Physics and Chemistry of the Earth最新文献

{"title":"Effectiveness of watershed management on water balance components-a review","authors":"Saeedreza Moazeni ,&nbsp;Ali Salajegheh ,&nbsp;Shahram Khalighi-Sigaroodi ,&nbsp;Ali Golkarian ,&nbsp;Artemi Cerda","doi":"10.1016/j.pce.2026.104277","DOIUrl":"10.1016/j.pce.2026.104277","url":null,"abstract":"<div><div>Watersheds serve as fundamental units of hydrological systems, playing a crucial role in freshwater supply, biodiversity conservation, and ecosystem sustainability. This study conducts a bibliometric and qualitative analysis of global research (1979–2024) to evaluate the effectiveness of watershed management interventions on key water-balance components, including runoff, infiltration, evapotranspiration, and groundwater recharge. Findings indicate a significant increase in publications, particularly after 2011. We used VOSviewer and Bibliometrix to map publication trends, co-occurrence and co-citation networks, leading authors and journals, and thematic clusters. Prominent studies focus on Land-use impacts on water resources, the application of hydrological models, and the role of vegetation in water regulation. The United States, China, and Canada lead research efforts in this field. Our synthesis classifies interventions into biological, mechanical and managerial types and summarizes reported effects on runoff, infiltration, evapotranspiration (ET), soil moisture, surface storage and groundwater recharge. Results show a marked increase in publications after 2011 and concentration of research activity in the United States, China and Canada. Among water-balance components, runoff (n = 447) and ET (n = 308) dominate the literature, whereas soil moisture (n = 66), surface storage (n = 40) and baseflow (n = 38) are comparatively underrepresented. Hydrological process models were the most used estimation approach (n = 440), followed by remote sensing (n = 140), groundwater models (n = 82) and machine-learning methods (n = 38). From the reviewed studies we synthesise typical outcomes: mechanical structures (e.g., terraces, check-dams) are frequently associated with reductions in surface runoff (up to ∼25 % in reported cases) and context-dependent increases in recharge (reported ranges of ∼40–70 %), while biological measures (e.g., afforestation) often improve infiltration but can elevate ET in water-limited environments. We identify recurrent methodological shortcomings — inconsistent reporting of uncertainty, limited reproducibility of bibliometric settings, and scarce comparative field studies — and propose a focused research agenda: transparent bibliometric reporting, prioritized monitoring of underexplored components (soil moisture, baseflow), development of hybrid process–data modeling frameworks, and targeted, context-specific evaluations of interventions under climate variability. This synthesis provides a state-of-the-art overview and a structured set of priorities to guide future watershed management research and policy.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"143 ","pages":"Article 104277"},"PeriodicalIF":4.1,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Emerging contaminants in soil: Source-oriented probabilistic health risks of potentially toxic elements and environmental impacts in Southern India","authors":"Muthusamy Subramanian ,&nbsp;Gopal Veeramalai ,&nbsp;Mohana Perumal ,&nbsp;Prasenjeet Chakraborty ,&nbsp;Ribu Cherian ,&nbsp;Ramamoorthy Ayyamperumal","doi":"10.1016/j.pce.2026.104291","DOIUrl":"10.1016/j.pce.2026.104291","url":null,"abstract":"<div><div>Potentially toxic elements (PTEs) in roadside soils pose serious risks to human health and the environment, with anthropogenic activities being the primary source of contamination. In this study, we evaluated the contamination levels, ecological risks, and human health hazards of heavy metals (Fe, Mn, Cr, Cu, Pb, Ni, Co, Zn) in soil samples collected from Tuticorin, Southern India. Elemental concentrations were determined using Energy Dispersive X-ray Fluorescence (EDXRF) and validated with HISS-1 standard reference material, showing recoveries of 92.3–100 %. Geochemical indices (Igeo, EF, CF, CD, PLI, PERF, PERI, and GUFI) revealed that Cu (Igeo = 2.17; EF = 5.55; CF = 7.70), Zn (Igeo = 1.22; EF = 2.77; CF = 3.71), and Pb (Igeo = 0.60; EF = 1.95; CF = 2.48) were the most enriched metals, reflecting moderate to severe contamination, while Fe, Mn, Co, and Ni were largely of natural origin. Principal component and correlation analyses indicated two main sources of contamination: industrial emissions (thermal power plants, leather industry) and traffic/maritime activities. Ecological risk assessment identified hotspots at sampling points 14, 17, 27–29, and 31–33, primarily due to Cu, Zn, and Pb. Probabilistic human health risk assessment showed that children were exposed to elevated non-carcinogenic risks (THI = 1.90) and carcinogenic risks (TCRI = 4.51 × 10⁻⁴), mainly from Cr, Ni, Zn, and Mn, while adults experienced lower risks (THI = 0.26; TCRI = 5.02 × 10⁻⁵). These findings highlight the need for continuous monitoring, targeted mitigation, and regulatory measures to control heavy metal contamination in roadside soils and protect vulnerable populations, especially children.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104291"},"PeriodicalIF":4.1,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925382","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Synergistic stability appraisal of engineered slopes along NH-7 in the lesser Himalaya using multiparametric empirical indices","authors":"Tarique Zafeer,&nbsp;Tariq Siddique,&nbsp;Harsh Varshney,&nbsp;Atif Ahamad,&nbsp;Pirzada Mohammad Haris,&nbsp;Md Erfan Ali Mondal","doi":"10.1016/j.pce.2026.104268","DOIUrl":"10.1016/j.pce.2026.104268","url":null,"abstract":"<div><div>Slope instability along engineered slopes in the Himalaya is one of the major concerns for stakeholders. The present research delves into slope stability appraisal along a strategic and critical segment of National Highway 7 (NH-7) in Uttarakhand. The evaluation employs a synergistic geotechnical appraisal using kinematic analysis, Hazard Index (HI) and Continuous Slope Mass Rating (CoSMR). Fifteen slopes prone to failure were examined by determining the pertinent input geotechnical parameters. Kinematic analysis was applied to define the failure, which indicated that six slopes are prone to planar sliding, four are vulnerable to wedge sliding and five may offer toppling failures. Stability assessments incorporated the Factor of Safety (FoS) from kinematic principles and triggering factors to calculate HI. According to the geotechnical assessment, the FoS values for investigated slopes range from 0.17 to 2.82, whereas the CoSMR values range between 25 and 77, which corresponds to unstable to stable and the HI score varies from 1 to 8.8 and which corresponds to good to very poor slope conditions. Comparative analysis of HI and CoSMR for structurally controlled slopes demonstrates a strong correlation (R² = 0.9), highlighting the synergistic control of rock mass quality, structural orientation and external triggers on stability of cut slopes. These findings provide a rapid and cost-effective framework for slope stability assessment, enabling identification and prioritisation of critical road-cut slopes and guiding slope-specific mitigation measures. This multiple rock mass classification (RMC) approach lays a foundation for future advanced numerical simulation studies to quantify the failure mechanism.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104268"},"PeriodicalIF":4.1,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Temporal evolution and dominant drivers of groundwater sulfate in China","authors":"Qing Zhou ,&nbsp;Ke Xing ,&nbsp;Jun Zhang ,&nbsp;Yijun Yao","doi":"10.1016/j.pce.2026.104297","DOIUrl":"10.1016/j.pce.2026.104297","url":null,"abstract":"<div><div>China's groundwater is facing a significant threat from sulfate pollution. Here we analyzed 2804 regional surveys of groundwater sulfate levels in China from 1990 to 2020, examining distribution, trends, and drivers. This study identifies a sustained rise in sulfate, with median concentrations increasing from 53.7 mg/L in 1990 to 137.3 mg/L in 2020. A stark contrast is observed between regions: levels are generally higher in northern and northwestern China than in the south and southeast; nationally, 24.2 % of counties exceed the Chinese standard of 250 mg/L and 12.0 % exceed the threshold of 500 mg/L, whereas about 64.4 % remain at or below 150 mg/L. From 1990 to 2020, non-ferrous metal mining emerged as the dominant driver of changes in groundwater sulfate concentrations, with additional contributions from coal consumption and industrial discharges; domestic sources show a gradual rise, while agriculture acts as a diffuse but smaller contributor. This analysis underscores the need for region-specific policies and interventions—particularly in northern and northwestern provinces—to curb sulfate inputs, strengthen wastewater management in urbanizing areas, and protect vulnerable shallow aquifers.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104297"},"PeriodicalIF":4.1,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977184","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated source apportionment, risk assessment, and probabilistic modeling of potentially toxic elements in soils from an undeveloped metallogenic region, northwest China","authors":"Qiong Han ,&nbsp;Wenfeng Wang ,&nbsp;Tongyang Zhao ,&nbsp;Liuyuan Jin ,&nbsp;Yangshuang Wang ,&nbsp;Shiming Yang ,&nbsp;Yunhui Zhang","doi":"10.1016/j.pce.2026.104288","DOIUrl":"10.1016/j.pce.2026.104288","url":null,"abstract":"<div><div>While risk zones of potentially toxic elements (PTEs) are generally associated with intensive anthropogenic disturbances, undeveloped metallogenic regions (UMRs) feature naturally high PTE concentrations that have received limited scientific attention. To fill this gap, this study investigated the sources and quantified the contributions of PTEs in the soils of a UMR in Northwest China, while assessing their associated ecological-health risks. A total of 480 soil samples were collected for analysis. A multi-method framework combining geochemical baseline values (GBVs), Positive Matrix Factorization (PMF), integrated pollution indices, and probabilistic health risk modeling was established. The results revealed that Mercury (Hg) exhibited the highest exceedance rate (13.33 %) compared with other PTEs. PTEs were predominantly derived from natural sources: soil parent material (61.91 %), Pb enrichment (22.98 %), Cu mineralization (10.50 %), and hydrothermal processes (4.61 %), suggesting a primarily geogenic origin. Ecological risk assessment identified Hg as the main contributor to ecological risk. Overall, soil quality in the study area ranged from unpolluted to slightly polluted, with localized high-risk zones mainly distributed in the eastern region. Health risk assessment showed that chromium (Cr) and cobalt (Co) were the major contributors to non-carcinogenic risk, while Cr dominated the carcinogenic risk. Notably, 49.32 % of the samples posed unacceptable non-carcinogenic risks to children, 10.01 % and 0.90 % posed unacceptable carcinogenic risks to children and adults, respectively. These results highlighted that natural metallogenic processes alone could generate unacceptable risks to ecological systems and vulnerable populations. Based on these findings, long-term monitoring of Cr, Co, and Hg in soils near PTE-enriched areas is recommended to mitigate potential ecological and health threats. The study results can provide a scientific basis for the management of soil PTEs in high-background metallogenic regions worldwide.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104288"},"PeriodicalIF":4.1,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977767","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimating reference evapotranspiration in data-scarce regions: Comparative analysis and calibration of empirical models against the Penman-Monteith method in Malawi","authors":"Patsani Gregory Kumambala ,&nbsp;Lenard Kumwenda ,&nbsp;Deogratias M.M. Mulungu ,&nbsp;Sheila Kavwenje","doi":"10.1016/j.pce.2026.104269","DOIUrl":"10.1016/j.pce.2026.104269","url":null,"abstract":"<div><div>Accurate estimation of reference evapotranspiration (ET₀) is vital for irrigation planning and hydrological modelling yet remains challenging in data-scarce regions such as Malawi. This study evaluated three empirical ET₀ models—the Hargreaves–Samani (HS), Priestley–Taylor (PT), and Turc methods—against the FAO-56 Penman–Monteith (PM) standard across six meteorological stations representing Malawi's major agro-climatic zones. Daily temperature, humidity, wind speed, and sunshine data were obtained from the Department of Climate Change and Meteorological Services in Malawi and aggregated to monthly scales. Linear regression calibration aligned empirical estimates with FAO-56 p.m. values, while the F-test and Jarque–Bera test assessed statistical significance and residual normality. Results show that radiation-based models outperformed the temperature-based HS method. The calibrated Turc model achieved the highest accuracy (<span><math><mrow><msup><mi>R</mi><mn>2</mn></msup></mrow></math></span> = 0.79–0.94), followed by PT (<span><math><mrow><msup><mi>R</mi><mn>2</mn></msup></mrow></math></span> = 0.55–0.95), whereas HS showed lower accuracy (<span><math><mrow><msup><mi>R</mi><mn>2</mn></msup></mrow></math></span> = 0.40–0.74) and strong sensitivity to wind speed. Model performance declined from humid highlands to arid lakeshore regions, highlighting the influence of aerodynamic factors. Overall, the calibrated Turc and Priestley–Taylor models provide robust, low-data alternatives to FAO-56 p.m., particularly in humid and moderate-wind environments. These findings offer practical tools for irrigation management, drought monitoring, and water-resource assessment in Malawi and similar tropical data-limited settings.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104269"},"PeriodicalIF":4.1,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Thermo-mechanical response of Sakesar limestone under cyclic steam injection: Insights from the Potwar Basin, Pakistan","authors":"Syed Bilawal Ali Shah","doi":"10.1016/j.pce.2026.104285","DOIUrl":"10.1016/j.pce.2026.104285","url":null,"abstract":"<div><div>Thermal enhanced oil recovery (EOR) is an effective method for extending production from mature fractured carbonate reservoirs. This study examines the thermo-mechanical behavior of the Eocene Sakesar Formation in the Balkassar Oil Field, Potwar Basin, Pakistan, under cyclic steam stimulation (CSS) conditions representative of near-wellbore operations. An integrated experimental and numerical approach was employed, including uniaxial compressive and Brazilian tensile strength tests, joint shear tests, permeability measurements, micro-computed tomography (micro-CT), and coupled thermo-hydro-mechanical (THM) modeling. Limestone cores were subjected to ten heating–cooling cycles between 25 °C and 180 °C, with properties assessed after each cycle. Results show progressive, non-linear mechanical degradation, with uniaxial compressive strength decreasing from 47.5 MPa to 31.5 MPa and Young's modulus from 22.4 GPa to 16.7 GPa. In contrast, permeability increased irreversibly from 28 mD to 56 mD. Micro-CT analysis indicates that permeability enhancement is governed mainly by fracture network connectivity and coalescence rather than aperture enlargement alone, despite mean fracture aperture increasing from 0.15 mm to 0.39 mm. THM simulations reveal significant near-wellbore thermal stress redistribution and identify a high slip-tendency zone within ∼10 m of the wellbore during heating, suggesting potential shear reactivation of pre-existing fractures. These findings highlight the dual impact of cyclic thermal stimulation: improved injectivity through fracture connectivity evolution and increased geomechanical risk related to fracture reactivation and wellbore stability in compressional settings. This integrated framework supports safer and more efficient thermal EOR design for the Balkassar Oil Field and similar fractured carbonate reservoirs.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104285"},"PeriodicalIF":4.1,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hydrolytic pretreatment strategies of brewery mashing wastewater for improved succinic acid production","authors":"Haidong Zhu,&nbsp;Dafang Fu,&nbsp;Yixin Yan,&nbsp;Rajendra Prasad Singh","doi":"10.1016/j.pce.2026.104290","DOIUrl":"10.1016/j.pce.2026.104290","url":null,"abstract":"<div><div>Brewery mashing wastewater (BMW), rich in oligosaccharides and organic matter, was evaluated as a low-cost substrate for succinic acid (SA) production by <em>Actinobacillus succinogenes</em>. To enhance fermentability, BMW pretreatment was conducted in acid, alkaline, and enzymatic conditions, and their effects on sugar release, inhibitor formation, and fermentation performance were systematically compared. Pectinase hydrolysis (PH) proved most effective, releasing 26.47 g/L glucose and simultaneously minimizing inhibitor accumulation. Fermentation with pectinase hydrolysates achieved an SA titer of 16.19 g/L and a yield of 0.61 g SA/g glucose, exceeding reported values for food-waste hydrolysates (0.43 g SA/g glucose) and approaching yields typically obtained with refined substrates. In contrast, acid and alkaline hydrolysis generated high levels of furanic inhibitors, impairing glucose utilization and limiting SA titers (&lt;1.40 and &lt; 1.61 g/L, respectively). α-Amylase hydrolysis exhibited moderate saccharification efficiency, and glucose supplementation was required to achieve yields comparable to PH. These results highlight that pretreatment choice is pivotal in shaping fermentation outcomes by balancing sugar accessibility against inhibitor toxicity. Enzymatic hydrolysis, particularly PH, offers a scalable and environmentally friendly path to convert BMW into platform chemicals. This study provides both mechanistic insights and practical evidence supporting the integration of BMW into circular bioeconomy strategies.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104290"},"PeriodicalIF":4.1,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977169","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robust monthly precipitation forecasts using VMD-decomposed hybrid LSSVM–BiLSTM model optimized by Harris Hawks algorithm","authors":"Behnoush Farokhzadeh ,&nbsp;Sepideh Choobeh ,&nbsp;Rasool Imani","doi":"10.1016/j.pce.2025.104257","DOIUrl":"10.1016/j.pce.2025.104257","url":null,"abstract":"<div><div>Accurate prediction of monthly precipitation is crucial for sustainable water resources management in climate-sensitive regions such as the Urmia Lake Basin, northwestern Iran. This study proposes an optimized hybrid forecasting framework integrating Variational Mode Decomposition (VMD), Least Squares Support Vector Machine (LSSVM), Bidirectional Long Short-Term Memory (BiLSTM) networks, and the Harris Hawks Optimization (HHO) algorithm. The normalized monthly precipitation series (1980–2024) was decomposed into five Intrinsic Mode Functions (IMFs) and a residual component using VMD. LSSVM was applied to model the high- and medium-frequency IMFs, while BiLSTM captured the long-term residual trend. The HHO algorithm was employed to optimize model hyperparameters, minimizing reconstruction and prediction errors. Model evaluation revealed outstanding predictive performance, with Nash–Sutcliffe Efficiency (NSE) exceeding 0.97 and Root Mean Square Error (RMSE) below 1.6 mm across both training and testing phases, and strong correlation between observed and simulated rainfall. Bootstrap-based uncertainty analysis confirmed the model's stability and reliability. The optimized hybrid model was further applied to project monthly precipitation for 2030–2050, indicating an approximate 7 % increase in mean annual rainfall relative to the historical average (1980–2024). These results demonstrate that the VMD–HHO–LSSVM–BiLSTM hybrid framework effectively captures nonlinear, nonstationary rainfall dynamics across multiple temporal scales and provides a robust tool for hydrological forecasting and adaptive water management in the Urmia Lake Basin.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104257"},"PeriodicalIF":4.1,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Designing landslide mitigation strategies through developing knowledge aware deep learning model integrated explainable artificial intelligence","authors":"Inzamam Ul Haq ,&nbsp;Habibullah ,&nbsp;Shafiq Ahmed ,&nbsp;Swapan Talukdar ,&nbsp;Mansoor Ahmad ,&nbsp;Md Sarfaraz Asgher ,&nbsp;Imran Khan ,&nbsp;G.V. Ramana","doi":"10.1016/j.pce.2025.104266","DOIUrl":"10.1016/j.pce.2025.104266","url":null,"abstract":"<div><div>Landslides are a recurrent and damaging hazard in the Pir Panjal Range of Jammu and Kashmir, India, where steep slopes, fragile lithology and rapid human interventions coincide. This study develops an integrated landslide susceptibility mapping framework that combines machine learning (ML), deep learning (DL) and explainable artificial intelligence (XAI) to support risk-informed planning. An inventory of 1105 landslides and eighteen conditioning factors trains four ML models: Random Forest (RF), Support Vector Machine (SVM), Gradient Boosting Machine (GBM) and Categorical Boosting (CatBoost), evaluated with five-fold stratified cross-validation and a held-out test set. In parallel, a Bayesian-optimised DL model, in which a Variational Autoencoder (VAE) learns latent features that are classified by a Multilayer Perceptron (MLP), captures complex non-linear controls on slope failure. The best performing models achieve an area under the receiver operating characteristic curve (AUC-ROC) of about 0.96 and overall accuracy close to 0.90 on validation data, indicating high discriminatory skill. Spatially, High and Very High susceptibility zones together occupy roughly one quarter of the range and are concentrated in tehsils such as Gool Gulabgarh, Ramban, Thathri and Banihal, which emerge as hotspots for intervention. XAI diagnostics highlight vegetation condition, monsoon rainfall, lineament density and drainage density as dominant controls on susceptibility, while counterfactual experiments show that realistic improvements in vegetation cover, surface drainage and slope management can shift many locations from high to lower risk classes. The framework therefore provides accurate, interpretable and useful susceptibility maps for land-use regulation, infrastructure planning and landslide mitigation in data-constrained mountain environments.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"143 ","pages":"Article 104266"},"PeriodicalIF":4.1,"publicationDate":"2025-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146015539","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}