Pub Date : 2025-03-10DOI: 10.1016/j.jhydrol.2025.133061
Chuanyin Jiang , Xiaoguang Wang , Hervé Jourde
Increased leakage at dam sites due to the dissolution widening of fractures in the sub-surface soluble rocks, i.e., karstification, poses a great threat to the longevity of dam structures. The elevated hydraulic gradient induced by impounded water may significantly accelerate karstification, dramatically increasing leakage by several orders of magnitude. Many previous numerical studies on karstification and leakage at dam sites have overlooked the effects of stress-dependent aperture heterogeneity and anisotropy as well as vertical temperature variations. In this study, we quantified the effects of stress and temperature on leakage dynamics using a coupled thermo-hydro-chemical model incorporating stress-dependent initial aperture fields. Results indicate that stress-induced aperture fields play a primary role in dissolution behaviors compared to the temperature effect. Initial aperture anisotropy controls the preferential penetration directions of dissolution fronts, and anisotropic stress conditions may accelerate breakthrough by up to 40% compared to an isotropic stress condition. The consideration of temperature effect leads to a delayed breakthrough by 10%–16% due to mineral precipitation (chemical control) and elevated fluid viscosity (hydraulic control). The temperature effects are also dependent on the different dissolution pathways controlled by aperture anisotropy and become more pronounced under a low initial rate where breakthrough times may be further delayed by up to 30%. This study offers valuable implications for designing engineering strategies in limestone bedrock dam construction to mitigate leakage hazards and extend structural longevity.
{"title":"Dam leakage potential related to karstification in limestone bedrock: Effects of temperature and stress-induced anisotropy","authors":"Chuanyin Jiang , Xiaoguang Wang , Hervé Jourde","doi":"10.1016/j.jhydrol.2025.133061","DOIUrl":"10.1016/j.jhydrol.2025.133061","url":null,"abstract":"<div><div>Increased leakage at dam sites due to the dissolution widening of fractures in the sub-surface soluble rocks, i.e., karstification, poses a great threat to the longevity of dam structures. The elevated hydraulic gradient induced by impounded water may significantly accelerate karstification, dramatically increasing leakage by several orders of magnitude. Many previous numerical studies on karstification and leakage at dam sites have overlooked the effects of stress-dependent aperture heterogeneity and anisotropy as well as vertical temperature variations. In this study, we quantified the effects of stress and temperature on leakage dynamics using a coupled thermo-hydro-chemical model incorporating stress-dependent initial aperture fields. Results indicate that stress-induced aperture fields play a primary role in dissolution behaviors compared to the temperature effect. Initial aperture anisotropy controls the preferential penetration directions of dissolution fronts, and anisotropic stress conditions may accelerate breakthrough by up to 40% compared to an isotropic stress condition. The consideration of temperature effect leads to a delayed breakthrough by 10%–16% due to mineral precipitation (chemical control) and elevated fluid viscosity (hydraulic control). The temperature effects are also dependent on the different dissolution pathways controlled by aperture anisotropy and become more pronounced under a low initial rate where breakthrough times may be further delayed by up to 30%. This study offers valuable implications for designing engineering strategies in limestone bedrock dam construction to mitigate leakage hazards and extend structural longevity.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"657 ","pages":"Article 133061"},"PeriodicalIF":5.9,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143628937","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}
Pub Date : 2025-03-10DOI: 10.1016/j.jhydrol.2025.133053
Huabing Huang, Zhiwen Tao, Jiayin Zhan, Changpeng Wang
Flood susceptibility modeling is a typical imbalanced problem in which the amount of flood data is much smaller than that of non-flood data. To ensure balanced learning, only a small fraction of non-flood data is selected for machine learning. Traditional sampling methods, such as Random Sampling (RS) and Stratified Sampling (SS), neglect abundant information within non-flood data and its relationship with flood data. This neglect leads to insufficient binary classification performance and biased susceptibility estimation, both of which are influenced by sample contrast and diversity, respectively. Unfortunately, these two objectives cannot be achieved simultaneously due to the trade-off between sample contrast and diversity. This dual-objective optimization requires a trade-off between contrast and diversity in sample quality. This study proposed a Distance-Based Sampling (DBS) framework that connects non-flood samples to flood data using Euclidean distance. Ten DBS scenarios with varying contrast and diversity levels (from 0.0 to 1.0 to 0.9–1.0) were designed for systematic evaluation. The best DBS scenario was further compared with RS, SS, and Inverse-Occurrence Sampling (IOS). To derive robust results, four machine learning techniques—Artificial Neural Network (ANN), Random Forest (RF), Support Vector Machine (SVM), and eXtreme Gradient Boosting (XGBoost)—were employed in two heterogeneous urban districts in Guangzhou, China, along with multiple indices, i.e. Area Under the Curve (AUC), mean susceptibility, standard deviation of susceptibility, and flood density order. The main findings of this study were as follows: (1) High sample contrast led to excellent binary classification performance but resulted in an overestimation of flood susceptibility. (2) High sample diversity resulted in insufficient binary classification performance and an underestimation of flood susceptibility. (3) Under the DBS framework, the performance curve of the dual-objective problem is unimodal. The best performance was achieved at a trade-off between contrast and diversity, specifically in the DBS scenario 0.3–1.0. (4) The DBS scenario 0.3–1.0 outperformed RS, SS, and IOS. Finally, this study underscores the critical role of non-flood sample quality and the balance between sample contrast and diversity in flood susceptibility modeling. The proposed DBS framework is objective and flexible, and can be applied to negative sampling in susceptibility modeling for other hazards, such as landslides and wildfires.
{"title":"Contrast or Diversity: Non-Flood sampling in urban flood susceptibility modelling","authors":"Huabing Huang, Zhiwen Tao, Jiayin Zhan, Changpeng Wang","doi":"10.1016/j.jhydrol.2025.133053","DOIUrl":"10.1016/j.jhydrol.2025.133053","url":null,"abstract":"<div><div>Flood susceptibility modeling is a typical imbalanced problem in which the amount of flood data is much smaller than that of non-flood data. To ensure balanced learning, only a small fraction of non-flood data is selected for machine learning. Traditional sampling methods, such as Random Sampling (RS) and Stratified Sampling (SS), neglect abundant information within non-flood data and its relationship with flood data. This neglect leads to insufficient binary classification performance and biased susceptibility estimation, both of which are influenced by sample contrast and diversity, respectively. Unfortunately, these two objectives cannot be achieved simultaneously due to the trade-off between sample contrast and diversity. This dual-objective optimization requires a trade-off between contrast and diversity in sample quality. This study proposed a Distance-Based Sampling (DBS) framework that connects non-flood samples to flood data using Euclidean distance. Ten DBS scenarios with varying contrast and diversity levels (from 0.0 to 1.0 to 0.9–1.0) were designed for systematic evaluation. The best DBS scenario was further compared with RS, SS, and Inverse-Occurrence Sampling (IOS). To derive robust results, four machine learning techniques—Artificial Neural Network (ANN), Random Forest (RF), Support Vector Machine (SVM), and eXtreme Gradient Boosting (XGBoost)—were employed in two heterogeneous urban districts in Guangzhou, China, along with multiple indices, i.e. Area Under the Curve (AUC), mean susceptibility, standard deviation of susceptibility, and flood density order. The main findings of this study were as follows: (1) High sample contrast led to excellent binary classification performance but resulted in an overestimation of flood susceptibility. (2) High sample diversity resulted in insufficient binary classification performance and an underestimation of flood susceptibility. (3) Under the DBS framework, the performance curve of the dual-objective problem is unimodal. The best performance was achieved at a trade-off between contrast and diversity, specifically in the DBS scenario 0.3–1.0. (4) The DBS scenario 0.3–1.0 outperformed RS, SS, and IOS. Finally, this study underscores the critical role of non-flood sample quality and the balance between sample contrast and diversity in flood susceptibility modeling. The proposed DBS framework is objective and flexible, and can be applied to negative sampling in susceptibility modeling for other hazards, such as landslides and wildfires.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"656 ","pages":"Article 133053"},"PeriodicalIF":5.9,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143610873","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}
Efficient soil water management and prediction of contaminant transport requires a deep understanding of spatial and temporal variation in soil hydraulic properties (SHPs). The growing interest in using ground-penetrating radar (GPR) for large-scale and non-destructive estimation of SHPs demands more effective approaches. This study evaluates the potential for monitoring soil water content (SWC) changes and estimating field-saturated hydraulic conductivity (Kfs) by employing the Green-Ampt (GA) model using GPR time-lapse data. At two locations at a podzolic soil site in western Newfoundland, Canada, infiltration experiments were carried out on different days using the Beerkan method, which involved applying equal volumes of water over a short duration. A surface GPR system with a center frequency of 500 MHz was employed to monitor these experiments. The downward movement of the wetting zone during infiltration was monitored by collecting time-lapse GPR traces every 5 s. SWC changes estimated from GPR (GPR-SWC), and soil moisture probes (SMP-SWC) (installed to a depth of 0.20 m) were used as parameters in the GA model to estimate Kfs. Our findings show that GPR provided consistent information on the dielectric constant (r = 0.902) and SWC variations during the infiltration experiment at both locations. The average Kfs value (1.4 × 10-5 ± 5.4 × 10-6 m/s) estimated using the GPR-SWC in the GA model was in a similar magnitude to the theoretical value for the tested soil type and in close range to values measured by using conventional methods, although all approaches were significantly different. Further research is needed to validate this approach across various soil types and conditions.
{"title":"Estimating soil hydraulic conductivity from time-lapse ground-penetrating radar data in podzolic soils using the green-ampt model","authors":"Juwonlo Dahunsi, Sashini Pathirana, Mumtaz Cheema, Manokararajah Krishnapillai, Lakshman Galagedara","doi":"10.1016/j.jhydrol.2025.133059","DOIUrl":"10.1016/j.jhydrol.2025.133059","url":null,"abstract":"<div><div>Efficient soil water management and prediction of contaminant transport requires a deep understanding of spatial and temporal variation in soil hydraulic properties (SHPs). The growing interest in using ground-penetrating radar (GPR) for large-scale and non-destructive estimation of SHPs demands more effective approaches. This study evaluates the potential for monitoring soil water content (SWC) changes and estimating field-saturated hydraulic conductivity (<em>K<sub>fs</sub></em>) by employing the Green-Ampt (GA) model using GPR time-lapse data. At two locations at a podzolic soil site in western Newfoundland, Canada, infiltration experiments were carried out on different days using the Beerkan method, which involved applying equal volumes of water over a short duration. A surface GPR system with a center frequency of 500 MHz was employed to monitor these experiments. The downward movement of the wetting zone during infiltration was monitored by collecting time-lapse GPR traces every 5 s. SWC changes estimated from GPR (GPR-SWC), and soil moisture probes (SMP-SWC) (installed to a depth of 0.20 m) were used as parameters in the GA model to estimate <em>K<sub>fs</sub></em>. Our findings show that GPR provided consistent information on the dielectric constant (r = 0.902) and SWC variations during the infiltration experiment at both locations. The average <em>K<sub>fs</sub></em> value (1.4 × 10<sup>-5</sup> ± 5.4 × 10<sup>-6</sup> m/s) estimated using the GPR-SWC in the GA model was in a similar magnitude to the theoretical value for the tested soil type and in close range to values measured by using conventional methods, although all approaches were significantly different. Further research is needed to validate this approach across various soil types and conditions.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"657 ","pages":"Article 133059"},"PeriodicalIF":5.9,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143642466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-09DOI: 10.1016/j.jhydrol.2025.133057
Lei Tian , Wenjie Wang , Xiaogang Ma , Hongdong Zhang , Shuchen Guo , Kai Yang , Jie Wang , Linhua Wang
Snow plays a crucial role in land surface hydrological and energy processes. Accurately representing the snow-streamflow relationship is important for understanding how climate change affects alpine hydrology. However, most land surface models and hydrological models’ snow schemes overlook the influences of snow depth and topography, causing uncertainties in snow and related hydrological processes simulations. This issue is more pronounced on the Tibetan Plateau (TP) due to its shallow snow and complex topography. The challenge of how inadequate snow cover parameterization affects snow and streamflow simulations is a critical scientific question. This study targets the upstream areas of the Heihe River basin on the TP. Using multi-source observational datasets and the WRF-Hydro model, we incorporated seven pre-existing snow schemes that consider snow depth and topography into the WRF-Hydro to identify the optimized scheme. Comparing the results simulated with the default and optimized schemes, we quantified the improvement in the representation of the snow-streamflow relationship by considering snow depth and topography and revealed the influencing mechanisms of these two factors. Results show that the default scheme largely overestimates snow cover fraction (SCF). Accounting for snow depth alone reduces the monthly SCF bias by 6.20%. When both snow depth and topography are considered, the monthly SCF bias is reduced by 20.88%. Moreover, the default scheme underestimates the cold-season streamflow and overestimates the warm-season streamflow. The optimized scheme greatly enhances the accuracy of streamflow simulation, reducing the cold-season streamflow underestimation by 12.13% and lowering the warm-season streamflow overestimation by 8.84%. Furthermore, such incorporation reduces albedo overestimation, increases absorbed shortwave radiation, and changes land surface temperature (LST) and surface resistance (rs). LST and rs are key variables through which snow influences evapotranspiration and snow water equivalent, eventually altering the snow-streamflow relationship. These findings highlight the importance of considering snow depth and topography in numerical simulations for alpine areas and provide valuable scientific support for understanding the response of hydrological processes to snow change under climate warming.
{"title":"Refining snow-streamflow dynamics in a Tibetan Plateau basin by incorporating snow depth and topography","authors":"Lei Tian , Wenjie Wang , Xiaogang Ma , Hongdong Zhang , Shuchen Guo , Kai Yang , Jie Wang , Linhua Wang","doi":"10.1016/j.jhydrol.2025.133057","DOIUrl":"10.1016/j.jhydrol.2025.133057","url":null,"abstract":"<div><div>Snow plays a crucial role in land surface hydrological and energy processes. Accurately representing the snow-streamflow relationship is important for understanding how climate change affects alpine hydrology. However, most land surface models and hydrological models’ snow schemes overlook the influences of snow depth and topography, causing uncertainties in snow and related hydrological processes simulations. This issue is more pronounced on the Tibetan Plateau (TP) due to its shallow snow and complex topography. The challenge of how inadequate snow cover parameterization affects snow and streamflow simulations is a critical scientific question. This study targets the upstream areas of the Heihe River basin on the TP. Using multi-source observational datasets and the WRF-Hydro model, we incorporated seven pre-existing snow schemes that consider snow depth and topography into the WRF-Hydro to identify the optimized scheme. Comparing the results simulated with the default and optimized schemes, we quantified the improvement in the representation of the snow-streamflow relationship by considering snow depth and topography and revealed the influencing mechanisms of these two factors. Results show that the default scheme largely overestimates snow cover fraction (<em>SCF</em>). Accounting for snow depth alone reduces the monthly <em>SCF</em> bias by 6.20%. When both snow depth and topography are considered, the monthly <em>SCF</em> bias is reduced by 20.88%. Moreover, the default scheme underestimates the cold-season streamflow and overestimates the warm-season streamflow. The optimized scheme greatly enhances the accuracy of streamflow simulation, reducing the cold-season streamflow underestimation by 12.13% and lowering the warm-season streamflow overestimation by 8.84%. Furthermore, such incorporation reduces albedo overestimation, increases absorbed shortwave radiation, and changes land surface temperature (<em>LST</em>) and surface resistance (<em>r<sub>s</sub></em>). <em>LST</em> and <em>r<sub>s</sub></em> are key variables through which snow influences evapotranspiration and snow water equivalent, eventually altering the snow-streamflow relationship. These findings highlight the importance of considering snow depth and topography in numerical simulations for alpine areas and provide valuable scientific support for understanding the response of hydrological processes to snow change under climate warming.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"657 ","pages":"Article 133057"},"PeriodicalIF":5.9,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143654610","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}
Pub Date : 2025-03-09DOI: 10.1016/j.jhydrol.2025.133047
Zhangkang Shu , Junliang Jin , Lucas Menzel , Jianyun Zhang , Jianfeng Luo , Guoqing Wang , Ningbo Cui , Tiesheng Guan , Yanli Liu
Accurate estimation of terrestrial ecosystem water-heat fluxes is crucial for agricultural production, ecosystem monitoring, and eco-hydrological model development. The selection and optimization of parameterization schemes for surface resistance (rs), a pivotal parameter in the water-carbon cycle, significantly impact the uncertainty of actual evapotranspiration (ET) estimation in Penman-Monteith (PM) models. This study investigates the effects of rs scheme selection and soil water function optimization on the PM modeling. Three types of rs parameterization schemes—Katerji-Perrier (KP) based on atmosphere, Kelliher-Leuning (KL) based on vegetation-atmosphere, and Jarvis based on soil–vegetation–atmosphere—were evaluated alongside three soil moisture constraint functions (Jarvis1, Jarvis2, Jarvis3). The sensitivity of rs and ET to various environmental factors was also analyzed. Model evaluations were conducted at 100 FLUXNET2015 flux towers worldwide, covering ten ecosystems: croplands (CRO), closed shrublands (CSH), deciduous broadleaf forests (DBF), evergreen broadleaf forests (EBF), evergreen needleleaf forests (ENF), grasslands (GRA), mixed forests (MF), open shrublands (OSH), savannas (SAV), and woody savannas (WSA). Results indicated that the PM-Jarvis3 model, which includes wilting point constraint weights and a nonlinear soil moisture response, significantly improved the simulation performance of rs and ET. The effectiveness of rs models varied considerably across ecosystems, with no single model consistently providing optimal ET simulation at all sites. Specifically, the PM-Jarvis3 model demonstrated clear advantages in ENF, GRA, and compound ecosystems (e.g., CSH, OSH, SAV, WSA), while the PM-KL model excelled at some agricultural and forested sites (e.g., DBF, EBF, MF). The PM-KP model showed consistent performance with PM-Jarvis3 only in agro-ecosystems. Overall, PM-Jarvis3 emerged as the most robust model across all biomes. Vapor pressure deficit and net radiation were identified as critical sensitivity factors in all three ET models, with leaf area index being crucial in the PM-KL and PM-Jarvis3 models. According to the PM-Jarvis3 model, soil moisture and CO2 were significant factors for rs and ET simulation in most regions. Despite these findings, the sensitivity of rs and ET to environmental factors remains highly variable among ecosystems. Our improved PM-Jarvis3 model enhances the understanding of the influence of environmental factors on rs and ET simulations globally, providing valuable insights for terrestrial ecohydrological modeling.
{"title":"Evaluating the effectiveness of different surface resistance schemes coupled with Penman-Monteith model for estimating actual evapotranspiration − A global comparative study","authors":"Zhangkang Shu , Junliang Jin , Lucas Menzel , Jianyun Zhang , Jianfeng Luo , Guoqing Wang , Ningbo Cui , Tiesheng Guan , Yanli Liu","doi":"10.1016/j.jhydrol.2025.133047","DOIUrl":"10.1016/j.jhydrol.2025.133047","url":null,"abstract":"<div><div>Accurate estimation of terrestrial ecosystem water-heat fluxes is crucial for agricultural production, ecosystem monitoring, and eco-hydrological model development. The selection and optimization of parameterization schemes for surface resistance (<em>r<sub>s</sub></em>), a pivotal parameter in the water-carbon cycle, significantly impact the uncertainty of actual evapotranspiration (ET) estimation in Penman-Monteith (PM) models. This study investigates the effects of <em>r<sub>s</sub></em> scheme selection and soil water function optimization on the PM modeling. Three types of <em>r<sub>s</sub></em> parameterization schemes—Katerji-Perrier (KP) based on atmosphere, Kelliher-Leuning (KL) based on vegetation-atmosphere, and Jarvis based on soil–vegetation–atmosphere—were evaluated alongside three soil moisture constraint functions (Jarvis1, Jarvis2, Jarvis3). The sensitivity of <em>r<sub>s</sub></em> and ET to various environmental factors was also analyzed. Model evaluations were conducted at 100 FLUXNET2015 flux towers worldwide, covering ten ecosystems: croplands (CRO), closed shrublands (CSH), deciduous broadleaf forests (DBF), evergreen broadleaf forests (EBF), evergreen needleleaf forests (ENF), grasslands (GRA), mixed forests (MF), open shrublands (OSH), savannas (SAV), and woody savannas (WSA). Results indicated that the PM-Jarvis3 model, which includes wilting point constraint weights and a nonlinear soil moisture response, significantly improved the simulation performance of <em>r<sub>s</sub></em> and ET. The effectiveness of <em>r<sub>s</sub></em> models varied considerably across ecosystems, with no single model consistently providing optimal ET simulation at all sites. Specifically, the PM-Jarvis3 model demonstrated clear advantages in ENF, GRA, and compound ecosystems (e.g., CSH, OSH, SAV, WSA), while the PM-KL model excelled at some agricultural and forested sites (e.g., DBF, EBF, MF). The PM-KP model showed consistent performance with PM-Jarvis3 only in agro-ecosystems. Overall, PM-Jarvis3 emerged as the most robust model across all biomes. Vapor pressure deficit and net radiation were identified as critical sensitivity factors in all three ET models, with leaf area index being crucial in the PM-KL and PM-Jarvis3 models. According to the PM-Jarvis3 model, soil moisture and CO<sub>2</sub> were significant factors for <em>r<sub>s</sub></em> and ET simulation in most regions. Despite these findings, the sensitivity of <em>r<sub>s</sub></em> and ET to environmental factors remains highly variable among ecosystems. Our improved PM-Jarvis3 model enhances the understanding of the influence of environmental factors on <em>r<sub>s</sub></em> and ET simulations globally, providing valuable insights for terrestrial ecohydrological modeling.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"656 ","pages":"Article 133047"},"PeriodicalIF":5.9,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143591502","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}
Pub Date : 2025-03-09DOI: 10.1016/j.jhydrol.2025.133046
Shaobo Xue , Peng Li , Zhiwei Cui , Zhanbin Li
This study investigates the impact of various check dam configurations on the downstream river channel morphology and water–sediment relationships through the big indoor simulation experiments as 31.5 m long and 16.5 m wide. The findings demonstrate that the construction of check dams effectively regulates the erosion and sedimentation processes of the river channel. When compared to the no-dam scenario, the flow rates decreased by 43 %-69 % across the three dam configurations, while sediment concentrations reduced by 45 %-86 %. The construction of dams notably altered both the cross-sectional shape and micro-topography of the river channel. Specifically, the dual-dam configuration led to a 41.35 % decrease in slope and a 30.74 % reduction in roughness. Interestingly, the reduction in slope and roughness were more significant in the dual-dam configuration than in the sum of the changes caused by two individual dams. Additionally, a random forest algorithm was applied to rank the importance of various parameters influencing sediment concentration and particle size. Key factors such as runoff shear stress, roughness, runoff power, and slope were identified as having the greatest impact. These findings provide valuable theoretical and technical insights for the optimization design and management of check dams in the Loess Plateau region.
{"title":"The influence of different check dam configurations on the downstream river topography and water–sediment relationship","authors":"Shaobo Xue , Peng Li , Zhiwei Cui , Zhanbin Li","doi":"10.1016/j.jhydrol.2025.133046","DOIUrl":"10.1016/j.jhydrol.2025.133046","url":null,"abstract":"<div><div>This study investigates the impact of various check dam configurations on the downstream river channel morphology and water–sediment relationships through the big indoor simulation experiments as 31.5 m long and 16.5 m wide. The findings demonstrate that the construction of check dams effectively regulates the erosion and sedimentation processes of the river channel. When compared to the no-dam scenario, the flow rates decreased by 43 %-69 % across the three dam configurations, while sediment concentrations reduced by 45 %-86 %. The construction of dams notably altered both the cross-sectional shape and micro-topography of the river channel. Specifically, the dual-dam configuration led to a 41.35 % decrease in slope and a 30.74 % reduction in roughness. Interestingly, the reduction in slope and roughness were more significant in the dual-dam configuration than in the sum of the changes caused by two individual dams. Additionally, a random forest algorithm was applied to rank the importance of various parameters influencing sediment concentration and particle size. Key factors such as runoff shear stress, roughness, runoff power, and slope were identified as having the greatest impact. These findings provide valuable theoretical and technical insights for the optimization design and management of check dams in the Loess Plateau region.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"656 ","pages":"Article 133046"},"PeriodicalIF":5.9,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143579662","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}
The escalating challenge of unsustainable urban development worldwide has precipitated changes in land usage, contributing to increased impermeability of the urban landscape. This phenomenon exacerbates urban runoff, a critical environmental concern. In response, Low Impact Development (LID) techniques, recognized for their environmental efficacy, have emerged as pivotal in mitigating urban runoff. However, transforming the hydrological dynamics of urban watersheds into a more sustainable state necessitates substantial financial commitments from relevant authorities. Consequently, strategic LID planning becomes essential to maximize effectiveness while minimizing costs. This research introduces a novel, hybrid modeling strategy that integrates the Storm Water Management Model (SWMM) with the Multi-Objective Evolutionary Algorithm by Decomposition (MOEA/D) optimization algorithm. This approach aims to concurrently minimize runoff volume, peak flow rate, and implementation expenses. Focusing on a segment of Tehran Municipality’s urban stormwater system in District 11, the study evaluates four distinct LID scenarios. These scenarios encompass various configurations of Rain Barrels (RB), Bioretention Cells (BC), Green Roofs (GR), and Porous Pavements (PP). Utilizing the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method for comparative analysis, the study results identify the most efficacious scenario, S2_1, including RB and BC, which achieves a 19.34% reduction in runoff volume and a 46.53 % decrease in peak flow rate, all at the implementation cost of 123,169 USD. A close second, scenario S3_1 incorporating RB and PP, demonstrates a 17 % and 46.55 % reduction in runoff volume and peak flow at an expenditure of 107,017 USD, respectively. The proposed SWMM-MOEA/D model, in conjunction with TOPSIS, presents a valuable tool for LID planning and optimization, offering decision-makers and relevant entities a pragmatic approach to address the challenges of urban runoff management.
{"title":"Coupled SWMM-MOEA/D for multi-objective optimization of low impact development in urban stormwater systems","authors":"Kazem Javan , Saeed Banihashemi , Amirhossein Nazari , Abbas Roozbahani , Mariam Darestani , Hanieh Hossieni","doi":"10.1016/j.jhydrol.2025.133044","DOIUrl":"10.1016/j.jhydrol.2025.133044","url":null,"abstract":"<div><div>The escalating challenge of unsustainable urban development worldwide has precipitated changes in land usage, contributing to increased impermeability of the urban landscape. This phenomenon exacerbates urban runoff, a critical environmental concern. In response, Low Impact Development (LID) techniques, recognized for their environmental efficacy, have emerged as pivotal in mitigating urban runoff. However, transforming the hydrological dynamics of urban watersheds into a more sustainable state necessitates substantial financial commitments from relevant authorities. Consequently, strategic LID planning becomes essential to maximize effectiveness while minimizing costs. This research introduces a novel, hybrid modeling strategy that integrates the Storm Water Management Model (SWMM) with the Multi-Objective Evolutionary Algorithm by Decomposition (MOEA/D) optimization algorithm. This approach aims to concurrently minimize runoff volume, peak flow rate, and implementation expenses. Focusing on a segment of Tehran Municipality’s urban stormwater system in District 11, the study evaluates four distinct LID scenarios. These scenarios encompass various configurations of Rain Barrels (RB), Bioretention Cells (BC), Green Roofs (GR), and Porous Pavements (PP). Utilizing the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method for comparative analysis, the study results identify the most efficacious scenario, S2_1, including RB and BC, which achieves a 19.34% reduction in runoff volume and a 46.53 % decrease in peak flow rate, all at the implementation cost of 123,169 USD. A close second, scenario S3_1 incorporating RB and PP, demonstrates a 17 % and 46.55 % reduction in runoff volume and peak flow at an expenditure of 107,017 USD, respectively. The proposed SWMM-MOEA/D model, in conjunction with TOPSIS, presents a valuable tool for LID planning and optimization, offering decision-makers and relevant entities a pragmatic approach to address the challenges of urban runoff management.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"656 ","pages":"Article 133044"},"PeriodicalIF":5.9,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143610876","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-09DOI: 10.1016/j.jhydrol.2025.133042
Jinzhao Liu , Daniele Penna , Xiong Xiao , Li Guo , Guo Chen , Chong Jiang , Huawu Wu , Shengjie Wang , Zhiyun Jiang , Zhao Jin , Zhisheng An
A mechanistic understanding of the interacting processes governing the ecohydrological cycle is of paramount importance for comprehending the soil–plant-atmosphere continuum (SPAC) in the Critical Zone. The analysis of these processes may necessarily consider the different water types that characterize ecohydrological flux exchanges at the catchment scale but, so far, few studies have disentangled functional interactions among various water types within the Critical Zone. This study leveraged three years of isotope data (δ18O and δ2H) collected from twelve water sources, including precipitation, throughfall, snow, stream water, groundwater, dew water, frost water, mobile and less-mobile soil water, root water, stem water, and leaf water in two catchments with distinct land cover (forestland versus grassland) on the Chinese Loess Plateau (CLP). We infer the main ecohydrological processes controlling water exchange in the Critical Zone under the contrasting vegetation covers. Our results showed new interactions among the several investigated water types, and in particular highlighted that: i) The seasonal isotopic variation in precipitation played a critical role in the seasonal isotopic patterns observed in other water types; ii) Dew water significantly contributed to leaf water uptake, more in forestland (26 ± 6 %) than in grassland (16 ± 11 %). Snow and groundwater were more influential for root water of shrubs and grasses in forestland (59 ± 34 % and 16 ± 8 % for snow and groundwater, respectively) than in grassland (36 ± 26 % and 6 ± 6 %) and they were very important for stem water of trees in forestland (84 ± 14 % and 45 ± 22 % for snow and groundwater, respectively); iii) Isotopic values in mobile and less-mobile soil water differed significantly between forestland and grassland (p < 0.05), but those in plant water (root, stem, and leaf water) did not differ significantly (p > 0.05); and iv) There were dynamic exchanges between mobile and less-mobile soil water, and between groundwater and soil water on the CLP. All these observations allowed us to establish a new isotope-based conceptual model of the ecohydrological cycle in the Critical Zone of the CLP that provides the foundation for future research and sustainable water resource management in this region.
{"title":"Towards an isotope-based conceptual catchment model of the ecohydrological cycle in the Critical Zone on the Loess Plateau of China","authors":"Jinzhao Liu , Daniele Penna , Xiong Xiao , Li Guo , Guo Chen , Chong Jiang , Huawu Wu , Shengjie Wang , Zhiyun Jiang , Zhao Jin , Zhisheng An","doi":"10.1016/j.jhydrol.2025.133042","DOIUrl":"10.1016/j.jhydrol.2025.133042","url":null,"abstract":"<div><div>A mechanistic understanding of the interacting processes governing the ecohydrological cycle is of paramount importance for comprehending the soil–plant-atmosphere continuum (SPAC) in the Critical Zone. The analysis of these processes may necessarily consider the different water types that characterize ecohydrological flux exchanges at the catchment scale but, so far, few studies have disentangled functional interactions among various water types within the Critical Zone. This study leveraged three years of isotope data (δ<sup>18</sup>O and δ<sup>2</sup>H) collected from twelve water sources, including precipitation, throughfall, snow, stream water, groundwater, dew water, frost water, mobile and less-mobile soil water, root water, stem water, and leaf water in two catchments with distinct land cover (forestland versus grassland) on the Chinese Loess Plateau (CLP). We infer the main ecohydrological processes controlling water exchange in the Critical Zone under the contrasting vegetation covers. Our results showed new interactions among the several investigated water types, and in particular highlighted that: i) The seasonal isotopic variation in precipitation played a critical role in the seasonal isotopic patterns observed in other water types; ii) Dew water significantly contributed to leaf water uptake, more in forestland (26 ± 6 %) than in grassland (16 ± 11 %). Snow and groundwater were more influential for root water of shrubs and grasses in forestland (59 ± 34 % and 16 ± 8 % for snow and groundwater, respectively) than in grassland (36 ± 26 % and 6 ± 6 %) and they were very important for stem water of trees in forestland (84 ± 14 % and 45 ± 22 % for snow and groundwater, respectively); iii) Isotopic values in mobile and less-mobile soil water differed significantly between forestland and grassland (<em>p</em> < 0.05), but those in plant water (root, stem, and leaf water) did not differ significantly (<em>p</em> > 0.05); and iv) There were dynamic exchanges between mobile and less-mobile soil water, and between groundwater and soil water on the CLP. All these observations allowed us to establish a new isotope-based conceptual model of the ecohydrological cycle in the Critical Zone of the CLP that provides the foundation for future research and sustainable water resource management in this region.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"656 ","pages":"Article 133042"},"PeriodicalIF":5.9,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143591353","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}
Pub Date : 2025-03-09DOI: 10.1016/j.jhydrol.2025.133045
Maryam Amiri , Saeed Sharafi , Mehdi Mohammadi Ghaleni
Accurate evaluation of daily reference evapotranspiration (ETref) is essential for effective water resource management and drought mitigation, particularly in arid climates. However, developing countries frequently lack the necessary infrastructure for precise ETref assessment. Recent advancements have introduced various black box machine learning (ML) models, including the Adaptive Neuro-Fuzzy Inference System-Particle Swarm Optimization algorithm (ANF-PSO), Random Forest (RF), and Support Vector Machine (SVM), to predict daily ETref. Despite their effectiveness, these models suffer from a lack of interpretability, raising concerns about biases, fairness, and accountability in decision-making. Additionally, their performance varies significantly across different climatic conditions, limiting their general applicability. To address these challenges, this paper presents DIRECTORS, a novel daily ETref prediction model based on pattern mining. DIRECTORS leverages correlations among meteorological parameters and autonomously extracts climate-specific behavioral patterns without predefined pattern lengths. By utilizing these patterns and recent station behavior, DIRECTORS forecasts macroscopic daily ETref values and further refines these predictions using RF based on identified similar patterns. This innovative approach offers distinctive insights and solutions to the limitations of traditional ML models in daily ETref prediction. Extensive evaluation demonstrates DIRECTORS’ effectiveness and its potential to significantly enhance predictive accuracy, making it a valuable tool for water resource management and planning in varying environmental conditions.
{"title":"Enhancing daily reference evapotranspiration (ETref) prediction across diverse climatic zones: A pattern mining approach with DIRECTORS model","authors":"Maryam Amiri , Saeed Sharafi , Mehdi Mohammadi Ghaleni","doi":"10.1016/j.jhydrol.2025.133045","DOIUrl":"10.1016/j.jhydrol.2025.133045","url":null,"abstract":"<div><div>Accurate evaluation of daily reference evapotranspiration (ET<sub>ref</sub>) is essential for effective water resource management and drought mitigation, particularly in arid climates. However, developing countries frequently lack the necessary infrastructure for precise ET<sub>ref</sub> assessment. Recent advancements have introduced various black box machine learning (ML) models, including the Adaptive Neuro-Fuzzy Inference System-Particle Swarm Optimization algorithm (ANF-PSO), Random Forest (RF), and Support Vector Machine (SVM), to predict daily ET<sub>ref</sub>. Despite their effectiveness, these models suffer from a lack of interpretability, raising concerns about biases, fairness, and accountability in decision-making. Additionally, their performance varies significantly across different climatic conditions, limiting their general applicability. To address these challenges, this paper presents DIRECTORS, a novel daily ET<sub>ref</sub> prediction model based on pattern mining. DIRECTORS leverages correlations among meteorological parameters and autonomously extracts climate-specific behavioral patterns without predefined pattern lengths. By utilizing these patterns and recent station behavior, DIRECTORS forecasts macroscopic daily ET<sub>ref</sub> values and further refines these predictions using RF based on identified similar patterns. This innovative approach offers distinctive insights and solutions to the limitations of traditional ML models in daily ET<sub>ref</sub> prediction. Extensive evaluation demonstrates DIRECTORS’ effectiveness and its potential to significantly enhance predictive accuracy, making it a valuable tool for water resource management and planning in varying environmental conditions.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"657 ","pages":"Article 133045"},"PeriodicalIF":5.9,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143621354","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}
Pub Date : 2025-03-09DOI: 10.1016/j.jhydrol.2025.133054
Haoxuan Feng , Xuguang Xing , Jiahao Xing , Jianqiang Du , Dongwei Li
Increasing soil salinization and cracking pose threats to agricultural productivity worldwide and can lead to long-term adverse consequences on soil hydrology. However, the combined effects of salts and cracks on water–salt migration and distribution remain unclarified. Furthermore, a numerical approach for modeling infiltration flow in cracked saline soils has not been developed. Therefore, we aimed to investigate the effects of salinity, crack angle, and crack depth on water and salt flows during the infiltration process. By integrating Richards’ equation, the convection–dispersion equation, and cubic law, a two-dimensional numerical approach was proposed and a model based on finite-element theory was established to simulate the infiltration process and soil water and salt distribution in cracked soils. The experimental observations indicated that soil salts and cracks had profound effects on the infiltration process. Specifically, higher soil salinity reduced cumulative infiltration, whereas larger crack angles and smaller crack depths increased it. However, differences in salinity did not significantly affect wetting pattern morphology. Conversely, differences in crack patterns caused differences in wetting pattern morphology, but the differences in the morphological characteristics gradually diminished when the crack depth exceeded 5 cm. After infiltration, the final average soil moisture showed a tendency to decrease with an increase in soil salinity and decrease in crack angle, and changes in the crack depth caused marked changes in the soil water distribution. Additionally, large crack angles and small crack depths facilitated salt leaching. The proposed model was employed and validated through comparisons between experimental observations and numerical simulations, which showed its high accuracy in simulating the infiltration process and water and salt distribution in cracked saline soils, with R2 values of 0.996–0.999, 0.985–0.999, 0.257–0.999, and 0.985–0.999 for cumulative infiltration, wetting pattern morphology, and distribution of water and salts, respectively, in all treatments. Our findings elucidate the influence of soil salts and cracks on water flow and confirm the potential of using simulation to predict water infiltration in cracked saline soils.
{"title":"Performance and modeling of infiltration flow in cracked saline soils","authors":"Haoxuan Feng , Xuguang Xing , Jiahao Xing , Jianqiang Du , Dongwei Li","doi":"10.1016/j.jhydrol.2025.133054","DOIUrl":"10.1016/j.jhydrol.2025.133054","url":null,"abstract":"<div><div>Increasing soil salinization and cracking pose threats to agricultural productivity worldwide and can lead to long-term adverse consequences on soil hydrology. However, the combined effects of salts and cracks on water–salt migration and distribution remain unclarified. Furthermore, a numerical approach for modeling infiltration flow in cracked saline soils has not been developed. Therefore, we aimed to investigate the effects of salinity, crack angle, and crack depth on water and salt flows during the infiltration process. By integrating Richards’ equation, the convection–dispersion equation, and cubic law, a two-dimensional numerical approach was proposed and a model based on finite-element theory was established to simulate the infiltration process and soil water and salt distribution in cracked soils. The experimental observations indicated that soil salts and cracks had profound effects on the infiltration process. Specifically, higher soil salinity reduced cumulative infiltration, whereas larger crack angles and smaller crack depths increased it. However, differences in salinity did not significantly affect wetting pattern morphology. Conversely, differences in crack patterns caused differences in wetting pattern morphology, but the differences in the morphological characteristics gradually diminished when the crack depth exceeded 5 cm. After infiltration, the final average soil moisture showed a tendency to decrease with an increase in soil salinity and decrease in crack angle, and changes in the crack depth caused marked changes in the soil water distribution. Additionally, large crack angles and small crack depths facilitated salt leaching. The proposed model was employed and validated through comparisons between experimental observations and numerical simulations, which showed its high accuracy in simulating the infiltration process and water and salt distribution in cracked saline soils, with <em>R</em><sup>2</sup> values of 0.996–0.999, 0.985–0.999, 0.257–0.999, and 0.985–0.999 for cumulative infiltration, wetting pattern morphology, and distribution of water and salts, respectively, in all treatments. Our findings elucidate the influence of soil salts and cracks on water flow and confirm the potential of using simulation to predict water infiltration in cracked saline soils.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"656 ","pages":"Article 133054"},"PeriodicalIF":5.9,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143579664","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号