Environmental Modelling & Software最新文献

{"title":"The Bayesian operating characteristic curve for feature analysis applied to urban land cover change","authors":"Rodrigo Lopez-Farias ,&nbsp;S. Ivvan Valdez ,&nbsp;Carlos Lara-Alvarez ,&nbsp;Robert Gilmore Pontius Jr","doi":"10.1016/j.envsoft.2025.106688","DOIUrl":"10.1016/j.envsoft.2025.106688","url":null,"abstract":"<div><div>The <em>Total Operating Characteristic</em> curve (TOC) is a visual tool used to assess the performance of binary classifiers, introduced as an improvement over the <em>Receiver Operating Characteristic</em> (ROC) curve. The TOC provides a function that maps the number of <em>hits</em> plus <em>false alarms</em> (True Positives and False Positives) to <em>hits</em> (True Positives). Despite its broad adoption in model evaluation, especially for land use change studies, the TOC’s mathematical properties for explaining the probabilistic performance of classifiers and data distributions remain underexplored. To fill this gap, this article introduces the Bayesian Operating Characteristic (BOC) curve, a novel explanatory framework derived from the application of Bayes’ theorem and numerical analysis on the TOC curve. Such a framework establishes the computation of cumulative and density distribution functions that describe and identify non-linear relations between a rank variable and its binary outcome, specifying the rank’s interval with the maximal positive or negative impact in the classification. The proposal is validated by the analysis and identification of urban land change drivers, revealing different non-linear probabilistic relationships between rank variables and the binary outcome.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"194 ","pages":"Article 106688"},"PeriodicalIF":4.6,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145155450","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}

{"title":"From gauged to ungauged: Large-scale deep learning rainfall-runoff modelling for reliable streamflow estimation in India's diverse basins","authors":"Siddik Barbhuiya,&nbsp;Vivek Gupta","doi":"10.1016/j.envsoft.2025.106696","DOIUrl":"10.1016/j.envsoft.2025.106696","url":null,"abstract":"<div><div>Runoff estimation in India faces challenges due to diverse climate zones, complex physiographic conditions, and variable rainfall patterns, limiting traditional hydrological models and prompting exploration of advanced deep learning methods for improved streamflow prediction. Existing deep learning hydrological models struggle to estimate discharge at ungauged sites. In this study, we tested eight different deep learning models, four recurrent neural networks (GRU, CudaLSTM, EALSTM, ARLSTM) and four attention-based architectures (Transformer, Informer, Reformer, Linformer), across 144 watersheds in the Indian subcontinent (ISC). Our training and testing datasets combined meteorological forcing, catchment attributes, and observed discharge records. According to the results, ARLSTM improved prediction accuracy, achieving a median Nash–Sutcliffe Efficiency (NSE) of 0.71 on test basins. ARLSTM performs exceptionally well in specific regions: tropical monsoon areas (median NSE = 0.849), semi-arid regions (median NSE = 0.586), monsoon-influenced subtropical zones (NSE = 0.688), and tropical wet–dry climates (NSE = 0.539), especially in arid zones where traditional hydrological models often struggle. The assessments of high- and low-flow frequencies and durations, mean discharge, and runoff ratios underscore ARLSTM's capability to capture both extreme and average flow conditions. ARLSTM's reliance on lagged streamflow limits its use in ungauged basins. To address this issue, we developed a novel deep learning architecture, Ungauged Basin LSTM (UBLSTM), to predict the runoff values for any ungauged basin in India. UBLSTM matches the performance of ARLSTM, making it a better choice for areas in India that lack sufficient data or have ungauged basins across various climate zones.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"194 ","pages":"Article 106696"},"PeriodicalIF":4.6,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109559","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}

{"title":"Transfer learning using the global Caravan dataset for developing a local river streamflow prediction model","authors":"Almas Alzhanov ,&nbsp;Aliya Nugumanova ,&nbsp;Vsevolod Moreido","doi":"10.1016/j.envsoft.2025.106691","DOIUrl":"10.1016/j.envsoft.2025.106691","url":null,"abstract":"<div><div>Effective water resource and flood risk management depends on reliable streamflow forecasting. However, the accuracy of such forecasts is often limited by sparse monitoring networks and insufficient historical data. To address this issue, we explore the potential of a multi-basin training approach using the global Caravan hydrological dataset to improve local streamflow forecasting. As a case study, we focus on the Uba River basin in East Kazakhstan. The developed models are evaluated against two baselines: GR4J hydrological model and an LSTM model trained exclusively on local data. Results indicate that our approach enhances forecasting accuracy and outperforms the baseline models, with the best model achieving Nash-Sutcliffe efficiency value of 0.8187 compared to 0.72 of GR4J and 0.7602 of LSTM trained exclusively on local data. These findings indicate that multi-basin training with global datasets can enhance local streamflow forecasting in data-scarce regions.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"194 ","pages":"Article 106691"},"PeriodicalIF":4.6,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145093796","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}

{"title":"A lightweight dataflow-based software framework for building forest simulators","authors":"Tapio Lempinen ,&nbsp;Lauri Mehtätalo ,&nbsp;Annika Kangas ,&nbsp;Tero Heinonen ,&nbsp;Paulo Borges ,&nbsp;Jari Vauhkonen","doi":"10.1016/j.envsoft.2025.106661","DOIUrl":"10.1016/j.envsoft.2025.106661","url":null,"abstract":"<div><div>Simulators are critical tools for decision-making in forest management. We propose a dataflow-based model linking approach to increase the flexibility and modularity of forest simulator software while maintaining high computational efficiency. Our approach dynamically constructs a data model and model chains for simulation based on a model library, enabled treatments, and requested output variables. Models are framework- and language-independent pure functions, described through metadata in a domain-specific language. A case study with three model libraries demonstrates the applicability and efficiency of our approach. We observed a 96% speedup compared to an unoptimized real-world model implementation, while 95% of our code (measured by lines) was framework-independent and reusable. We observed a 15% slowdown compared to an optimized hand-written C implementation of a simpler model. We conclude that dataflow-based model linking can be used to build flexible and modular simulation software with a small runtime overhead.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"194 ","pages":"Article 106661"},"PeriodicalIF":4.6,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145093828","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}

{"title":"Backward erosion piping in numerical models: A literature review","authors":"E.M. van der Linde ,&nbsp;M. Wewer ,&nbsp;B.A. Robbins ,&nbsp;O. Colomés ,&nbsp;S.N. Jonkman ,&nbsp;J.P. Aguilar-López","doi":"10.1016/j.envsoft.2025.106681","DOIUrl":"10.1016/j.envsoft.2025.106681","url":null,"abstract":"<div><div>Backward erosion piping is a failure mechanism of dikes. Numerical modelling is crucial for design and assessment against BEP. Over 30 models have been developed, each with a different purpose and approach. This paper provides a comprehensive overview of the available numerical BEP models, highlighting their limitations, capabilities, and associated challenges. It discusses the different assumptions and their implications on the representation of BEP. Key challenges in the numerical modelling of BEP are (1) the flow (regime) inside the pipe, which is often simplified, even though the impact of this is relatively unknown. (2) The type of erosion (primary or secondary) differs per model, and even within a given type of erosion, approaches vary. (3) Overcoming the difference in scale is a trade-off between the computational effort and simplification. (4) Furthermore, validation of the physics in BEP modelling is difficult due to a of lack micro-scale experimental data.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"194 ","pages":"Article 106681"},"PeriodicalIF":4.6,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145044851","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}

{"title":"Augmented iterative tropospheric decomposition strategy for GNSS-based zenith tropospheric delay map generation","authors":"Xiangyang Song ,&nbsp;Giovanna Venuti ,&nbsp;Andrea Virgilio Monti-Guarnieri ,&nbsp;Marco Manzoni","doi":"10.1016/j.envsoft.2025.106669","DOIUrl":"10.1016/j.envsoft.2025.106669","url":null,"abstract":"<div><div>Global Navigation Satellite System (GNSS) permanent networks, deployed for geodetic purposes, provide valuable information on atmospheric water vapor content. The interaction between GNSS signals and the troposphere affects the signal propagation velocity, introducing an observable extra-path or delay along the zenith direction above each station, known as the Zenith Tropospheric Delay (ZTD). ZTDs can be used to correct Synthetic Aperture Radar (SAR) observations, which are influenced by similar propagation delays. To achieve this, ZTD maps with the same spatial resolution as the SAR observed images must be generated. In some cases a direct spatial interpolation of the delays is performed, in some others a tomographic approach is applied to derive a three-dimensional refractivity grid and then integrated to derive maps of delay along the SAR signal line of sight. This paper introduces a two-step procedure, called Augmented Iterative Tropospheric Decomposition (AITD), which can be ascribed to direct spatial interpolation techniques. It is derived from the well-established Iterative Tropospheric Decomposition (ITD) strategy, implemented in the Generic Atmospheric Correction Online Service for InSAR (GACOS), to interpolate dense and regular ZTDs derived from Numerical Weather Prediction Models (NWPMs). The AITD is comparable to the original decomposition approach in terms of prediction error accuracy (the prediction error Root-Mean-Square (RMS) of the two strategies differs by 1 mm, corresponding to approximately 10% of the ITD RMS, as assessed by a leave-one-out validation). However, it allows for the mitigation of interpolation artifacts introduced by the original approach when applied to sparse and not regularly distributed data. The augmented procedure allows to reduce the standard deviation of SAR phase screens by 45% and is computationally more efficient than ITD. It halves the processing time for GNSS-derived ZTDs and is nearly 20 times faster when applied to NWPM-derived ZTDs. The procedure enables the generation of high temporal resolution time series of maps as an additional product of GNSS network data processing, giving useful insight on the water vapor distribution for meteorological purposes.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"194 ","pages":"Article 106669"},"PeriodicalIF":4.6,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145046614","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}

{"title":"Leveraging artificial intelligence methods to map seagrass ecosystems in Italian Seas: Tackling human impact and climate change","authors":"Angelica Bianconi ,&nbsp;Sebastiano Vascon ,&nbsp;Elisa Furlan ,&nbsp;Andrea Critto","doi":"10.1016/j.envsoft.2025.106678","DOIUrl":"10.1016/j.envsoft.2025.106678","url":null,"abstract":"<div><div>Marine coastal ecosystems (MCEs) are crucial for human health, playing a key role in climate change adaptation. However, MCEs are globally threatened by environmental and human pressures. This study applies Graph Neural Networks (GNNs) to model seagrass distribution in the Italian Seas using a dataset of 2244 spatial units with environmental, climatic, and anthropogenic factors harmonised at 4 km resolution. GNN models, including Graph Convolutional and Attention Networks, were benchmarked against traditional machine learning methods: Random Forest, Support Vector Machine, and Multi-Layer Perceptron. GNNs achieved comparable overall accuracy (91%) but delivered more spatially consistent predictions and higher F1-scores (0.89) for the minority class (seagrass presence). Sensitivity analysis identified climatic and human variables as key drivers of seagrass distribution. These insights support the implementation of blue Nature-based Solutions (NbS) to protect and restore seagrass habitats, aiding biodiversity conservation and climate change mitigation while guiding effective policymaking.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"194 ","pages":"Article 106678"},"PeriodicalIF":4.6,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145093797","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}

{"title":"Enhancing the modularity and interoperability of hydrologic models: A demonstration with the Structure for Unifying Multiple Modeling Alternatives (SUMMA)","authors":"Sean J. Trim ,&nbsp;Martyn P. Clark ,&nbsp;Ashley E. Van Beusekom ,&nbsp;Kyle Klenk ,&nbsp;Wouter J.M. Knoben ,&nbsp;Raymond J. Spiteri","doi":"10.1016/j.envsoft.2025.106668","DOIUrl":"10.1016/j.envsoft.2025.106668","url":null,"abstract":"<div><div>We present a general approach to improve the modularity and interoperability of hydrologic and land models, demonstrated through refactoring the Structure for Unifying Multiple Modeling Alternatives (SUMMA) model. We first present a general hierarchical structure to organize the model source code, representing the horizontal variability across the domain (i.e., the “hydrofabric” of sub-basins) as well as the vertical architecture where each sub-basin is represented by vertical columns that extend from the top of the vegetation canopy to the depth of active groundwater. We then present a flexible strategy to solve the coupled conservation equations for water and energy in each vertical column, refactoring SUMMA’s operator-splitting methods, non-linear solvers, and flux calculators. Modularity was improved using internal subroutines and classes, and interoperability was improved using the initialize–update–finalize sequence at fine granularity. These refactoring developments support the use of Basic Model Interface (BMI) functions, improving component reusability, extensibility, and maintainability.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"194 ","pages":"Article 106668"},"PeriodicalIF":4.6,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145154843","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}

{"title":"A web-based platform for integrated real-time water Information: A prototype system for Missouri","authors":"Ayesha Siddiqua ,&nbsp;Bong-Chul Seo ,&nbsp;Steven Corns ,&nbsp;Joel Burken ,&nbsp;Robert R. Holmes Jr. ,&nbsp;Chang-Soo Kim","doi":"10.1016/j.envsoft.2025.106687","DOIUrl":"10.1016/j.envsoft.2025.106687","url":null,"abstract":"<div><div>The study demonstrates a framework to develop a web-based platform that presents real-time water data from observations, model estimations, and model predictions over a pilot domain in Missouri, United States. The Missouri Water Information System (MoWIS) is a prototype water information portal that provides current hydrologic (rainfall, drought, and stream) conditions and future stream predictions based on national radar and stream monitoring networks. The primary data sources are USGS and NOAA APIs for river stage/streamflow observations and forecasts. The platform visualizes these data within a geospatial context through a map-based environment, accessible upon user interaction, to facilitate intuitive interpretation and spatial analysis. With a modular architecture designed for system scalability, the platform supports fully automated operations and responsive performance by offloading intensive data processing and modeling tasks to external servers. Since the platform leverages nationally available web services, the framework is readily transferable to other geographic regions.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"194 ","pages":"Article 106687"},"PeriodicalIF":4.6,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145046610","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}

{"title":"One-dimensional simulation of land subsidence in vertically-heterogeneous highly compressible aquitards coupled with data assimilation via ensemble Kalman filter","authors":"Berenice Zapata-Norberto ,&nbsp;Eric Morales-Casique ,&nbsp;Graciela S. Herrera","doi":"10.1016/j.envsoft.2025.106690","DOIUrl":"10.1016/j.envsoft.2025.106690","url":null,"abstract":"<div><div>This study presents a methodology for calibrating a nonlinear groundwater flow and consolidation model in highly compressible, heterogeneous aquitards, focusing on vertical heterogeneity. Inspired by conditions in the Mexico basin, where the nature of the aquitard sediments, along with pore pressure monitoring through piezometers, plays a significant role. The model combines a nonlinear one-dimensional groundwater flow algorithm with an Ensemble Kalman Filter (EnKF) for data assimilation, correcting hydraulic head (<em>h</em>) and vertical hydraulic conductivity (<em>K</em>) distributions. Four reference cases were tested, and three data assimilation strategies were explored: (a) only <em>h</em> measurements, (b) only <em>K</em> measurements, and (c) both. Results show that all strategies provide satisfactory parameter estimations and settlement predictions, with the combined approach yielding the highest accuracy. While the method successfully simulates subsidence, its effectiveness diminishes if data assimilation only occurs in the initial simulation phase. This methodology has strong potential for predicting subsidence in real-world heterogeneous aquitards.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"194 ","pages":"Article 106690"},"PeriodicalIF":4.6,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145046612","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}