Environmental Modelling & Software最新文献_第2页

A python framework for differentiable hydrological modeling and research workflow automation 可微分水文建模和研究工作流自动化的Python框架

IF 4.6 2区环境科学与生态学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Environmental Modelling & Software

Pub Date : 2026-03-01 Epub Date: 2026-01-24 DOI: 10.1016/j.envsoft.2026.106895

Wenyu Ouyang , Shuolong Xu , Yikai Chai , Laihong Zhuang , Zhihong Liu , Lei Ye , Xinzhuo Wu , Yong Peng , Chi Zhang

This study introduces a Python-based framework for constructing differentiable hydrological models with a modular design to streamline research workflows. The framework integrates five key modules: hydrodataset and hydrodatasource for data preprocessing, hydromodel and torchhydro for traditional and differentiable modeling, and HydroDHM for orchestrating integrated workflows. The data modules automate preparation of diverse datasets, including open-access and proprietary resources. Hydromodel supports process-based model calibration and evaluation, while torchhydro enables neural network integration for differentiable models. HydroDHM coordinates these components through a unified interface for configuring and executing end-to-end modeling pipelines. Case studies in CAMELS basins demonstrate that differentiable models achieve comparable streamflow simulation performance to traditional approaches. By decoupling data handling from model development and providing uv-installable (and pip-compatible) modules, the framework ensures reproducibility, scalability, and adaptability across diverse hydrological contexts.

本研究介绍了一个基于python的框架，用于构建具有模块化设计的可微分水文模型，以简化研究工作流程。该框架集成了五个关键模块：用于数据预处理的hydrodataset和hydrodatasource，用于传统和可微分建模的hydromodel和torchhydro，以及用于编排集成工作流的HydroDHM。数据模块自动准备各种数据集，包括开放访问和专有资源。Hydromodel支持基于过程的模型校准和评估，而torchhydro支持可微分模型的神经网络集成。HydroDHM通过一个统一的接口来协调这些组件，用于配置和执行端到端建模管道。骆驼盆地的案例研究表明，可微分模型与传统方法相比具有相当的流量模拟性能。通过将数据处理与模型开发分离，并提供uv安装（和pip兼容）模块，该框架确保了在不同水文环境下的再现性、可扩展性和适应性。

{"title":"A python framework for differentiable hydrological modeling and research workflow automation","authors":"Wenyu Ouyang , Shuolong Xu , Yikai Chai , Laihong Zhuang , Zhihong Liu , Lei Ye , Xinzhuo Wu , Yong Peng , Chi Zhang","doi":"10.1016/j.envsoft.2026.106895","DOIUrl":"10.1016/j.envsoft.2026.106895","url":null,"abstract":"<div><div>This study introduces a Python-based framework for constructing differentiable hydrological models with a modular design to streamline research workflows. The framework integrates five key modules: hydrodataset and hydrodatasource for data preprocessing, hydromodel and torchhydro for traditional and differentiable modeling, and HydroDHM for orchestrating integrated workflows. The data modules automate preparation of diverse datasets, including open-access and proprietary resources. Hydromodel supports process-based model calibration and evaluation, while torchhydro enables neural network integration for differentiable models. HydroDHM coordinates these components through a unified interface for configuring and executing end-to-end modeling pipelines. Case studies in CAMELS basins demonstrate that differentiable models achieve comparable streamflow simulation performance to traditional approaches. By decoupling data handling from model development and providing uv-installable (and pip-compatible) modules, the framework ensures reproducibility, scalability, and adaptability across diverse hydrological contexts.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"198 ","pages":"Article 106895"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146047987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Modeling hydrologic response to wildfires in the Pacific Northwest with a modified calibration technique 用改进的校准技术模拟太平洋西北地区野火的水文响应

IF 4.6 2区环境科学与生态学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Environmental Modelling & Software

Pub Date : 2026-03-01 Epub Date: 2026-01-24 DOI: 10.1016/j.envsoft.2026.106896

Hyunwoo Kang , Cameron E. Naficy , Kevin D. Bladon

The 2020 Labor Day fires in the Western Cascades of Oregon, USA, burned extensive forested areas, which altered hydrologic processes, water quality, aquatic ecosystems, and drinking water resources. Understanding wildfire severity effects on hydrologic processes is crucial for improved water resource management. Our study assessed wildfire severity impacts on hydrology using a modified calibration method for the Soil and Water Assessment Tool (SWAT) model. Calibration incorporated evapotranspiration and leaf area index to represent vegetation loss and hydrologic impacts. We also integrated a wildfire module to simulate fire effects on soil and vegetation parameters. This improved modeling approach effectively captured post-fire hydrologic behavior, especially increased high streamflows and reduced evapotranspiration, with greater changes linked to higher burn severity. These findings emphasize the importance of considering fire severity in hydrologic modeling, aiding proactive management and mitigation strategies to protect water supply and enhance ecosystem resilience in wildfire-prone regions.

2020年劳动节，美国俄勒冈州西部瀑布发生火灾，烧毁了大片森林地区，改变了水文过程、水质、水生生态系统和饮用水资源。了解野火严重程度对水文过程的影响对于改善水资源管理至关重要。本研究使用改良的土壤和水评估工具（SWAT）模型校准方法评估了野火严重程度对水文的影响。校正采用蒸散和叶面积指数来表示植被损失和水文影响。我们还集成了一个野火模块来模拟火灾对土壤和植被参数的影响。这种改进的建模方法有效地捕获了火灾后的水文行为，特别是增加的高流量和减少的蒸散量，这些变化与更高的烧伤严重程度有关。这些发现强调了在水文建模中考虑火灾严重程度的重要性，有助于主动管理和缓解战略，以保护供水，增强野火易发地区的生态系统恢复能力。

{"title":"Modeling hydrologic response to wildfires in the Pacific Northwest with a modified calibration technique","authors":"Hyunwoo Kang , Cameron E. Naficy , Kevin D. Bladon","doi":"10.1016/j.envsoft.2026.106896","DOIUrl":"10.1016/j.envsoft.2026.106896","url":null,"abstract":"<div><div>The 2020 Labor Day fires in the Western Cascades of Oregon, USA, burned extensive forested areas, which altered hydrologic processes, water quality, aquatic ecosystems, and drinking water resources. Understanding wildfire severity effects on hydrologic processes is crucial for improved water resource management. Our study assessed wildfire severity impacts on hydrology using a modified calibration method for the Soil and Water Assessment Tool (SWAT) model. Calibration incorporated evapotranspiration and leaf area index to represent vegetation loss and hydrologic impacts. We also integrated a wildfire module to simulate fire effects on soil and vegetation parameters. This improved modeling approach effectively captured post-fire hydrologic behavior, especially increased high streamflows and reduced evapotranspiration, with greater changes linked to higher burn severity. These findings emphasize the importance of considering fire severity in hydrologic modeling, aiding proactive management and mitigation strategies to protect water supply and enhance ecosystem resilience in wildfire-prone regions.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"198 ","pages":"Article 106896"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146047988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A stepwise back-correction function for precipitation representation in hydrologic models 水文模型降水表示的逐步反校正函数

IF 4.6 2区环境科学与生态学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Environmental Modelling & Software

Pub Date : 2026-03-01 Epub Date: 2026-02-10 DOI: 10.1016/j.envsoft.2026.106908

Dany A. Hernandez , Jorge A. Guzman , Sandra R. Villamizar , Maria L. Chu , Camila Ribeiro , Carlos R. de Mello

This study addresses how spatial and temporal uncertainties in precipitation limit calibration of hydrological models. Adjusting model parameters alone cannot compensate for poorly represented precipitation at the model's lower resolution. A reanalysis framework that integrates traditional calibration with a stepwise precipitation back correction approach was introduced. Using a composite exponential error function, the method derives precipitation correction factors from mismatches between observed and simulated streamflow. The approach was tested with three hydrological models—SWAT, MIKE-SHE, and MHD—across watersheds in the United States and Brazil. The workflow involved an initial standard calibration, followed by iterative precipitation correction without altering model parameters, and a final recalibration incorporating the corrected precipitation. Results showed 10–18% improvements in KGE while maintaining PBIAS below 10% at most stations. The study highlights the value of constraining water balance to avoid unrealistic corrections and demonstrates how addressing precipitation uncertainties enhances model performance across diverse hydrological settings.

本研究解决了降水的时空不确定性如何限制水文模型的校准。仅调整模式参数不能补偿模式分辨率较低时表现不佳的降水。介绍了一种将传统定标与逐级降水反演方法相结合的再分析框架。该方法利用复合指数误差函数，从实测与模拟流量的不匹配中导出降水校正因子。该方法在美国和巴西的流域用swat、MIKE-SHE和mhd三种水文模型进行了测试。工作流程包括最初的标准校准，随后是不改变模型参数的迭代降水校正，以及结合校正后的降水的最终重新校准。结果显示，大多数站点的KGE改善了10-18%，而PBIAS保持在10%以下。该研究强调了限制水平衡以避免不切实际的修正的价值，并展示了解决降水不确定性如何提高模型在不同水文设置中的性能。

{"title":"A stepwise back-correction function for precipitation representation in hydrologic models","authors":"Dany A. Hernandez , Jorge A. Guzman , Sandra R. Villamizar , Maria L. Chu , Camila Ribeiro , Carlos R. de Mello","doi":"10.1016/j.envsoft.2026.106908","DOIUrl":"10.1016/j.envsoft.2026.106908","url":null,"abstract":"<div><div>This study addresses how spatial and temporal uncertainties in precipitation limit calibration of hydrological models. Adjusting model parameters alone cannot compensate for poorly represented precipitation at the model's lower resolution. A reanalysis framework that integrates traditional calibration with a stepwise precipitation back correction approach was introduced. Using a composite exponential error function, the method derives precipitation correction factors from mismatches between observed and simulated streamflow. The approach was tested with three hydrological models—SWAT, MIKE-SHE, and MHD—across watersheds in the United States and Brazil. The workflow involved an initial standard calibration, followed by iterative precipitation correction without altering model parameters, and a final recalibration incorporating the corrected precipitation. Results showed 10–18% improvements in KGE while maintaining PBIAS below 10% at most stations. The study highlights the value of constraining water balance to avoid unrealistic corrections and demonstrates how addressing precipitation uncertainties enhances model performance across diverse hydrological settings.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"198 ","pages":"Article 106908"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146152965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The importance of system interactions in hydrodynamic models of parts of complex interconnected deltas 系统相互作用在复杂互联三角洲部分水动力模型中的重要性

IF 4.6 2区环境科学与生态学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Environmental Modelling & Software

Pub Date : 2026-03-01 Epub Date: 2026-01-06 DOI: 10.1016/j.envsoft.2025.106838

Niels M. Welsch, Jord J. Warmink, Suzanne J.M.H. Hulscher, Denie C.M. Augustijn

Climate change affects river deltas worldwide. Hydrodynamic models are used to study these effects. However, choosing the spatial scale and boundary conditions for these models is complex due to interconnectivity within river deltas. We study how boundary conditions of a model covering only part of such systems are impacted by changing conditions outside of the domain. We couple different components of the Dutch river delta into a model covering the complete delta, and force it with a range of river discharges and sea levels. Results show that the impact depends on the distance to the boundaries, as well as the relative (upstream) discharge in the considered rivers. As these differences are found to propagate far upstream, these findings underline the importance of choosing appropriate downstream boundaries when modelling water levels in parts of interconnected systems influenced by changing conditions outside the modelled domain (e.g. sea level rise or changing hydrographs).

气候变化影响着全世界的河流三角洲。水动力模型用于研究这些效应。然而，由于河流三角洲内部的相互联系，这些模型的空间尺度和边界条件的选择是复杂的。我们研究了仅覆盖此类系统的一部分的模型的边界条件如何受到域外变化条件的影响。我们将荷兰河三角洲的不同组成部分结合成一个覆盖整个三角洲的模型，并将其与一系列河流流量和海平面相结合。结果表明，影响取决于与边界的距离，以及所考虑河流的相对（上游）流量。由于发现这些差异向上游传播得很远，这些发现强调了在对受模拟域外变化条件（例如海平面上升或变化的水文曲线）影响的互联系统部分的水位进行建模时选择适当的下游边界的重要性。

引用次数: 0

Advancing water level prediction using clustering-based machine learning techniques in data-scarce regions 在数据稀缺地区使用基于聚类的机器学习技术推进水位预测

IF 4.6 2区环境科学与生态学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Environmental Modelling & Software

Pub Date : 2026-03-01 Epub Date: 2026-01-27 DOI: 10.1016/j.envsoft.2026.106899

SangHyun Lee, Taeil Jang

Accurate and scalable water level forecasting is essential for effective water resources management, particularly in regions with limited long-term records. We present a clustering-based framework for one- and three-day-ahead water level prediction in the Saemangeum Watershed, South Korea. Twenty-five monitoring stations were grouped into six hydrologically similar clusters using k-means clustering with wavelet-entropy features. Within each cluster, multilayer perceptron (MLP) models were trained using two strategies: (1) training only at the centroid station and (2) training at the station with the longest record in each cluster. The longest-record strategy showed strong agreement with observations, achieving mean Nash–Sutcliffe efficiency and root-mean-square error values of 0.97 and 0.06 for one-day-ahead forecasts, and 0.83 and 0.14 for three-day-ahead forecasts across all stations. By training one MLP per cluster and transferring it to all member stations, the framework reduces computational cost and provides a practical solution for large-scale water level forecasting in data-scarce environments.

准确和可扩展的水位预报对于有效的水资源管理至关重要，特别是在长期记录有限的地区。我们提出了一个基于聚类的框架，用于预测韩国新万金流域1天和3天前的水位。采用具有小波熵特征的k-均值聚类方法将25个监测站划分为6个水文相似的聚类。在每个聚类中，多层感知器（MLP）模型使用两种策略进行训练：(1)仅在质心站进行训练；(2)在每个聚类中记录最长的站进行训练。记录时间最长的策略与观测结果表现出很强的一致性，在所有台站中，提前一天预报的平均纳什-萨特克利夫效率和均方根误差分别为0.97和0.06，提前三天预报的误差分别为0.83和0.14。通过每个集群训练一个MLP并将其传递到所有成员站，该框架降低了计算成本，并为数据稀缺环境下的大规模水位预测提供了实用的解决方案。

{"title":"Advancing water level prediction using clustering-based machine learning techniques in data-scarce regions","authors":"SangHyun Lee, Taeil Jang","doi":"10.1016/j.envsoft.2026.106899","DOIUrl":"10.1016/j.envsoft.2026.106899","url":null,"abstract":"<div><div>Accurate and scalable water level forecasting is essential for effective water resources management, particularly in regions with limited long-term records. We present a clustering-based framework for one- and three-day-ahead water level prediction in the Saemangeum Watershed, South Korea. Twenty-five monitoring stations were grouped into six hydrologically similar clusters using k-means clustering with wavelet-entropy features. Within each cluster, multilayer perceptron (MLP) models were trained using two strategies: (1) training only at the centroid station and (2) training at the station with the longest record in each cluster. The longest-record strategy showed strong agreement with observations, achieving mean Nash–Sutcliffe efficiency and root-mean-square error values of 0.97 and 0.06 for one-day-ahead forecasts, and 0.83 and 0.14 for three-day-ahead forecasts across all stations. By training one MLP per cluster and transferring it to all member stations, the framework reduces computational cost and provides a practical solution for large-scale water level forecasting in data-scarce environments.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"198 ","pages":"Article 106899"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SPAR-TC: A framework for accounting spatial representativeness in triple collocation SPAR-TC：三重搭配中空间代表性的核算框架

IF 4.6 2区环境科学与生态学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Environmental Modelling & Software

Pub Date : 2026-03-01 Epub Date: 2026-01-12 DOI: 10.1016/j.envsoft.2026.106874

Diksha Gupta, C.T. Dhanya

Triple collocation (TC) has been widely used to overcome the rarity of “ground truth” in geophysical measurements. While TC assumes all systems observe the same underlying geophysical variable, it does not inherently correct for spatial representativeness errors due to different spatial measurement systems. To address this, we propose the Spatially Representative Triple Collocation (SPAR-TC), which accounts for the spatial variability of the “ground truth” across different spatial scales. A synthetic soil moisture experiment assessed SPAR-TC sensitivity to spatial heterogeneity and sample size, followed by a real-world application with remotely sensed precipitation data. Results showed that SPAR-TC provides more reliable estimates of “true” error variance compared with traditional TC, especially in spatially heterogeneous regions. Both methods yield comparable dataset rankings; however, SPAR-TC provides error variance estimates more consistent with ground-based observations. Hence, SPAR-TC offers robust framework for addressing spatial representativeness errors and improves error quantification for datasets with differing spatial support.

为了克服地球物理测量中“地面真值”的稀缺性，三重配置（TC）被广泛应用。虽然TC假设所有系统都观测到相同的潜在地球物理变量，但它并不能固有地纠正由于不同空间测量系统而导致的空间代表性误差。为了解决这个问题，我们提出了空间代表性三重搭配（SPAR-TC），它解释了“地面真值”在不同空间尺度上的空间变异性。综合土壤湿度试验评估了SPAR-TC对空间异质性和样本量的敏感性，随后进行了遥感降水数据的实际应用。结果表明，SPAR-TC比传统TC提供了更可靠的“真实”误差方差估计，特别是在空间异质性区域。两种方法都会产生可比较的数据集排名；然而，SPAR-TC提供的误差方差估计值与地面观测值更为一致。因此，SPAR-TC为解决空间代表性误差提供了强大的框架，并改进了具有不同空间支持的数据集的误差量化。

{"title":"SPAR-TC: A framework for accounting spatial representativeness in triple collocation","authors":"Diksha Gupta, C.T. Dhanya","doi":"10.1016/j.envsoft.2026.106874","DOIUrl":"10.1016/j.envsoft.2026.106874","url":null,"abstract":"<div><div>Triple collocation (TC) has been widely used to overcome the rarity of “ground truth” in geophysical measurements. While TC assumes all systems observe the same underlying geophysical variable, it does not inherently correct for spatial representativeness errors due to different spatial measurement systems. To address this, we propose the Spatially Representative Triple Collocation (SPAR-TC), which accounts for the spatial variability of the “ground truth” across different spatial scales. A synthetic soil moisture experiment assessed SPAR-TC sensitivity to spatial heterogeneity and sample size, followed by a real-world application with remotely sensed precipitation data. Results showed that SPAR-TC provides more reliable estimates of “true” error variance compared with traditional TC, especially in spatially heterogeneous regions. Both methods yield comparable dataset rankings; however, SPAR-TC provides error variance estimates more consistent with ground-based observations. Hence, SPAR-TC offers robust framework for addressing spatial representativeness errors and improves error quantification for datasets with differing spatial support.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"198 ","pages":"Article 106874"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956764","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ADAPT: A novel IoT-driven analytical data assimilation method based on phase-space tuning for long-sequence water quality forecasting ADAPT：一种基于相空间调整的物联网驱动的分析数据同化方法，用于长序列水质预测

IF 4.6 2区环境科学与生态学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Environmental Modelling & Software

Pub Date : 2026-03-01 Epub Date: 2026-01-20 DOI: 10.1016/j.envsoft.2026.106882

Mingzhuang Sun , Zhili Li , Guangtao Fu , Haifeng Jia

Water quality models often suffer from performance degradation due to parameter obsolescence caused by external environmental changes. To address this, this study proposes a novel framework named Analytical Data Assimilation via Phase-space Tuning (ADAPT). Unlike traditional data assimilation methods that directly update state variables, ADAPT dynamically calibrates model parameters by establishing a robust link between parameters and water quality dynamics using Aquaformer, a Transformer-based deep learning model driven by phase-space reconstruction. The method was validated through digital-twin and real-world experiments on the Diannong River, China. Results demonstrate that ADAPT significantly outperforms the Ensemble Kalman Filter, reducing prediction errors by 36.26 % at monitored sites and 54.66 % at unmonitored sites. ADAPT exhibits superior transferability and stable error control, effectively overcoming the limitations of traditional methods in spatial generalization. This study provides a reliable, physics-informed solution for high-frequency auto-calibration in smart water management systems.

由于外部环境变化引起的参数过时，水质模型的性能往往会下降。为了解决这个问题，本研究提出了一个新的框架，名为通过相空间调谐的分析数据同化（ADAPT）。与直接更新状态变量的传统数据同化方法不同，ADAPT通过使用Aquaformer（一种基于transformer的深度学习模型，由相空间重建驱动）在参数和水质动态之间建立鲁棒联系，从而动态校准模型参数。该方法通过数字孪生和实际实验在中国滇农河进行了验证。结果表明，ADAPT显著优于Ensemble Kalman Filter，在监测站点和非监测站点的预测误差分别降低了36.26%和54.66%。ADAPT具有良好的可转移性和稳定的误差控制，有效克服了传统空间泛化方法的局限性。该研究为智能水管理系统中的高频自动校准提供了可靠的物理信息解决方案。

{"title":"ADAPT: A novel IoT-driven analytical data assimilation method based on phase-space tuning for long-sequence water quality forecasting","authors":"Mingzhuang Sun , Zhili Li , Guangtao Fu , Haifeng Jia","doi":"10.1016/j.envsoft.2026.106882","DOIUrl":"10.1016/j.envsoft.2026.106882","url":null,"abstract":"<div><div>Water quality models often suffer from performance degradation due to parameter obsolescence caused by external environmental changes. To address this, this study proposes a novel framework named Analytical Data Assimilation via Phase-space Tuning (ADAPT). Unlike traditional data assimilation methods that directly update state variables, ADAPT dynamically calibrates model parameters by establishing a robust link between parameters and water quality dynamics using Aquaformer, a Transformer-based deep learning model driven by phase-space reconstruction. The method was validated through digital-twin and real-world experiments on the Diannong River, China. Results demonstrate that ADAPT significantly outperforms the Ensemble Kalman Filter, reducing prediction errors by 36.26 % at monitored sites and 54.66 % at unmonitored sites. ADAPT exhibits superior transferability and stable error control, effectively overcoming the limitations of traditional methods in spatial generalization. This study provides a reliable, physics-informed solution for high-frequency auto-calibration in smart water management systems.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"198 ","pages":"Article 106882"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Climate Risk STAC: A living metadata catalog of geospatial data for climate risk assessments 气候风险STAC：用于气候风险评估的地理空间数据的活元数据目录

IF 4.6 2区环境科学与生态学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Environmental Modelling & Software

Pub Date : 2026-03-01 Epub Date: 2026-02-10 DOI: 10.1016/j.envsoft.2026.106906

Lena Reimann , Dirk Eilander , Timothy Tiggeloven , Milana Vuckovic , Matti Kummu , Andrea Vajda , Jeremy S. Pal , Maurizio Mazzoleni , Fredrik Wetterhall , Jeroen C.J.H. Aerts

Climate risks are increasing globally due to climate change, driven by intensifying climate hazards and changes in socioeconomic conditions that drive exposure and vulnerability. Climate Risk Assessments (CRAs) constitute a tool to understand such risks based on the analysis of geospatial datasets. However, CRA data are often scattered across different platforms, thereby inhibiting their Findability, Accessibility, Interoperability, and Reusability (FAIR). To make CRA data FAIR, we develop Climate Risk STAC, a living metadata catalog of open-access geospatial datasets that is hosted in a collaborative environment for continuous development. Climate Risk STAC (version 1.0) currently includes 214 metadata entries from nine different hazards, five types of exposed elements, and seven vulnerability categories. All data entries can be explored in a user-friendly browser which eases data selection. We encourage contributions of new datasets to maintain a growing, community-led catalog that reflects state-of-the-art CRA concepts and data.

由于气候变化，气候风险正在全球范围内增加，其驱动因素是气候灾害加剧以及社会经济条件的变化，这些变化导致了气候暴露和脆弱性。气候风险评估（CRAs）是一种基于地理空间数据集分析了解此类风险的工具。然而，CRA数据通常分散在不同的平台上，从而限制了它们的可查找性、可访问性、互操作性和可重用性（FAIR）。为了使CRA数据公平，我们开发了气候风险STAC，这是一个开放获取地理空间数据集的活元数据目录，托管在一个协作环境中，以供持续开发。气候风险STAC（1.0版本）目前包括214个元数据条目，涉及9种不同的危害、5种暴露元素和7种脆弱性类别。所有数据条目都可以在用户友好的浏览器中进行探索，从而简化数据选择。我们鼓励新数据集的贡献保持反映最先进的CRA概念和数据的一个增长的、社区领导的目录。

{"title":"Climate Risk STAC: A living metadata catalog of geospatial data for climate risk assessments","authors":"Lena Reimann , Dirk Eilander , Timothy Tiggeloven , Milana Vuckovic , Matti Kummu , Andrea Vajda , Jeremy S. Pal , Maurizio Mazzoleni , Fredrik Wetterhall , Jeroen C.J.H. Aerts","doi":"10.1016/j.envsoft.2026.106906","DOIUrl":"10.1016/j.envsoft.2026.106906","url":null,"abstract":"<div><div>Climate risks are increasing globally due to climate change, driven by intensifying climate hazards and changes in socioeconomic conditions that drive exposure and vulnerability. Climate Risk Assessments (CRAs) constitute a tool to understand such risks based on the analysis of geospatial datasets. However, CRA data are often scattered across different platforms, thereby inhibiting their Findability, Accessibility, Interoperability, and Reusability (FAIR). To make CRA data FAIR, we develop Climate Risk STAC, a living metadata catalog of open-access geospatial datasets that is hosted in a collaborative environment for continuous development. Climate Risk STAC (version 1.0) currently includes 214 metadata entries from nine different hazards, five types of exposed elements, and seven vulnerability categories. All data entries can be explored in a user-friendly browser which eases data selection. We encourage contributions of new datasets to maintain a growing, community-led catalog that reflects state-of-the-art CRA concepts and data.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"198 ","pages":"Article 106906"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146152964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Development of a web-based tool for rapid flood inundation modeling 开发基于网络的快速洪水淹没建模工具

IF 4.6 2区环境科学与生态学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Environmental Modelling & Software

Pub Date : 2026-03-01 Epub Date: 2026-01-13 DOI: 10.1016/j.envsoft.2026.106876

Dawei Xiao , Binjie Yuan , Zhengxu Guo , Wanhong Yang , Jingchao Jiang , Min Chen , Guonian Lv , Junzhi Liu

To address the growing risk of floods under global climate change, management agencies need flood inundation modeling to support decision-making and emergency response. However, traditional desktop-based modeling remains a complex and time-consuming process, making it difficult for users to perform rapid flood simulations. To overcome this limitation, this study developed a web-based rapid flood modeling tool based on the LISFLOOD-FP model. Each key step involved in the modeling process—such as data preparation, preprocessing, model run and calibration, and postprocessing— was encapsulated into an automated executable workflow. These workflows were deployed on servers, published as web services, and invoked from a web-based interface, significantly streamlining and simplifying the modeling process. Four flood events in the upper Missouri River Basin were successfully simulated to showcase the tool's capability. This user-friendly web-based tool enables users to conduct flood inundation modeling quickly, thereby lowering user barriers and facilitating timely flood risk mitigation.

为了应对全球气候变化下日益增长的洪水风险，管理机构需要洪水淹没建模来支持决策和应急响应。然而，传统的基于桌面的建模仍然是一个复杂且耗时的过程，使得用户难以执行快速的洪水模拟。为了克服这一局限性，本研究基于LISFLOOD-FP模型开发了基于web的快速洪水建模工具。建模过程中涉及的每个关键步骤（如数据准备、预处理、模型运行和校准以及后处理）都被封装到一个自动化的可执行工作流中。这些工作流部署在服务器上，作为web服务发布，并从基于web的接口调用，显著地简化了建模过程。成功地模拟了密苏里河上游流域的四次洪水事件，以展示该工具的能力。这个用户友好的基于web的工具使用户能够快速进行洪水淹没建模，从而降低用户的障碍，并促进及时减轻洪水风险。

{"title":"Development of a web-based tool for rapid flood inundation modeling","authors":"Dawei Xiao , Binjie Yuan , Zhengxu Guo , Wanhong Yang , Jingchao Jiang , Min Chen , Guonian Lv , Junzhi Liu","doi":"10.1016/j.envsoft.2026.106876","DOIUrl":"10.1016/j.envsoft.2026.106876","url":null,"abstract":"<div><div>To address the growing risk of floods under global climate change, management agencies need flood inundation modeling to support decision-making and emergency response. However, traditional desktop-based modeling remains a complex and time-consuming process, making it difficult for users to perform rapid flood simulations. To overcome this limitation, this study developed a web-based rapid flood modeling tool based on the LISFLOOD-FP model. Each key step involved in the modeling process—such as data preparation, preprocessing, model run and calibration, and postprocessing— was encapsulated into an automated executable workflow. These workflows were deployed on servers, published as web services, and invoked from a web-based interface, significantly streamlining and simplifying the modeling process. Four flood events in the upper Missouri River Basin were successfully simulated to showcase the tool's capability. This user-friendly web-based tool enables users to conduct flood inundation modeling quickly, thereby lowering user barriers and facilitating timely flood risk mitigation.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"198 ","pages":"Article 106876"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145961785","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel interpretable ozone forecasting approach based on deep learning with masked residual connections 一种基于深度学习的残差连接可解释臭氧预测方法

IF 4.6 2区环境科学与生态学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Environmental Modelling & Software

Pub Date : 2026-03-01 Epub Date: 2026-01-15 DOI: 10.1016/j.envsoft.2026.106878

P. Reina-Jiménez , M.J. Jiménez-Navarro , G. Asencio-Cortés , F. Martínez-Álvarez , M. Martínez-Ballesteros

Air pollution is a growing threat, especially in low- and middle-income countries, causing over 4 million premature deaths annually. Ground-level ozone is a major concern, demanding accurate and interpretable prediction systems for effective public health management. However, existing time-series forecasting methods struggle to capture both linear and nonlinear dependencies in atmospheric data. This study introduces ResSelNet, a novel Residual Selection Network that integrates masked residual connections and embedded feature selection within a unified deep learning architecture. The model dynamically determines the optimal processing depth for each feature, allowing linear relationships to bypass nonlinear transformations while capturing complex patterns when necessary. Applied to five monitoring stations across Andalusia (Spain), ResSelNet consistently outperformed state-of-the-art baselines, achieving 8%–12% lower RMSE and MAE than LSTM and Transformer models. Beyond accuracy, the framework improves interpretability and robustness, revealing the hierarchical relevance of meteorological and pollutant variables. ResSelNet therefore offers an effective and explainable solution for multi-horizon environmental time-series forecasting.

空气污染是一个日益严重的威胁，特别是在低收入和中等收入国家，每年造成400多万人过早死亡。地面臭氧是一个主要问题，需要准确和可解释的预测系统，以便进行有效的公共卫生管理。然而，现有的时间序列预报方法难以捕捉大气数据中的线性和非线性依赖关系。本研究介绍了一种新的残差选择网络ResSelNet，它在统一的深度学习架构中集成了屏蔽残差连接和嵌入式特征选择。该模型动态确定每个特征的最佳处理深度，允许线性关系绕过非线性转换，同时在必要时捕获复杂模式。ResSelNet应用于安达卢西亚（西班牙）的五个监测站，始终优于最先进的基线，与LSTM和Transformer模型相比，RMSE和MAE降低了8%-12%。除了准确性之外，该框架还提高了可解释性和鲁棒性，揭示了气象和污染物变量的层次相关性。因此，ResSelNet为多水平环境时间序列预测提供了一个有效且可解释的解决方案。

{"title":"A novel interpretable ozone forecasting approach based on deep learning with masked residual connections","authors":"P. Reina-Jiménez , M.J. Jiménez-Navarro , G. Asencio-Cortés , F. Martínez-Álvarez , M. Martínez-Ballesteros","doi":"10.1016/j.envsoft.2026.106878","DOIUrl":"10.1016/j.envsoft.2026.106878","url":null,"abstract":"<div><div>Air pollution is a growing threat, especially in low- and middle-income countries, causing over 4 million premature deaths annually. Ground-level ozone is a major concern, demanding accurate and interpretable prediction systems for effective public health management. However, existing time-series forecasting methods struggle to capture both linear and nonlinear dependencies in atmospheric data. This study introduces ResSelNet, a novel Residual Selection Network that integrates masked residual connections and embedded feature selection within a unified deep learning architecture. The model dynamically determines the optimal processing depth for each feature, allowing linear relationships to bypass nonlinear transformations while capturing complex patterns when necessary. Applied to five monitoring stations across Andalusia (Spain), ResSelNet consistently outperformed state-of-the-art baselines, achieving 8%–12% lower RMSE and MAE than LSTM and Transformer models. Beyond accuracy, the framework improves interpretability and robustness, revealing the hierarchical relevance of meteorological and pollutant variables. ResSelNet therefore offers an effective and explainable solution for multi-horizon environmental time-series forecasting.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"198 ","pages":"Article 106878"},"PeriodicalIF":4.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145995200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0