Groundwater is vital to local human life and agricultural irrigation, and the quality of the water is critical to human health. As a result, it is critical to investigate the hydrochemical evolution and water quality of groundwater in the Sanjiang Plain. There were 259 samples obtained. Furthermore, hydrogeochemical simulation was performed to highlight groundwater’s hydrochemical features, evolution process, and water quality. The analytical results show that the groundwater in the study area is somewhat alkaline with a mean TDS of 285.94 mgL⁻¹ and the primary contributing ions being Ca²⁺ and HCO₃⁻. The closer the concentration of TDS and NO₃⁻ is to the city, the higher the concentration, indicating that the chemical composition of the water body has been affected by certain human activities. The Piper diagram, Gibbs diagram, and correlation analysis results demonstrate that the chemical type of groundwater is mostly HCO₃-Ca and the hydrochemistry is primarily regulated by weathering and carbonate and silicate dissolution. According to the entropy-weighted water quality index, the groundwater quality in this location is pretty acceptable. This study could help strengthen groundwater quality monitoring based on local conditions, identify the source of nitrate, provide data support for the safe use of local water resources, and serve as a reference for global water chemical evolution and water quality evaluation in cold regions.

Riverine islands are widespread alluvium wetlands developed in large rivers, and bacterial communities are crucial to their ecological function, yet their assembly processes are rarely addressed. The ecosystem services provided by the middle and the lower Yangtze are primarily threatened by pollution discharge from agricultural land use, and resource overutilization (e.g., embankments), respectively. Here, we assessed bacterial community assembly processes and their drivers within riverine islands in the middle Yangtze River (MR islands) and those in the lower reach (LR islands). A significant distance–decay relationship was observed, although the turnover rate was lower than that of the terrestrial ecosystem with less connectivity. Deterministic and stochastic processes jointly shaped community patterns, and the influence of stochastic increased from 26% in MR islands to 59% for those in LR islands. Meanwhile, the bacterial community in MR islands was controlled more by inorganic nitrogen availability, whereas those in LR islands were governed by pH and EC, although those factors explained a limited fraction of variation in the bacterial community. Potential indicator taxa (affiliated with Nocardioides and Lysobacter ) characterized the waterway transport pollution. Overall, our study demonstrated that bacterial community dissimilarity and the importance of dispersal limitation increased concurrently along the flow direction, while distinct local factors further determined bacterial community compositions by selecting habitat-specificity taxa and particularly metabolism function. These findings enhanced our understanding of the mechanisms driving changes in bacterial communities of riverine islands subject to increased anthropogenic impacts. IMPORTANCE: Rivers are among the most threatened ecosystems globally and face multiple stressors related to human activity. However, linkages between microbial diversity patterns and assembly processes in rivers remain unclear, especially in riverine islands developed in large rivers. Our findings reveal that distinct factors result in divergent bacterial community compositions and functional profiles in the riverine islands in the middle Yangtze and those in the lower Yangtze, with substantial differentiation in deterministic and stochastic processes that jointly contribute to bacterial community assemblages. Additionally, keystone species may play important metabolic roles in coping with human-related disturbances. This study provides an improved understanding of relationships between microbial diversity patterns and ecosystem functions under environmental changes in large river ecosystems.

Karst rocky desertification (KRD) represents one of the most severe forms of land degradation globally. Soils in these regions are characterized by structural instability, severe nutrient depletion, and fragmented microbial networks, all of which collectively impose formidable constraints on natural biogeochemical recovery. In this study, we experimentally induced biological soil crusts (BSCs) through the use of moss, algae, and their co-inoculation, which represents a composite BSCs construction strategy, under both controlled and field conditions. We specifically examined how inoculation strategy regulates carbon–nitrogen coupling, nutrient pools, enzyme activities, dissolved organic matter transformation, and microbial community reorganization over 90 days. Results showed that co-inoculation acted as the dominant driver of early-stage ecosystem restoration. It increased Chl a up to tenfold, enhanced total organic carbon, and elevated nitrate concentrations by up to fivefold under controlled conditions, accompanied by a 40–43% rise in urease activity. Coordinated increases in sucrase, urease, and alkaline phosphatase indicated enhanced microbial C-N-P acquisition. Concurrently, DOM shifted toward greater microbial contribution with early-stage humification signals. Microbial richness more than doubled, communities transitioned from unicellular to filamentous nitrogen-fixing cyanobacteria, and saprotrophic Ascomycota became dominant, alongside enrichment of nitrogen-reduction and complex carbon-degradation pathways. Collectively, these findings demonstrate that induced BSCs promote early-stage nutrient accumulation and microbial community reorganization in soils within KRD areas. Moss-cyanobacterial co-inoculation consistently outperformed single inoculation treatments under both indoor and field conditions. The study underscores the potential of induced BSCs technology as a promising ecological strategy for the early restoration of degraded soils within KRD area.

River sediments are vital organic carbon reservoirs, yet the formation mechanisms of mineral-associated organic carbon (MAOC) in karst spring-fed rivers remain poorly understood. Here, we investigated sediments, water chemistry, submerged macrophytes, and sediment-extracted dissolved organic matter (DOM) from five typical karst spring‐fed rivers in the Guilin and Jianshui regions of Southwest China. Results showed that the total organic carbon (TOC) in sediments averaged 32.21 ± 24.07 g/kg, and MAOC averaged 5.45 ± 2.68 g/kg. While regional TOC and MAOC stocks were similar, the dominant MAOC fractions showed clear regional divergence. Guilin sediments were characterized by higher iron‐bound organic carbon (Fe‐OC), whereas Jianshui sediments showed greater calcium‐bound organic carbon (Ca‐OC). Random forest analysis (RF) identified δ13Cw, BIX, and SpCw as the main predictors of MAOC variation; SpCw, DOw, SR, and SUVA254 as key predictors of Fe-OC; and δ13Cw, DOC, and SUVA254 as important predictors of Ca-OC. Piecewise structural equation modeling (pSEM) further supported a statistical pathway linking water chemistry, submerged macrophytes, and sediment-extracted DOM properties and MAOC composition. Specifically, high dissolved oxygen together with aromatic‐rich DOM in Guilin were linked to Fe‐OC enrichment, whereas high ionic strength and low‐molecular‐weight DOM in Jianshui were associated with Ca‐OC enrichment. These findings indicate that similar MAOC stocks in karst spring-fed river sediments can be maintained through contrasting Fe- and Ca-associated carbon allocation patterns. Our study highlights the need to move beyond bulk carbon stock assessment and consider carrier-specific MAOC composition when evaluating the stability and management potential of karst sedimentary carbon pool.