Journal of Hydrology最新文献

{"title":"An improved dynamics framework for accurate multi-step ahead daily streamflow prediction with spatial–temporal and global information","authors":"Longxia Qian, Lili Deng, Yong Zhao, Suzhen Dang, Hongrui Wang","doi":"10.1016/j.jhydrol.2025.134325","DOIUrl":"https://doi.org/10.1016/j.jhydrol.2025.134325","url":null,"abstract":"Accurate and reliable multi-step ahead streamflow forecasting is important for water resource management and flood prevention. To alleviate the temporal lag in multi-step prediction and improve peak prediction capability, this research develops a multi-head self-attention-spatiotemporal skip-connection model (MHSA-STSM), which is based on nonlinear dynamic systems and deep learning approaches. MHSA-STSM comprises a temporal module constructed from a convolutional neural network (CNN), a spatiotemporal module fashioned from a multi-head self-attention mechanism, along with a skip connection that links directly to the original input; these modules enable MHSA-STSM to effectively amalgamate temporal, spatiotemporal, and global information within the data. By learning the mapping between the original attractors and the delay attractors, MHSA-STSM can extract spatiotemporal features from the original attractors, thereby enabling the prediction of future values for the target variable. MHSA-STSM is applied to make a multi-step forecast of daily streamflow in rivers in the states of Maine, USA. For a five-step forecast, the highest R value of MHSA-STSM is 0.960, which is 1.05%–11.10% higher than CNN, multi-head self-attention mechanism-Long Short-Term Memory (MHSA-LSTM) and STSM; the lowest R value of 0.792 is at the USGS1047000 station, which shows a 91.4% improvement over the average of CNN, MHSA-LSTM and STSM; the RMSE and MAPE values of MHSA-STSM are 10.76%–102.50% and 19.26%–305.51% lower than those of three comparative models; the NSE of MHSA-STSM is significantly greater than that of the other models, and is as high as 0.920 at USGS 01013500 station. Moreover, sensitivity experiments on the prediction step length are performed for the model. It is found that MHSA-STSM performed excellently in five-step, seven-step, and ten-step predictions and can effectively alleviate the time lag issue. The R value ranges from 0.960 to 0.938, with NSE from 0.920 to 0.836. As the step length increases from 5 to 10, the R value decreases by only 2.3%, and the NSE decreases by 9.1%, demonstrating high stability, while the performance of other models significantly declines. Therefore, MHSA-STSM can effectively capture the spatiotemporal information embedded in high-dimensional data and make accurate multi-step predictions of daily streamflow.","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"7 1","pages":""},"PeriodicalIF":6.4,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145209758","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":"Determination of dissolved inorganic carbon export and the controlling factors in small mountainous Rivers, Taiwan","authors":"Man-Ching Choi, Pei-Hao Chen, Chi-Wang Tsui, Jr-Chuan Huang, Jun-Yi Lee, Li-Chin Lee","doi":"10.1016/j.jhydrol.2025.134324","DOIUrl":"https://doi.org/10.1016/j.jhydrol.2025.134324","url":null,"abstract":"Riverine dissolved inorganic carbon (DIC), mainly sourced from rock weathering and soil respiration, constitutes the majority of the riverine total dissolved carbon transported to the ocean. Taiwanese rivers exhibit extremely high erosion and are hypothesized to export disproportionately high DIC. Yet, the magnitude, composition, and drivers of DIC export in subtropical mountainous rivers remain poorly understood. This study analyzed DIC compositions, concentrations, and yields across 43 rivers in Taiwan. Considering physio-geographic factors, the main influences and spatial patterns of DIC distribution were identified. The results revealed an average DIC concentration of ∼ 17.28 mg-C L<ce:sup loc=\"post\">−1</ce:sup>, with HCO<ce:inf loc=\"post\">3</ce:inf><ce:sup loc=\"post\">–</ce:sup> accounting for over 90 % of DIC. The average DIC yield of 27.65 ton-C km<ce:sup loc=\"post\">−2</ce:sup> yr<ce:sup loc=\"post\">−1</ce:sup> is tenfold greater than the global average (2.58 ton-C km<ce:sup loc=\"post\">−2</ce:sup> yr<ce:sup loc=\"post\">−1</ce:sup>). Stepwise regression showed that proportion of agricultural land was positively correlated, and proportion of sandstone, shale, and argillite (SSA) was negatively correlated with DIC concentration and yield, respectively. Concentration–discharge (C–Q, C = aQ<ce:sup loc=\"post\">b</ce:sup>) analysis indicated that the intercept (<ce:italic>a</ce:italic>) was positively associated with agricultural land use and negatively with SSA coverage. The slope (<ce:italic>b</ce:italic>) increased with catchment slope, emphasizing the role of landscape controls. These findings underscore that SSA (via rock weathering) and agricultural land (via soil respiration) substantially elevates DIC sources, while the warm, wet climate and high catchment slope (via physical erosion) promote carbonate dissolution. This study provides a piece of the missing puzzle in elucidating the significance of DIC export from subtropical mountainous rivers within the global riverine carbon budget.","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"69 1","pages":""},"PeriodicalIF":6.4,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145209759","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":"Multi-lithofacies alluvial characterization via airborne electromagnetic-borehole fusion using ordinary interval kriging and geologic constraints","authors":"Yuqi Song ,&nbsp;Frank T.-C. Tsai ,&nbsp;Burke J. Minsley ,&nbsp;Wade H. Kress","doi":"10.1016/j.jhydrol.2025.134209","DOIUrl":"10.1016/j.jhydrol.2025.134209","url":null,"abstract":"<div><div>Airborne electromagnetic (AEM) survey provides extensive spatial coverage and detailed resolution at the near surface and can be used to develop hydrogeological models. However, utilization of AEM data is not straightforward because AEM resistivity is an indirect measurement for inferring sediment types. This study develops an ordinary interval kriging (OIK) algorithm and a resistivity-to-multi-lithofacies (R2ML) data fusion workflow for multi-lithofacies alluvial characterization. OIK utilizes irregular interval data to construct three-dimensional (3D) resistivity fields from one-dimensional inverted AEM resistivity models. The R2ML workflow maps the resistivity field generated from OIK into a multi-facies lithological model, incorporating geologic constraints derived from well logs and geological observations. The numerical and real-world cases demonstrate that OIK is computationally efficient, accounts for 3D anisotropy, and minimizes the smoothing effect, thereby preserving resistivity contrasts and reducing interpolation uncertainty. The methodology is applied to lithologic characterization of the Mississippi River Valley alluvial aquifer (MRVA) in the Shellmound area, Mississippi, U.S. A frequency-domain AEM survey was conducted to support groundwater studies for the managed aquifer recharge (MAR) to the MRVA. The resulting lithological model, including four types of lithofacies—clay, very fine sand, fine-medium sands, and graveliferous sands, illustrates the geomorphological processes of the MRVA and implies potential MAR. The alignment between the lithological model and existing geological and hydrogeological investigations demonstrates that OIK and R2ML workflow effectively capture the subsurface architecture of the MRVA. The methods have broad applicability for characterizing alluvial aquifers through AEM-borehole data fusion, supporting sustainable groundwater management.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"663 ","pages":"Article 134209"},"PeriodicalIF":6.3,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145155414","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":"Hydrological response and stability of landslide with cracks under intermittent rainfall: integrating physical modeling, numerical simulation, and field investigations","authors":"Chenyu Zhang ,&nbsp;Shimei Wang ,&nbsp;Li Wang ,&nbsp;Yong Chen ,&nbsp;Yuanyuan He ,&nbsp;Xiaofeng Li ,&nbsp;Kun Fang","doi":"10.1016/j.jhydrol.2025.134316","DOIUrl":"10.1016/j.jhydrol.2025.134316","url":null,"abstract":"<div><div>Intermittent rainfall significantly impacts the landslide stability with cracks in the Three Gorges Reservoir Area (TGRA). This study integrates physical modeling, numerical simulation, and field investigations to investigate hydrological response and landslide stability with cracks under intermittent rainfall. Firstly, landslide physical test models were constructed considering crack locations and depths to investigate deformation processes and hydrological response of the landslides under intermittent rainfall. Subsequently, through the verified numerical simulation method and field investigation, the influence of different crack depths on the seepage field, displacement field and stability of Tanjiawan landslide is explored. The results show that, in contrast to the progressive shallow failure mode of the landslides without cracks, landslides with cracks exhibit failure modes of “local instability − progressive collapse-deep sliding” and “shallow toe sliding − deep progressive collapse” with the middle-rear crack and middle-front cracks respectively, due to the preferential seepage channels. As the crack depth increases, landslide deformation exhibits a trend of nonlinear accelerated growth with a distinct threshold effect. When the crack depth ratio exceeds 40%, the variation rate of displacement increases sharply, indicating an accelerated trend of landslide instability. Intermittent rainfall drives a sequential process in deep soil mass characterized by “stepwise increase in moisture content − gradual accumulation of residual pore water pressure − progressive reduction in effective stress”, leading to progressive degradation of soil strength and serving as a key factor in landslide instability. This study provides theoretical reference into the instability mechanisms of landslides with cracks and scientific support for landslide prevention and mitigation in the TGRA.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"663 ","pages":"Article 134316"},"PeriodicalIF":6.3,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145155347","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":"A multivariate copula framework quantifies the augmenting joint risks of compound high temperature and dry/wet events under climate change in China","authors":"Min Liu ,&nbsp;Tao Li ,&nbsp;Peter Goethals ,&nbsp;Fengjiao Song ,&nbsp;Jianying Ma ,&nbsp;Wei Sun","doi":"10.1016/j.jhydrol.2025.134321","DOIUrl":"10.1016/j.jhydrol.2025.134321","url":null,"abstract":"<div><div>As climate change intensifies, the urgency to accurately quantify the risks associated with compound climate events increases, yet the multi-dimensional integrated risk assessment of these events remains very weak. Here, we assessed the occurrence characteristics of compound high temperature and dry events (CHTDE) and compound high temperature and wet events (CHTWE) in the historical and future using daily data from the Coupled Model Intercomparison Project Phase 6 (CMIP6), and constructed a three-dimensional copula model to investigate their joint occurrence risks in China. In future emission scenarios, the precipitation threshold line are projected to shift northward, and the extent of high temperatures will continuously expand. These changes directly influenced the spatial distribution of two types of compound events (CEs). Under the highest emission scenario, China is likely to experiences an increase in the total/maximum duration, frequency, and magnitude of both types of CEs, with CHTDE and CHTWE rising by 25%-55% and 92%-677%, respectively. If emissions were controlled under the SSP245 scenario, the occurrence of CHTDE could decrease overall by 11.5%-30%. Specifically, 87% and 93% of regions will experience more frequent, larger, and more intense of CHTDE and CHTWE. Historically, CEs were concentrated in July and August, but it is likely to occur more frequently across a broader range of months in future. The three-dimensional copula analysis highlights the complex interdependencies among the duration, frequency, and intensity of CEs under different climate scenarios. Our findings further emphasize the inrease in severity of CEs in future scenarios, with a noticeable reduction in their return periods, compared to two-dimensional copula analysis. We quantitatively assessesed the occurrence risk of CEs, explored the mechanisms of their occurrence and persistence, and emphasized that CHTWE should warrant greater attention.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"663 ","pages":"Article 134321"},"PeriodicalIF":6.3,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145155350","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":"A soil erosion-based framework for assessing mercury nonpoint pollution in mining watersheds","authors":"Xianwu Zheng ,&nbsp;Kouping Chen ,&nbsp;Bowen Luo ,&nbsp;Jichun Wu ,&nbsp;Huali Chen ,&nbsp;Lei Xiang","doi":"10.1016/j.jhydrol.2025.134309","DOIUrl":"10.1016/j.jhydrol.2025.134309","url":null,"abstract":"<div><div>In the Wanshan mining watershed, erosion is an important surface process transporting mercury contaminants from soil to river, posing substantial pollution risk to ecosystem. The erosion-driven mercury pollution is influenced by rainfall, soil erodibility, vegetation coverage, topography, mercury pollutant, and human activity. Quantifying the contribution of these factors is vital to evaluate erosion-driven mercury pollution. Previous studies over-relied on expert-based subjective method to assess factor contributions, yielding discrepancies between mercury pollution risk and observed fluvial contamination. Here, this study applied the CRITIC method to quantify objective factor contributions. Results indicate that rainfall, human activity, and mercury pollutant predominantly determine pollution risk, with contribution of 28 %, 20 %, and 18 %, respectively. Based on objective contribution results, we develop a novel risk index to evaluate mercury pollution. The evaluated mercury pollution risk and observed fluvial mercury contamination show a consistent distribution pattern. These findings provide effective reference for pollution control within mining watersheds.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"663 ","pages":"Article 134309"},"PeriodicalIF":6.3,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145154856","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":"Near real-time satellite soil moisture estimation via residual learning integrated with sensor networks","authors":"Soumita Sengupta,&nbsp;Hone-Jay Chu","doi":"10.1016/j.jhydrol.2025.134302","DOIUrl":"10.1016/j.jhydrol.2025.134302","url":null,"abstract":"<div><div>Soil moisture (SM) is crucial for climate dynamics, hydrological processes, agricultural productivity, drought and flood management. However, real-time SM monitoring remains challenging due to sparse in-situ observations. This study presents a novel sensor driven residual learning framework that integrates multi-source data—including in-situ measurements (COSMOS-UK), satellite information (SMAP, AMSR2/GCOM-W1, SMOPS, and MODIS), and meteorological variables to generate high-precision, near real-time SM estimates across the United Kingdom (UK). The methodology employs a two-stage machine learning approach: the first stage utilizes an ensemble model to generate initial SM estimates, while the second stage applies residual learning informed by automated sensor networks to refine these estimates by correcting systematic deviations observed in the UK. Unlike conventional approaches that rely on historical time-series data, this framework demonstrates that reliable SM estimation can be achieved using single-time satellite observations with in-situ data, enabling near real-time monitoring. Initial SM estimates achieved an R² of 0.75 across 40 stations, with 37 stations achieving &gt;70 % relative accuracy. Interestingly, residual analysis within the model revealed comparatively large residuals in central and southern UK regions, and the final refined SM estimations through residual learning improved the R² to 0.94. This computationally efficient, scalable framework offers a robust solution for data-sparse regions, advancing near real-time hydrological forecasting, drought assessment, and climate resilience strategies.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"663 ","pages":"Article 134302"},"PeriodicalIF":6.3,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145154857","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":"Improvement of CMIP6 water vapor accuracy by the digital twin innovation based on GNSS","authors":"Ao Guo ,&nbsp;Yan Xu ,&nbsp;Nan Jiang ,&nbsp;Yubo Wang ,&nbsp;Jiangteng Wang ,&nbsp;Tianhe Xu ,&nbsp;Harald Schuh","doi":"10.1016/j.jhydrol.2025.134305","DOIUrl":"10.1016/j.jhydrol.2025.134305","url":null,"abstract":"<div><div>The accelerating pace of global warming poses unprecedented challenges to climate prediction and environmental sustainability. The ensuing development of the sixth phase of the Coupled Model Intercomparison Project (CMIP6) has empowered climate research into a new era, enabling simulation and projection of the global atmosphere. However, the Global Climate Models (GCMs) database is built upon physical models, inevitably with limitations of deficient observational restraints, insufficient regional simulation capacities and low spatio-temporal resolution. In contrast, the Global Navigation Satellite System (GNSS) is characterized by high precision, high temporal resolution and all-weather availability. Therefore, we propose a GNSS-integrated approach that leverages the high-precision feature of GNSS observations to enhance CMIP6 water vapor accuracy and demonstrate the improved performances of the digital twin of atmospheric Precipitable Water Vapor (PWV) over the Turkey with comprehensive validations. The results show that the Root Mean Square Errors (RMSEs) of CMIP6 water vapor improved from CNN, XGBoost and LSTM algorithm digital twins are 4.57 mm, 4.04 mm and 4.93 mm against GNSS-PWV and 5.40 mm, 5.66 mm and 5.41 mm against ERA5-PWV, which are improved by 22.27 %, 18.51 % and 22.12 %, respectively. Spatio-temporal analysis reveals the pronounced improvements during winter and in mid-altitude regions. Notably, low RMSEs were recorded in the eastern and central inland areas (improved by 50 % upon XGBoost). Across all digital twin implementations, this study pioneers GNSS into CMIP6 water vapor correction, improving the accuracy of future water vapor projections from GCMs obviously. These breakthroughs promote the contribution of GNSS in meteorology and geodesy for climate research.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"663 ","pages":"Article 134305"},"PeriodicalIF":6.3,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145155421","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":"MamGA: a deep neural network architecture for dual-channel parallel monthly runoff prediction based on mamba and depth-gated attention layer","authors":"Wen-chuan Wang ,&nbsp;Wei-can Tian ,&nbsp;Ming-lei Ren ,&nbsp;Dong-mei Xu","doi":"10.1016/j.jhydrol.2025.134304","DOIUrl":"10.1016/j.jhydrol.2025.134304","url":null,"abstract":"<div><div>Monthly runoff prediction is crucial in water resource management, involving both short-term hydrological dynamics and long-term planning. It has a decisive impact on flood prevention, resource allocation, and ecological protection. In the context of increasing uncertainties in runoff due to climate change and human activities, accurate monthly runoff forecasting becomes even more essential. Therefore, this paper proposes a novel dual-channel parallel monthly runoff prediction deep neural network architecture—MamGA—built on the significant application value of deep neural networks in runoff prediction. The architecture first introduces the Mamba model, which employs a selection mechanism to achieve selective information propagation and suppression, effectively enhancing the processing capability of global feature information while reducing the computational complexity of modelling long sequences. Furthermore, this paper incorporates a Depth-gated Attention Layer that combines bidirectional depth-gated modules and linear attention mechanisms to address the shortcomings of the Mamba network in unidirectional information processing. Integrating an Embedded Coding layer and a Sequential Decoding layer constructs an efficient coding and decoding system, further strengthening the model’s ability to capture global features and temporal information. To validate the effectiveness and advancement of the MamGA model, this study selected the Manwan Station (MW), Xiaowan Station (XW) in China, and the Thunder Creek Station (TC) in the United States as experimental subjects. Five evaluation metrics were employed for comparative analysis against nine benchmark models. The experimental results indicate that the MamGA model exhibits significant superiority across all cases. For instance, at the MW station, compared to the Long Short-Term Memory (LSTM) model, the MamGA model reduced the Mean Absolute Error (MAE) and Normalized Root Mean Square Error (NRMSE) by 33.08% and 23.93%, respectively. Meanwhile, the Nash Efficiency Coefficient (NSE), correlation coefficient (R), and Kling-Gupta Efficiency (KGE) improved by 8.41%, 3.93%, and 8.36%, respectively, with both R and NSE exceeding 0.9. The MamGA model also demonstrated significant performance improvements at other stations compared to the competing models. The study suggests that the MamGA model, as an advanced tool for monthly runoff prediction, can significantly enhance the accuracy of runoff forecasting, providing robust support for the optimal allocation and management of water resources.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"663 ","pages":"Article 134304"},"PeriodicalIF":6.3,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145155348","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":"Simulating and predicting lake dynamics by fusing HBV modeling, machine learning approach and remote sensing data","authors":"Muhammad Naeem ,&nbsp;Yongqiang Zhang ,&nbsp;Ning Ma ,&nbsp;Zixuan Tang ,&nbsp;Ping Miao ,&nbsp;Xiaoqiang Tian ,&nbsp;Congcong Li ,&nbsp;Qi Huang ,&nbsp;Zhenwu Xu ,&nbsp;Longhao wang ,&nbsp;Zhen Huang","doi":"10.1016/j.jhydrol.2025.134303","DOIUrl":"10.1016/j.jhydrol.2025.134303","url":null,"abstract":"<div><div>This study provides a comprehensive analysis of the hydrological dynamics and land use changes in the Hongjiannao Lake Basin from 1990 to 2023, with projections extending to 2060. By integrating advanced hydrological modeling Hydrologiska Byrans Vattenbalansavdelning (HBV), a machine learning algorithm Random Forest (RF), Cellular Automata (CA) Markov, and remote sensing data, this research offers a robust framework for understanding the interactions between climate change, anthropogenic activities, and ecosystem responses. The historical analysis revealed remarkable fluctuations in the lake’s area, including a 25.5 % reduction between 2000 and 2011, followed by a recovery from 2012 to 2023. The lake area increased by 26.2 % during the recovery phase, highlighting a partial reversal of decline. Projections indicate that, under various future climate scenarios, the lake area could increase by 29 % by 2060, showcasing the resilience of the ecosystem despite ongoing climate and anthropogenic pressures. The RF model demonstrated strong predictive capabilities, with R² values of 0.92 during 1990–2013 calibration and 0.76 during 2014–2023 validation, coupled with root mean square errors of 0.12 km² and 0.26 km², respectively. Additionally, the CA-Markov model predicted vegetation growth and urbanization, highlighting potential for significant landscape changes. These findings stress the need for water management strategies to preserve the lake’s ecological health, advocating for the integration of climate, land use, and hydrological factors in management plans for sustainable conservation and restoration in semi-arid regions.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"663 ","pages":"Article 134303"},"PeriodicalIF":6.3,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145118373","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}