Engineering Geology最新文献

{"title":"Uniaxial compressive strength and crack propagation of debris-ice mixtures: Insights from experiments and particle-based modeling","authors":"Guoqing Cui ,&nbsp;Jianlin Chang ,&nbsp;Markov Alexey ,&nbsp;Galina Kozinets ,&nbsp;Siqi Zhang ,&nbsp;Pinlu Cao","doi":"10.1016/j.enggeo.2025.108253","DOIUrl":"10.1016/j.enggeo.2025.108253","url":null,"abstract":"<div><div>The compressive strength and crack propagation behavior of debris-ice mixtures are crucial for assessing cold-region geohazards, particularly for interpreting the initiation mechanisms of rock-ice avalanches. This study reveals the failure characteristics under uniaxial compression through experiments and discrete element modeling (PFC^3D), providing key parameters for the initiation dynamics model of rock-ice avalanches through strength degradation and crack network evolution. The numerical simulation results strongly agree with the experimental results, with a maximum deviation of 6.32 % in the compressive strength and elastic modulus. The findings reveal that rock debris significantly redirect crack propagation by inducing bypass trajectories or crack arrest, rather than allowing direct fracture paths. The compressive strength increases with enlarging rock debris volume fraction but decreases with rising temperature and debris size. The loading rate governs failure mode transitions from plastic deformation (shear-slip dominated) to brittle fracture (tensile-penetration dominated), with peak strength occurring at a critical rate of 1 mm/min in the ductile-to-brittle transition regime. Furthermore, reduced debris porosity enhances the interfacial cementation strength, improving the integrity of debris-ice composite. Crack evolution progresses through three distinct phases: initial nucleation, slow nonlinear propagation, and rapid linear expansion, with tensile cracks dominating 56.9 % of the total fractures. The results provide new insights into the failure mechanisms of multiphase geological materials, offering practical guidance for engineering geology applications, particularly in understanding the initiation and evolution of rock-ice avalanches in periglacial environments, as well as for slope stability prediction and hazard mitigation in cold regions.</div></div>","PeriodicalId":11567,"journal":{"name":"Engineering Geology","volume":"355 ","pages":"Article 108253"},"PeriodicalIF":6.9,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654093","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-source data analysis for conceptual modelling of slow landslide mechanisms: Application to the Pianello hillslope in the Daunia Appenines","authors":"Nunzio Losacco ,&nbsp;Federica Cotecchia ,&nbsp;Francesca Santaloia ,&nbsp;Claudia Vitone ,&nbsp;Giuseppe Palladino","doi":"10.1016/j.enggeo.2025.108255","DOIUrl":"10.1016/j.enggeo.2025.108255","url":null,"abstract":"<div><div>A thorough diagnosis of landslide processes at the slope scale is the prerequisite for the design and implementation of effective and sustainable mitigation measures. This paper describes the advanced application of the first, phenomenological stage of an already published stagewise methodological approach for landslide hazard assessment. The presented application aims at deriving the conceptual model of landslide mechanisms through the integration of different thematic models built upon big sets of cross-disciplinary, multi-source data, while providing the systematization and fine-tuning of the previously proposed methodology. The employed workflow is particularly apt to be implemented through digital tools that would boost the process of conceptual and numerical modelling of landslides and other geotechnical processes, particularly in complex geo-hydro-mechanical contexts.</div><div>The methodology is applied to an urbanised site in the Southeastern Apennines, the Pianello hillslope in the town of Bovino, a prototype case representative of many other sites affected by ancient slow-moving landslides in structurally complex formations. The slopes are characterized by a complex geo-structural and hydro-geological setting, and by a highly heterogeneous, heavily tectonised soil, composed of fractured rock layers embedded in a predominant clayey matrix.</div><div>For the test site, low shear strength, both intrinsic and induced by previous shearing, and high piezometric heads, are recognized as internal factors predisposing the slope to failure. Besides the erosion exerted by the stream at the toe of the hillslope, the reactivation of existing landslides is ascribed to porewater pressure oscillations at depth, induced by high cumulative rainfall over a seasonal timescale. Thanks to the integration of different models and data, the study highlights that the interplay between predisposing and triggering factors influencing the landslide activity is strongly related to the complexity of the geo-hydro-mechanical setting. The insight into deep slow-moving landslide processes acquired through this advanced application of the methodology to the prototype case of the Pianello hillslope can be easily generalised to many other cases in similar geo-hydro-mechanical contexts.</div></div>","PeriodicalId":11567,"journal":{"name":"Engineering Geology","volume":"355 ","pages":"Article 108255"},"PeriodicalIF":6.9,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144663181","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Investigation on the breach mechanism of ice-containing blockage dams: Insights into the impact of ice melting on overtopping erosion","authors":"Bo Pang ,&nbsp;Zongji Yang ,&nbsp;Zhaoying Wu ,&nbsp;Zhiyong Huang","doi":"10.1016/j.enggeo.2025.108250","DOIUrl":"10.1016/j.enggeo.2025.108250","url":null,"abstract":"<div><div>With global climate warming, ice avalanches have become increasingly frequent, often resulting in the formation of blocked dams. The presence of ice complicates the failure processes of these dams. Limited observations of breach processes and insufficient consideration of ice as a dam component impede the understanding of these processes. In this paper, a series of flume experiments were carried out to examine the effects of ice melting on overtopping breach mechanisms in dams. The results demonstrate that higher ice content enhances dam settlement and porosity, reducing structural stability. Dam height decreases from 30 cm (ice-free) to 18.5 cm at 60 % ice content, while void ratio increases from 0.82 to 1.73. These changes accelerate overtopping onset and reduce water storage capacity. The breach process can be divided into three stages: backward erosion, accelerated erosion, and attenuation-rebalancing. Both the duration of each stage and the total breach time decrease with increasing ice content. Peak discharge surges by 48.1 % (from 5.2 L/s to 7.7 L/s) at 30 % ice content and occurs 26.7 % earlier (from 172 s to 126 s). Beyond 30 % ice content, reduced water storage suppresses peak discharge (4.3 L/s at 60 % ice content). Ice melting creates new flow paths and enlarges pores in the dam, accelerating fine sediment transport and increasing dam heterogeneity. It also enhances erodibility index (from 0.0157 to 0.0516 cm·Pa⁻¹·s⁻¹) and reduces the critical shear stress (from 20.5 to 8.2 Pa) of dam materials. The ice phase modulates overtopping erosion process by controlling the erodibility of the dam materials and shear stress exerted by water flow. According to the ice content, three modes of erosion dynamics are revealed: enhanced amplification mode, attenuated amplification mode, and inhibitory breach mode. This study sheds light on the combined effects of internal ice melting and scour erosion in dams, providing valuable scientific insights for disaster prevention in glacial regions and paving the way for further research on the role of ice in dam breaches.</div></div>","PeriodicalId":11567,"journal":{"name":"Engineering Geology","volume":"355 ","pages":"Article 108250"},"PeriodicalIF":6.9,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144634319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Equations for modeling the shrinkage behavior of unsaturated soils","authors":"Yao Li,&nbsp;Sai K. Vanapalli","doi":"10.1016/j.enggeo.2025.108247","DOIUrl":"10.1016/j.enggeo.2025.108247","url":null,"abstract":"<div><div>The relationship between the soil volume and water content (i.e., either gravimetric or volumetric) is referred to as the soil shrinkage curve (SSC). Several researchers have proposed models for predicting the SSCs for fine-grained soils in the literature. However, a valid theoretical framework that can be reliably extended for modeling shrinkage curves is still lacking. In this study, a framework is proposed for modeling the SSCs by extending the assumption that the soil volume change under matric suction is strongly related to variation of pore size distribution. The assumption for modeling SSCs is consistent with the well-established method for modeling the soil water characteristics curves. The equations based on the proposed framework were found to be valuable in the reliable prediction of the shrinkage curves of various soils. The relationships between the proposed SSC equations and the widely used models from the literature are also discussed. The present study provides a new approach for modeling the SSCs that can be used for interpreting and modeling the complex hydro-mechanical behaviors of unsaturated soils.</div></div>","PeriodicalId":11567,"journal":{"name":"Engineering Geology","volume":"355 ","pages":"Article 108247"},"PeriodicalIF":6.9,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144663182","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Slope-specific rainfall thresholds for regional landslide early warning systems","authors":"Xiaoxu Xie ,&nbsp;Juan Du ,&nbsp;Kunlong Yin ,&nbsp;Renato Macciotta ,&nbsp;Shuhao Liu ,&nbsp;Jun Jiang ,&nbsp;Haoran Yang","doi":"10.1016/j.enggeo.2025.108249","DOIUrl":"10.1016/j.enggeo.2025.108249","url":null,"abstract":"<div><div>The increasing frequency and intensity of extreme rainfall events have amplified the demand for effective landslide early warning systems. However, traditional regional models often neglect the spatial variability of rainfall thresholds, resulting in reduced warning efficiency. This study proposes a region-based threshold calculation framework that incorporates slope-specific geoenvironmental characteristics and historical rainfall conditions, aiming to bridge the gap between regional-scale and slope-scale early warning approaches. The method first establishes base rainfall thresholds and quantifies the spatial variability of geoenvironmental factors (SVGF) using a widely adopted regional thresholding technique and the information value method for landslide susceptibility assessment. It then constructs the spatial variability of historical rainfall (SVHR) based on the maximum historical effective cumulative rainfall. Finally, slope-specific thresholds are derived by adjusting the base curves using variability coefficients obtained from SVGF and SVHR. The proposed framework was validated in six rainfall-induced landslide-prone counties in northeastern Chongqing, China. Results show that the method outperformed existing models in both the modeling dataset (accuracy = 79.07 %) and the prediction dataset (accuracy = 75.39 %). During extreme rainfall events, the average hit rate improved by 46.10 %, and the maximum AUC reached 0.9282—surpassing all other models. By extending traditional threshold frameworks to support slope-specific adaptation, the proposed method effectively integrates with existing thresholding and susceptibility models. It offers a technically sound and adaptable solution for landslide early warning, with considerable promise for practical application.</div></div>","PeriodicalId":11567,"journal":{"name":"Engineering Geology","volume":"355 ","pages":"Article 108249"},"PeriodicalIF":6.9,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654094","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Three-dimensional deformation inversion of mountain excavation and city construction areas using multi-track Sentinel-1 datasets and constraints of strain model and surface-parallel flow","authors":"Guangrong Li ,&nbsp;Chaoying Zhao ,&nbsp;Jiuyuan Li ,&nbsp;Qin Zhang ,&nbsp;Liquan Chen ,&nbsp;Xiaojie Liu ,&nbsp;Ming Yan ,&nbsp;Jianqi Lou ,&nbsp;Baohang Wang","doi":"10.1016/j.enggeo.2025.108251","DOIUrl":"10.1016/j.enggeo.2025.108251","url":null,"abstract":"<div><div>Mountain Excavation and City Construction (MECC) projects are common in many mountainous areas to create construction land, but ground deformation during and after MECC projects has not been fully studied. Synthetic Aperture Radar Interferometry (InSAR) is an advanced spaceborne technology to monitor large coverage ground deformation, but existing studies have neglected its horizontal deformation especially in the north-south direction due to its insensibility. As the original terrain of the filling area in the MECC areas is “V” shaped, the filling body can be regarded as a man-made “landslide” with the original slope surface as the sliding subsurface, resulting in the three-dimensional (3D) ground deformation including horizontal and vertical compaction. In this study, an InSAR technique for 3D deformation inversion of MECC areas is proposed by multitrack Sentinel-1 datasets. Firstly, surface height changes and line-of-sight (LOS) deformation are dynamically estimated in the MECC area. Then, the 3D deformation is estimated using strain model (SM) and surface-parallel flow (SPF) constraints. The results of the MECC area in Lanzhou city, China, show that the surface height changes in the MECC area range from −100 m to 75 m, along with the deformation rates in the east-west, north-south, and vertical directions amounting to 110, 80, and 180 mm/year, respectively. The magnitude of vertical deformation is positively correlated to the thickness of the filling loess, with the correlation coefficient of 0.7. Furthermore, the magnitude of the horizontal deformation is positively correlated to the gradient of the thickness of the filling loess, with the correlation coefficient of 0.7 and 0.36 in the east-west and north-south directions, respectively. Finally, the finite difference model (FDM) is employed to simulate the consolidation process in the high fill areas. The real experiments and FDM revealed that the high fill areas experienced both vertical and horizontal deformation, where horizontal consolidation of the filling loess causes the deformation fluctuations in the InSAR LOS direction at the bottom of the ravine.</div></div>","PeriodicalId":11567,"journal":{"name":"Engineering Geology","volume":"355 ","pages":"Article 108251"},"PeriodicalIF":6.9,"publicationDate":"2025-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144613313","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Physics-based and data-driven long-term evolution of a landslide: From inversion to prediction","authors":"Wenjie Du ,&nbsp;Xiaodong Fu ,&nbsp;Qian Sheng ,&nbsp;Jian Chen ,&nbsp;Yongqiang Zhou ,&nbsp;Shaojie Zheng","doi":"10.1016/j.enggeo.2025.108252","DOIUrl":"10.1016/j.enggeo.2025.108252","url":null,"abstract":"<div><div>Leveraging digital technology to improve public service capabilities in disaster prevention and mitigation is a key focus in geological hazard research. This research expands the utilization of numerical simulations for geological disaster forecasting, including studies on current-state inversion, future prediction techniques, and data integration. A geological hazard simulation framework endowed with long-term predictive capabilities was developed using the Material Point Method (MPM), with sensing data of the prototype system state and digital twin models constitute the core data components. A disaster-scenario twin model was established by extracting key elements from disaster scenarios, serving as the data storage medium for digital simulations. State sensing of the disaster prototype scenario was used to conduct system state inversion and optimize the selection of physico-mechanical parameters. The twin model was then updated using physico-mechanical models of geotechnical materials to forecast the future development of the disaster. Taking the Taihe landslide as a case study, the prototype system state was derived from monitoring data, along with the time-dependent characteristics of strength degradation of geotechnical materials. Inversion of disaster state and predictive simulations were performed using the twin model driven by data and physical principles. The landslide deformation direction, range, and local collapse locations between the twin model and the disaster prototype were compared to further achieve the constraints and optimization of the input parameters of the twin model. Building on this, long-term stability simulations for the landslide were performed, integrating the strength evolution of geotechnical materials to incorporate time-span characteristics in the landslide twin model for long-term behavior. The evolution of future disasters of the Taihe landslide over the next decade was forecasted. A disaster-scenario digital twin model, extending traditional numerical simulation in disaster fields, was developed. By integrating data and physics-driven approaches, it creates a simulation framework with temporal generalization capability. This advances the utilization of digital technologies in engineering geology and improves the accuracy of disaster prediction.</div></div>","PeriodicalId":11567,"journal":{"name":"Engineering Geology","volume":"355 ","pages":"Article 108252"},"PeriodicalIF":6.9,"publicationDate":"2025-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144613302","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep undrained failure mechanism of high-fill slopes on saturated grounds: centrifuge tests versus finite element limit analysis","authors":"Zhou Yang ,&nbsp;Xiaohui Cheng ,&nbsp;Xuedong Zhang ,&nbsp;Qiang Ma ,&nbsp;Ruifeng Ma","doi":"10.1016/j.enggeo.2025.108248","DOIUrl":"10.1016/j.enggeo.2025.108248","url":null,"abstract":"<div><div>The failure of high-fill slopes in the construction of airports, roads, and dam embankments may lead to significant human and economic losses. Rapid placement of fill with limited consolidation on saturated, low-permeability ground can trigger deep-seated landslides under undrained conditions. Prior studies on this deep undrained failure mechanism have primarily focused on low backfilled embankments. In cases with fill heights exceeding 20 m, it remains challenging to assess the stability of high-fill slopes on saturated grounds. To investigate the failure mechanism of the deep-seated landslides, a pair of centrifuge model tests based on the “modeling of models” method were conducted on a backfill slope over reconstituted saturated grounds. A similar failure mechanism of deep retrogressive flowslides was observed in both models. The landslides originated in the saturated ground beneath the slope face and foot, where increased gravity induced high excess pore water pressure, reducing the effective stress nearly to the level of static liquefaction. When these two plastic failure zones, located beneath the slope face and toe, merged into a deep flowslide surface, the deeply situated plastic failure caused the high backfills above to slide downward. Finite element limit analysis (FELA) and limit equilibrium method calculations, based on three undrained strength parameters obtained from soil mechanics laboratory and in situ miniature vane shear tests, were performed to validate the failure mechanism observed in the model tests. Finally, the failure assessment of a high-fill airport embankment was performed based on the validated FELA method. The results indicate a high risk of deep-seated undrained failure mechanisms in the original ground, with the depth of such failures closely related to the relative strength of the original ground compared with the backfills.</div></div>","PeriodicalId":11567,"journal":{"name":"Engineering Geology","volume":"355 ","pages":"Article 108248"},"PeriodicalIF":6.9,"publicationDate":"2025-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144613314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Data-driven machine learning approach for stress history evaluation in cohesive soils using cone penetration test data","authors":"Daeun Gwak ,&nbsp;Taeseo Ku","doi":"10.1016/j.enggeo.2025.108246","DOIUrl":"10.1016/j.enggeo.2025.108246","url":null,"abstract":"<div><div>Accurately assessing the geostatic stress history is crucial for predicting the deformation characteristics and engineering properties of soils, as it is influenced by various geotechnical and geological factors such as varying loads, groundwater fluctuations, and environmental conditions. Although a traditional laboratory method using consolidation tests still provides a direct reference of stress history, it often has clear limitations, particularly with silts and sands, and is also time-consuming. As a result, alternative indirect approaches, such as analyzing field test data from the Cone Penetration Test (CPT), have also been developed and widely adopted due to their advantages such as fast and easy practical application and continuous profiling. However, despite these advantages, concerns remain regarding the reliability of CPT-based stress history estimation. This study proposes a robust data-driven approach to enhance stress history prediction using CPT data, addressing the existing reliability concerns. An extensive investigation was conducted by applying advanced machine learning techniques, including Deep Neural Network (DNN), Support Vector Regression (SVR), Random Forest (RF), eXtreme Gradient Boosting (XGB), and Light Gradient Boosting Machine (LGBM). The approach utilizes a large-scale and high-quality global database compiled from CPT testing sites worldwide, focusing on key measurable parameters such as cone tip resistance, porewater pressure, and depth. This study also addresses essential methodological steps, such as data preprocessing, hyperparameter tuning, and 5-fold cross-validation. The results demonstrate that the machine learning-based models achieve remarkably improved accuracy in predicting the overconsolidation ratio (OCR) and preconsolidation stress (σ_p’) compared to conventional methods.</div></div>","PeriodicalId":11567,"journal":{"name":"Engineering Geology","volume":"355 ","pages":"Article 108246"},"PeriodicalIF":6.9,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144613316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"OWHK: Operational volumetric water content forecasting model for shallow rainfall-induced landslides in Hong Kong","authors":"Kyrillos Ebrahim ,&nbsp;Ridwan Taiwo ,&nbsp;Tarek Zayed","doi":"10.1016/j.enggeo.2025.108228","DOIUrl":"10.1016/j.enggeo.2025.108228","url":null,"abstract":"<div><div>Rainfall-induced landslides result from complex hydrological and geotechnical interactions, with one of the key challenges being the accurate estimation of infiltrated rainfall. This study introduces a unique operational model for predicting shallow volumetric water content (VWC), a critical parameter for assessing rainwater infiltration. Using Hong Kong as a case study, the proposed approach overcomes spatiotemporal limitations in existing predictive models by accounting for the randomness of rainfall-triggering mechanisms. The methodology integrates a unique data preparation technique, independence-oriented time series windowing, with the predictive power of deep learning (DL), specifically, Long Short-Term Memory (LSTM) networks, and deterministic seepage modeling via GeoStudio SEEP/W. The model is developed using field data from 15 sensors across three sites in Hong Kong (Pa Mei, Tung Chung, and Tsing Shan) at two depths (0.5 m and 1.5 m), complemented by a numerically validated case at Fei Ngo Shan Reservoir. 33 field samples were collected from eleven different locations to validate initial hypotheses. This study is among the initial systematic evaluations comparing the accuracy of operational versus site-specific models. The outcome is OWHK (Operational VWC Forecasting Model for Shallow Layers in Hong Kong), a user-friendly tool demonstrating predictive performance with mean absolute error (MAE) below 0.6 %, coefficient of determination (R²) exceeding 0.92, and a mean Monte Carlo dropout standard deviation (SD) of 0.3 %.</div></div>","PeriodicalId":11567,"journal":{"name":"Engineering Geology","volume":"355 ","pages":"Article 108228"},"PeriodicalIF":6.9,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144613315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}