Journal of Hydrology-Regional Studies最新文献_第4页

Runoff variation and attribution analysis in northern Chinese watersheds using an ABCD-snowmelt hydrological model 基于abcd -融雪水文模型的中国北方流域径流变化及归因分析

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies

Pub Date : 2026-01-10 DOI: 10.1016/j.ejrh.2026.103115

Bo Zhang , Limin Duan , Yixuan Wang , Jin Sun , Shaojie Chu , Guixin Zhang , Javed Iqbal , Tingxi Liu

Study region

Nine representative watersheds across northern China, encompassing mountain forest, steppe, and sandy coarse sand watersheds.

Study focus

To quantify climate change and human activities’ impacts on runoff dynamics in arid/semi-arid regions, using the Mann–Kendall test and an ABCD-snowmelt hydrological model to disentangle their relative influences.

New hydrological insights for the region

With the exception of a transient increase in dry-season runoff in the Miandu, all watersheds exhibited significant declines in annual, wet-season, and dry-season runoff. At the monthly scale, runoff dynamics in March and April were strongly affected by spring snowmelt, with climate change emerging as the dominant driver in steppe and sandy coarse sand watersheds. Seasonal analysis indicated that spring and summer runoff variations were largely controlled by climate change, while winter runoff reflected comparable contributions from climate change and human activities. At the annual scale, runoff changes in mountain forest watersheds were primarily attributed to climate change (73–86 %), whereas steppe (43–65 %) and sandy coarse sand watersheds (62–80 %) were more strongly influenced by human activities. This study identifies the spatiotemporal heterogeneity of runoff responses to climate change and human activities across watersheds with different underlying surface conditions. The findings provide a clear scientific basis for developing targeted ecological restoration measures and sustainable water-resource management strategies.

研究区域中国北方9个代表性流域，包括山林、草原、沙质粗沙流域。为了量化气候变化和人类活动对干旱/半干旱区径流动态的影响，采用Mann-Kendall检验和abcd -融雪水文模型来厘清两者的相对影响。除绵渡干季径流量短暂增加外，所有流域的年径流量、湿季径流量和干季径流量均显著下降。月尺度上，春季融雪对3、4月径流动态的影响较大，气候变化是草原和沙质粗沙流域径流动态的主要驱动因素。季节分析表明，春季和夏季径流变化主要受气候变化控制，冬季径流受气候变化和人类活动的影响相当。在年尺度上，山林流域径流变化主要受气候变化的影响（73% ~ 86% %），而草原流域（43 ~ 65% %）和砂质粗砂流域（62 ~ 80% %）受人类活动的影响更大。研究了不同下垫面条件下流域径流对气候变化和人类活动响应的时空异质性。研究结果为制定有针对性的生态恢复措施和可持续水资源管理战略提供了明确的科学依据。

{"title":"Runoff variation and attribution analysis in northern Chinese watersheds using an ABCD-snowmelt hydrological model","authors":"Bo Zhang , Limin Duan , Yixuan Wang , Jin Sun , Shaojie Chu , Guixin Zhang , Javed Iqbal , Tingxi Liu","doi":"10.1016/j.ejrh.2026.103115","DOIUrl":"10.1016/j.ejrh.2026.103115","url":null,"abstract":"<div><h3>Study region</h3><div>Nine representative watersheds across northern China, encompassing mountain forest, steppe, and sandy coarse sand watersheds.</div></div><div><h3>Study focus</h3><div>To quantify climate change and human activities’ impacts on runoff dynamics in arid/semi-arid regions, using the Mann–Kendall test and an ABCD-snowmelt hydrological model to disentangle their relative influences.</div></div><div><h3>New hydrological insights for the region</h3><div>With the exception of a transient increase in dry-season runoff in the Miandu, all watersheds exhibited significant declines in annual, wet-season, and dry-season runoff. At the monthly scale, runoff dynamics in March and April were strongly affected by spring snowmelt, with climate change emerging as the dominant driver in steppe and sandy coarse sand watersheds. Seasonal analysis indicated that spring and summer runoff variations were largely controlled by climate change, while winter runoff reflected comparable contributions from climate change and human activities. At the annual scale, runoff changes in mountain forest watersheds were primarily attributed to climate change (73–86 %), whereas steppe (43–65 %) and sandy coarse sand watersheds (62–80 %) were more strongly influenced by human activities. This study identifies the spatiotemporal heterogeneity of runoff responses to climate change and human activities across watersheds with different underlying surface conditions. The findings provide a clear scientific basis for developing targeted ecological restoration measures and sustainable water-resource management strategies.</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"63 ","pages":"Article 103115"},"PeriodicalIF":5.0,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926556","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Drought resistance of dams based on propagation analysis: A case study of various multipurpose dams in South Korea 基于繁殖分析的水坝抗旱性：以韩国多用途水坝为例

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies

Pub Date : 2026-01-09 DOI: 10.1016/j.ejrh.2026.103106

Chaelim Lee , Sinuk Kang , Sangdan Kim

Study region

Nine major multipurpose dam basins in South Korea

Study focus

The foundation for quantitatively evaluating the drought resistance of dams is still insufficient. In this study, we targeted multipurpose dams in Korea, identified meteorological drought (MD), hydrological drought in the natural state (HDn, drought under the assumption that no dam exists), and hydrological drought in the adjusted state (HDa, drought under the assumption that the dam is in a regulated state), and then sought to quantify the drought resistance of dams based on propagation analysis.

New hydrological insights from the region

When drought events propagated from HDn to HDa, we found that the number of drought events decreased, the duration shortened, and the intensity of droughts tended to weaken. Among the nine dams analyzed, six exhibited high levels of drought propagation resistance. However, due to differences in dam design parameters, operational practices, and the topographical characteristics of the catchment areas, the remaining two dams showed relatively low resistance, and the final dam actually exacerbated drought conditions. The methodology proposed in this study, which involves estimating drought propagation time and delay time and analyzing the α-shape through bivariate frequency analysis, can serve as a standard tool for quantitatively evaluating the drought resistance of dams.

研究区域韩国9个主要的多用途水坝流域研究重点定量评价水坝抗旱性的基础仍然不足。本研究以韩国多用途大坝为研究对象，识别气象干旱（MD）、自然状态水文干旱（HDn，假设无大坝存在的干旱）和调节状态水文干旱（HDa，假设大坝处于调节状态下的干旱），并基于繁殖分析对大坝抗旱性进行量化。当干旱事件从HDn向HDa传播时，我们发现干旱事件的数量减少，持续时间缩短，干旱强度趋于减弱。在分析的9座大坝中，有6座大坝表现出较高的抗旱繁殖能力。然而，由于大坝设计参数、运行实践和流域地形特征的差异，其余两座大坝的阻力相对较低，最终大坝实际上加剧了干旱状况。本文提出的方法包括估算干旱传播时间和延迟时间，并通过双变量频率分析分析α-形状，可作为定量评价大坝抗旱性的标准工具。

{"title":"Drought resistance of dams based on propagation analysis: A case study of various multipurpose dams in South Korea","authors":"Chaelim Lee , Sinuk Kang , Sangdan Kim","doi":"10.1016/j.ejrh.2026.103106","DOIUrl":"10.1016/j.ejrh.2026.103106","url":null,"abstract":"<div><h3>Study region</h3><div>Nine major multipurpose dam basins in South Korea</div></div><div><h3>Study focus</h3><div>The foundation for quantitatively evaluating the drought resistance of dams is still insufficient. In this study, we targeted multipurpose dams in Korea, identified meteorological drought (MD), hydrological drought in the natural state (HDn, drought under the assumption that no dam exists), and hydrological drought in the adjusted state (HDa, drought under the assumption that the dam is in <em>a</em> regulated state), and then sought to quantify the drought resistance of dams based on propagation analysis.</div></div><div><h3>New hydrological insights from the region</h3><div>When drought events propagated from HDn to HDa, we found that the number of drought events decreased, the duration shortened, and the intensity of droughts tended to weaken. Among the nine dams analyzed, six exhibited high levels of drought propagation resistance. However, due to differences in dam design parameters, operational practices, and the topographical characteristics of the catchment areas, the remaining two dams showed relatively low resistance, and the final dam actually exacerbated drought conditions. The methodology proposed in this study, which involves estimating drought propagation time and delay time and analyzing the α-shape through bivariate frequency analysis, can serve as a standard tool for quantitatively evaluating the drought resistance of dams.</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"63 ","pages":"Article 103106"},"PeriodicalIF":5.0,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spatiotemporal analysis of lower tail precipitation events over the Eastern Middle East using the complex network to capture drought patterns 利用复杂网络捕捉干旱模式的中东东部低尾降水事件时空分析

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies

Pub Date : 2026-01-09 DOI: 10.1016/j.ejrh.2026.103105

Mohamad Haytham Klaho , Mohammadali Alijanian , Hamed Yazdian , Somayeh Abedian

Study region

Iran and the countries around it.‎

Study focus

Precipitation deficits, particularly lower tail precipitation events (LTPEs), are a critical driver of drought; nevertheless, their spatiotemporal dynamics remain insufficiently examined. This study investigates LTPEs in Iran using a complex network (CN) framework, employing 30 years (1994–2023) of monthly precipitation data from the Climate Prediction Center. The 10th percentile criterion was employed to identify LTPEs, followed by event synchronization to assess connectivity. Undirected and directed complex networks were established to assess degree centrality, betweenness centrality, clustering coefficient, mean synchronized distance, network divergence, and directional fluxes.

New hydrological insights for the region

The results of all undirected coefficients indicate that Saudi Arabia, southeastern Iran, and Pakistan suffer from severe drought, revealing localized ‎drought clusters in these zones, highly vulnerable to the reduction of LTPEs. The mean synchronized distance (MSD) ‎reveals that eastern, central, and western Iran are substantially impacted by localized droughts, while ‎increased MSD values in Saudi Arabia, southeastern Iran, and Pakistan imply a heightened risk of extensive drought ‎propagation to other zones. In addition, the results of directed coefficients reveal Saudi Arabia and Turkmenistan as critical pathways ‎for drought propagation. The CN approach effectively illustrates the synchronization, distribution, and ‎directional spread of drought based on LTPEs, providing novel insights into drought dynamics in a particularly ‎vulnerable area.‎

研究伊朗及其周边国家。降水不足，特别是低尾降水事件（LTPEs），是干旱的关键驱动因素；然而，它们的时空动态仍然没有得到充分的研究。本研究使用一个复杂网络（CN）框架，利用气候预测中心提供的30年（1994-2023）月降水数据，对伊朗的ltpe进行了研究。采用第10百分位标准来识别ltpe，然后使用事件同步来评估连通性。建立无向和有向复杂网络，评估度中心性、中间中心性、聚类系数、平均同步距离、网络散度和方向通量。所有无向系数的结果表明，沙特阿拉伯、伊朗东南部和巴基斯坦遭受严重干旱，揭示了这些地区的局部干旱集群，高度容易受到ltpe减少的影响。平均同步距离（MSD）显示，伊朗东部、中部和西部地区受到局部干旱的严重影响，而沙特阿拉伯、伊朗东南部和巴基斯坦的MSD值增加意味着干旱向其他地区广泛传播的风险增加。此外，有向系数结果显示，沙特阿拉伯和土库曼斯坦是干旱传播的关键途径。CN方法有效地说明了基于ltpe的干旱的同步、分布和定向传播，为特别脆弱地区的干旱动态提供了新的见解

{"title":"Spatiotemporal analysis of lower tail precipitation events over the Eastern Middle East using the complex network to capture drought patterns","authors":"Mohamad Haytham Klaho , Mohammadali Alijanian , Hamed Yazdian , Somayeh Abedian","doi":"10.1016/j.ejrh.2026.103105","DOIUrl":"10.1016/j.ejrh.2026.103105","url":null,"abstract":"<div><h3>Study region</h3><div>Iran and the countries around it.‎</div></div><div><h3>Study focus</h3><div>Precipitation deficits, particularly lower tail precipitation events (LTPEs), are a critical driver of drought; nevertheless, their spatiotemporal dynamics remain insufficiently examined. This study investigates LTPEs in Iran using a complex network (CN) framework, employing 30 years (1994–2023) of monthly precipitation data from the Climate Prediction Center. The 10th percentile criterion was employed to identify LTPEs, followed by event synchronization to assess connectivity. Undirected and directed complex networks were established to assess degree centrality, betweenness centrality, clustering coefficient, mean synchronized distance, network divergence, and directional fluxes.</div></div><div><h3>New hydrological insights for the region</h3><div>The results of all undirected coefficients indicate that Saudi Arabia, southeastern Iran, and Pakistan suffer from severe drought, revealing localized ‎drought clusters in these zones, highly vulnerable to the reduction of LTPEs. The mean synchronized distance (MSD) ‎reveals that eastern, central, and western Iran are substantially impacted by localized droughts, while ‎increased MSD values in Saudi Arabia, southeastern Iran, and Pakistan imply a heightened risk of extensive drought ‎propagation to other zones. In addition, the results of directed coefficients reveal Saudi Arabia and Turkmenistan as critical pathways ‎for drought propagation. The CN approach effectively illustrates the synchronization, distribution, and ‎directional spread of drought based on LTPEs, providing novel insights into drought dynamics in a particularly ‎vulnerable area.‎</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"63 ","pages":"Article 103105"},"PeriodicalIF":5.0,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Mid-late Holocene hydrological regime on the Sanjiang Plain (Northeast China) in response to East Asian summer monsoon based on a multi-proxy record 基于多代理记录的全新世中晚期三江平原对东亚夏季风的水文响应

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies

Pub Date : 2026-01-08 DOI: 10.1016/j.ejrh.2026.103111

Dongxue Han , Jinxin Cong , Guoping Wang , Lin Li , Chuanyu Gao

Study region

The Sanjiang Plain, which is located in mid-high latitudes and monsoon marginal regions.

Study focus

A comprehensive study of long-term hydrological record from climate-sensitive regions is necessary for understanding wetland development and degradation in the context of global warming. This study aims to reconstruct wetland hydrological variations based on multi-proxy during the mid-late Holocene, and to explore how wetland evolution responded to East Asian summer monsoon (EASM).

New hydrological insights for the region

Water level fluctuations in Zhuaji Wetland closely followed changes in EASM intensity during the mid-late Holocene. The grain-size median diameter, sand content and the first principal component scores based on diatom collectively indicated a declining water level caused by the weakened intensity of EASM after the Holocene Climatic Optimum. The diatom record clearly documented key climatic events such as the warm and humid Holocene Climatic Optimum and Medieval Climate Anomaly, as well as the cold and dry 4.2 ka event and Little Ice Age, due to its high sensitivity to hydroclimatic variations. The most dramatic diatom compositional change was the increase in dry-indicating species, Hantzschia amphioxys and Pinnularia subcapitata, which began approximately 300 years ago. Sand content also peaked since 0.3 cal ka BP in Zhuaji Wetland. Intensified human disturbance, primarily through agricultural activities, had led to soil erosion, wetland degradation and diatom diversity destruction on the Sanjiang Plain.

研究区域：三江平原，地处中高纬度和季风边缘地区。研究重点气候敏感区长期水文记录的综合研究是理解全球变暖背景下湿地发展与退化的必要条件。本研究旨在重建全新世中晚期湿地水文变化，探讨湿地演化对东亚夏季风的响应。全新世中后期EASM强度的变化为诸集湿地区域水位波动提供了新的水文信息。粒度中值直径、含沙量和硅藻第一主成分得分共同表明，全新世气候最适期后，由于EASM强度减弱，水位下降。由于硅藻对水文气候变化的高度敏感性，硅藻记录清楚地记录了温暖湿润的全新世气候最佳期和中世纪气候异常期、寒冷干燥的4.2 ka事件和小冰期等关键气候事件。最显著的硅藻组成变化是在大约300年前开始的指示干旱的种类，如amphioxys和Pinnularia subcapitata的增加。砂含量在0.3 cal ka BP后达到峰值。以农业活动为主的人为干扰加剧，导致三江平原土壤侵蚀、湿地退化和硅藻多样性破坏。

{"title":"Mid-late Holocene hydrological regime on the Sanjiang Plain (Northeast China) in response to East Asian summer monsoon based on a multi-proxy record","authors":"Dongxue Han , Jinxin Cong , Guoping Wang , Lin Li , Chuanyu Gao","doi":"10.1016/j.ejrh.2026.103111","DOIUrl":"10.1016/j.ejrh.2026.103111","url":null,"abstract":"<div><h3>Study region</h3><div>The Sanjiang Plain, which is located in mid-high latitudes and monsoon marginal regions.</div></div><div><h3>Study focus</h3><div>A comprehensive study of long-term hydrological record from climate-sensitive regions is necessary for understanding wetland development and degradation in the context of global warming. This study aims to reconstruct wetland hydrological variations based on multi-proxy during the mid-late Holocene, and to explore how wetland evolution responded to East Asian summer monsoon (EASM).</div></div><div><h3>New hydrological insights for the region</h3><div>Water level fluctuations in Zhuaji Wetland closely followed changes in EASM intensity during the mid-late Holocene. The grain-size median diameter, sand content and the first principal component scores based on diatom collectively indicated a declining water level caused by the weakened intensity of EASM after the Holocene Climatic Optimum. The diatom record clearly documented key climatic events such as the warm and humid Holocene Climatic Optimum and Medieval Climate Anomaly, as well as the cold and dry 4.2 ka event and Little Ice Age, due to its high sensitivity to hydroclimatic variations. The most dramatic diatom compositional change was the increase in dry-indicating species, <em>Hantzschia amphioxys</em> and <em>Pinnularia subcapitata</em>, which began approximately 300 years ago. Sand content also peaked since 0.3 cal ka BP in Zhuaji Wetland. Intensified human disturbance, primarily through agricultural activities, had led to soil erosion, wetland degradation and diatom diversity destruction on the Sanjiang Plain.</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"63 ","pages":"Article 103111"},"PeriodicalIF":5.0,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Improving evapotranspiration estimation by integrating process-based biophysical variables into a deep learning approach 通过将基于过程的生物物理变量集成到深度学习方法中来改进蒸散估算

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies

Pub Date : 2026-01-08 DOI: 10.1016/j.ejrh.2026.103114

Mingming Xie , Jianyun Zhang , Zhenxin Bao , Linus Zhang , Zheng Duan , Guoqing Wang , Cuishan Liu , Feifei Yuan , Xiaoxiang Guan

Study Region

103 FLUXNET2015 flux towers distributed across diverse climatic and ecological regions.

Study Focus

Accurate estimation of evapotranspiration (ET) is critical for understanding regional ecohydrological processes. Physically based models such as the Penman-Monteith-Leuning (PML) model are robust but often constrained by fixed parameterization schemes, while data-driven approaches such as Long Short-Term Memory (LSTM) networks can capture nonlinearities but depend heavily on training data. To address these limitations, this study developed a hybrid model (PML-LSTM) by integrating biophysical variables from PML simulation into LSTM network. Model performance was systematically evaluated against standalone PML and LSTM across three modelling levels: local (site), type (vegetation type), and group (forest and non-forest).

New Hydrological Insights for the Region

The PML-LSTM model achieved superior performance, with median NSE values of 0.851, 0.913, and 0.933 during validation, surpassing both the PML model (0.843, 0.788, 0.766) and the LSTM model (0.818, 0.879, 0.873). Integrating biophysical information in the PML-LSTM model improved ET estimation accuracy and model generalization, leading to more robust spatiotemporal performance under leave-one-out cross-validation and extreme weather extrapolation experiments. Distinct model behaviors emerged under varying sample conditions: the PML model exhibited greater robustness under data-scarce conditions at the local level, while the LSTM and PML-LSTM models benefited from larger training datasets. This study highlights the potential of combining process-based and data-driven approaches to improve ET estimation and provides insights for regional ecohydrological modelling.

FLUXNET2015通量塔分布在不同的气候和生态区域。研究重点蒸散发（ET）的准确估算是理解区域生态水文过程的关键。基于物理的模型，如Penman-Monteith-Leuning （PML）模型具有鲁棒性，但通常受到固定参数化方案的限制，而数据驱动的方法，如长短期记忆（LSTM）网络可以捕获非线性，但严重依赖于训练数据。为了解决这些限制，本研究通过将PML模拟的生物物理变量集成到LSTM网络中，开发了一个混合模型（PML-LSTM）。针对独立的PML和LSTM在三个建模级别上对模型性能进行了系统评估：局部（站点）、类型（植被类型）和组（森林和非森林）。PML-LSTM模型的NSE中位数分别为0.851、0.913和0.933，优于PML模型（0.843、0.788、0.766）和LSTM模型（0.818、0.879、0.873）。将生物物理信息整合到PML-LSTM模型中，提高了蒸散发估算的精度和模型的泛化能力，在留一交叉验证和极端天气外推实验中具有更强的时空鲁棒性。不同的样本条件下出现了不同的模型行为：PML模型在局部数据稀缺条件下表现出更强的鲁棒性，而LSTM和PML-LSTM模型受益于更大的训练数据集。该研究强调了将基于过程的方法和数据驱动的方法结合起来改进蒸散发估算的潜力，并为区域生态水文建模提供了见解。

{"title":"Improving evapotranspiration estimation by integrating process-based biophysical variables into a deep learning approach","authors":"Mingming Xie , Jianyun Zhang , Zhenxin Bao , Linus Zhang , Zheng Duan , Guoqing Wang , Cuishan Liu , Feifei Yuan , Xiaoxiang Guan","doi":"10.1016/j.ejrh.2026.103114","DOIUrl":"10.1016/j.ejrh.2026.103114","url":null,"abstract":"<div><h3>Study Region</h3><div>103 FLUXNET2015 flux towers distributed across diverse climatic and ecological regions.</div></div><div><h3>Study Focus</h3><div>Accurate estimation of evapotranspiration (ET) is critical for understanding regional ecohydrological processes. Physically based models such as the Penman-Monteith-Leuning (PML) model are robust but often constrained by fixed parameterization schemes, while data-driven approaches such as Long Short-Term Memory (LSTM) networks can capture nonlinearities but depend heavily on training data. To address these limitations, this study developed a hybrid model (PML-LSTM) by integrating biophysical variables from PML simulation into LSTM network. Model performance was systematically evaluated against standalone PML and LSTM across three modelling levels: local (site), type (vegetation type), and group (forest and non-forest).</div></div><div><h3>New Hydrological Insights for the Region</h3><div>The PML-LSTM model achieved superior performance, with median NSE values of 0.851, 0.913, and 0.933 during validation, surpassing both the PML model (0.843, 0.788, 0.766) and the LSTM model (0.818, 0.879, 0.873). Integrating biophysical information in the PML-LSTM model improved ET estimation accuracy and model generalization, leading to more robust spatiotemporal performance under leave-one-out cross-validation and extreme weather extrapolation experiments. Distinct model behaviors emerged under varying sample conditions: the PML model exhibited greater robustness under data-scarce conditions at the local level, while the LSTM and PML-LSTM models benefited from larger training datasets. This study highlights the potential of combining process-based and data-driven approaches to improve ET estimation and provides insights for regional ecohydrological modelling.</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"63 ","pages":"Article 103114"},"PeriodicalIF":5.0,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Changes and driving factors of compound droughts in China based on the soil-atmosphere compound drought index 基于土壤-大气复合干旱指数的中国复合干旱变化及其驱动因素

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies

Pub Date : 2026-01-07 DOI: 10.1016/j.ejrh.2026.103109

Wei Zhang , Anzhou Zhao , Xianfeng Liu , Lidong Zou , Yahui Wang , Xiaoran Han

Study region

China

Study focus

Soil-atmosphere compound droughts, driven by high vapor pressure deficit (VPD) and low soil moisture (SM), significantly threaten vegetation growth. However, there remains a critical gap in quantitatively assessing the spatiotemporal patterns and driving mechanisms of these compound drought events in China. This study constructs a Soil-Atmosphere Compound Drought Index (SACDI) using SM and VPD data spanning 1982–2020, aiming to characterize the spatiotemporal evolution and driving factors of soil-atmosphere compound droughts across China.

New hydrological insights for the region

This study yields several new hydrological insights into compound droughts in China. The identification of drought-fluctuation-wet evolutionary phases (1982–1991: slope=−0.0235/a, p < 0.05; 1992–2006: slope=0.0015/a, p > 0.05; 2007–2020: slope=0.0529/a, p < 0.01) reveals that compound droughts follow long-term hydroclimatic cycles, with negative slopes reflecting drought intensification and positive slopes reflecting alleviation. Persistent hotspots in the Continental River Basin (CRB), Songhua and Liaohe River Basin (SLRB), and Huai River Basin (HRB) indicate basin-specific hydrological vulnerability, affecting river discharge, soil moisture, and water management. Most basins face frequent short compound droughts, whereas Northeast China and southern CRB endure prolonged hydrological stress, highlighting spatial heterogeneity. Attribution analysis shows temperature (22.92 %), atmospheric circulation (22.09 %), and precipitation (17.73 %) as key drivers, implying that warming and circulation anomalies could exacerbate hydrological risks. These findings provide a solid basis for basin-targeted strategies to mitigate compound drought impacts on regional water systems.

土壤-大气复合干旱是由高水汽压亏缺（VPD）和低土壤水分（SM）驱动的，严重威胁植被生长。然而，在定量评估中国这些复合干旱事件的时空格局和驱动机制方面仍存在重大空白。利用1982-2020年的SM和VPD数据，构建了中国土壤-大气复合干旱指数（SACDI），以表征中国土壤-大气复合干旱的时空演变特征及其驱动因素。本研究为中国复合干旱提供了几个新的水文见解。干旱-波动-湿润演化阶段的识别（1982-1991年：坡度= - 0.0235/a, p <； 0.05；1992-2006年：坡度=0.0015/a, p >； 0.05；2007-2020年：坡度=0.0529/a, p <； 0.01）表明，复合干旱遵循长期水文气候循环，负坡度反映干旱加剧，正坡度反映干旱缓解。大陆河流域（CRB）、松花河和辽河流域（SLRB）和淮河流域（HRB）的持续热点表明流域特有的水文脆弱性，影响河流流量、土壤湿度和水管理。大部分流域面临频繁的短期复合干旱，而东北和CRB南部则面临长期的水文压力，空间异质性突出。归因分析显示，气温（22.92 %）、大气环流（22.09 %）和降水（17.73 %）是主要驱动因素，表明增温和环流异常会加剧水文风险。这些发现为以流域为目标的策略提供了坚实的基础，以减轻干旱对区域水系统的复合影响。

{"title":"Changes and driving factors of compound droughts in China based on the soil-atmosphere compound drought index","authors":"Wei Zhang , Anzhou Zhao , Xianfeng Liu , Lidong Zou , Yahui Wang , Xiaoran Han","doi":"10.1016/j.ejrh.2026.103109","DOIUrl":"10.1016/j.ejrh.2026.103109","url":null,"abstract":"<div><h3>Study region</h3><div>China</div></div><div><h3>Study focus</h3><div>Soil-atmosphere compound droughts, driven by high vapor pressure deficit (VPD) and low soil moisture (SM), significantly threaten vegetation growth. However, there remains a critical gap in quantitatively assessing the spatiotemporal patterns and driving mechanisms of these compound drought events in China. This study constructs a Soil-Atmosphere Compound Drought Index (SACDI) using SM and VPD data spanning 1982–2020, aiming to characterize the spatiotemporal evolution and driving factors of soil-atmosphere compound droughts across China.</div></div><div><h3>New hydrological insights for the region</h3><div>This study yields several new hydrological insights into compound droughts in China. The identification of drought-fluctuation-wet evolutionary phases (1982–1991: slope=−0.0235/a, <em>p</em> < 0.05; 1992–2006: slope=0.0015/a, <em>p</em> > 0.05; 2007–2020: slope=0.0529/a, <em>p</em> < 0.01) reveals that compound droughts follow long-term hydroclimatic cycles, with negative slopes reflecting drought intensification and positive slopes reflecting alleviation. Persistent hotspots in the Continental River Basin (CRB), Songhua and Liaohe River Basin (SLRB), and Huai River Basin (HRB) indicate basin-specific hydrological vulnerability, affecting river discharge, soil moisture, and water management. Most basins face frequent short compound droughts, whereas Northeast China and southern CRB endure prolonged hydrological stress, highlighting spatial heterogeneity. Attribution analysis shows temperature (22.92 %), atmospheric circulation (22.09 %), and precipitation (17.73 %) as key drivers, implying that warming and circulation anomalies could exacerbate hydrological risks. These findings provide a solid basis for basin-targeted strategies to mitigate compound drought impacts on regional water systems.</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"63 ","pages":"Article 103109"},"PeriodicalIF":5.0,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926519","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spatiotemporal evolution and attribution analysis of groundwater drought in the North China Plain: GGDI constructed based on downscaled GRACE GWSA 华北平原地下水干旱时空演变与归因分析——基于GRACE GWSA的GGDI构建

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies

Pub Date : 2026-01-07 DOI: 10.1016/j.ejrh.2026.103103

Yijun Li , Xueting Zhong , Huili Gong , Beibei Chen , Chaofan Zhou , Hao Li , Xincheng Wang , Lewei Xu

Study Region

North China Plain (NCP), China.

Study Focus

To address the limitations of low-resolution data accuracy and insufficient quantitative characterization of driving mechanisms in groundwater drought assessment in the NCP, this study proposes a systematic research framework: downscaling GRACE data using Geographically Weighted Regression (GWR) to reconstruct 0.05° Groundwater Storage Anomalies (GWSA) for 2002–2023 and construct the GRACE Groundwater Drought Index (GGDI). Furthermore, combining Random Forest (RF) with Geographic Detector (GD) reveals the quantitative contributions of natural and anthropogenic factors and their interactions (2003–2022). This framework overcomes the limitations of traditional research in spatiotemporal resolution and mechanistic interpretation, providing a new methodological approach for regional groundwater drought research.

New Hydrological Insights for the Region

(1) From 2002–2023, the 0.05° GWSA exhibited higher accuracy than the 0.25°, revealing a −17.81 ± 3.31 mm/yr decline. (2) Drought was mild in the northeast and severe in the southwest. The 0.05° GWSA first detected relief in approximately 4 % of the region (e.g., northern Beijing and northwestern Cangzhou). Fifteen drought events were identified during the study period, showing increasing frequency and intensity. The GGDI rebounded in 2016 and 2021 due to rainfall, the South-to-North Water Diversion, and groundwater-withdrawal restrictions. (3) Groundwater extraction was the dominant driver (contributing 45 %), and its interaction with precipitation markedly amplified drought risk (q = 0.83), particularly in fine-grained aquifer areas.

中国华北平原研究区。研究重点针对华北盆地地下水干旱评价中数据分辨率低、驱动机制定量表征不足等问题，提出了一个系统的研究框架：利用地理加权回归（GWR）降尺度重构2002-2023年0.05°地下水库容异常（GWSA），构建GRACE地下水干旱指数（GGDI）。此外，将随机森林（RF）与地理探测器（GD）相结合，揭示了自然因子和人为因子的定量贡献及其相互作用（2003-2022）。该框架克服了传统研究在时空解析和机理解释上的局限性，为区域地下水干旱研究提供了新的方法途径。(1) 2002-2023年，0.05°GWSA的精度高于0.25°，下降幅度为- 17.81 ± 3.31 mm/yr。(2)东北干旱轻微，西南干旱严重。0.05°GWSA首先探测到约4. %的区域（如北京北部和沧州西北部）出现起伏。在研究期间确定了15次干旱事件，其频率和强度都在增加。由于降雨、南水北调和地下水开采限制，2016年和2021年GGDI出现反弹。(3)地下水开采是主要驱动因素（贡献45 %），其与降水的相互作用显著放大了干旱风险（q = 0.83），特别是在细粒含水层地区。

{"title":"Spatiotemporal evolution and attribution analysis of groundwater drought in the North China Plain: GGDI constructed based on downscaled GRACE GWSA","authors":"Yijun Li , Xueting Zhong , Huili Gong , Beibei Chen , Chaofan Zhou , Hao Li , Xincheng Wang , Lewei Xu","doi":"10.1016/j.ejrh.2026.103103","DOIUrl":"10.1016/j.ejrh.2026.103103","url":null,"abstract":"<div><h3>Study Region</h3><div>North China Plain (NCP), China.</div></div><div><h3>Study Focus</h3><div>To address the limitations of low-resolution data accuracy and insufficient quantitative characterization of driving mechanisms in groundwater drought assessment in the NCP, this study proposes a systematic research framework: downscaling GRACE data using Geographically Weighted Regression (GWR) to reconstruct 0.05° Groundwater Storage Anomalies (GWSA) for 2002–2023 and construct the GRACE Groundwater Drought Index (GGDI). Furthermore, combining Random Forest (RF) with Geographic Detector (GD) reveals the quantitative contributions of natural and anthropogenic factors and their interactions (2003–2022). This framework overcomes the limitations of traditional research in spatiotemporal resolution and mechanistic interpretation, providing a new methodological approach for regional groundwater drought research.</div></div><div><h3>New Hydrological Insights for the Region</h3><div>(1) From 2002–2023, the 0.05° GWSA exhibited higher accuracy than the 0.25°, revealing a −17.81 ± 3.31 mm/yr decline. (2) Drought was mild in the northeast and severe in the southwest. The 0.05° GWSA first detected relief in approximately 4 % of the region (e.g., northern Beijing and northwestern Cangzhou). Fifteen drought events were identified during the study period, showing increasing frequency and intensity. The GGDI rebounded in 2016 and 2021 due to rainfall, the South-to-North Water Diversion, and groundwater-withdrawal restrictions. (3) Groundwater extraction was the dominant driver (contributing 45 %), and its interaction with precipitation markedly amplified drought risk (q = 0.83), particularly in fine-grained aquifer areas.</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"63 ","pages":"Article 103103"},"PeriodicalIF":5.0,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Integrated analysis of groundwater storage dynamics and drought migration in the Tarim River Basin 塔里木河流域地下水蓄水动态与干旱迁移的综合分析

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies

Pub Date : 2026-01-07 DOI: 10.1016/j.ejrh.2025.103099

Jing Wang , Guotao Dong , Hongbo Ling , Na Tang , Zhe Gao

Study region

The Tarim River Basin (TRB), an arid inland river basin in northwestern China.

Study focus

This study investigates groundwater drought dynamics in the TRB from 2002 to 2024 using groundwater storage anomalies (GWSA). By integrating GRACE total water storage, GLDAS soil moisture, and JRC surface water datasets, a Groundwater Drought Index (GGDI) was developed to analyze spatiotemporal drought evolution, long-term trends, seasonal variability, and drought centroid migration under combined climatic and anthropogenic influences.

New hydrological insights for the region

Results reveal a significant long-term groundwater decline of −8.6 mm yr⁻¹ (p < 0.01), equivalent to a cumulative loss of 198 mm (23.4 billion m³), indicating persistent imbalance between recharge and abstraction, particularly in irrigated midstream and downstream areas. Over 65 % of the basin experienced moderate to extreme groundwater drought for more than a decade. Spring and winter show heightened vulnerability due to early irrigation withdrawals, frozen-soil constraints, and delayed recharge. A southeastward shift of drought centroids is associated with glacier retreat and increasing groundwater extraction, reflecting reorganization of recharge–discharge processes. The GGDI proves effective for groundwater drought assessment and water management in data-scarce arid regions.

研究区域塔里木河流域是中国西北干旱内陆河流域。利用地下水储量异常（GWSA）对2002 - 2024年青藏高原地下水干旱动态进行了研究。综合GRACE总储水量、GLDAS土壤水分和JRC地表水数据，建立了地下水干旱指数（GGDI），分析了气候和人为影响下的干旱时空演变、长期趋势、季节变化和干旱质心迁移。结果显示，地下水长期显著下降- 8.6 mm yr⁻¹ （p <； 0.01），相当于累积损失198 mm（234亿m³），表明补给和抽取之间持续不平衡，特别是在灌溉中游和下游地区。超过65% %的流域经历了十多年的中度至极端地下水干旱。春季和冬季由于灌溉提前回灌、冻土约束和补给延迟，脆弱性增加。干旱质心向东南移动与冰川退缩和地下水开采增加有关，反映了补给-排放过程的重组。在数据匮乏的干旱区，GGDI对地下水干旱评价和水资源管理是有效的。

{"title":"Integrated analysis of groundwater storage dynamics and drought migration in the Tarim River Basin","authors":"Jing Wang , Guotao Dong , Hongbo Ling , Na Tang , Zhe Gao","doi":"10.1016/j.ejrh.2025.103099","DOIUrl":"10.1016/j.ejrh.2025.103099","url":null,"abstract":"<div><h3>Study region</h3><div>The Tarim River Basin (TRB), an arid inland river basin in northwestern China.</div></div><div><h3>Study focus</h3><div>This study investigates groundwater drought dynamics in the TRB from 2002 to 2024 using groundwater storage anomalies (GWSA). By integrating GRACE total water storage, GLDAS soil moisture, and JRC surface water datasets, a Groundwater Drought Index (GGDI) was developed to analyze spatiotemporal drought evolution, long-term trends, seasonal variability, and drought centroid migration under combined climatic and anthropogenic influences.</div></div><div><h3>New hydrological insights for the region</h3><div>Results reveal a significant long-term groundwater decline of −8.6 mm yr⁻¹ (p < 0.01), equivalent to a cumulative loss of 198 mm (23.4 billion m³), indicating persistent imbalance between recharge and abstraction, particularly in irrigated midstream and downstream areas. Over 65 % of the basin experienced moderate to extreme groundwater drought for more than a decade. Spring and winter show heightened vulnerability due to early irrigation withdrawals, frozen-soil constraints, and delayed recharge. A southeastward shift of drought centroids is associated with glacier retreat and increasing groundwater extraction, reflecting reorganization of recharge–discharge processes. The GGDI proves effective for groundwater drought assessment and water management in data-scarce arid regions.</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"63 ","pages":"Article 103099"},"PeriodicalIF":5.0,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926557","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The May 2023 flood in Emilia-Romagna and 50-year trends in extreme precipitation based on ERA5-Land 基于ERA5-Land的2023年5月艾米利亚-罗马涅洪水与极端降水50年趋势

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies

Pub Date : 2026-01-07 DOI: 10.1016/j.ejrh.2025.103070

Francesco Di Paola, Domenico Cimini, Maria Pia De Natale, Donatello Gallucci, Sabrina Gentile, Edoardo Geraldi, Salvatore Larosa, Saverio Teodosio Nilo, Elisabetta Ricciardelli, Mariassunta Viggiano, Michele Volini, Filomena Romano

Study region

Emilia-Romagna, northern Italy.

Study focus

This study analyses the extreme precipitation events of May 2023 in Emilia-Romagna, which caused fatalities and widespread damage, including flooding and landslides, with economic losses exceeding €8 billion. These events are characterised using rain gauge observations together with ERA5 and ERA5-Land reanalyses, and long-term (1974–2023) precipitation trends are derived from ERA5-Land to provide a multi-decadal context.

New hydrological insights for the region

At the scale of individual ERA5-Land grid points, over the 50-year reference period, statistical evidence indicates mean increases of 5 % for accumulations lasting up to 2 h and of 10 % for 1-h extreme accumulations, the latter consistent with the thermodynamic Clausius–Clapeyron rate given a concurrent 2-m air-temperature increase of 1.3 °C. In terms of frequency, statistical evidence indicates a 12 % reduction for 1-h accumulations and an average increase of 26 % for extreme accumulations with durations up to 3 h. The analysis also reveals a reduction in wet hours and a more fragmented distribution of precipitation in time and space. For events defined as extreme in terms of regional spatiotemporal extent, duration, persistence and accumulated precipitation, the mean values of these characteristics decrease by 16 % without significant changes in frequency, whereas events defined as extreme only in terms of mean precipitation increase by 9 % in magnitude and 61 % in frequency.

研究区域：意大利北部的艾米利亚-罗马涅。本研究分析了2023年5月艾米利亚-罗马涅的极端降水事件，该事件造成了包括洪水和山体滑坡在内的人员死亡和广泛的破坏，经济损失超过80亿欧元。利用雨量计观测以及ERA5和ERA5- land再分析来描述这些事件的特征，并从ERA5- land得到长期（1974-2023）降水趋势，以提供多年代际背景。在单个ERA5-Land网格点的尺度上，统计证据表明，在50年的参考期内，持续2 h的累积平均增加5 %，1 h的极端累积平均增加10 %，后者与热力学克劳usius - clapeyron速率一致，同时气温升高2 m，为1.3°C。在频率方面，统计证据表明，1小时累积减少了12 %，持续时间高达3 小时的极端累积平均增加了26 %。分析还显示，降雨时数减少，降水在时间和空间上的分布更加碎片化。以区域时空范围、持续时间、持续时间和累积降水量定义为极端的事件，这些特征的平均值降低了16% %，但频率没有显著变化，而仅以平均降水量定义为极端的事件，其强度增加了9% %，频率增加了61 %。

{"title":"The May 2023 flood in Emilia-Romagna and 50-year trends in extreme precipitation based on ERA5-Land","authors":"Francesco Di Paola, Domenico Cimini, Maria Pia De Natale, Donatello Gallucci, Sabrina Gentile, Edoardo Geraldi, Salvatore Larosa, Saverio Teodosio Nilo, Elisabetta Ricciardelli, Mariassunta Viggiano, Michele Volini, Filomena Romano","doi":"10.1016/j.ejrh.2025.103070","DOIUrl":"10.1016/j.ejrh.2025.103070","url":null,"abstract":"<div><h3>Study region</h3><div>Emilia-Romagna, northern Italy.</div></div><div><h3>Study focus</h3><div>This study analyses the extreme precipitation events of May 2023 in Emilia-Romagna, which caused fatalities and widespread damage, including flooding and landslides, with economic losses exceeding €8 billion. These events are characterised using rain gauge observations together with ERA5 and ERA5-Land reanalyses, and long-term (1974–2023) precipitation trends are derived from ERA5-Land to provide a multi-decadal context.</div></div><div><h3>New hydrological insights for the region</h3><div>At the scale of individual ERA5-Land grid points, over the 50-year reference period, statistical evidence indicates mean increases of 5 % for accumulations lasting up to 2 h and of 10 % for 1-h extreme accumulations, the latter consistent with the thermodynamic Clausius–Clapeyron rate given a concurrent 2-m air-temperature increase of 1.3 °C. In terms of frequency, statistical evidence indicates a 12 % reduction for 1-h accumulations and an average increase of 26 % for extreme accumulations with durations up to 3 h. The analysis also reveals a reduction in wet hours and a more fragmented distribution of precipitation in time and space. For events defined as extreme in terms of regional spatiotemporal extent, duration, persistence and accumulated precipitation, the mean values of these characteristics decrease by 16 % without significant changes in frequency, whereas events defined as extreme only in terms of mean precipitation increase by 9 % in magnitude and 61 % in frequency.</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"63 ","pages":"Article 103070"},"PeriodicalIF":5.0,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926518","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A hybrid physical-machine learning approach with uncertainty propagation for quantifying the Gorgan Bay-Caspian Sea hydrodynamics 一种具有不确定性传播的物理-机器混合学习方法用于量化戈尔根湾-里海流体动力学

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies

Pub Date : 2026-01-06 DOI: 10.1016/j.ejrh.2025.103086

Zahra Kazempour, Mohammad Danesh-Yazdi

Study region

The Gorgan Bay (GB) biosphere reserve is facing endangerment due to sustained water storage and quality loss. GB is hydraulically connected to the Caspian Sea, where the bay-sea flow exchange is key for protecting GB’s ecosystems from environmental threats.

Study focus

Generally, quantification of bay-sea hydrological exchange is challenging due to limited measurements or estimation uncertainty. In this study, we combined physics, machine learning, and a forward uncertainty propagation approach to quantity bay-sea interaction and the associated uncertainty.

New hydrological insights

Bathymetric modeling using Landsat’s visible bands within an ANN architecture achieved testing accuracy of R² = 0.84 and RMSE = 4.2 cm. Physical modeling of fluxes in GB revealed that monthly flow reverses to bay-to-sea during October-February due to low Caspian Sea/GB levels which enable river currents to reverse flow in the bay channels. Strong correlation between cumulative GB’s volume change and the Caspian Sea’s water level (r = 0.9) further confirmed sea-level fluctuations as the primary external control. Finally, results revealed that bay-sea flow exchange exhibits a statistically significant decreasing trend between October 1998- September 1999 and October 2023-September 2024. Critically, the estimation uncertainty remains high (CV = 2.7), originating from uncertainty in the storage volume estimation (i.e., ±50 cm error in water level and ±10 cm error sedimentation rate). The findings demonstrated that recession of the Caspian Sea and channel sedimentation threaten the sustainability of GB.

研究区高根湾（Gorgan Bay， GB）生物圈保护区因持续蓄水和水质下降而面临濒危。巴林通过水力与里海相连，在里海湾海水流交换是保护巴林生态系统免受环境威胁的关键。一般来说，由于测量的限制或估算的不确定性，海湾-海洋水文交换的量化具有挑战性。在这项研究中，我们结合了物理、机器学习和前向不确定性传播方法来研究海湾-海洋相互作用及其相关的不确定性。新的水文见解在人工神经网络架构中使用Landsat可见波段的水深建模实现了R²= 0.84和RMSE = 4.2 cm的测试精度。GB通量的物理模拟显示，在10月至2月期间，由于里海/GB水位较低，使河流在海湾通道中逆流，每月的流量向海湾-海洋方向逆转。累积GB体积变化与里海水位之间的强相关性（r = 0.9）进一步证实了海平面波动是主要的外部控制因素。结果表明，1998年10月~ 1999年9月和2023年10月~ 2024年9月，海湾-海流交换呈统计学上显著的减少趋势。至关重要的是，估算的不确定性仍然很高（CV = 2.7），这源于存储体积估算的不确定性（即水位误差±50 cm和沉降率误差±10 cm）。研究结果表明，里海的退缩和航道沉积威胁着里海的可持续性。

{"title":"A hybrid physical-machine learning approach with uncertainty propagation for quantifying the Gorgan Bay-Caspian Sea hydrodynamics","authors":"Zahra Kazempour, Mohammad Danesh-Yazdi","doi":"10.1016/j.ejrh.2025.103086","DOIUrl":"10.1016/j.ejrh.2025.103086","url":null,"abstract":"<div><h3>Study region</h3><div>The Gorgan Bay (GB) biosphere reserve is facing endangerment due to sustained water storage and quality loss. GB is hydraulically connected to the Caspian Sea, where the bay-sea flow exchange is key for protecting GB’s ecosystems from environmental threats.</div></div><div><h3>Study focus</h3><div>Generally, quantification of bay-sea hydrological exchange is challenging due to limited measurements or estimation uncertainty. In this study, we combined physics, machine learning, and a forward uncertainty propagation approach to quantity bay-sea interaction and the associated uncertainty.</div></div><div><h3>New hydrological insights</h3><div>Bathymetric modeling using Landsat’s visible bands within an ANN architecture achieved testing accuracy of R² = 0.84 and RMSE = 4.2 cm. Physical modeling of fluxes in GB revealed that monthly flow reverses to bay-to-sea during October-February due to low Caspian Sea/GB levels which enable river currents to reverse flow in the bay channels. Strong correlation between cumulative GB’s volume change and the Caspian Sea’s water level (r = 0.9) further confirmed sea-level fluctuations as the primary external control. Finally, results revealed that bay-sea flow exchange exhibits a statistically significant decreasing trend between October 1998- September 1999 and October 2023-September 2024. Critically, the estimation uncertainty remains high (CV = 2.7), originating from uncertainty in the storage volume estimation (i.e., ±50 cm error in water level and ±10 cm error sedimentation rate). The findings demonstrated that recession of the Caspian Sea and channel sedimentation threaten the sustainability of GB.</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"63 ","pages":"Article 103086"},"PeriodicalIF":5.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0