Tengxu Zhang , Zhuohao Wang , Liangke Huang , Lin He , Chaolong Yao
{"title":"A novel XGBoost-based approach for reconstruction terrestrial water storage variations with GNSS in the Northeastern Tibetan Plateau","authors":"Tengxu Zhang , Zhuohao Wang , Liangke Huang , Lin He , Chaolong Yao","doi":"10.1016/j.jhydrol.2025.133255","DOIUrl":"10.1016/j.jhydrol.2025.133255","url":null,"abstract":"<div><div>Accurately estimating terrestrial water storage (TWS) variations is essential for ensuring the sustainable management of global water resources. The Global Navigation Satellite System (GNSS) offers a promising approach for monitoring TWS changes with high spatial and temporal resolution. However, its application is significantly constrained by the sparse and uneven distribution of GNSS stations. In this study, we build upon traditional GNSS inversion techniques by employing the Extreme Gradient Boosting Machine Learning (XGBML) model to simulate crustal deformation caused by hydrological loading. The simulation is conducted on a <span><math><mrow><mn>0</mn><mo>.</mo><msup><mn>5</mn><mo>°</mo></msup><mo>×</mo><mn>0</mn><mo>.</mo><msup><mn>5</mn><mo>°</mo></msup></mrow></math></span> grid across the Northeastern Tibetan Plateau (NETP). This study compared TWS variations derived from the XGBML simulations and traditional inversion methods with data from the Gravity Recovery and Climate Experiment (GRACE) satellite and the Global Land Data Assimilation System (GLDAS). The Pearson Correlation Coefficients (PCC) between TWS changes derived from the XGBML inversion technique and those from GRACE and GLDAS data were 0.72 and 0.50, respectively, representing improvements of 8.82 % and 11.10 % compared to the conventional inversion approach. Furthermore, GNSS-DSI, GRACE-DSI, and SPEI were integrated to analyze hydrological drought events in the study area, revealing that precipitation and temperature are important drivers of hydrological drought in the NETP. These findings highlight the effectiveness of the XGBML model in simulating GNSS vertical displacements induced by hydrological loading and demonstrate its potential as a novel tool for identifying water storage variations in regions with uneven GNSS station distribution.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"659 ","pages":"Article 133255"},"PeriodicalIF":5.9,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143807981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jiang Sun , Renata Graf , Dariusz Wrzesiński , Yi Luo , Senlin Zhu , Fabio Di Nunno , Roohollah Noori , Francesco Granata
{"title":"Impact of extreme atmospheric heat events on river thermal dynamics and heatwaves","authors":"Jiang Sun , Renata Graf , Dariusz Wrzesiński , Yi Luo , Senlin Zhu , Fabio Di Nunno , Roohollah Noori , Francesco Granata","doi":"10.1016/j.jhydrol.2025.133292","DOIUrl":"10.1016/j.jhydrol.2025.133292","url":null,"abstract":"<div><div>Atmospheric hot temperature extremes (HTEs) can impact river water temperatures, but how HTEs affect river thermal dynamics and heatwaves is not well understood, especially at a regional scale. In this study, we used numerical modelling and field observations to quantify the contribution of HTEs to variations in river water temperatures and river heatwaves in 70 Polish rivers covering 125 gauges between 1966 and 2020. During the study time period, HTEs duration and both cumulative and maximum intensities over the studied rivers increased significantly, at average rates of 0.379 days per year, 0.582°C per year, and 0.037 °C per year, respectively. Our results showed that HTEs can accelerate river warming; despite only accounting for 4.9 % of the total days, HTEs are responsible for 25.8 %, 16.9 %, 23.7 %, 32.8 %, and 38.3 % of river warming trends at annual, spring, summer, autumn, and winter time scales. Moreover, HTEs are important drivers of both duration and severe heatwave events on the studied rivers. The results showed that though HTEs contribute to 17.8 % of the occurrence of river heatwaves, they contribute greatly (84.3 %) to the occurrence of severe heatwave events. As the first study on this topic, our findings underscore the critical role played by short-term extreme atmospheric heat events in shaping long-term river thermal dynamics.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"659 ","pages":"Article 133292"},"PeriodicalIF":5.9,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143822406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hui Cheng , Fugang Wang , Yilong Yuan , Heng Li , Hailong Tian , Qingcheng He
{"title":"Ignoring the pre-Darcy flow phenomenon in low permeability media may lead to great deviation in contaminant transport prediction","authors":"Hui Cheng , Fugang Wang , Yilong Yuan , Heng Li , Hailong Tian , Qingcheng He","doi":"10.1016/j.jhydrol.2025.133287","DOIUrl":"10.1016/j.jhydrol.2025.133287","url":null,"abstract":"<div><div>This study focuses on the impact of pre-Darcy flow on solute transport in low-permeability media. Most existing studies on solute transport in low-permeability media typically neglect the effects of pre-Darcy flow and adopt Darcy’s law to describe fluid flow for simplification, which may lead to inaccurate estimations of solute transport behavior and concentration distribution. Based on experiments and numerical simulations, we evaluated the applicability of Darcy’s law in simulating solute transport in low-permeability media. The results indicate that the use of Darcy’s law overestimates solute transport rates in low-permeability media. As an alternative approach, the pre-Darcy flow model improves the agreement between simulation and experimental results. The maximum absolute percentage errors of solute concentration obtained from simulations using Darcy’s law and the pre-Darcy model are 18.22% and 6.76%, respectively. We also analyzed the effects of the nonlinearity between flow velocity and hydraulic gradient, permeability, and dispersion coefficient on solute transport. The deviation caused by Darcy’s law increases with the enhancement of velocity-hydraulic gradient nonlinearity. A decrease in permeability and dispersion coefficient also amplifies the deviation. Among these influencing factors, the nonlinearity between velocity and hydraulic gradient has the most significant impact on simulation results, followed by permeability and then dispersion coefficient. The smaller the hydraulic gradient, the greater the potential deviation caused by neglecting pre-Darcy flow. In a case study on radionuclide transport in a geological nuclear waste repository considered in this research, neglecting the pre-Darcy flow effect resulted in a maximum concentration deviation of up to 20.41% when the hydraulic gradient ranged from 1.6 to 3.3.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"659 ","pages":"Article 133287"},"PeriodicalIF":5.9,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143835262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Husamettin Taysi , Y.C. Ethan Yang , Sudershan Gangrade , Taher Chegini , Shih-Chieh Kao , Hong-Yi Li
{"title":"Enhancing 2D hydrodynamic flood models through machine learning and urban drainage integration","authors":"Husamettin Taysi , Y.C. Ethan Yang , Sudershan Gangrade , Taher Chegini , Shih-Chieh Kao , Hong-Yi Li","doi":"10.1016/j.jhydrol.2025.133258","DOIUrl":"10.1016/j.jhydrol.2025.133258","url":null,"abstract":"<div><div>Two-dimensional hydrodynamic flood models are commonly employed for simulating flood extent and inundation depth. However, the influence of urban drainage network (UDN) is frequently overlooked in these models, potentially compromising their accuracy. Furthermore, the expensive computational costs and longer processing times make them challenging for large-scale hydrodynamic simulation. To address these challenges, this paper develops a machine learning (ML)-driven emulator for an open-source flood model, the Two-dimensional Runoff Inundation Toolkit for Operational Needs (TRITON). A TRITON-ML Emulator (TR-Emulator) that utilizes Convolutional Long Short-Term Memory is developed to capture the spatiotemporal features of flood events based on the outputs from TRITON. We further enhance the emulator by integrating UDN parameters (TR-UDN), such as the flow capacity of drainage pipes, pipe size, and pipe length, via an ML stacking technique to improve the water surface elevation (WSE) simulation. Hurricane Harvey 2017 in Houston, TX is used as the case study. We compare WSE results from TRITON, TR-Emulator, TR-UDN, and the United States Geological Survey (USGS) observations to evaluate the performance of these models. The results indicate that the TR-Emulator effectively replicates the WSE simulated by TRITON. Additionally, TR-UDN performs well in capturing WSE patterns and peak flows, aligning more closely with USGS observations, except in areas with milder slopes where conveyance discrepancies are observed. We further test the generalizability of our ML-based models using another smaller event. This paper shows that the TR-Emulator is effective for users and engineers to emulate a 2D hydrodynamic model, and the enhanced version of the TR-Emulator, TR-UDN, can be an efficient tool for predicting WSEs during urban flooding.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"659 ","pages":"Article 133258"},"PeriodicalIF":5.9,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143816438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Impact of extreme storms on the effect of soil and water conservation measures","authors":"Ge Li , Jinfeng Wu , Zhentao Cong , Dawen Yang","doi":"10.1016/j.jhydrol.2025.133290","DOIUrl":"10.1016/j.jhydrol.2025.133290","url":null,"abstract":"<div><div>The Loess Plateau of China has been widely concerned as the main sediment source area of the Yellow River. Soil and water conservation measures, including check dams in gully and vegetation restoration in hillslope, have been constructed on the Loess Plateau since 1950s, which have significantly reduced the sediment loads also the runoff generation. The impact of extreme storms on the effect of soil and water conservation measures in the Loess Plateau is a very important issue since the extreme storms have significantly increased in this area under climate changes. An event-based hydrological model (OpenLISEM) was used to analyze the impact of extreme storms on the effect of soil and water conservation measures in the Huangfuchuan catchment, a typical hilly-gully basin in the Loess Plateau. The results indicated that (1) an increase in rainfall intensity significantly reduces the effect of check dams and vegetation restoration. The reduction rate of SWC measures has decreased from 75% to 36% for runoff and 81% to 48% for sediment under the extreme storms with rainfall return periods from 2a to 10,000a in this catchment; (2) the runoff reduction capacity of vegetation restoration is higher than that of check dams under low-intensity extreme storms but lower under high-intensity extreme storms, while the sediment reduction capacity of vegetation restoration is always higher than that of check dams; (3) the sedimentation of check dams will increase the impact of extreme storms on check dams. The impacts of vegetation restoration and check dams on runoff and sediment will exhibit varying trends under the extreme storms, providing useful information for the assessment of these measures.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"659 ","pages":"Article 133290"},"PeriodicalIF":5.9,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143834619","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Weikang Chen , Yuan Wang , Xin Peng , Qiqian Wu , Josep Peñuelas , Yan Peng , Zimin Li , Petr Heděnec , Chaoxiang Yuan , Fuzhong Wu , Kai Yue
{"title":"A synthesis on the spatial patterns and driving factors of water-holding capacity of forest litter layer across China","authors":"Weikang Chen , Yuan Wang , Xin Peng , Qiqian Wu , Josep Peñuelas , Yan Peng , Zimin Li , Petr Heděnec , Chaoxiang Yuan , Fuzhong Wu , Kai Yue","doi":"10.1016/j.jhydrol.2025.133272","DOIUrl":"10.1016/j.jhydrol.2025.133272","url":null,"abstract":"<div><div>Forest litter layers play a crucial role in regulating hydrological processes and conserving ecosystem water, yet their spatial patterns and key drivers at regional scales remain unclear. By analyzing data from 1062 sampling sites across China using machine learning, we identified the distribution and controlling factors of forest litter mass, litter water-holding rate (LWHR), and litter water-holding amount (LWHA). Our findings reveal that (1) the mean storage of undecomposed and semi-decomposed litter was 6.2 and 10.3 t ha<sup>−1</sup>, respectively, with a mean LWHR of 247.8 % and LWHA of 34.3 t ha<sup>−1</sup>; (2) LWHR varied significantly among forest types, being highest in deciduous forests (285.7 %) and lowest in evergreen needleleaf forests (223.5 %); (3) LWHA was greatest in deciduous needleleaf forests (59.7 t ha<sup>−1</sup>) and lowest in bamboo forests (17.4 t ha<sup>−1</sup>); (4) both LWHR and LWHA decreased with latitude, indicating regional differences in litter water retention; and (5) climate and topography were the primary drivers of litter water-holding capacity, influencing litter production, accumulation, and decomposition. Our findings provide a scientific basis for forest management strategies aimed at enhancing water conservation, particularly in regions vulnerable to climate change and water shortages. The results also offer a reference for global forest ecosystems, highlighting the importance of litter layers in sustaining water resources and informing policy decisions on forest conservation and watershed management.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"659 ","pages":"Article 133272"},"PeriodicalIF":5.9,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143822370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shaoyang Geng , Yuhui Zhou , Mei Geng , Zhigang Wen , Fangfang Bai , Yuanyuan Bai
{"title":"New insights into the identification and characterization of Darcy-to-Forchheimer flow transitions in rough fractures","authors":"Shaoyang Geng , Yuhui Zhou , Mei Geng , Zhigang Wen , Fangfang Bai , Yuanyuan Bai","doi":"10.1016/j.jhydrol.2025.133252","DOIUrl":"10.1016/j.jhydrol.2025.133252","url":null,"abstract":"<div><div>The fluid flow in rough fractures (RFs) undergoes a transition from Darcy flow to non-Darcy (nDarcy) flow. Accurately identifying the transition in flow regimes and selecting appropriate flow equations are crucial for comprehensively characterizing flow within fractures in engineering applications. This study employs the spatial frequency method to construct 54 three-dimensional RFs with varying apertures and roughness levels, simulating the complete flow transition process from Darcy to nDarcy flow. A systematic investigation was conducted on the identification of flow regime transitions and the characterization of the non-Darcy coefficient (<em>β</em>) within RFs. We deduce that the fundamental genesis of nDarcy flow in RFs is the collision between the fluid and the RFs walls, and utilizing <em>E</em> = 0.01 as the criterion for Darcy-to-Forchheimer flow transitions allows for a more accurate determination of its occurrence. A more specific formula to predict the critical Reynolds number (<em>R<sub>ec</sub></em>) at the transition of flow regimes was derived. Furthermore, given that the <em>β</em> is influenced by inertia, it varies across different Reynolds numbers (<em>R<sub>e</sub></em>). Based on 275 sets of <em>β</em> for RFs, a unified predictive model for the <em>β</em> was proposed considering the fractures geometric characteristics and inertia effects. The proposed model could predict the <em>β</em> for different fracture apertures, roughness levels, and <em>R<sub>e</sub></em>. The findings provide new insights into nDarcy flow in RFs and lay the groundwork for the broader application of the Forchheimer equation, and achieved better results in engineering applications.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"659 ","pages":"Article 133252"},"PeriodicalIF":5.9,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143800339","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rebecca Gustine , Christine M. Lee , Yonas Demissie , Jennifer Adam
{"title":"Historical occurrence of and shift in snow drought drivers in global mountain ranges","authors":"Rebecca Gustine , Christine M. Lee , Yonas Demissie , Jennifer Adam","doi":"10.1016/j.jhydrol.2025.133270","DOIUrl":"10.1016/j.jhydrol.2025.133270","url":null,"abstract":"<div><div>Snow is a critical form of water storage, but climate change is reducing snowpack and affecting water resources. Snow droughts are periods of abnormally low snowpack that are driven by low precipitation or high temperatures, or both. Globally, we do not have a good understanding of how snow drought drivers have changed over time. Here, we used a multivariate snow drought index to identify the dominant driver of snow droughts between 1980 and 2021 in all global mountain ranges and ones that are highly dependent on winter precipitation for water. We then identified whether those drivers shifted during the study period. Finally, we identified what similarities exist based on climatological areas by using snow classification types, including forest, montane, maritime, tundra, prairie, and ephemeral, as a proxy. We found that in both the Northern and Southern Hemispheres, dry snow droughts (driven by low precipitation) are the most common. Furthermore, in both the Northern and Southern Hemispheres, more mountain range area shifted to having temperature be the main driver of snow droughts in the historical record. In the Northern Hemisphere, dry snow droughts affected the largest amount of area in tundra, boreal, prairie, and ice snow types. In the Southern Hemisphere, warm snow droughts affected the largest amount of area in all snow types except for tundra. This work identified common snow drought drivers and patterns that exist across similar geographical areas (i.e., mountain ranges) and climatological areas and areas that are most vulnerable to warm and dry snow droughts.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"659 ","pages":"Article 133270"},"PeriodicalIF":5.9,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143829728","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jun Mao, Jingwei Wu, Yawen Liu, Chenyao Guo, Chunan Xiao, Yan Lu, Liya Zhao, Renjie Zhang, Hanyi Zhang
{"title":"A model for simulating evaporation from seasonally frozen saline soil","authors":"Jun Mao, Jingwei Wu, Yawen Liu, Chenyao Guo, Chunan Xiao, Yan Lu, Liya Zhao, Renjie Zhang, Hanyi Zhang","doi":"10.1016/j.jhydrol.2025.133259","DOIUrl":"10.1016/j.jhydrol.2025.133259","url":null,"abstract":"<div><div>Evaporation from soil significantly influences soil physicochemical characteristics, water resource management, and crop growth. Therefore, studying soil evaporation under saline conditions is essential for preventing soil salinization, improving water-use efficiency and promoting sustainable agriculture. In this study, a numerical model tailored for calculating evaporation from saline soil in seasonally frozen regions is introduced. The model comprehensively accounts for the changes in the upper boundaries of the soil due to salinity changes and incorporates three components: (1) decreased osmotic potential (DOP) at the soil surface, (2) increased salt resistance (ISR) to the diffusion of water vapour from the salt crust, and (3) increased soil albedo (ISA) due to the presence of the salt crust. The proposed model (M1) is validated through a field soil column test conducted in a seasonally frozen region and with data from two published laboratory-based studies of evaporation from saline soil. We compare the performance of M1 with that of both a model (M2) that neglects salinity effects and the widely-used SHAW model. The M1 simulation results are in good agreement with the measured values across all three cases. The average relative errors (<em>ARE</em>s) for evaporation calculated with M1 remain below 10 % in all cases, indicating its high accuracy. Neglecting or inadequately considering the role of salinity results in significant errors in simulating soil evaporation and water–heat–salt transport, with the maximum <em>ARE</em> for evaporation reaching as high as 130 %. Salt inhibits both soil evaporation and water–salt migration, whereas it initially promotes and then restrains the soil temperature. The relative impacts of salt on soil evaporation and water-heat-salt transport processes are ranked as follows: ISA > ISR > DOP. In this study, a theoretical framework is established to provide a practical tool for simulating soil evaporation and water–heat–salt interactions in saline soil.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"659 ","pages":"Article 133259"},"PeriodicalIF":5.9,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143848549","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Local scour around bridge abutments in vegetated beds under ice-covered flow conditions – An experimental study and mathematical assessment using machine learning methods","authors":"Sanaz Sediqi, Jueyi Sui, Guowei Li","doi":"10.1016/j.jhydrol.2025.133257","DOIUrl":"10.1016/j.jhydrol.2025.133257","url":null,"abstract":"<div><div>Local scour around bridge abutments is a critical process that can cause bridge failures, posing substantial environmental risks. Accurate estimation of the maximum scour depth around bridge abutments is crucial for bridge design. In the current study, extensive experiments have been conducted in a large-scale outdoor flume to investigate local scour around the bridge in the presence of vegetation in the channel bed and ice cover on the water surface. Different layout vegetation patterns and densities, water surface cover conditions, abutment shapes, and particle size of bed material are considered. Based on data collected from laboratory experiments, this study employs machine learning methods, including Artificial Neural Networks (ANN), Support Vector Machines (SVM), Multiple Linear Regression (MLR), and Gene Expression Programming (GEP), to predict the maximum scour depth around the abutments, with the GEP model showing the highest accuracy. Results revealed deeper scour with higher Froude numbers, rectangular abutments, and increased ice roughness, while coarser sediments and dense, staggered vegetation reduced scour. Sensitivity analysis using the Partial Mutual Information (PMI) and SHAP (SHapley Additive exPlanations) captured the effects of key variables affecting the maximum scour depth around abutments, including flow Froude number (Fr), the ratio of median grain size of bed material to flow depth (d<sub>50</sub>/H), the standard deviation of sediments (σ<sub>g</sub>), the ratio of ice cover roughness coefficient to that of channel bed (n<sub>I</sub>/n<sub>B</sub>), abutment shape factor (K<sub>s</sub>), and vegetation roughness density (λ). Two predictive formulas were developed for use in vegetated, ice-covered channels.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"659 ","pages":"Article 133257"},"PeriodicalIF":5.9,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143816237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}