Current Issue

2026, Volume 33 Issue 1
25 January 2026 Previous Issue   

---

For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Select

Characteristics of surface water-groundwater interaction in the fractured riverbank of the Gezhouba Reservoir area WEN Zhang, LI Yiming, GUO Xulei, WAN Tan, LUO Qingshu, ZHOU Hong 2026, 33(1): 1-13.  DOI: 10.13745/j.esf.sf.2025.10.5 Abstract ( 27 )   HTML ( 4 )   PDF (11014KB) ( 17 )   Surface water-groundwater interaction is a critical driver of basin-scale water cycling and solute transport. Its mechanisms hinge on fine-scale characterization of aquifer structure and hydrodynamics within riverbank zones. While previous studies have concentrated on alluvial banks, the exchange processes in fractured-bedrock riverbanks remain poorly understood due to geological heterogeneity and complex flow paths. Taking a representative fractured riverbank on the south shore of the Gezhouba Reservoir as a test site, we established a borehole monitoring network to characterize the groundwater response to Yangtze River stage fluctuations and developed a three-dimensional (3-D) coupled fracture-karst numerical model to quantify water exchange within the dual-layer fractured system. Key findings include: (1) The groundwater system consists of a dual-layer fractured aquifer intersected by three large karst conduits, with groundwater dynamics strongly regulated by river stage; (2) Discharge from the deeper layer exceeds that from the shallow layer; (3) The large conduits serve as the dominant pathways for groundwater storage, migration, and exchange with the Yangtze River, with discharge rates three orders of magnitude higher than the fractured aquifer matrix; (4) These conduits mediate inter-layer water transfer between the shallow and deep fractured aquifers. By quantifying exchange fluxes and conduit effects in the fractured-karst riverbank system, this study provides a transferable framework and technical paradigm for refined water-resource management at bedrock riverbanks and to support conservation efforts in the Yangtze River. Figures and Tables | References | Related Articles | Metrics

Select

Adsorption-desorption behavior and release flux of nitrogen and phosphorus in sediments of the reservoir drawdown zone under water level fluctuations WU Mohan, SU Xiaosi, SONG Tiejun, HAO Yuan 2026, 33(1): 14-24.  DOI: 10.13745/j.esf.sf.2025.10.35 Abstract ( 18 )   HTML ( 3 )   PDF (3643KB) ( 17 )   The drawdown zone, formed by water-level fluctuations in reservoirs, represents a critical terrestrial-aquatic interface. The release of nitrogen and phosphorus from its sediments significantly influences reservoir water quality. However, the alternating redox conditions caused by periodic water-level fluctuations lead to poorly understood mechanisms governing the release of these nutrients and their contribution to water quality. This study was conducted in a large-scale reservoir in Jilin Province. Water and sediment samples from the drawdown zone were collected during the pre-flood, flood, and post-flood periods to investigate the release patterns of nitrogen and phosphorus in response to water-level fluctuations and their impact on reservoir water quality. The results showed that during the water-level fluctuation cycle, the concentrations of ammonia nitrogen (${\mathrm{NH}}_{4}^{+}$-N) and phosphate (${\mathrm{PO}}_{4}^{3-}$-P) in the reservoir water first increased and then decreased. In the sediments, ammonia nitrogen content initially decreased and then increased, while inorganic phosphorus content showed an opposite trend. Adsorption-desorption experiments revealed that at pH 5, the sediments had a high adsorption capacity for both ammonia nitrogen and phosphate. In contrast, at pH 9, the desorption of nitrogen and phosphorus was enhanced. Notably, during the post-flood period, the sediments exhibited significant desorption behavior. Mass balance calculations indicated that the release fluxes of nitrogen and phosphorus from the sediments were 324.15 t and 8.18 t, accounting for 47.22% and 57.72% of the variations in inorganic nitrogen and phosphorus concentrations in the surface water, respectively. This study provides crucial scientific evidence for identifying internal pollution mechanisms in reservoirs, predicting water quality risks, and supporting sustainable water resource management and drinking water safety. Figures and Tables | References | Related Articles | Metrics

Select

Study on the critical thresholds of identifying the transitions of dominant controlling factors for CO2 production in vertical hyporheic zones ZHOU Shiyu, YANG Yiqun, DAI Junyi, GAO Di, LI Shenyan, WANG Lichun 2026, 33(1): 25-38.  DOI: 10.13745/j.esf.sf.2025.10.10 Abstract ( 14 )   HTML ( 3 )   PDF (7534KB) ( 3 )   Riverine carbon emissions play a crucial role in the global carbon cycle, where the hyporheic zone (HZ) serves as a key component that significantly influences CO2 emission. The production of CO2 within the HZ is strongly affected by dynamic factors such as river temperature (T) and dissolved oxygen (DO). However, most existing models assumed that T and DO were time-invariant, leading to an inaccurate capture of the temporal variability of these processes, which ultimately caused significant uncertainty in riverine carbon emission estimation. To address this problem, this study developed a coupled model integrating physical and biogeochemical processes by including the periodic variations of T and DO for a representative bedform-induced HZ. The model was numerically solved using COMSOL Multiphysics, and the dynamic impacts of river T and DO on CO2 production within the HZ were analyzed using the Damköhler number (Da) and correlation coefficients. Our results indicate that both T and DO competitively regulate CO2 production rate. Moreover, when the mean residence time of HZ is less than 15.7 h, a critical temperature threshold (Tc) exists, above or below which the dominant role in regulating CO2 production rate shifts between T and DO. Specifically, when the mean river temperature is below Tc, the CO2 production is primarily controlled by T fluctuations, whereas above Tc, it is dominated by DO fluctuations. Notably, this shift in dominant role disappears when the mean residence time exceeds 15.7 h. Our study reveals how T and DO fluctuations impact dynamics of CO2 production in the HZ, and could improve the assessment of the HZ’s contribution to the global carbon cycle, particularly for predicting how the river ecosystem responds to climate change. Figures and Tables | References | Related Articles | Metrics

Select

Impact mechanisms of groundwater dynamics on root water uptake of Salix matsudana in Mu Us Sandy Land, China QIAO Gang, YIN Lihe, XU Yong, ZHANG Jun, SHI Changchun, YU Kun 2026, 33(1): 39-49.  DOI: 10.13745/j.esf.sf.2025.10.21 Abstract ( 19 )   HTML ( 2 )   PDF (3584KB) ( 6 )   Salix matsudana is widely distributed in the Mu Us Sandy Land. Although significant progress has been made in understanding the water sources for willow growth, research on the response mechanisms of single-plant root water uptake to groundwater level dynamics remains limited at the site scale. This knowledge gap impedes a comprehensive understanding of vegetation-groundwater feedback mechanisms. To address this, our study combined root excavation, in-situ monitoring, and controlled irrigation experiments. By monitoring meteorological factors, plant physiological parameters, soil moisture at different depths, and groundwater levels—complemented by an analysis of root vertical distribution—we investigated changes in vadose zone moisture and root water uptake strategies during groundwater fluctuations. Key findings reveal: (1) The root system of Salix matsudana exhibits three distinct water-absorption zones: A shallow fibrous root zone (0.2-0.8 m depth) concentrated in the upper vadose zone, primarily absorbing moisture from precipitation infiltration; A fine and lateral root zone (2.0-3.8 m depth) that absorbs deep vadose moisture and capillary water; and A deep vertical root zone (4.3-4.8 m depth) that penetrates to the water table, primarily extracting groundwater. (2) When utilizing deep vadose moisture derived from groundwater, the intensity of root water uptake tripled. (3) The critical pressure head threshold for water uptake from deep vadose moisture derived from groundwater was identified at -10.7 m. Based on these findings and a synthesis of relevant studies, we elucidate the ecohydrological mechanisms underlying the influence of groundwater dynamics on these distinct root absorption zones and discuss their adaptive significance. Finally, we propose recommendations for implementing a “multi-layered water absorption / dynamic partitioning” strategy in Salix matsudana plantations to enhance resilience to extreme drought events. These findings advance the understanding of groundwater dynamics and willow water uptake mechanisms, contributing to eco-hydrological theory in arid regions. They also provide scientific support for global desertification control efforts. Figures and Tables | References | Related Articles | Metrics

Select

Exploring the spatio-temporal evolution and driving mechanisms of vegetation net primary productivity in the Ziya River Basin ZHANG Yongcheng, CHEN Kang, GUI Yingjia, ZHANG Xiaochen 2026, 33(1): 50-62.  DOI: 10.13745/j.esf.sf.2025.10.14 Abstract ( 25 )   HTML ( 6 )   PDF (7917KB) ( 9 )   Understanding the spatio-temporal evolution and driving mechanisms of vegetation net primary productivity (NPP) in the Ziya River Basin is crucial for elucidating the regional carbon budget and promoting the sustainable development of its ecological environment. Based on MODIS NPP data from 2001 to 2022, this study comprehensively analyzed the effects of climate, soil, topography, and human activities. We utilized the Theil-Sen median trend analysis, Mann-Kendall significance test, and Hurst exponent method to examine the spatio-temporal variation characteristics of vegetation NPP. Furthermore, the optimal parameters-based geographical detector (OPGD) model was applied to quantitatively assess the impact of various factors and their interactions on vegetation NPP dynamics. The results showed that: (1) During the study period, the annual mean NPP in the basin fluctuated between 224.86 and 371.21 gC·m-2·a-1, exhibiting a significant increasing trend with a rate of 4.53 gC·m-2·a-1. (2) Spatially, areas with an annual average NPP higher than 400 gC·m-2·a-1 were primarily concentrated in the western mountainous regions, whereas areas with NPP lower than 300 gC·m-2·a-1 were mainly distributed in the eastern plains. (3) Over the study period, the changes in annual NPP were predominantly characterized by extremely significant and significant increases, accounting for 67.21% and 18.25% of the total area, respectively. In contrast, only 2.03% of the area showed a decreasing trend, which was sporadically distributed around urban districts and was associated with accelerated urbanization. Predictions based on the Hurst index suggest that future vegetation NPP changes are likely to be predominantly opposite to the past trend, with 88.30% of the area expected to experience a reversal from the previous increasing trajectory. (4) Soil temperature, air temperature, elevation, and potential evapotranspiration were identified as the primary factors influencing vegetation NPP. The interactions between these factors exhibited synergistic effects, yielding a greater explanatory power for the variation in vegetation NPP than any single factor alone. These findings contribute to a deeper understanding of the characteristics and driving mechanisms of vegetation NPP in the Ziya River Basin and provide a scientific basis for regional carbon sequestration, emission reduction, and ecological conservation efforts. Figures and Tables | References | Related Articles | Metrics

Select

Distribution prediction of natural low-quality groundwater in the plains of Henan Province based on machine learning YU Furong, LI Rui, LI Zhiping, WU Lin, LIU Zhongpei 2026, 33(1): 63-79.  DOI: 10.13745/j.esf.sf.2025.10.33 Abstract ( 14 )   HTML ( 1 )   PDF (8885KB) ( 15 )   As a crucial drinking water source for over two billion people worldwide, groundwater quality is intrinsically linked to human health and ecosystem integrity. Geogenic groundwater contamination (GGC), characterized by excessive levels of arsenic (As), fluoride (F), and iodine (I), originates from natural geological processes. The distribution of GGC, influenced by geological structures, hydrogeochemistry, and anthropogenic activities, exhibits regional patterns with local complexities. Research into its formation mechanisms and control strategies is therefore critical for ensuring water security. Using Henan Province as a case study, this research employed methods including Gibbs diagrams to analyze groundwater hydrochemical characteristics and their controlling factors, thereby identifying the origins of GGC. The correlation between GGC and the distribution of endemic diseases was investigated. Furthermore, machine learning models were introduced to achieve accurate spatial prediction of GGC. Subsequently, health risk control zones were proposed. The results indicate that: (1) GGC in the study area is concentrated in the Eastern Henan Plain and the regions along the Yellow River, with contamination levels in phreatic water being significantly higher than those in confined water; (2) Weakly alkaline and reducing environments represent key hydrogeochemical conditions for GGC formation, where rock weathering and dissolution combined with intense evaporation govern the enrichment of characteristic ions; (3) A spatial correlation exists between GGC distribution and endemic disease areas; (4) Arsenic, fluoride, and iodine in groundwater all exhibit significant spatial aggregation. Notably, the high-high (HH) clusters, indicating areas with co-occurrence of high arsenic, fluoride, and iodine, show strong agreement with the high-risk zones predicted by the machine learning models. Based on these findings, scientifically delineating protection zones in key regions such as Puyang, Xinxiang, Zhoukou, Kaifeng, and Shangqiu cities holds significant practical importance for ensuring local residents’ drinking water safety. Figures and Tables | References | Related Articles | Metrics

Select

Earthquake hydrogeology: Water rock interaction from a disaster per-spective SHI Zheming, WANG Guangcai, YAN Rui, Qi Zhiyu 2026, 33(1): 80-94.  DOI: 10.13745/j.esf.sf.2025.10.12 Abstract ( 17 )   HTML ( 2 )   PDF (1312KB) ( 11 )   Earthquake hydrogeology studies the interaction between earthquakes and groundwater. Unlike traditional hydrogeology, it mainly focuses on the evolution of groundwater hydrological processes caused by earthquake-induced aquifer property changes. This paper provides a comprehensive review of groundwater co-seismic and post-seismic responses, and the characterization of earthquake-induced hydrogeological property changes, with a focus on advancements over the past two decades. The development of linear poroelastic theory has established a theoretical foundation for quantitatively interpreting the dynamic relationships between crustal movements (e. g., earthquakes) and groundwater. Subsurface fluid precursor anomalies have served as valuable supplementary tools in recent earthquake forecasting efforts. Notable developments in this area include enhanced monitoring of groundwater geochemical indicators and the construction of large-scale geochemical observation networks. Nevertheless, key future challenges involve: ① establishing mechanistic models of precursor anomalies that couple physical and chemical groundwater dynamics, and ② developing precursor signal identification methodologies such as machine learning. Earthquake-induced permeability changes of aquifer systems, leading to dynamic changes in water exchange and hydrochemistry, are the most plausible mechanism explaining co-seismic and post-seismic groundwater responses. Aquifer parameter identification based on groundwater responses to periodic signals (e. g., Earth tides and barometric pressure) offers new pathways for the continuous acquisition of hydrogeological parameters. However, existing analytical models for tidal and barometric responses are often plagued by the problem of non-uniqueness in parameter estimation. Reducing uncertainty in computed results through novel models and methods remains a critical future direction. Finally, to advance the understanding of earthquake-groundwater system interactions, establishing fault-zone observatories that integrate thermal springs and groundwater monitoring wells is essential. Equipped to measure water level, temperature, flow rate, chemical composition, crustal deformation, and seismic waves, these observatories, built upon prior research, are essential for deepening the theoretical foundations of earthquake hydrogeology. Figures and Tables | References | Related Articles | Metrics

Select

Microbial community structure and environmental driving mechanisms in high-salinity groundwater of the Yellow River Delta ZHI Chuanshun, HU Xiaonong, BAI Jing, MU Hui, WU Xiancang, YANG Fan, CHANG Wenbo, WEI Ruchun, LI Yuxi 2026, 33(1): 95-106.  DOI: 10.13745/j.esf.sf.2025.10.2 Abstract ( 12 )   HTML ( 3 )   Shallow groundwater in the Yellow River Delta is characterized by high salinity due to historical seawater intrusion and evaporative concentration, which has led to complex hydrochemical evolution. However, the succession of microbial communities under these conditions remains poorly understood. In this study, 25 shallow groundwater samples, two Yellow River water samples, and one seawater sample were collected. Hydrochemical analysis, stable isotope tracing, and metagenomic sequencing were employed to identify salinity sources and to investigate the composition, succession patterns, and controlling factors of microbial communities across a salinity gradient. Results showed that groundwater total dissolved solids (TDS) ranged from 1.0-35.0 g/L, dominated by the Cl-Na water type. Combined evidence from δD and δ18O and Br-/Cl- ratios indicated the dual influence of seawater mixing and evaporative concentration. Microbial communities were dominated by bacteria (67.7%-98.6% relative abundance), followed by archaea (1.0%-32.0%), while eukaryotes and viruses were generally less abundant. Major bacterial groups included Pseudomonadota and Candidatus Omnitrophota, with the former being enriched in high-salinity samples and the latter preferring lower salinity environments. Among archaea, Candidatus Woesearchaeota was widespread, while Candidatus Lokiarchaeota and Candidatus Thermoprofundales showed ecological dominance under high-salinity conditions. Principal coordinates analysis (PCoA) and linear discriminant analysis effect size (LEfSe) revealed distinct microbial community structures across different salinity levels. Bacteria displayed greater sensitivity to salinity variation, whereas archaea exhibited higher salinity tolerance. Redundancy analysis (RDA) indicated that TDS, ${\mathrm{SO}}_{4}^{2-}$, Fe2+, and ${\mathrm{NH}}_{4}^{+}$ were the main factors shaping microbial communities in high-salinity groundwater, while ${\mathrm{NO}}_{3}^{-}$, pH, and ORP were more influential in low-salinity environments. This study reveals the succession patterns and environmental drivers of microbial communities in high-salinity groundwater, providing theoretical insight into microbial responses and ecological functions under salinity stress in coastal aquifer systems. Figures and Tables | References | Related Articles | Metrics

Select

Spatial evolution and genetic mechanisms of manganese in shallow groundwater of the North Shandong Plain GUO Jiju, CAO Wengeng, LU Chongsheng, WANG Zhe, ZHU Jingsi, WANG Yanyan, LI Xiangzhi, MA Cuiyan 2026, 33(1): 107-120.  DOI: 10.13745/j.esf.sf.2025.10.23 Abstract ( 13 )   HTML ( 1 )   PDF (15792KB) ( 4 )   Manganese (Mn), a widespread neurotoxic metallic element in groundwater, poses significant health risks when its concentration exceeds the permissible limit (> 0.1 mg·L-1), potentially inducing Parkinsonian-like neurological disorders. Although previous studies have revealed pronounced spatial heterogeneity of Mn in the shallow groundwater of the lower Yellow River alluvial plain, a systematic understanding of the hydrogeochemical mechanisms controlling its enrichment across different geomorphic units is still lacking. To address this gap, this study investigates the regional distribution of Mn and its controlling factors, complemented by a detailed characterization of representative hydrogeological cross-sections to elucidate the key processes governing high-Mn groundwater. The results indicate that the regional groundwater Mn concentration ranges from below the detection limit to 12.0 mg·L-1, with a mean of 0.65 mg·L-1 and an exceedance rate of 96.6%. Along a representative transect, the average Mn concentration increases significantly from the paleochannel highland (0.27 mg·L-1) to the fluvial plain (0.83 mg·L-1) and reaches its peak in the coastal plain (2.53 mg·L-1). This spatial pattern is accompanied by a gradual evolution of hydrochemical facies, from HCO3-Na-Mg and HCO3-Cl-Na-Mg types to predominantly Cl-HCO3-Na-Mg, and finally to the Cl-Na type. The primary sources of Mn are identified as manganese hydroxides and rhodochrosite within the sediments. The regional distribution of Mn is primarily influenced by a series of water-rock interactions, including silicate weathering, dissolution of evaporite minerals, cation exchange, and redox processes. The contributions of these natural sources exhibit significant spatial variations across different hydrogeochemical zones, mirroring the evolution of Mn concentrations. Integrated analysis reveals that the spatial heterogeneity of shallow groundwater Mn is collectively controlled by pH-driven dissolution and adsorption, salinity-driven ion exchange, and the release of Mn from primary sediments under the prevailing depositional background. Figures and Tables | References | Related Articles | Metrics

Select

Distribution, source and transformation of nitrate in water bodies of an ion-adsorption rare earth mining areas: A case study of the Zudong rare earth mine in the Jiangxi Province WEI Chunyi, YU Shengpin, BAI Ximin, LIU Haiyan, WANG Zhen, GE Qin, CHEN Gongxin, ZHOU Zhongkui, SUN Zhanxue, GUO Huaming 2026, 33(1): 121-134.  DOI: 10.13745/j.esf.sf.2025.10.7 Abstract ( 14 )   HTML ( 1 )   PDF (5837KB) ( 11 )   Mining of ion-adsorption rare earth elements (REEs) has caused severe nitrogen pollution in water and soil. However, the distribution, migration and transformation, and pollution sources of nitrate (${\mathrm{NO}}_{3}^{-}$-N) in water bodies affected by such drainage remains insufficiently studied. This study investigated the origins and transformations of ${\mathrm{NO}}_{3}^{-}$-N in surface water and groundwater downstream of the Zudong ion-adsorption REE mining area in southern Jiangxi, China, using hydrochemical analysis and a multi-isotope (δ18O-H2O,δ15N-${\mathrm{NO}}_{3}^{-}$ and δ18O-${\mathrm{NO}}_{3}^{-}$) approach. The contribution of various sources was quantitatively assessed using the MixSIAR model. The results showed that surface water and groundwater were weakly acidic and were characterized by low TDS values. Surface water was predominantly of the SO4-Ca type, whereas 80% of the groundwater samples were of the HCO3-Ca type. The concentrations of total nitrogen (TN), ${\mathrm{NO}}_{3}^{-}$-N, and ${\mathrm{NH}}_{4}^{+}$-N in surface water were significantly higher than those in groundwater, indicating that nitrogen pollution in surface water was closely related to mining activities. Spatially, nitrogen sources were closely related to mining activities that utilized ammonium-nitrogen-rich brine, which posed a significant nitrogen pollution risk to surface water. Distributions of land-use types indicated that nitrogen sources in surface water differed from those in groundwater. Nitrogen in surface water primarily originated from mine drainage in forested areas, whereas nitrogen in groundwater was mainly derived from agricultural activities on cultivated land. The compositions of δ18O-H2O, δ15N-${\mathrm{NO}}_{3}^{-}$, δ18O-${\mathrm{NO}}_{3}^{-}$ and fractionation coefficients revealed that nitrification was the dominant process in both surface water and groundwater. End-member analysis based on measured and reconstructed δ15N-${\mathrm{NO}}_{3}^{-}$ and δ18O-${\mathrm{NO}}_{3}^{-}$ values suggested that ${\mathrm{NO}}_{3}^{-}$-N in SW1 was primarily derived from ammonium nitrogen discharged from mining activities. In contrast, ${\mathrm{NO}}_{3}^{-}$-N in SW2 was influenced by both mining and agricultural activities, with contributions from ammonium nitrogen, soil nitrogen, and sewage manure. ${\mathrm{NO}}_{3}^{-}$-N in groundwater mainly originated from soil nitrogen and sewage manure. Quantitative assessment using the MixSIAR model indicated that mine drainage contributed over 50% (mean value) of the ${\mathrm{NO}}_{3}^{-}$-N in SW1 (surface water nearest to the mining area). Of this contribution, 65% to 94% originated from the native ${\mathrm{NO}}_{3}^{-}$-N present in the mine drainage; while in surface water farther from the mining area (SW2), mine drainage contributed approximately 30% of the ${\mathrm{NO}}_{3}^{-}$-N. Uncertainty analysis (UI90) showed that the contribution of atmospheric precipitation was the most stable, while the contributions of mine drainage, sewage and manure, and soil nitrogen exhibited significant uncertainty. This research elucidates the formation mechanism of ${\mathrm{NO}}_{3}^{-}$-N pollution in ion-adsorption REE mining areas and provides a scientific basis for the precise prevention and control of nitrogen pollution. Figures and Tables | References | Related Articles | Metrics

Select

Study on the blocking effect of the loess cover layer on water-salt transport in the Yungang Grottoes OUYANG Kaigao, YAN Hongbin, JIANG Xiaowei, LI Na, ZHANG Shaoyou, NIU Ran, YANG Xi, TANG Xulin, CHI Huaqing, WAN Li 2026, 33(1): 135-142.  DOI: 10.13745/j.esf.sf.2025.10.31 Abstract ( 18 )   HTML ( 2 )   PDF (5099KB) ( 4 )   Atmospheric precipitation infiltration into rock and soil is a common natural phenomenon. However, understanding the response of different materials to infiltration remains limited, particularly for sandstone where in-situ moisture monitoring is challenging due to significant differences in hydrological parameters between materials like sandstone and loess. This study, conducted at the Yungang Grottoes in northern China, employed frequency domain reflectometry (FDR) to monitor moisture content at various depths. Monitoring sites included exposed sandstone atop Cave 4, where infiltration potentially causes internal seepage, and loess atop Cave 9, where infiltration does not lead to seepage inside the cave. The responses of these two materials to rainfall events were compared. Monitoring results indicated that shallow sandstone (10 cm depth) responded rapidly to rainfall events as small as 10 mm, with moisture content increasing significantly, nearly reaching saturation. In contrast, similar rainfall intensity only increased the saturation of shallow loess (10 cm depth) to approximately 0.75, while the deep loess (220 cm depth) showed almost no response. Mineral composition analysis using X-ray diffraction (XRD) on weathering products from the cave walls revealed that decades of acid rain infiltration into Cave 4 led to ${\mathrm{SO}}_{4}^{2-}$ accumulation and the formation of efflorescent salts, which dominate salt weathering. Analysis of weathering product samples collected in the 1960s, alongside recent samples from Caves 5-10, suggests that the overlying loess layer has prevented acid rain infiltration into these caves. This study demonstrates that the loess cover layer exhibits low permeability, a finding that can inform future decision-making for the conservation of the cultural relics within the caves. Figures and Tables | References | Related Articles | Metrics

Select

Mg isotopes for tracing water-rock interactions in terrestrial water: Research progress and prospects WANG Rui, JIANG Xiaowei, JI Taotao 2026, 33(1): 143-151.  DOI: 10.13745/j.esf.sf.2025.10.18 Abstract ( 15 )   HTML ( 1 )   PDF (3119KB) ( 10 )   Magnesium (Mg) is a major constituent of terrestrial water, and significant mass-dependent fractionation of Mg isotopes can occur during various water-rock interactions. Consequently, Mg isotopes have considerable potential for tracing these processes. This paper systematically summarizes the Mg isotopic compositions of the key reservoirs that define the initial Mg isotopic composition of natural waters, including silicates, carbonates, evaporites, and rainwater. Silicates typically exhibit high and variable δ26Mg values, whereas carbonates typically have low and relatively uniform δ26Mg values. Evaporites display a wide range of δ26Mg values depending on their mineralogy, and the δ26Mg values of rainwater are often shaped by local environmental conditions. Subsequently, geochemical processes that remove Mg from water, such as clay formation, carbonate precipitation, adsorption, cation exchange, and plant uptake, fractionate Mg isotopes. Specifically, the formation of secondary minerals like montmorillonite, adsorption onto solid surfaces, and plant uptake preferentially incorporate heavy Mg isotopes (e.g., 26Mg) into the solid/biological phase, leaving the surrounding water enriched in the light 24Mg. Conversely, the formation of chlorite, cation exchange, and carbonate precipitation can preferentially incorporate lighter Mg isotopes, leaving the water heavier. The Mg isotopic signatures of rivers and groundwater are further influenced by hydrological setting and water-rock interaction timescales. In river water, which has high renewal rates, the Mg isotopic signature is primarily shaped by silicate and carbonate dissolution, clay formation, and cation exchange. In contrast, groundwater, which typically involves longer water-rock interaction timescales, can also be significantly influenced by adsorption. Furthermore, in regional groundwater systems with long flow paths, Mg isotopes serve as effective tracers for identifying multiple water-rock interaction processes in both carbonate- and silicate-dominated aquifers; the nature of these reactions is primarily controlled by aquifer mineralogy. Future research in Mg isotope hydrogeochemistry should prioritize: (1) integrating Mg with K isotopes to better identify adsorption processes in river systems, and (2) characterizing Mg isotopic variations along flow paths in aquifers of different lithologies and scales. Figures and Tables | References | Related Articles | Metrics

Select

Evolution and state assessment of groundwater contamination plumes ZHAO Yongsheng 2026, 33(1): 152-162.  DOI: 10.13745/j.esf.sf.2025.10.4 Abstract ( 11 )   HTML ( 1 )   PDF (1306KB) ( 17 )   The evolutionary patterns and existing states of groundwater contaminant plumes serve as the fundamental basis for formulating pollution control strategies, as their different developmental stages and spatial features directly influence the precision and effectiveness of remediation measures. Therefore, elucidating the dynamic evolution mechanisms of contaminant plumes and their governing factors holds significant importance for achieving scientific groundwater pollution management. The evolution process of groundwater contaminant plumes exhibits notable spatiotemporal heterogeneity and complexity. The duration and characteristics of each evolutionary stage are jointly influenced by multiple factors, including contaminant input flux, pollutant properties, aquifer lithology, groundwater flow dynamics, and seasonal variations. The intricate interactions among these factors pose substantial challenges to accurately assessing and predicting the evolutionary stages and existing states of contaminant plumes. Currently, there remains a lack of systematic and comprehensive quantitative assessment methodologies internationally. Although numerical models for groundwater contamination can predict the spatiotemporal distribution of pollutant concentrations, their predictive results cannot directly determine the developmental stages and evolutionary patterns of contaminant plumes. Moreover, numerical modeling imposes stringent requirements on data and parameters for characterizing hydrogeological conditions and contaminant properties. Most groundwater contamination sites fail to meet the extensive data prerequisites necessary for numerical simulations. To address these challenges, this study focuses on typical groundwater contaminants (including heavy metals and organic pollutants) prevalent in China and establishes a scientific framework for determining contaminant plume formation. It proposes that the evolutionary process of groundwater contaminant plumes is fundamentally governed by the dynamic equilibrium of key factors, including pollutant release mass, storage capacity of geological media, contaminant migration flux in groundwater, and destructive removal processes (e.g., microbial degradation). Building upon this, a multidimensional evaluation index system encompassing pollution sources, hydrogeological conditions, and pollutant interaction processes was developed, categorizing plume evolution into “expansion-maturity-recession” phases. Additionally, a quantitative assessment method for plume states was established. The research outcomes provide theoretical and technical support for identifying plume developmental stages and implementing differentiated remediation strategies. These findings hold practical significance for advancing precision control of groundwater pollution. Figures and Tables | References | Related Articles | Metrics

Select

Current status and future prospects of sulfur-based autotrophic denitrification technology in engineering applications CHEN Nan, CHEN Fangxin, PENG Tong, LI Yeping, SUN Daxin, YUAN Yuan, LIU Chengtian, MEI Duoduo, ZHAN Yongheng, WANG Xiaotong, FENG Chuanping 2026, 33(1): 163-178.  DOI: 10.13745/j.esf.sf.2025.10.24 Abstract ( 24 )   HTML ( 1 )   PDF (2839KB) ( 10 )   Biological denitrification is a key technology for nitrate removal. However, conventional heterotrophic denitrification requires external organic carbon sources, which leads to high operational costs, excessive sludge production, and significant carbon emissions. In the context of the “Dual Carbon Goals,” selecting cost-effective and environmentally sustainable biological treatment technologies has become a crucial strategy for low-carbon nitrate wastewater remediation. Sulfur-autotrophic denitrification (SAD) utilizes elemental sulfur or sulfides as electron donors to support microbial nitrate reduction. Compared to traditional heterotrophic processes, SAD significantly reduces carbon emissions and the need for external carbon sources. This review systematically examines the principles and engineering applications of SAD, compares the nitrogen removal performances of composite sulfur-based materials and high-sulfur-content substrates, and discusses the functional roles of sulfur-autotrophic filter media in practical implementations. By elucidating differences in reactor configurations, material properties, contact mechanisms, and microbial community enrichment strategies, this work highlights the performance characteristics of various reactor types in real-world scenarios. Based on an analysis of system operational features, the review summarizes full-scale applications of autotrophic denitrification in municipal, industrial, and ecological contexts. It explores the current use of SAD in areas such as municipal wastewater treatment, high-salinity effluents, and constructed wetlands, and provides a quantitative evaluation of its operational costs and environmental benefits. Finally, the review outlines future research directions for SAD, emphasizing material development and genetic engineering-based targeted regulation to standardize processes and overcome engineering bottlenecks. The ultimate goal is to establish a low-carbon, high-efficiency sulfur-autotrophic nitrogen removal system, offering valuable insights for the engineering application of SAD technology. Figures and Tables | References | Related Articles | Metrics

Select

Early warning of groundwater pollution in oasis area of Hami Basin JIANG Feng, ZHOU Jinlong, ZHOU Yinzhu, LIU Jiangtao, ZENG Yanyan, LIU Yu, DING Qizhen 2026, 33(1): 179-192.  DOI: 10.13745/j.esf.sf.2025.10.22 Abstract ( 11 )   HTML ( 1 )   PDF (13279KB) ( 4 )   Early warning of groundwater pollution is of critical importance for identifying key prevention and control areas and ensuring the safety of groundwater resources. Using the oasis area of the Hami Basin in Xinjiang as a case study, a comprehensive evaluation system for pollution risk was constructed, covering the inherent vulnerability, pollution load, and functional value of groundwater. The reasonableness of the model was verified through sensitivity analysis. Subsequently, the entropy weight TOPSIS method was used to evaluate the current groundwater quality status, and the trend in water quality change was analyzed using data from typical monitoring wells. A game theory-based comprehensive weighting method was then introduced to integrate the three indicators—water quality status, change trend, and pollution risk—for early warning level classification. The results show that: 1) The risk of groundwater pollution in the study area is generally low, with higher-risk and high-risk areas accounting for only 9.7%. The high-risk areas are characterized by a vadose zone with lithology dominated by sand and gravel, which has strong permeability, coupled with intensive human activities, high-intensity groundwater exploitation, and a combination of non-point and point source pollution. 2) The assessment of water quality status shows that Class IV and Class V water account for 10.1%, indicating that groundwater quality is mainly affected by a combination of hydrogeological conditions and human activities. The overall trend of groundwater quality is improving, with areas of deterioration accounting for only 5.4%. Industrial and agricultural pollution along with domestic sewage discharge were the main driving factors for deterioration. 3) The early warning levels are dominated by “no warning” and “light warning”, with high and heavy warning areas comprising 5.3%. These high-level warning areas were categorized into natural factor-dominated, human activity-dominated, and compound influence-dominated types. The groundwater pollution early warning method developed in this study, which is based on the game theory-based comprehensive weighting, effectively improved the objectivity and spatial continuity of the early warning classification, and can provide a theoretical basis and methodological support for regional groundwater pollution prevention and control zoning, as well as the identification of key remediation areas. Figures and Tables | References | Related Articles | Metrics

Select

Enhanced sulfate removal from groundwater by iron-carbon micro-electrolysis coupled with polyvinyl alcohol-sodium alginate immobilization: Functional community regulation and closed-loop sulfur recovery ZHANG Boaiqi, ZHAO Chaorui, ZHU Kun, GUO Qiuzhi, CHEN Nan, FENG Chuanping, HU Yutian 2026, 33(1): 193-206.  DOI: 10.13745/j.esf.sf.2025.10.34 Abstract ( 20 )   HTML ( 2 )   PDF (6141KB) ( 9 )   To address the escalating sulfate pollution in groundwater (primarily from mining, chemical industries, and agricultural non-point sources), this study developed and validated a closed-loop treatment strategy integrating iron-carbon micro-electrolysis with PVA-SA immobilization for sulfur reduction, fixation, and resource recovery. An up-flow packed column simulating aquifer conditions was constructed to compare the long-term sulfate removal and sulfur fixation performance of three fillers: quartz sand (Control), PVA-SA-embedded maifanite (Ma), and PVA-SA-embedded iron-carbon-maifanite (FeC-Ma). Key results demonstrated: (1) The effluent sulfate concentration in FeC-Ma stabilized at 136.1 mg/L (average removal: 90%), significantly outperforming the Control (230.3 mg/L, 80.4%); (2) Iron-carbon micro-electrolysis effectively fixed sulfides as FeS precipitates within the gel matrix, reducing effluent sulfide to 2.1 mg/L (Phase I, HRT=2 d; Control: 182.5 mg/L) with 70% fixation efficiency sustained for 40 days; (3) Metagenomic analysis revealed that during sulfate reduction, FeC-Ma enriched complete-oxidizing sulfate-reducing bacteria (e.g., Desulfococcus), upregulated dissimilatory reduction genes (aprA/B-dsrA/B), and suppressed competitive methanogens (Methanothrix decreased by 52%), jointly driving sulfur precipitation; (4) Independent sulfur-autotrophic denitrification using fixed sulfur as electron donor achieved efficient nitrate reduction (effluent ${\mathrm{NO}}_{3}^{-}$-N<5 mg/L) with 82.1% sulfur recovery. This technology provides an efficient solution for groundwater sulfate remediation and sulfur resource recovery. Figures and Tables | References | Related Articles | Metrics

Select

Effective modeling framework and pertaining key breakthroughs for efficient prediction of VOCs transport and diffusion fluxes in fractured media QU Cixiao, WANG Mingyu 2026, 33(1): 207-221.  DOI: 10.13745/j.esf.sf.2025.10.20 Abstract ( 13 )   HTML ( 2 )   PDF (7054KB) ( 1 )   Groundwater organic pollution poses a long-term threat to global sustainable development and water security. Fractured bedrock aquifers (including fractured porous sedimentary rocks, igneous rocks, and karstic carbonate rocks) are widely distributed and represent a major category of contaminated media by volatile organic compounds (VOCs). Quantifying the transport and diffusion processes of VOCs within fractured media is critical for environmental system protection and water resource security. However, at larger spatial scales, such modeling becomes highly complex, and achieving fast, convenient, and precise predictions of contaminant transport and diffusion fluxes remains a major challenge. Given this context, this paper analyzes the complexity of modeling VOCs transport and diffusion fluxes in fractured media and the challenges in achieving efficient and precise prediction. Furthermore, it proposes and elaborates on a comprehensive modeling framework for the efficient prediction of VOCs transport and diffusion fluxes in fractured media. It also explores corresponding methodologies for developing efficient statistical surrogate models and intelligent predictive models through illustrative case studies. Finally, the paper identifies key challenges warranting further research and proposes potential pathways to overcome these bottlenecks for the efficient and accurate modeling of VOCs transport and diffusion fluxes in fractured media. These include: a dimension-reduction equivalent modeling approach for efficient prediction; an integrated modeling framework leveraging knowledge graphs for deep, multi-dimensional fusion; artificial intelligence modeling driven by simulation data and enhanced by constitutive/statistical relationships; and the development of an intelligent software system, tailored to specific application scenarios and built upon the proposed integrated modeling framework, to deliver efficient and precise VOCs flux predictions in practice. The comprehensive modeling framework, the specific effective modeling methodologies, and the proposed pathways for future in-depth study will help address the challenge of efficient and accurate prediction and quantification of key interfacial fluxes of multiphase VOCs contaminants in complex groundwater environmental systems. This work is expected to provide crucial support for efficient groundwater remediation and optimal risk management. Figures and Tables | References | Related Articles | Metrics

Select

Transport and retention mechanisms of micro/nano plastics during two-phase flow LIU Qiming, YANG Zhibing, WU Ting 2026, 33(1): 222-235.  DOI: 10.13745/j.esf.sf.2025.10.9 Abstract ( 12 )   HTML ( 1 )   PDF (11594KB) ( 1 )   The transport and retention of micro/nano-plastics (M/NPs) in porous media is a critical concern in soil and groundwater contamination. However, limitations in visualization techniques have hindered a deeper understanding of the microscopic mechanisms governing M/NP transport under two-phase flow conditions. Using a three-dimensional visualization apparatus, we directly observed and quantified the distribution characteristics of M/NPs within pore spaces, with a specific focus on their retention at grain-grain contacts, solid surfaces, and fluid-fluid interfaces during imbibition. The effects of M/NP size and imbibition rate on their spatial distribution were further investigated. Our results reveal that M/NP size predominantly governs the selection of retention sites by altering the M/NP-solid and M/NP-fluid interface interaction energies. Conversely, the magnitude of interaction energy between M/NPs determines their potential to form aggregated clusters. The flow rate primarily influences the overall retention by altering the solid-wetting phase interfacial area. Moreover, distinct mechanisms govern retention at different locations: straining dominates at grain-grain contacts, while interfacial adsorption prevails at solid surfaces and fluid-fluid interfaces. Under high flow rates, M/NPs retained on solid surfaces can be readily detached due to enhanced shear forces. In contrast, the size of aggregated M/NP clusters at grain-grain contacts increases with flow velocity. This study elucidates the synergistic mechanisms by which M/NP size and flow velocity govern M/NP migration and retention, providing a theoretical basis for predicting the transport and fate of contaminants in multiphase subsurface environments. Figures and Tables | References | Related Articles | Metrics

Select

Effect of silicon-doped ferrihydrite on the kinetics of Cr(VI) reduction by gallic acid and Pseudomonas aeruginosa in groundwater ZHANG Mengfan, CAI Xuyi 2026, 33(1): 236-249.  DOI: 10.13745/j.esf.sf.2025.10.13 Abstract ( 17 )   HTML ( 2 )   PDF (5586KB) ( 3 )   Polyphenols and Pseudomonas aeruginosa are commonly used remediation agents for reducing and immobilizing Cr(VI) with minimal risk of secondary pollution. The kinetics of this process are influenced by various environmental factors, yet the impact of these factors on the remediation and transformation of Cr(VI) remains poorly understood. This study focuses on the effect of silicon-doped ferrihydrite particles-ubiquitous in groundwater environments-on the kinetics of Cr(VI) reduction by polyphenols and bacteria. Through a series of controlled laboratory batch experiments involving binary, ternary, and quaternary reaction systems, we investigated the influence of the extent of silicon incorporation in ferrihydrite on this process. In ternary systems (comprising silicon-doped ferrihydrite, either polyphenols or bacteria, and Cr(VI)), rapid complexation and reduction of Cr(VI) by polyphenols significantly enhanced the initial removal efficiency (up to 87%), though sustained removal subsequently relied on the adsorption capacity of silicon-doped ferrihydrite. Bacteria facilitated Cr(VI) reduction and stabilization through the secretion of extracellular polymeric substances (EPS), thereby mitigating the inhibitory effect of silicon doping on Cr(VI) adsorption by ferrihydrite. In the quaternary system (silicon-doped ferrihydrite-polyphenols-bacteria-Cr(VI)), the synergistic interaction between polyphenols and bacteria partially counteracted the suppression caused by silicon doping, maintaining a higher Cr(VI) removal efficiency. Within this system, polyphenols were degraded or oxidized, while silicon-doped ferrihydrite provided a habitat for bacterial colonization. Figures and Tables | References | Related Articles | Metrics

Select

In-situ bioremediation of acidic uranium-contaminated groundwater: Development and challenges LIU Yajie, LI Jiang, WANG Xuegang, KE Pingchao, SUN Zhanxue 2026, 33(1): 250-268.  DOI: 10.13745/j.esf.sf.2025.10.36 Abstract ( 13 )   HTML ( 3 )   PDF (3188KB) ( 2 )   Acidic uranium-contaminated groundwater represents a significant and urgent environmental issue worldwide. Due to its cost-effectiveness and sustainability, in situ bioremediation has emerged as a promising mitigation strategy. This review systematically analyzes the technical principles, existing challenges, and future directions for in situ bioremediation of this environmental challenge. We begin by tracing the genesis of acidic uranium-contaminated water from uranium mining and milling activities, elucidating microbial reduction mechanisms, and outlining the evolution of in situ remediation technologies. Key uranium immobilization mechanisms - including bioreduction (the core pathway), biomineralization (for stable sequestration), biosorption (enabling rapid retention), and bioaccumulation (with resource recovery potential - are critically examined. This review also analyzes critical constraints such as extreme acidity (pH<4.0), the presence of competitive electron acceptors (e.g., nitrate), oligotrophic conditions, dynamic microbial community succession, and the stability risks of U(IV), proposing targeted countermeasures for each. Future research should prioritize: (1) developing self-adaptive microbial consortia, (2) enhancing long-term stability through the promotion of crystalline U(IV) phases, and (3) establishing synergistic systems that integrate multiple mechanisms (e.g., bio-chemical coupling and smart electron donor delivery). These innovations are crucial for advancing the efficacy and longevity of in situ bioremediation in complex acidic environments. Figures and Tables | References | Related Articles | Metrics

Select

Advances in characterization techniques of deep geothermal reservoir fracture structures by integrating microseismic and hydrological data XU Tianfu, LI Siyuan, JIANG Zhenjiao 2026, 33(1): 269-282.  DOI: 10.13745/j.esf.sf.2025.10.3 Abstract ( 16 )   HTML ( 1 )   PDF (9431KB) ( 2 )   To achieve high-resolution characterization of the induced fracture network, we developed a methodology for fracture network characterization by fusing microseismic and hydraulic data, taking into account the realistic conditions of enhanced geothermal systems. Synthetic tests and field applications demonstrated that the method significantly enhanced the robustness and accuracy of fracture characterization. This study reviews the development of the method, including the sequential and joint inversion of microseismic and hydraulic data for permeability estimation in equivalent continuous media, and the sequential inversion for explicit fracture imaging in discrete fracture networks. We compare and analyze the advantages and disadvantages of these different approaches. Finally, we summarize the remaining challenges and outline directions for future research, aiming to establish a more advanced methodology for high-resolution fracture characterization and risk mitigation in subsurface engineering. Figures and Tables | References | Related Articles | Metrics

Select

The distribution, genesis and utilization directions of geothermal resources in Xinjiang: A summary of the Third Xinjiang Scientific Expedition Program to geothermal science in Xinjiang KONG Yanlong, DONG Yuanzhuo, YE Cantao, CHENG Yuanzhi, NAI Weihua, PANG Zhonghe, MA Weibin, ZHANG Jian, GONG Yulie, LI Yiman, JIANG Guangzheng, WANG Yingchun, Gulbostan TURSUN, TIAN Xiaoming, LI Lengxue, SHI Deyang, ZHANG Chengkai, TIAN Yinan, LI Bin, CHEN Feng, ZHANG Lei, WANG Ke, REN Yaqian, ZHANG Weizun, LUO Guanzhong, DUAN Jiabin, CHEN Yakui, ZHANG Xiaolei, WANG Jiyang 2026, 33(1): 283-295.  DOI: 10.13745/j.esf.sf.2025.10.27 Abstract ( 15 )   HTML ( 1 )   PDF (5543KB) ( 7 )   Based on the theoretical understanding of a “cold crust and cold mantle” geological structure, Xinjiang has long been considered relatively poor in geothermal resources, with its overall exploration and utilization significantly lagging behind eastern China. As part of the Third Xinjiang Scientific Expedition Program, the geothermal study involved supplementary field surveys alongside the collection and systematization of existing geothermal research data from across Xinjiang. The expedition newly identified 11 geothermal sites and established two models of geothermal resource occurrence, estimating the region’s geothermal potential to be equivalent to over 1.8 billion tons of standard coal. Survey results indicate that Xinjiang’s geothermal resources are primarily distributed in intermountain uplifted areas, showing significant overlap with population centers. The region’s relatively small population implies a relatively high per capita geothermal resource endowment, underscoring its considerable development potential. The distribution of geothermal resources in Xinjiang exhibits a distinct pattern: generally poorer endowment in the north versus richer endowment in the south. The North Xinjiang Tianshan and Altay Mountain regions host widespread medium-low temperature hydrothermal resources formed by deep fracture circulation, while the South Xinjiang Pamir Plateau, influenced by the India-Eurasia plate collision, contains the only high-temperature geothermal system so far discovered in Xinjiang. However, the overall development and utilization of these resources remains limited, characterized by singular applications such as bathing, healthcare, and tourism, concentrated mainly in the more accessible and economically developed North Xinjiang; development in the south is comparatively scarce. This scientific expedition has clarified the distribution and status of Xinjiang’s geothermal resources. Based on these findings and aligned with Xinjiang’s territorial spatial planning and ‘Dual Carbon’ goals, the research team proposes the following development pathways: promoting “geothermal power generation + comprehensive utilization” in southern Xinjiang areas such as Tashkurgan County; advancing “Geothermal +” integrated energy development in northern Xinjiang areas like Wenquan County; and enhancing the utilization of shallow to medium-depth geothermal energy from rock and soil layers throughout the Urumqi-Changji-Shihezi urban agglomeration. Figures and Tables | References | Related Articles | Metrics

Select

Challenges and progress in fracture flow research of deep hydrogeology DONG Yanhui, WANG Liheng, ZHANG Qian, ZHOU Zhichao, WEN Dongguang, LI Shouding, WAN Li 2026, 33(1): 296-312.  DOI: 10.13745/j.esf.sf.2025.10.15 Abstract ( 17 )   HTML ( 1 )   PDF (9681KB) ( 8 )   Deep hydrogeology has emerged as a frontier field of research in the earth sciences, focusing on bedrock fracture aquifer systems at depths ranging from hundreds to thousands of meters. As China’s demands for deep resource development, environmental protection, and subsurface space utilization continue to grow, understanding and managing deep groundwater systems have become key scientific and technological challenges that support national strategic goals. This paper systematically reviews the research progress in deep hydrogeology, with an emphasis on key scientific issues, research methodologies, and engineering practices related to fluid flow in deep fractured media. First, it identifies five core scientific issues: the origin and age of deep groundwater, the mechanisms of interaction between deep and shallow hydrological cycles, the characterization of highly heterogeneous aquifer systems, the interactions between deep fluids and engineering projects, and the influence of the deep biosphere. Second, it systematically reviews key methods and technologies for investigating fluid flow and transport in deep fractured media, including high-precision laboratory observation and testing techniques, field experimentation and long-term monitoring approaches, multi-scale numerical simulation and multi-field coupling modeling technologies, as well as multidisciplinary integration and artificial intelligence research paradigms. Third, within typical engineering contexts, the paper discusses research advances related to four types of deep hydrogeological challenges: (1) fluid flow in deep low-permeability fractured rocks, using the geological disposal of high-level radioactive waste as an example, to analyze the characteristics of deep groundwater systems and hydrological circulation patterns in the pre-selected Beishan area; (2) solute transport in deep low-permeability fractured rocks, using shale gas hydraulic fracturing as an example, to explore the hydrogeological controls on the upward migration of deep fluids; (3) reactive transport of deep fluids, using deep carbonate reservoir space as an example, to reveal the dominant role of deep hydrothermal fluids in reservoir modification; and (4) Coupled thermo-hydrological processes in deep fractures, using the development of Enhanced Geothermal Systems as a case study, to analyze the governing mechanisms for reservoir fracture evolution and the sustainability of heat extraction under multi-physics coupling. Finally, the paper outlines six future research directions, including the systematic development of deep observation technologies, high-precision modeling of fracture flow, enhancement of multi-field coupling simulation capabilities, the integration of geophysical imaging with hydrogeology, high-resolution hydrochemical analysis, and the integration of artificial intelligence with big data analytics. This comprehensive summary of the theories, methods, and applications of deep hydrogeology aims to provide theoretical guidance and technical support for deep geological engineering projects such as the geological disposal of high-level radioactive waste, shale gas development, geological carbon sequestration, enhanced geothermal systems, and the exploitation of deep and ultra-deep oil and gas resources. This review also aims to provide a scientific basis for the sustainable management and environmental safety assessment of deep groundwater resources. Figures and Tables | References | Related Articles | Metrics

Select

Advances in groundwater nonlinear seepage in fractured media under conditions of high in-situ stress and temperature XU Lin, MA Haichun, WANG Jingping, ZHANG Qing, HUANG Yihang, QIAN Jiazhong, WANG Wanlin 2026, 33(1): 313-327.  DOI: 10.13745/j.esf.sf.2025.10.29 Abstract ( 11 )   HTML ( 1 )   PDF (1622KB) ( 2 )   With the growing demands in geothermal energy development, geological disposal of nuclear waste, and deep resource exploration, the seepage behavior of groundwater under conditions of high in-situ stress and temperature has become a research hotspot. Accurately modeling this behavior, however, remains a major challenge. This requires the development of multi-physical field coupling models that are applicable to high in-situ stress and temperature environments and capable of accurately describing the dynamic processes of water-rock interaction, fracture evolution, and fluid property variation. This paper comprehensively reviewed recent advances in fundamental theories, experimental studies, and numerical simulations related to groundwater seepage under conditions of high in-situ stress and temperature. First, it discussed the applicability and limitations of the classical cubic law for modeling fracture flow, along with extensions to non-Darcy and generalized Darcy flow theories, with a particular emphasis on the coupling effects of in-situ stress and temperature on permeability. It summarized various experimental findings on stress-thermal coupling and its influence on seepage parameters. Numerical simulations that analyzed the evolution of fracture geometry and flow paths under in-situ stress conditions were examined. Additionally, the study assessed the effects of mineral thermal expansion, thermal cracking, and changes in fracture roughness on flow pathways in high-temperature environments. It also evaluated the predictive capability of coupled thermal-hydraulic-mechanical (THM) models and data-driven approaches. Finally, a critical review of current theoretical models was provided, revealing their limitations in capturing the multi-field coupling mechanisms, fracture-scale effects, and long-term evolution under high in-situ stress and temperature. Future research directions were also proposed. Figures and Tables | References | Related Articles | Metrics

Select

Groundwater research in the Tibetan Plateau: Current understanding and key knowledge gaps GUO Huaming, GAO Zhipeng, HU Yalu, XING Shiping, LI Yao, JIANG Xiaowei, PENG Jianbing 2026, 33(1): 328-341.  DOI: 10.13745/j.esf.sf.2025.10.16 Abstract ( 16 )   HTML ( 1 )   PDF (4875KB) ( 9 )   The Tibetan Plateau, known as the “Asian Water Tower”, harbors groundwater resources that are crucial for sustaining the flow of major Asian rivers, maintaining regional ecological balance, and ensuring water security for hundreds of millions of people. However, groundwater in this region remains poorly understood compared to surface water and glaciers-permafrost. Existing studies have primarily utilized GRACE satellite data to estimate groundwater storage changes and explored local groundwater dynamics under freeze-thaw conditions. This paper systematically reviews the types, distribution, and utilization of groundwater on the Tibetan Plateau, provides a preliminary assessment of groundwater storage and resources, and reveals the interactions between groundwater and surface water in glacial-permafrost environments. Key unresolved scientific challenges are identified, which include: (1) the mechanisms and fluxes of shallow solid-liquid water transformation, (2) the depth and patterns of groundwater circulation, and (3) the storage capacity and variation mechanisms of groundwater. Addressing these issues will deepen the theoretical understanding of groundwater storage, occurrence patterns, circulation characteristics, and water-heat coupling processes under unique meteorological, hydrological, topographic, geological, tectonic, and hydrogeological conditions, further enriching the theoretical framework of hydrogeology. Figures and Tables | References | Related Articles | Metrics

Select

Research progress and prospect of karst eco-hydrology in China JIANG Zhongcheng, LUO Weiqun, WU Zeyan, ZHANG Cheng, ZOU Shengzhang 2026, 33(1): 342-353.  DOI: 10.13745/j.esf.sf.2025.10.26 Abstract ( 14 )   HTML ( 2 )   PDF (4054KB) ( 5 )   Since the late 20th century, research on the correlation between epikarst water utilization and the comprehensive treatment of rocky desertification in Southwest China has promoted the formation and development of eco-hydrology in the country. Notable research advancements include: (1) Studies on karst ecosystems with high dependency on groundwater: These ecosystems commonly feature dimorphic root systems, comprising both shallow and deep roots. Some deep roots extend over 100 meters and are completely dependent on karst groundwater, though the degree of dependence varies with hydrogeological conditions. (2) Investigation of the regulatory function of ecosystems on karst water resources: Ecology plays a regulatory role in the water resources of the epikarst zone, with forest ecosystems being of great value for conserving epikarst water resources. This makes epikarst springs the principal source of drinking water for residents in the karst mountainous areas of Southwest China. Thus, ecological regulation is essential for the sustainable utilization of karst water resources. (3) Analysis of the critical zone structure and eco-hydrological models: This research has elucidated the structure and characteristics of the critical karst zone, leading to the development of coupled eco-hydrological models for karst regions. These models facilitate better water resource management and help effectively prevent soil erosion and water leakage. Key future research directions include the six water cycles within the karst critical zone, establishing monitoring networks for karst eco-hydrological processes, and understanding the response mechanisms of karst eco-hydrology under extreme climate conditions and human disturbances. Figures and Tables | References | Related Articles | Metrics

Select

Development characteristics and formation mechanism of Ordovician paleokarst in the Huainan Coalfield XU Guangquan, YANG Tingting, WANG Chuanbing, CHENG Haiyan, ZHU Changhuai, ZHOU Jisheng, HE Shifang 2026, 33(1): 354-368.  DOI: 10.13745/j.esf.sf.2025.10.32 Abstract ( 20 )   HTML ( 2 )   PDF (62430KB) ( 7 )   The distribution and development of paleokarst systems of the North China Coalfield are the key factors which control deep karst fluid storage and transport. Studying their characteristics, controlling factors, and genesis is essential for assessing karst aquifer water abundance and permeability, and also is of the significance for karst water inrush during mining. Taking the Ordovician paleokarst of the Huainan Coalfield, situated in the southern North China Coalfield as objective, and with such methods as field geological survey, core drilling and microscopic identification, geophysical exploration, and rock geochemistry, the distribution, development characteristics, and controlled factors of Ordovician paleokarst are systematically investigated and studied, combined with regional sedimentary history, structural evolution history, and paleohydrology, the development characteristics and formation process of Ordovician paleokarst are discussed. Results reveal that the Ordovician paleokarst of the Huainan Coalfield is mainly divided into three types: sedimentary karst, epigenetic karst, and tectonic zone karst, which are five forms: breccia, cracks, pores, caves, and karst collapse columns, and which is mostly distributed within 0-50 m of the Ordovician weathering crust, and which internal fillings can be divided into three types: rock debris, chemical bonding materials, and mixed materials. The degree of ancient karst development is controlled by factors including carbonate rock lithology and structure, diagenetic geological processes, geological structural evolution, and paleohydrology. The formation of Ordovician paleokarst has evolved from the consolidated stage during carbonate deposition, to the stage of weathering and erosion owing to plate uplift, and to the stage of tectonic activity during the Late Yanshanian Movement. Besides, the formation process of paleokarst also is related to the leaching and dissolved effects, paleo-atmospheric freshwater, and deep hydrothermal activity. Therefore, different types of paleokarst are the results of multi-stage karst reconstruct. Figures and Tables | References | Related Articles | Metrics

Select

Carbon cycling in the karst groundwater system PU Junbing 2026, 33(1): 369-383.  DOI: 10.13745/j.esf.sf.2025.10.28 Abstract ( 14 )   HTML ( 1 )   PDF (2112KB) ( 7 )   Karst terrain covers approximately 20 million km2 globally, with karst groundwater accounting for about 26.4% of the world’s groundwater resources. In karst regions, the CO2-H2O-CaCO3 system drives carbonate rock dissolution, a key surface process that facilitates substantial atmospheric CO2 uptake. A significant portion of this absorbed CO2 enters karst aquifers, making these systems a critical component of carbon sequestration within the Earth’s surface system. The karst carbon cycle is fundamentally driven by hydrological processes, where the spatial distribution and regimes of water bodies significantly influence its pathways, efficiency, and ultimate sink effect. This paper synthesizes the fundamental characteristics of karst groundwater systems and reviews recent advances in understanding their carbon cycle, focusing on three key aspects: carbon sources, carbon dynamics, and carbon sequestration. Karst groundwater systems are characterized by multi-spatial-scale heterogeneity, structural instability of aquifers, high hydrological variability, frequent groundwater-surface water exchange, low self-purification capacity, and high vulnerability to contamination. These features collectively give rise to unique hydrological processes that control the pathways and efficiency of the system’s carbon cycle. Carbon within these systems exists primarily in four forms: Dissolved Inorganic Carbon (DIC), Particulate Inorganic Carbon (PIC), Dissolved Organic Carbon (DOC), and Particulate Organic Carbon (POC). The concentrations of these carbon species result from a dynamic equilibrium between endogenous and exogenous sources. Their relative contributions are co-determined by regional climatic, geological, and hydrological conditions, as well as ecological and biological factors. The dynamics of DIC are modulated by the CO2 effect, dilution effect, and piston effect, exhibiting significant spatial variability due to differences in hydrogeological conditions and the intensity of water-rock-gas interactions. In contrast, variations in Organic Carbon (OC) are primarily governed by source effects and hydrological effects. Key carbon sequestration mechanisms within karst groundwater systems include carbon assimilation by autotrophic microorganisms, the Biological Carbon Pump (BCP) effect, and the Microbial Carbon Pump (MCP) effect, all of which show considerable potential for long-term carbon storage. Future research should prioritize developing distributed coupled water-carbon cycle models, establishing comprehensive monitoring and analytical frameworks for the entire carbon pathway (including uptake, transformation, and sequestration) within karst subterranean river systems, and formulating artificial regulation strategies to enhance the carbon sequestration capacity of karst groundwater systems. Addressing these challenges will provide essential scientific insights for deepening our understanding of karst carbon cycling, accurately assessing regional carbon budgets, safeguarding karst ecosystem security, and supporting global “carbon neutrality” goals. Figures and Tables | References | Related Articles | Metrics

Select

Research progress and prospect of groundwater circulation in coastal zones and its environmental effects HAN Dongmei, CAO Guoliang, XIAO Yi, SONG Xianfang 2026, 33(1): 384-404.  DOI: 10.13745/j.esf.sf.2025.10.17 Abstract ( 24 )   HTML ( 1 )   PDF (3361KB) ( 6 )   Coastal groundwater circulation is a critical component of hydrodynamic processes and eco-environmental evolution in the land-sea interaction zone. Its dynamic evolution is driven by both natural and anthropogenic factors,exerting profound impacts on coastal resources,ecosystems,and sustainable development. This paper systematically reviews research progress on coastal groundwater circulation and its environmental effects,with a focus on the following aspects: multi-scale processes of the coastal water cycle,including surface water-groundwater interactions,tidal-driven groundwater flow dynamics,and hydrological exchange at the land-sea interface; driving factors of groundwater circulation evolution,encompassing climate change (e.g.,altered precipitation patterns,sea-level rise),human activities (e.g.,population and economic growth,land-use changes,water resource exploitation in watersheds),and geological-morphological controls on water flow; environmental effects induced by these hydrogeological processes are also examined,including nutrient and pollutant transport that influences nearshore water quality,ecological threats from seawater intrusion and soil salinization,the contribution of submarine groundwater discharge (SGD) to marine carbon/nitrogen cycles,and land subsidence hazards caused by over-extraction. The specific characteristics of these effects within China’s coastal zones are highlighted. Future research should prioritize the development of coupled multi-process models,high-resolution monitoring,management strategies that coordinate human and natural systems,and ecological impact assessments under carbon neutrality goals,thereby supporting the development of strategies for coastal water resource security and ecological conservation. Figures and Tables | References | Related Articles | Metrics

Select

Prediction of fracture distribution in karst area based on machine learning method: Taking Fangshan area in Beijing as a case study QIAO Xiaojuan, LUO Chengke, CHAI Xinyu, YU Wenjin 2026, 33(1): 405-418.  DOI: 10.13745/j.esf.sf.2025.10.30 Abstract ( 15 )   HTML ( 1 )   PDF (5190KB) ( 5 )   The development of karst fissures is characterized by high dimensionality, nonlinearity, and spatial heterogeneity. Accurately characterizing the spatial distribution of fissures remains a challenging issue in the study of karst development patterns. Data-driven machine learning modeling methods can effectively capture the implicit nonlinear and discontinuous characteristics within fissure systems, thereby significantly improving the efficiency and accuracy of fissure identification and characterization. This study takes the Zhangfang area of Fangshan, Beijing, as its research subject. Based on detailed field-measured fissure data, it systematically integrates multi-source datasets including surface topographic information, regional tectonic background, stratigraphic lithology distribution, and groundwater levels. Using a machine learning framework, a comprehensive quantitative feature system was constructed, covering multiple dimensions such as spatial fault influence, stratigraphic lithological characteristics, variations in groundwater depth, and high-precision topographic derivatives (e.g., slope, curvature). The study focused on comparing three machine learning methods—Support Vector Regression, Extreme Gradient Boosting, and Random Forest-to predict the development and spatial distribution of karst fissures in the study area. The results indicate that the prediction model based on Random Forest performed the best. The simulation results for fissure density, joint orientation, and dip angle were most consistent with measured statistical data, demonstrating the most robust model performance, strong generalization capability, and high method applicability. It exhibited unique advantages in representing complex geological processes such as multi-phase fissure development. The findings of this study reveal that integrating data-driven models with in-depth geological mechanism analysis is an effective approach to overcoming the challenges in the quantitative characterization and prediction of complex karst systems. Figures and Tables | References | Related Articles | Metrics

Select

Research on groundwater level prediction of northern karst spring of China based on LSTM-Attention neural network HUANG Linxian, XU Zhenghe, ZHI Chuanshun, LI Shuang, LIU Zhizheng, XING Liting, ZHU Henghua, WANG Xiaowei, BI Wenwen, HU Xiaonong 2026, 33(1): 419-431.  DOI: 10.13745/j.esf.sf.2025.10.1 Abstract ( 16 )   HTML ( 1 )   PDF (6399KB) ( 4 )   Karst groundwater serves as a critical water source in northern karst regions of China, and accurately forecasting its water level dynamics is essential for the scientific management and conservation of groundwater resources. However, the inherent strong heterogeneity and anisotropy of karst aquifer systems often result in non-stationary and nonlinear groundwater fluctuations, posing significant challenges for reliable prediction and frequently leading to considerable errors. This study proposes a multivariate groundwater level forecasting model for the Baotu Spring, based on the integration of an attention mechanism and a long short-term memory (LSTM) neural network. The model is trained and validated using daily data from 2013 to 2024, including precipitation (as a proxy for recharge), vapor pressure, air temperature, and groundwater extraction volume (as indicators of discharge). The main findings are as follows: (a) The BEAST (Bayesian Estimator of Abrupt Change, Seasonality, and Trend) algorithm was applied to the 1958-2024 Baotu Spring groundwater level time series, identifying four abrupt change points. Based on these, the time series was segmented into four distinct stages. (b) Cross-correlation analysis revealed a time lag of approximately 2-3 months between precipitation and groundwater level variations, suggesting a strong dynamic coupling between the two variables. (c) The proposed model incorporates multiple input variables (precipitation, vapor pressure, temperature, and extraction volume). The interplay among these variables enhances model interpretability and contributes to improved forecasting accuracy. (d) Daily temperature data were smoothed using a sinusoidal fitting function, which effectively reduces measurement noise and improves prediction performance. (e) Compared to standard LSTM and gated recurrent unit (GRU) models, the LSTM-Attention model demonstrates superior predictive capability. By leveraging the attention mechanism to prioritize influential features, it achieved a root mean square error (RMSE) of 0.13 m and a coefficient of determination (R2) of 0.94. In summary, the LSTM-Attention-based forecasting model exhibits robust accuracy and stability, offering valuable insights into the precise prediction of groundwater dynamics in karst environments. Figures and Tables | References | Related Articles | Metrics

Select

Error correction method for groundwater numerical models considering parameter uncertainty HU Litang, GAN Lin, SUN Jianchong, LIU Hongbo, TIAN Lei, SHEN Qi 2026, 33(1): 432-443.  DOI: 10.13745/j.esf.sf.2025.10.25 Abstract ( 12 )   HTML ( 1 )   PDF (5431KB) ( 1 )   Uncertainty in model structure and parameters remains a critical bottleneck, limiting the accuracy of groundwater simulations and their application in high-precision management. To address this, we proposed an error correction method for simulated hydraulic heads based on observation well data. The method involves three steps: correcting simulated errors at observation wells, interpolating the spatial error distribution using inverse distance weighting, and applying the resulting error field to compensate other grid cells in the model, thereby enhancing overall accuracy. Using an in-house developed numerical groundwater flow model, we designed three heterogeneous scenarios with variable flow rates and different numbers of pumping and observation wells. These scenarios were used to systematically evaluate the method’s performance under homogeneous parameter assumptions. Results indicated a significant improvement in the accuracy of simulated heads at the observation wells. The degree of improvement was inversely proportional to the mean hydraulic conductivity and directly proportional to its variance. Under conditions of multi-well pumping and injection with strong heterogeneity and variable flow rates, the root mean square error decreased from 1.5 m to 0.25 m. Furthermore, analysis of scenarios with different numbers of calibration points showed that while the method significantly improved accuracy at those specific locations, its enhancement of the overall spatial head field was relatively limited. The presented approach offers a practical and effective solution for improving the accuracy of calibrated groundwater models and shows considerable potential for broader application. Figures and Tables | References | Related Articles | Metrics

Select

Research progress in geophysical methods on groundwater contamination MAO Deqiang, MENG Jian, ZHAI Kexiang, ZENG Zihao, LIU Shiliang 2026, 33(1): 444-469.  DOI: 10.13745/j.esf.sf.2025.10.8 Abstract ( 14 )   HTML ( 3 )   PDF (8883KB) ( 5 )   Accurate characterization and monitoring of the spatiotemporal distribution of contaminants in groundwater have become crucial objectives for effective groundwater management. However, the elusive nature and heterogeneity of groundwater contamination constrain the precise characterization of contamination distribution and migration paths. Geophysical methods offer advantages such as being non-invasive, low-cost, and efficient, while providing continuous information, and have thus emerged as important tools for characterizing and monitoring groundwater contamination. This review summarizes mature geophysical methods in the field of groundwater contamination, outlining their basic principles, models, and research examples. Through integrated column experiments and micro-scale imaging techniques, interpretive models for geophysical responses in porous media under contaminant influence were established. A series of petrophysical models were developed to relate hydrogeological parameters to geophysical data, demonstrating the potential for direct inversion of contaminant concentrations from field-scale geophysical measurements. Structure-constrained inversion methods that incorporate prior information were developed to reduce uncertainty in the results. Novel numerical simulation approaches were introduced to integrate and couple simulations of multi-source data. To address the problems of uncertainty in geophysical methods, various mechanistic models, inversion methods, and multi-source data fusion approaches have been developed. Future research should focus on the pore-scale mechanisms of contaminant migration and establish a unified petrophysical database of geophysical responses to groundwater contamination. Moreover, by combining new technologies such as artificial intelligence and data assimilation, it will be possible to describe, predict, and manage contaminated groundwater sites with greater accuracy and comprehensiveness. Figures and Tables | References | Related Articles | Metrics

Select

Analyses of groundwater storage changes in the Eastern Tibetan Plateau based on gravimetric satellites and baseflow separation LIU Suyi, HAN Ning, HUANG Zhiyong, ZHENG Longqun, ZHANG Chong, GONG Huili, PAN Yun 2026, 33(1): 470-482.  DOI: 10.13745/j.esf.sf.2025.10.37 Abstract ( 10 )   HTML ( 1 )   PDF (7838KB) ( 2 )   Under the background of global warming and intensifying human activities, quantitatively analyzing the spatiotemporal evolution of groundwater storage on the Tibetan Plateau is crucial for understanding the changing mechanisms of the “Asian Water Tower” water cycle. This study combines data from the Gravity Recovery and Climate Experiment (GRACE) and GRACE-Follow On (GRACE-FO) satellites, global land surface models, and a global hydrology model to estimate groundwater storage changes in the Eastern Tibetan Plateau. The results are compared and validated with those obtained from baseflow separation. For the period 2003-2022, GRACE/GRACE-FO inversion results indicate that terrestrial water storage (TWS) changes were dominated by soil moisture storage (SMS; 48.45%), followed by groundwater storage (GWS; 32.69%). Specifically, SMS was the dominant component of TWS change in three sub-basins (Upper Yangtze, Yalong River, and Dadu River, accounting for 52.7% of the area), whereas GWS was the major contributor in the remaining seven sub-basins (47.3% of the area). Over the eastern Tibetan Plateau, GWS exhibited a significant increasing trend ((2.11±0.57)mm/a). Among the 10 sub-basins, seven exhibited increasing trends in GWS changes derived from both baseflow separation and GRACE inversion, with a correlation coefficient of 0.78 between the two methods. However, the increasing trends derived from baseflow separation are significantly lower, possibly due to: (1) the continuously reducing catchment area during baseflow recession, (2) a systematic underestimation of baseflow by the separation algorithm, and (3) errors inherent in GRACE/GRACE-FO data processing. Multivariate regression analysis reveals that precipitation, air temperature, and downward shortwave radiation jointly drive the increasing trend in groundwater storage across the study area. Figures and Tables | References | Related Articles | Metrics

Select

Frontier advances and challenges of machine learning in groundwater science KANG Jinzheng, MO Shaoxing, KANG Xueyuan, DANG Jingxuan, CHENG Chijitai, XU Peijie, SHI Xiaoqing 2026, 33(1): 483-499.  DOI: 10.13745/j.esf.sf.2025.10.19 Abstract ( 12 )   HTML ( 1 )   PDF (4036KB) ( 5 )   Groundwater is a vital component of the global freshwater supply and plays a critical role in ensuring the water resource security,maintaining ecosystem stability,and supporting the sustainable development of human society. As a deeply buried,dynamically evolving,and highly heterogeneous natural system,groundwater poses persistent challenges in observation,modeling,and management,primarily due to the sparsity and uneven distribution of monitoring points and the strong nonlinearity of subsurface physical processes. In recent years,the rapid advancement of artificial intelligence—particularly machine learning—has demonstrated significant potential in feature extraction,high-dimensional nonlinear modeling,and optimization,thereby driving groundwater science into a new paradigm that integrates data-driven approaches with intelligent modeling. This paper provides a comprehensive review of machine learning applications in addressing three core scientific problems in groundwater research: element characterization,process modeling,and management optimization. Key topics include aquifer structure identification,groundwater level,storage and quality analysis,forward simulation,inverse modeling,physics-informed modeling,and groundwater resource management. Based on these developments,a generalized research framework is proposed,structured around the cycle of “scientific problem-data-driven modeling-model selection-result interpretation.” Given existing challenges such as limited incorporation of physical constraints,weak generalization capability,and poor model interpretability,this study also discusses future directions,including the integration of multi-source heterogeneous data,embedding of physical priors,and the use of causal inference methods. This work aims to provide a conceptual and methodological foundation for the application of machine learning in groundwater science and to outline pathways toward developing intelligent and sustainable groundwater systems. Figures and Tables | References | Related Articles | Metrics

Select

Physics-informed neural networks with hard constraints for hydraulic conductivity field inversion SHU Wei, JIANG Jianguo, WU Jichun 2026, 33(1): 500-510.  DOI: 10.13745/j.esf.sf.2025.10.38 Abstract ( 10 )   HTML ( 1 )   PDF (5901KB) ( 4 )   In recent years, Physics-Informed Neural Networks (PINNs) have been widely applied to numerically solve partial differential equations and in computational fluid dynamics, having demonstrated promising potential for groundwater modeling. Existing studies typically handle boundary conditions in groundwater models using soft-constraint algorithms, which satisfy physical constraints approximately by minimizing the boundary condition error. However, the application of hard-constraint algorithms, which can enhance solving accuracy and stability by design, remains limited in groundwater modeling. To address this gap, this paper introduces a hard-constraint PINNs algorithm that simultaneously enforces constant-head and no-flow boundary conditions. Using the inverse problem of estimating the hydraulic conductivity field in a two-dimensional confined aquifer as a case study, we demonstrate the superior performance of the hard-constraint approach over its soft-constraint counterpart. The results show that the proposed method reduces the average relative inversion error by 75% compared to soft-constraint PINNs and by 60% compared to hard-constraint PINNs considering constant-head boundaries alone. Furthermore, the proposed method reduces the number of required training samples and alleviates the need for extensive manual hyperparameter tuning, thereby improving training efficiency and robustness. Therefore, the hard-constraint PINNs method proves to be a highly accurate and efficient approach for reconstructing hydraulic conductivity fields, demonstrating significant potential for broader application. Figures and Tables | References | Related Articles | Metrics

Select

Integrating numerical simulation and machine learning for identification of groundwater potential zone and its governing factors in the Minqin Basin, Northwest China ZHOU Feiran, YIN Ziyue, SUN Xiaomin, SONG Jian, YANG Yun, WU Jianfeng 2026, 33(1): 511-522.  DOI: 10.13745/j.esf.sf.2025.10.6 Abstract ( 11 )   HTML ( 1 )   PDF (9146KB) ( 5 )   Oases are critical for maintaining ecological security in arid regions, and their stability is heavily dependent on sustainable groundwater resources. The delineation of groundwater potential zones (GPZs) offers decision-making support for optimal water resource allocation in arid areas. This study focuses on the Minqin Basin, a typical inland river oasis in northwest China, and proposes an integrated approach that combines numerical simulation with machine learning to assess groundwater potential in arid regions. Precise spatial distributions of groundwater depth and two key parameters (hydraulic conductivity and specific yield) were derived from groundwater numerical simulation. Considering 18 influencing factors across five categories (meteorology, hydrology, land use, topography, and geology), six machine learning models were employed to evaluate the spatial characteristics of groundwater potential. Results indicate that the LightGBM (Accuracy: 87.87%; F1 score: 0.716; AUC: 0.943) achieved the best performance, followed by XGBoost and Random Forest, whereas the Support Vector Machine, K-Nearest Neighbors, and BP Neural Network models demonstrated inferior performance. Subsequently, feature importance analysis using tree-based models (Random Forest, XGBoost, and LightGBM) revealed that groundwater depth (weight: 17.1%-18.5%) was the primary controlling factor. Secondary factors include potential evapotranspiration (12.5%-14.2%), precipitation (8.6%-12.5%), NDVI (6.2%-12.8%), and surface elevation (6.7%-11.4%). The proposed methodology establishes a multi-parameter assessment framework for groundwater potential in arid oasis areas, offering a scientific basis for sustainable groundwater management in water-scarce regions. Figures and Tables | References | Related Articles | Metrics

Select

Dynamic characterization of heat transfer processes in low-permeability media using ERT during thermal tracer tests ZHAO Yongsheng, WANG Jinguo, QIAO Fei, LIU Ruitong, CHEN Zhou 2026, 33(1): 523-533.  DOI: 10.13745/j.esf.sf.2025.10.11 Abstract ( 8 )   HTML ( 1 )   PDF (5266KB) ( 1 )   Thermal tracer technology is a critical method for characterizing hydraulic and thermophysical properties of porous media, with extensive applications in hydrogeology. However, conventional thermal tracer tests rely on point-scale temperature data from monitoring wells, which can not resolve the spatial architecture of heat transport processes. To overcome the spatial resolution limitations of point-scale temperature monitoring, this study integrated electrical resistivity tomography (ERT) into thermal tracer tests in low-permeability media. We combined ERT with thermal tracer tests to investigate its capacity to map heat transport dynamics with high spatial resolution. We also developed a quantitative resistivity-temperature inversion model for clay formations. The model’s reliability was verified through numerical forward modeling, and key influencing factors were analyzed to demonstrate its field applicability. Key findings include: In heat conduction-dominated zones (monitoring wells), ERT-derived temperatures closely matched groundwater temperature trends and amplitudes. In local thermal convection zones (e.g., near the injection well), ERT captured trends identical to point measurements, confirming its reliability in low-permeability settings. ERT successfully identified preferential flow paths and thermally sensitive zones, demonstrating high sensitivity to subsurface heterogeneity and substantially improving the spatial resolution of heat transport visualization. The developed resistivity-temperature inversion framework provides a novel approach for thermal tracing in low-permeability media, effectively overcoming the spatial resolution constraints of conventional methods. Figures and Tables | References | Related Articles | Metrics