basin using dual isotopes and hydrochemistry combined with a Bayesian mixing model: application in a typical mining city. Environ. Pollut. 2020;267:115651. doi: 10.1016/j.envpol.2020.115651. [DOI] [PubMed] [Google Scholar]
  12. Guo J. Zhang Y. Huang H. Yang F. Deciphering soil bacterial community structure in subsidence area caused by underground coal mining in arid and semiarid area. Appl. Soil Ecol. 2021;163:103916. doi: 10.1016/j.apsoil.2021.103916. [DOI] [Google Scholar]
  13. Li J. Huang Y. Wang G. Hu B. Long-term water and salt transport patterns during soil reconstruction in coal mining subsidence areas using activated coal gangue. Process Saf. Environ. Prot. 2025;197:107000. doi: 10.1016/j.psep.2025.107000. [DOI] [Google Scholar]
  14. Ma K. Zhang Y. Ruan M. Guo J. Chai T. Land Subsidence in a Coal Mining Area Reduced Soil Fertility and Led to Soil Degradation in Arid and Semi-Arid Regions. Int. J. Environ. Res. Public Health. 2019;16(20):3929. doi: 10.3390/ijerph16203929. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Song W. Li J. Li X. Xu D. Min X. Effects of land reclamation on soil organic carbon and its components in reclaimed coal mining subsidence areas. Sci. Total Environ. 2023;908:168523. doi: 10.1016/j.scitotenv.2023.168523. [DOI] [PubMed] [Google Scholar]
  16. Li Y. Chen L. Wen H. Zhou T. Zhang T. Gao X. 454 pyrosequencing analysis of bacterial diversity revealed by a comparative study of soils from mining subsidence and reclamation areas. J. Microbiol. Biotechnol. 2014;24(3):313–323. doi: 10.4014/jmb.1309.09001. [DOI] [PubMed] [Google Scholar]
  17. Ashkanani Z. Mohtar R. Enezi S. A. Smith P. K. Calabrese S. Ma X. et al., AI-assisted systematic review on remediation of contaminated soils with PAHs and heavy metals. J. Hazard. Mater. 2024;468:133813. doi: 10.1016/j.jhazmat.2024.133813. [DOI] [PubMed] [Google Scholar]
  18. Zhang L. Zhu Y. Zhang Y. Zhong J. Li J. Yang S. et al., Characteristics, source analysis, and health risk assessment of potentially toxic elements pollution in soil of dense molybdenum tailing ponds area in central China. Environ. Geochem. Health. 2024;46(4):129–144. doi: 10.1007/s10653-024-01886-8. [DOI] [PubMed] [Google Scholar]
  19. Hou D. Jia X. Wang L. McGrath S. P. Zhu Y. G. Hu Q. et al., Global soil pollution by toxic metals threatens agriculture and human health. Science. 2025;388(6744):316–321. doi: 10.1126/science.adr5214. [DOI] [PubMed] [Google Scholar]
  20. Luo B. Chen K. Zheng X. Wu J. Li P. Chen H. Reactive transport and retention of Cd, Pb, and Zn under coexisting multimetal-xanthate conditions in porous media. Water Res. 2025;288(Pt B):124713. doi: 10.1016/j.watres.2025.124713. [DOI] [PubMed] [Google Scholar]
  21. Yang J. Sun P. Zhang X. Wei X. Huang Y. Du W. et al., Source apportionment of PAHs in roadside agricultural soils of a megacity using positive matrix factorization receptor model and compound-specific carbon isotope analysis. J. Hazard. Mater. 2021;403:123592. doi: 10.1016/j.jhazmat.2020.123592. [DOI] [PubMed] [Google Scholar]
  22. Zhang M. Chen X. Zhu Y. Hu J. Hu M. Muhammad I. et al., The distribution of polycyclic aromatic hydrocarbons pollution and the characteristics of microbial diversity in the soil of coal mining areas. Environ. Geochem. Health. 2025;47(9):3289–3306. doi: 10.1007/s10653-025-02676-6. [DOI] [PubMed] [Google Scholar]
  23. Kong J. Huang F. Wei R. Zhang X. Zhu G. Guo Q. Study on the variation mechanism of Zn isotope in polluted farmland soil. J. Hazard. Mater. 2024;479:135561. doi: 10.1016/j.jhazmat.2024.135561. [DOI] [PubMed] [Google Scholar]
  24. Wei R. Liu Y. Kang F. Tian L. Wei Q. Li Z. et al., Impact of Rhizosphere Biostimulation on Cd Transport and Isotope Fractionation in Cd-Tolerant and Hyperaccumulating Plants Based on MC-ICP-MS and NanoSIMS. Environ. Sci. Technol. 2024;58(43):19408–19418. doi: 10.1021/acs.est.4c03674. [DOI] [PubMed] [Google Scholar]
  25. Ali M. Song X. Ding D. Wang Q. Zhang Z. Tang Z. Bioremediation of PAHs and heavy metals co-contaminated soils: Challenges and enhancement strategies. Environ. Pollut. 2021;295:118686. doi: 10.1016/j.envpol.2021.118686. [DOI] [PubMed] [Google Scholar]
  26. Thavamani P. Malik S. Beer M. Megharaj M. Naidu R. Microbial activity and diversity in long-term mixed contaminated soils with respect to polyaromatic hydrocarbons and heavy metals. J. Environ. Manage. 2012;99:10–17. doi: 10.1016/j.jenvman.2011.12.030. [DOI] [PubMed] [Google Scholar]
  27. Newsome L. Falagán C. The Microbiology of Metal Mine Waste: Bioremediation Applications and Implications for Planetary Health. GeoHealth. 2021;5(10):e2020GH000380. doi: 10.1029/2020gh000380. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Tom R. Jürgen C. Axel V. G. Nathalie V. Mohamed M. Adam W. et al., Soil microbial community structure and functionality changes in response to long-term metal and radionuclide pollution. Environ. Microbiol. 2021;23(3):1670–1683. doi: 10.1111/1462-2920.15394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. van der Heijden M. G. A. Bardgett R. D. van Straalen N. M. The unseen majority: soil microbes as drivers of plant diversity and productivity in terrestrial ecosystems. Ecol. Lett. 2008;11(3):296–310. doi: 10.1111/j.1461-0248.2007.01139.x. [DOI] [PubMed] [Google Scholar]
  30. Lisa J. Caroline D. T. Astrid H. Lieven C. Bart V. Jane D. Exploring the microbial response as a potential bio-indicator for soil health: Insights from a controlled incubator experiment. Agric., Ecosyst. Environ. 2023;356:108634. doi: 10.1016/j.agee.2023.108634. [DOI] [Google Scholar]
  31. Trivedi P. Delgado-Baquerizo M. Anderson I. Singh B. Singh B. Response of soil properties and microbial communities to agriculture: Implications for primary productivity and soil health indicators. Front. Plant Sci. 2016;7:990. doi: 10.3389/fpls.2016.00990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Debarati B. Debjani S. Arnab B. Verma B. C. Sushmita M. Biswanath D. A review on effective soil health bio-indicators for ecosystem restoration and sustainability. Front. Microbiol. 2022;13:938481. doi: 10.3389/fmicb.2022.938481. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Jia J. de Goede R. Li Y. Zhang J. Wang G. Zhang J. et al., Unlocking soil health: Are microbial functional genes effective indicators? Soil Biol. Biochem. 2025;204:109768. doi: 10.1016/j.soilbio.2025.109768. [DOI] [Google Scholar]
  34. Xi B. Yu H. Li Y. Dang Q. Tan W. Wang Y. et al., Insights into the effects of heavy metal pressure driven by long-term treated wastewater irrigation on bacterial communities and nitrogen-transforming genes along vertical soil profiles. J. Hazard. Mater. 2021;403:123853. doi: 10.1016/j.jhazmat.2020.123853. [DOI] [PubMed] [Google Scholar]
  35. Xue K. van Nostrand J. D. Vangronsveld J. Witters N. Janssen J. O. Kumpiene J. et al., Management with willow short rotation coppice increase the functional gene diversity and functional activity of a heavy metal polluted soil. Chemosphere. 2015;138:469–477. doi: 10.1016/j.chemosphere.2015.06.062. [DOI] [PubMed] [Google Scholar]
  36. Wu Y. Song Q. Wu J. Zhou J. Zhou L. Wu W. Field study on the soil bacterial associations to combined contamination with heavy metals and organic contaminants. Sci. Total Environ. 2021;778:146282. doi: 10.1016/j.scitotenv.2021.146282. [DOI] [PubMed] [Google Scholar]
  37. Shao B. Xie Y. Zhang L. Ruan Y. Liang B. Zhang R. et al., Versatile nitrate-respiring heterotrophs are previously concealed contributors to sulfur cycle. Nat. Commun. 2025;16(1):1202. doi: 10.1038/s41467-025-56588-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Song L. Wang Y. Zhang R. Yang S. Microbial Mediation of Carbon, Nitrogen, and Sulfur Cycles During Solid Waste Decomposition. Microb. Ecol. 2022;86(1):311–324. doi: 10.1007/s00248-022-02056-y. [DOI] [PubMed] [Google Scholar]
  39. Sun Y. Chen H. Chen X. Hisano M. Chen X. Reich P. B. Rising global temperatures reduce soil microbial diversity over the long term. Proc. Natl. Acad. Sci. U. S. A. 2025;122(35):e2426200122. doi: 10.1073/pnas.2426200122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Ghosh A. K. Bhattacharyya P. Pal R. Effect of arsenic contamination on microbial biomass and its activities in arsenic contaminated soils of Gangetic West Bengal, India. Environ. Int. 2004;30(4):491–499. doi: 10.1016/j.envint.2003.10.002. [DOI] [PubMed] [Google Scholar]
  41. Li Y. Lin H. Gao P. Yang N. Xu R. Sun X. et al., Variation in the diazotrophic community in a vertical soil profile contaminated with antimony and arsenic. Environ. Pollut. 2021;291:118248. doi: 10.1016/j.envpol.2021.118248. [DOI] [PubMed] [Google Scholar]
  42. Xu R. Sun X. Lin H. Han F. Xiao E. Li B. et al., Microbial adaptation in vertical soil profiles contaminated by an antimony smelting plant. FEMS Microbiol. Ecol. 2020;96(11):fiaa188. doi: 10.1093/femsec/fiaa188. [DOI] [PubMed] [Google Scholar]
  43. Li X. Du S. Hu S. Dong D. Jiang D. Cao C. et al., Simulation of surface water-groundwater interaction in coal mining subsidence areas: A case study of the Kuye River Basin in China. J. Hydrol. 2025;659:133243. doi: 10.1016/j.jhydrol.2025.133243. [DOI] [Google Scholar]
  44. Cora C. C. Seijo A. R. Rodríguez P. P. Calviño D. F. Miguel V. S. Effect of heavy metal pollution on soil microorganisms: Influence of soil physicochemical properties. A systematic review. Eur. J. Soil Biol. 2025;124:103706. doi: 10.1016/j.ejsobi.2024.103706. [DOI] [Google Scholar]
  45. Shi J. Du Y. Zou J. Ma S. Mao W. Li W. et al., Mechanisms of microbial-driven changes in soil ecological stoichiometry around gold mines. J. Hazard. Mater. 2024;465:133239. doi: 10.1016/j.jhazmat.2023.133239. [DOI] [PubMed] [Google Scholar]
  46. Xu R. Li J. Li X. Zhang J. Song W. Effect of Coal Mining Subsidence on Soil Enzyme Activity in Mining Areas with High Underground Water Levels. Water. 2024;16(12):1704. doi: 10.3390/w16121704. [DOI] [Google Scholar]
  47. Behnert A. Schiffer M. Muller-Deile J. Beck Jr L. H. Mahler M. Fritzler M. J. Antiphospholipase A(2) receptor autoantibodies: a comparison of three different immunoassays for the diagnosis of idiopathic membranous nephropathy. J. Immunol. Res. 2014;2014:143274. doi: 10.1155/2014/143274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Liu T. Wang X. Wang S. Zhu E. Hall S. J. Feng X. Iron-Driven Fast Decomposition of Soil Carbon Under Periodic Anoxia. Glob. Change Biol. 2025;31(4):e70184. doi: 10.1111/gcb.70184. [DOI] [PubMed] [Google Scholar]
  49. Zhang Z. Bo F. A. Soil redox dynamics under dynamic hydrologic regimes - A review. Sci. Total Environ. 2020;763:143026. doi: 10.1016/j.scitotenv.2020.143026. [DOI] [PubMed] [Google Scholar]
  50. Zhao Y. Liu Y. Cao S. Hao Q. Liu C. Li Y. Anaerobic oxidation of methane driven by different electron acceptors: A review. Sci. Total Environ. 2024;946:174287. doi: 10.1016/j.scitotenv.2024.174287. [DOI] [PubMed] [Google Scholar]
  51. Elumalai P. Yi X. Chen Z. Rajasekar A. Brazil D. P. T. C. Hassaan M. A. et al., Detection of Neonicotinoids in agriculture soil and degradation of thiacloprid through photo degradation, biodegradation and photo-biodegradation. Environ. Pollut. 2022;306:119452. doi: 10.1016/j.envpol.2022.119452. [DOI] [PubMed] [Google Scholar]
  52. Gauff R. P. M. Davoult D. Greff S. Bohner O. Coudret J. Jacquet S. et al., Pollution gradient leads to local adaptation and small-scale spatial variability of communities and functions in an urban marine environment. Sci. Total Environ. 2022;838(Pt 1):155911. doi: 10.1016/j.scitotenv.2022.155911. [DOI] [PubMed] [Google Scholar]
  53. Petruzzelli G. Pezzarossa B. Pedron F. The Fate of Chemical Contaminants in Soil with a View to Potential Risk to Human Health: A Review. Environments. 2025;12(6):183. doi: 10.3390/environments12060183. [DOI] [Google Scholar]
  54. Ye Z. Xu D. Zhong J. Gao S. Wang J. Zhang Y. et al., Influence of Soil Colloids on the Transport of Cd2+and Pb2+ under Different pH and Ionic Strength Conditions. Agronomy. 2024;14(2):352. doi: 10.3390/agronomy14020352. [DOI] [Google Scholar]
  55. Cao X. Liu H. Zhang R. Wen Y. Ma L. Xu Z. et al., Composition, Predicted Functions, and Co-occurrence Networks of Bacteria and Fungi in Hummock Wetlands of Northeastern Inner Mongolia, China. Microb. Ecol. 2025;88(1):34. doi: 10.1007/s00248-025-02534-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Shen C. Gunina A. Luo Y. Wang J. He J. Yakov K. et al., Contrasting patterns and drivers of soil bacterial and fungal diversity across a mountain gradient. Environ. Microbiol. 2020;22(8):3287–3301. doi: 10.1111/1462-2920.15090. [DOI] [PubMed] [Google Scholar]
  57. Yang X. Huang X. Cheng J. Cheng Z. Yang Q. Hu L. et al., Diversity-triggered bottom-up trophic interactions impair key soil functions under lindane pollution stress. Environ. Pollut. 2022;314:120293. doi: 10.1016/j.envpol.2022.120293. [DOI] [PubMed] [Google Scholar]
  58. Zheng B. Hui N. Jumpponen A. Lu C. Pouyat R. Szlavecz K. et al., Urbanization leads to asynchronous homogenization of soil microbial communities across biomes. Environ. Sci. Ecotechnol. 2025;25:100547. doi: 10.1016/j.ese.2025.100547. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Boie F. Shaheen S. M. Kandeler E. Rinklebe J. Controlled redox potentials and flooding duration affect microbial community composition and biomass in an arable soil. Soil Biol. Biochem. 2025;211:109962. doi: 10.1016/j.soilbio.2025.109962. [DOI] [Google Scholar]
  60. Lacroix E. M. Aeppli M. Boye K. Brodie E. Fendorf S. Keiluweit M. et al., Consider the Anoxic Microsite: Acknowledging and Appreciating Spatiotemporal Redox Heterogeneity in Soils and Sediments. ACS Earth Space Chem. 2023;7(9):1592–1609. doi: 10.1021/acsearthspacechem.3c00032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Xu Z. Lu Z. Zhang L. Fan H. Wang Y. Li J. et al., Red mud based passivator reduced Cd accumulation in edible amaranth by influencing root organic matter metabolism and soil aggregate distribution. Environ. Pollut. 2021;275:116543. doi: 10.1016/j.envpol.2021.116543. [DOI] [PubMed] [Google Scholar]
  62. Groth P. D. Jones B. S. Mary F. Jones H. D. Merritt T. Mark W. Abundance of microbial genes associated with nitrogen cycling as indices of biogeochemical process rates across a vegetation gradient in Alaska. Environ. Microbiol. 2012;14(4):993–1008. doi: 10.1111/j.1462-2920.2011.02679.x. [DOI] [PubMed] [Google Scholar]
  63. Li C. Hou J. Datry T. Adyel T. M. Zhou W. Wu J. et al., How river drying influences greenhouse gas emissions: insights from species and gene shifts. ISME Commun. 2025;5(1):ycaf187. doi: 10.1093/ismeco/ycaf187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Ma X. Xu H. Cao Z. Shu L. Zhu R. Will climate change cause the global peatland to expand or contract? Evidence from the habitat shift pattern of Sphagnum mosses. Glob. Change Biol. 2022;28(21):6419–6432. doi: 10.1111/gcb.16354. [DOI] [PubMed] [Google Scholar]
  65. Rocca J. D. Hall E. K. Lennon J. T. Evans S. E. Waldrop M. P. Cotner J. B. et al., Relationships between protein-encoding gene abundance and corresponding process are commonly assumed yet rarely observed. ISME J. 2015;9(8):1693–1699. doi: 10.1038/ismej.2014.252. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

RA-OLF-D6RA03694E-s001
RA-OLF-D6RA03694E-s001.pdf (559.6KB, pdf)

Data Availability Statement

The data that support the findings will be available in the NCBI Sequence Read Archive at https://www.ncbi.nlm.nih.gov/sra/PRJNA1438108 following an embargo from the date of publication to allow for commercialization of research findings.

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d6ra03694e.

Articles from RSC Advances are provided here courtesy of Royal Society of Chemistry

RESOURCES