Environmental Modelling & Software最新文献

{"title":"Development of a self-supervised deep learning framework for chlorophyll-a retrieval in data-scarce inland waters","authors":"Bongseok Jeong ,&nbsp;Jihoon Shin ,&nbsp;YoonKyung Cha","doi":"10.1016/j.envsoft.2025.106817","DOIUrl":"10.1016/j.envsoft.2025.106817","url":null,"abstract":"<div><div>Deep learning and remote sensing-based chlorophyll-a (Chl-a) monitoring face challenges due to the optical complexity of inland waters and the scarcity of labeled data<strong>.</strong> To address these limitations, this study develops a self-supervised learning-based deep learning (SSL-DL) framework that leverages both labeled and unlabeled data. Three SSL-DL models are developed: a predictive SSL-DL model, which learns weak labels (incomplete labels); a generative SSL-DL model, which reconstructs input reflectance to capture underlying features; and an integrated SSL-DL model, which combines both. The models are applied to Sentinel-2 imagery of Daecheong and Paldang Lakes in South Korea. Results indicate that SSL-DL models outperform baseline models, with the integrated SSL-DL model achieving the highest test NSE (improvements of 0.1–0.36 over baselines in Daecheong Lake, improvements of 0.03–0.58 in Paldang Lake). The findings highlight the significance of SSL-DL in overcoming data limitations and enhancing scalability, demonstrating the potential for broader environmental remote sensing applications.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"197 ","pages":"Article 106817"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145732503","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":"An ensemble method for integrating rainfall forecast products based on average mutual information decomposition","authors":"Wei Ding ,&nbsp;Minglei Ren ,&nbsp;Yawei Ning ,&nbsp;Xuan Li ,&nbsp;Jinnan Zhang ,&nbsp;Huicheng Zhou","doi":"10.1016/j.envsoft.2026.106864","DOIUrl":"10.1016/j.envsoft.2026.106864","url":null,"abstract":"<div><div>Accurate rainfall forecasting is vital for reservoir operation, shifting focus from minimizing missed detections in flood season to reducing false alarms post-flood. Conventional Multi-Model Ensemble (MME) methods use static weights, ignoring forecast skill variations across rainfall intensities. This study proposes MME-MID, a novel framework using Normalized Average Mutual Information Decomposition to dynamically weight forecasts by categorized rainfall intensities. Two decomposition approaches are applied: (1) Uncertainty decomposition prioritizes higher-accuracy categories to reduce post-flood false alarms; (2) Information decomposition enhances detection of high-intensity events during flood season. Validation in Dahuofang and Huanren basins shows MME-MID reduces miss ratio by 5.5 % in flood season and false alarm ratio by 11 % post-flood versus conventional MME, without increasing overall error, significantly improving decision reliability.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"197 ","pages":"Article 106864"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921051","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":"Deterministic discharge time series for morphodynamic assessments of river interventions","authors":"S.J.H.A. Gradussen ,&nbsp;A.L. de Jongste ,&nbsp;V. Chavarrías","doi":"10.1016/j.envsoft.2025.106800","DOIUrl":"10.1016/j.envsoft.2025.106800","url":null,"abstract":"<div><div>When modelling the morphological response to river interventions, deterministic discharge time series are commonly used as a cost-effective alternative for approximating the most likely morphological changes resulting from Monte Carlo (MC) analyses. The conventional deterministic approach cycles long-term discharge statistics through a Cycled Annual Hydrograph (CyAH), neglecting inter-annual discharge variability. We introduce a new deterministic method, the Multiple Annual Hydrograph (MuAH), that accounts for such inter-annual discharge variations. Using a one-dimensional model, we simulate morphodynamic changes resulting from a hypothetical intervention in an alluvial river to demonstrate the application of these deterministic time series and to evaluate their performance. We find that MuAH time series result in both long-term (quasi-static) evolution and seasonal (dynamic) morphological changes that more closely match MC results and natural discharge time series, compared with the CyAH approach. This enables more accurate assessments of morphological change induced by river interventions when using deterministic time series.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"197 ","pages":"Article 106800"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145593069","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":"Three-dimensional narrative visualization in virtual geographic scenes for enhancing textual information driven by knowledge and semantics","authors":"Yukun Guo ,&nbsp;Jun Zhu ,&nbsp;Zhihao Guo ,&nbsp;Jianlin Wu ,&nbsp;Jinbin Zhang","doi":"10.1016/j.envsoft.2025.106844","DOIUrl":"10.1016/j.envsoft.2025.106844","url":null,"abstract":"<div><div>The narrative visualization of geographic textual information significantly facilitates information dissemination and enhances public understanding. However, current visualization approaches suffer from inaccurate text parsing, inefficient visualization construction processes, and unintuitive visual outputs. To address these issues, this paper proposes a 3D narrative visualization method that integrates semantic information and knowledge with virtual geographic scenes. The method includes the construction of a geographic narrative knowledge graph, the design of a multi-level spatiotemporal narrative visualization model, and a collaborative narrative visualization strategy using large language models and knowledge graphs. Experimental analyses were conducted using texts describing natural disasters and social events in Luding County. Results showed effectiveness scores consistently above 3.5, cognitive accuracy improvements of up to 16 %, and cognitive processing time reduced by approximately half. These findings verify that the proposed method effectively transforms textual geographic information into narrative visualizations, significantly improving public comprehension and cognitive efficiency, thus demonstrating its practical potential for broader applications in geographic information communication.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"197 ","pages":"Article 106844"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823152","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":"Taming the non-linearity: An iterative conceptual routing model for improving flood peak prediction","authors":"Ekant Sarkar ,&nbsp;Akshay Kadu ,&nbsp;S.L. Kesav Unnithan ,&nbsp;Basudev Biswal","doi":"10.1016/j.envsoft.2025.106843","DOIUrl":"10.1016/j.envsoft.2025.106843","url":null,"abstract":"<div><div>The magnitude and timing of flood peaks are strongly influenced by the non-linear relationship between flow velocity and discharge. Traditional routing models struggle to account for the variation in flow velocity, particularly with time, due to computational constraints. This study proposes a novel Iterative Routing Model (IRM) that updates flow velocity as a function of streamflow magnitude. The IRM was applied to seven gauging stations in the Godavari River Basin, India. It outperformed two other models used in this study in simulating peak discharge and timing, with the lowest average absolute deviations of 29.98 % and 0.2 days, respectively. The IRM achieved the highest median NSE (0.78) and KGE (0.79) values across all stations. Moreover, the calibrated Manning's roughness from the proposed model appears more realistic compared to that given by other models. Overall, our findings highlight the potential of the proposed model to improve flood peak predictions in large river basins.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"197 ","pages":"Article 106843"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823153","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":"Producing Earth science data for impact: Improved commercial cloud usability of archive model data","authors":"Amy McNally ,&nbsp;Lucas Sterzinger ,&nbsp;Ian Carroll","doi":"10.1016/j.envsoft.2025.106846","DOIUrl":"10.1016/j.envsoft.2025.106846","url":null,"abstract":"<div><div>This overview introduces concepts, challenges, and solutions for analysis of Earth System Model data in cloud computing environments. We highlight how NASA's Earth Science Data System is migrating data to the cloud, but existing data formats are not yet optimized for cloud performance. Specifically, there is significant performance degradation in the on-cloud analysis due to fragmented metadata and small data chunk size. Using data from two NASA Land Data Assimilation Systems we present a case study comparing on-premises vs. commercial cloud workflows. The case study demonstrates that re-chunking archival data may be necessary, as well as consolidating metadata and generating metadata sidecar files, for enhanced cloud performance. We also recommend resources and tools for users interested in cloud-based analysis and shifts in practices for data producers. These changes will allow for successful cloud migration of NASA Earthdata and improve data discoverability and usability for critical Earth science research and applications.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"197 ","pages":"Article 106846"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823151","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 reliable deep ensemble hybrid model for urban air quality health index forecasting in maritime Canada","authors":"Mehdi Jamei ,&nbsp;Gurjit S. Randhawa ,&nbsp;Mumtaz Ali ,&nbsp;Masoud Karbasi ,&nbsp;Ismail Olumegbon ,&nbsp;Saad Javed Cheema ,&nbsp;Travis J. Esau ,&nbsp;Qamar U. Zaman ,&nbsp;Aitazaz A. Farooque","doi":"10.1016/j.envsoft.2025.106837","DOIUrl":"10.1016/j.envsoft.2025.106837","url":null,"abstract":"<div><div>Accurate Air Quality Health Index (AQHI) forecasting is crucial for safeguarding public health and informing policy decisions in coastal urban regions of Maritime Canada. This study introduces a graph-enhanced deep ensemble model that integrates Robust Empirical Mode Decomposition (REMD), Deep Ensemble Random Vector Functional Link (DeepERVFL), graph-based feature selection, and Borda Count multi-criteria decision making for multi-weekly AQHI forecasting. Forecast uncertainty is quantified using bootstrap resampling to ensure confidence in the results. Benchmarking against Recursive LSTM and Histogram-Based Gradient Boosting Ensemble (HBGBE) models shows the superior performance of the REMD-DeepERVFL framework, with BORDA scores of 0.940 (T+1) and 1.06 (T+3) in Halifax, 0.797 (T+3) in Charlottetown, and 0.931 (T+3) in St. John's. The framework supports air-quality early warning systems, public health communication, and climate-health monitoring, offering timely and reliable information. This hybrid approach provides a robust, scalable, and uncertainty-aware solution for regional AQHI forecasting in Atlantic Canada.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"197 ","pages":"Article 106837"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823143","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":"RegreSSM: A novel software tool for downscaling the SMAP L3 soil moisture operational product utilizing the Ts/VI feature space and Sentinel-3 data","authors":"George P. Petropoulos ,&nbsp;Spyridon E. Detsikas ,&nbsp;Vasileios Anagnostopoulos ,&nbsp;Christina Lekka","doi":"10.1016/j.envsoft.2025.106836","DOIUrl":"10.1016/j.envsoft.2025.106836","url":null,"abstract":"<div><div>Herein we present RegreSSM, a software tool that enables the downscaling of SMAP L3 Surface Soil Moisture (SSM) operational product from 36 km to 1 km by the fusion of optical and thermal data retrieved from Sentinel-3 platform. The downscaling method is based on the well-established properties of the Ts/VI feature space. Most of the existing soil moisture downscaling methods are computationally complex, require advanced expertise, and lack standalone tools suitable for operational or non-expert use. To address these limitations, this study proposes a simple and accessible framework for generating high-resolution SSM maps using only land surface temperature and vegetation cover as inputs. The tool has been developed in python programming language as a stand-alone application and can be executed in any operational system. The application offers automated and reproducible workflows for spatiotemporal matching and processing of SMAP L3 SSM products and Sentinel-3 dataset. The software tool's practical application is demonstrated over the Iberian Peninsula, where validation of the SMAP L3 product performed for all calendar year 2022 using in-situ observations from the REMEDHUS operational network stations. Results showed a satisfactory retrieval of SSM with a small average bias of 0.01 m³/m³, a MAD of 0.06 m³/m³, a RMSD of 0.07 m³/m³, and a satisfactory R² of 0.63, confirming the ability of the proposed downscaling framework and RegreSSM to retrieve SSM at the 1 km spatial resolution. Results obtained herein were also compared to the validation metrics reported for operational RS-based SSM products, with typically reported uncertainty of 0.04 m³/m³. The availability of RegreSSM to the SSM users' community consists an important step towards the standardization of downscaling procedures as well as bridging the spatial gap of existing operational SM products to the requirements of the fine-scale applications. It also contributes towards advancing the deployment of geo-processing tools utilizing the synergies between state-of-the-art methods and RS data available today from the most sophisticated satellites in orbit.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"197 ","pages":"Article 106836"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145785085","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":"Development of an interactive web-based tool for flood risk analysis and climate–resilient road drainage design: RiskDRAIN","authors":"Mohammad Fereshtehpour ,&nbsp;Rashid Bashir ,&nbsp;Neil F. Tandon","doi":"10.1016/j.envsoft.2026.106867","DOIUrl":"10.1016/j.envsoft.2026.106867","url":null,"abstract":"<div><div>As climate change intensifies extreme rainfall, traditional design storm methods based on stationary assumptions are increasingly inadequate, often leading to misdesigned drainage infrastructure. To address this and manage projection uncertainties, we developed RiskDRAIN, a web-based application designed for the risk-based adjustment of projected design storms. RiskDRAIN, which stands for <strong>Risk</strong>-based <strong>D</strong>esign for <strong>R</strong>esilient <strong>A</strong>daptation to <strong>I</strong>nfrastructure <strong>N</strong>eeds, integrates risk analysis with Canadian downscaled CMIP6 projections (CanDCS-M6). The framework incorporates Intensity-Duration-Frequency (IDF) curves derived from the IDF-CC tool, considering two projection techniques (Clausius-Clapeyron scaling and Equidistance Quantile Matching) with GEV and Gumbel distributions, covering a range of emission pathways (SSP2-4.5, SSP5-8.5) and future horizons. Through an interactive interface, users refine design storms by evaluating provincial and site-specific risks derived from hazard exposure and multi-dimensional vulnerability (socio-economic, transportation, and environmental). Validated through a highway drainage case study, RiskDRAIN empowers practitioners with a data-driven platform for cost-effective, climate-resilient infrastructure planning.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"197 ","pages":"Article 106867"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956766","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 LLM-based agent for the construction of typhoon knowledge graphs","authors":"Yi Huang ,&nbsp;Yongqi Xia ,&nbsp;Ran Tao ,&nbsp;Donglai Jiao ,&nbsp;Xiangqiang Min ,&nbsp;Jieying Zheng ,&nbsp;Yuting Jiang ,&nbsp;Wenjun Wu ,&nbsp;Peijun Du","doi":"10.1016/j.envsoft.2026.106856","DOIUrl":"10.1016/j.envsoft.2026.106856","url":null,"abstract":"<div><div>Conventional knowledge graphs (KGs) struggle to integrate fragmented typhoon disaster data due to error accumulation and inadequate modeling of complex spatiotemporal relationships. To overcome this, we propose TyphoonKGent, an agent driven by large language models (LLMs), which employs hierarchical knowledge representation to structurally encode typhoon evolution and impacts. It decomposes KG construction into specialized tasks (role-playing, spatiotemporal completion, entity alignment, lifecycle determination, event identification) with domain-optimized Chain-of-Thought (CoT) generation to enhance LLM reasoning for geospatial tasks. Built via efficient LoRA-based fine-tuning of distilled LLaMA/Qwen models, TyphoonKGent improves accuracy by 30 % over non-finetuned baseline models and outperforms DeepSeek-R1 under 3-shot inference by 2 %–5 %. Furthermore, visualization applications confirm its effectiveness in trajectory analysis, impact mapping, and real-time decision support. The proposed TyphoonKGent enables end-to-end KG construction, cross-domain adaptability via customizable CoTs, task-specific fine-tuning, and scalable dynamic updates for disaster management.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"197 ","pages":"Article 106856"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956770","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}