To mitigate air pollution, source apportionment is a key element for the design of effective measures. However, source apportionment often involves complex model chains only accessible to expert users. In this paper we present a new web-application, the Concawe NO2 source apportionment viewer. It allows experts and non-expert users to evaluate the contributions of different sectors and the impact of measures in the road transport sector on current and future NO2 pollution in the EU27+UK in a fast and user-friendly way. The methodology behind the viewer was described in a previous paper byDegraeuwe et al. (2024). Here we describe the user interface and give some examples; the contribution of different sectors to the NO2 concentrations in the 3136 monitoring stations, and the impact of specific transport policies (e.g., Euro 7/VII standard, urban access regulations) on the NO2 concentrations in 948 European cities.
{"title":"The Concawe NO2 source apportionment viewer: A web-application to mitigate NO2 pollution from traffic and other sources","authors":"Bart Degraeuwe , Robin Houdmeyers , Stijn Janssen , Wouter Lefebvre , Athanasios Megaritis","doi":"10.1016/j.envsoft.2024.106315","DOIUrl":"10.1016/j.envsoft.2024.106315","url":null,"abstract":"<div><div>To mitigate air pollution, source apportionment is a key element for the design of effective measures. However, source apportionment often involves complex model chains only accessible to expert users. In this paper we present a new web-application, the Concawe NO<sub>2</sub> source apportionment viewer. It allows experts and non-expert users to evaluate the contributions of different sectors and the impact of measures in the road transport sector on current and future NO<sub>2</sub> pollution in the EU27+UK in a fast and user-friendly way. The methodology behind the viewer was described in a previous paper byDegraeuwe et al. (2024). Here we describe the user interface and give some examples; the contribution of different sectors to the NO<sub>2</sub> concentrations in the 3136 monitoring stations, and the impact of specific transport policies (e.g., Euro 7/VII standard, urban access regulations) on the NO<sub>2</sub> concentrations in 948 European cities.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"185 ","pages":"Article 106315"},"PeriodicalIF":4.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142935447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.envsoft.2025.106326
Zeynep Özcan , Merih Aydınalp Köksal , Emre Alp
The synergies and conflicts between the energy and water systems, necessitate the collaboration between these sectors. Effective management of the interdependent energy and water systems requires a nexus approach that acknowledges these interconnections, as opposed to regarding them as distinct systems. We applied an integrated modeling approach for evaluating the Water-Energy Nexus based on a variety of criteria as water consumption, energy production, and CO2 emissions. According to the simulations, 96% reduction in water savings can be achieved when wet cooling systems of the thermal power plant (TPP) are converted to dry. Moreover, if the TPPs are shut down to reduce CO2 emissions, the hydroelectric power plants can only cover 16% of the total electricity production. Hence, securing energy while reducing CO2 emissions is a challenging task. Despite producing only 10–15% of total energy, HPPs account for 70–100% of total water consumption in all scenarios.
{"title":"An integrated modeling approach to assess water-energy nexus in a semi-arid watershed","authors":"Zeynep Özcan , Merih Aydınalp Köksal , Emre Alp","doi":"10.1016/j.envsoft.2025.106326","DOIUrl":"10.1016/j.envsoft.2025.106326","url":null,"abstract":"<div><div>The synergies and conflicts between the energy and water systems, necessitate the collaboration between these sectors. Effective management of the interdependent energy and water systems requires a nexus approach that acknowledges these interconnections, as opposed to regarding them as distinct systems. We applied an integrated modeling approach for evaluating the Water-Energy Nexus based on a variety of criteria as water consumption, energy production, and CO<sub>2</sub> emissions. According to the simulations, 96% reduction in water savings can be achieved when wet cooling systems of the thermal power plant (TPP) are converted to dry. Moreover, if the TPPs are shut down to reduce CO<sub>2</sub> emissions, the hydroelectric power plants can only cover 16% of the total electricity production. Hence, securing energy while reducing CO<sub>2</sub> emissions is a challenging task. Despite producing only 10–15% of total energy, HPPs account for 70–100% of total water consumption in all scenarios.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"185 ","pages":"Article 106326"},"PeriodicalIF":4.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142975673","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}
Pub Date : 2025-02-01DOI: 10.1016/j.envsoft.2024.106286
Daryn Sagel, Bryan Quaife
The increasing frequency and severity of wildfires highlight the need for accurate fire and plume spread models. We introduce an approach that effectively isolates and tracks fire and plume behavior across various spatial and temporal scales and image types, identifying physical phenomena in the system and providing insights useful for developing and validating models. Our method combines image segmentation and graph theory to delineate fire fronts and plume boundaries. We demonstrate that the method effectively distinguishes fires and plumes from visually similar objects. Results demonstrate the successful isolation and tracking of fire and plume dynamics across various image sources, ranging from synoptic-scale (– m) satellite images to sub-microscale (– m) images captured close to the fire environment. Furthermore, the methodology leverages image inpainting and spatio-temporal dataset generation for use in statistical and machine learning models.
{"title":"Fire dynamic vision: Image segmentation and tracking for multi-scale fire and plume behavior","authors":"Daryn Sagel, Bryan Quaife","doi":"10.1016/j.envsoft.2024.106286","DOIUrl":"10.1016/j.envsoft.2024.106286","url":null,"abstract":"<div><div>The increasing frequency and severity of wildfires highlight the need for accurate fire and plume spread models. We introduce an approach that effectively isolates and tracks fire and plume behavior across various spatial and temporal scales and image types, identifying physical phenomena in the system and providing insights useful for developing and validating models. Our method combines image segmentation and graph theory to delineate fire fronts and plume boundaries. We demonstrate that the method effectively distinguishes fires and plumes from visually similar objects. Results demonstrate the successful isolation and tracking of fire and plume dynamics across various image sources, ranging from synoptic-scale (<span><math><mrow><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mn>4</mn></mrow></msup></mrow></math></span>–<span><math><mrow><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mn>5</mn></mrow></msup></mrow></math></span> m) satellite images to sub-microscale (<span><math><mrow><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mn>0</mn></mrow></msup></mrow></math></span>–<span><math><mrow><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mn>1</mn></mrow></msup></mrow></math></span> m) images captured close to the fire environment. Furthermore, the methodology leverages image inpainting and spatio-temporal dataset generation for use in statistical and machine learning models.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"185 ","pages":"Article 106286"},"PeriodicalIF":4.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825321","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}
Pub Date : 2025-02-01DOI: 10.1016/j.envsoft.2024.106291
Inès Astrid Tougma , Marijn Van de Broek , Johan Six , Thomas Gaiser , Maire Holz , Isabel Zentgraf , Heidi Webber
Most cropping system models simulate conceptual soil organic matter (SOM) pools, such as active, passive and slow pools that cannot be measured, complicating model calibration. In reality, SOM can be described in terms of quantifiable pools of particulate organic matter (POM) and mineral-associated organic matter (MAOM) which respond differently to management and climate. We present the AMPSOM model, integrated in a cropping system modelling framework (SIMPLACE). AMPSOM simulates carbon and nitrogen dynamics in MAOM and POM in response to crop growth and management, as well as soil texture, water and nitrogen content and temperature. It also simulates the radiocarbon isotope (14C) of soil organic carbon (SOC) to constrain the turnover time of slowly cycling SOC pools. Model calibration and evaluation were performed for thirty six sandy and loamy arable soils in Brandenburg, Germany. Results show that AMPSOM can reproduce observed patterns of SOC and nitrogen stocks in POM and MAOM along depth profiles across different soil types.
{"title":"AMPSOM: A measureable pool soil organic carbon and nitrogen model for arable cropping systems","authors":"Inès Astrid Tougma , Marijn Van de Broek , Johan Six , Thomas Gaiser , Maire Holz , Isabel Zentgraf , Heidi Webber","doi":"10.1016/j.envsoft.2024.106291","DOIUrl":"10.1016/j.envsoft.2024.106291","url":null,"abstract":"<div><div>Most cropping system models simulate conceptual soil organic matter (SOM) pools, such as active, passive and slow pools that cannot be measured, complicating model calibration. In reality, SOM can be described in terms of quantifiable pools of particulate organic matter (POM) and mineral-associated organic matter (MAOM) which respond differently to management and climate. We present the AMPSOM model, integrated in a cropping system modelling framework (SIMPLACE). AMPSOM simulates carbon and nitrogen dynamics in MAOM and POM in response to crop growth and management, as well as soil texture, water and nitrogen content and temperature. It also simulates the radiocarbon isotope (<sup>14</sup>C) of soil organic carbon (SOC) to constrain the turnover time of slowly cycling SOC pools. Model calibration and evaluation were performed for thirty six sandy and loamy arable soils in Brandenburg, Germany. Results show that AMPSOM can reproduce observed patterns of SOC and nitrogen stocks in POM and MAOM along depth profiles across different soil types.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"185 ","pages":"Article 106291"},"PeriodicalIF":4.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Large-scale land-use change simulations are crucial for understanding land dynamics, investigating climate change, and shaping policy regulations. However, conducting fine-resolution land-use change simulations on a large scale is challenging due to high computational demands. Conversely, land-use change simulations with coarse-resolution data distort spatial details, thereby reducing simulation performance. Parallel computing can reduce computational demands but requires significant computational resources. Mixed-cell CA models offer a solution to balance simulation performance and computational intensity. The comparison experiments using various resolution land use datasets demonstrate that mixed-cell CA models, even those with coarse-resolution data, achieve results comparable to those of pure-cell CA models using fine-resolution data, but with significantly reduced simulation time. This highlights the efficiency of mixed-cell CA models in achieving comparable performance with lower computational intensity. Additionally, this study provides a measurement method for the uncertainty of mixed-cell CA models. In summary, this study reveals the unique advantages of mixed-cell CA models in efficient large-scale land use simulations, thereby providing valuable insights and guidance for future land use management and policy decisions.
{"title":"Balancing simulation performance and computational intensity of CA models for large-scale land-use change simulations","authors":"Zhewei Liang , Xun Liang , Xintong Jiang , Tingyu Li , Qingfeng Guan","doi":"10.1016/j.envsoft.2024.106293","DOIUrl":"10.1016/j.envsoft.2024.106293","url":null,"abstract":"<div><div>Large-scale land-use change simulations are crucial for understanding land dynamics, investigating climate change, and shaping policy regulations. However, conducting fine-resolution land-use change simulations on a large scale is challenging due to high computational demands. Conversely, land-use change simulations with coarse-resolution data distort spatial details, thereby reducing simulation performance. Parallel computing can reduce computational demands but requires significant computational resources. Mixed-cell CA models offer a solution to balance simulation performance and computational intensity. The comparison experiments using various resolution land use datasets demonstrate that mixed-cell CA models, even those with coarse-resolution data, achieve results comparable to those of pure-cell CA models using fine-resolution data, but with significantly reduced simulation time. This highlights the efficiency of mixed-cell CA models in achieving comparable performance with lower computational intensity. Additionally, this study provides a measurement method for the uncertainty of mixed-cell CA models. In summary, this study reveals the unique advantages of mixed-cell CA models in efficient large-scale land use simulations, thereby providing valuable insights and guidance for future land use management and policy decisions.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"185 ","pages":"Article 106293"},"PeriodicalIF":4.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825337","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}
Pub Date : 2025-02-01DOI: 10.1016/j.envsoft.2025.106332
Lei Yao , Jiangjiang Zhang , Chenglong Cao , Feifei Zheng
Rainfall-runoff (RR) modeling is crucial for flood preparedness and water resource management. Accurate RR model predictions depend on effective parameter estimation and uncertainty quantification using observed data through data assimilation (DA). Traditional DA methods often struggle with challenges such as non-Gaussianity and equifinality. To address these challenges, this study introduces two ensemble smoother methods, i.e., ESDL with a deep learning-based update, and ESLU with a local ensemble update, aiming to enhance the calibration of RR models. To demonstrate the effectiveness of our proposed methods, we conduct a comprehensive analysis involving various DA techniques applied to parameter estimation of RR models. We compare these methods with traditional approaches, evaluating deep neural network architectures, iteration numbers, and measurement errors. The results unequivocally showcase the consistent reliability of ESDL and ESLU, especially the latter one, across diverse scenarios, establishing them as promising approaches for the effective calibration and uncertainty quantification of RR models.
{"title":"Parameter estimation and uncertainty quantification of rainfall-runoff models using data assimilation methods based on deep learning and local ensemble updates","authors":"Lei Yao , Jiangjiang Zhang , Chenglong Cao , Feifei Zheng","doi":"10.1016/j.envsoft.2025.106332","DOIUrl":"10.1016/j.envsoft.2025.106332","url":null,"abstract":"<div><div>Rainfall-runoff (RR) modeling is crucial for flood preparedness and water resource management. Accurate RR model predictions depend on effective parameter estimation and uncertainty quantification using observed data through data assimilation (DA). Traditional DA methods often struggle with challenges such as non-Gaussianity and equifinality. To address these challenges, this study introduces two ensemble smoother methods, i.e., ES<sub>DL</sub> with a deep learning-based update, and ES<sub>LU</sub> with a local ensemble update, aiming to enhance the calibration of RR models. To demonstrate the effectiveness of our proposed methods, we conduct a comprehensive analysis involving various DA techniques applied to parameter estimation of RR models. We compare these methods with traditional approaches, evaluating deep neural network architectures, iteration numbers, and measurement errors. The results unequivocally showcase the consistent reliability of ES<sub>DL</sub> and ES<sub>LU</sub>, especially the latter one, across diverse scenarios, establishing them as promising approaches for the effective calibration and uncertainty quantification of RR models.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"185 ","pages":"Article 106332"},"PeriodicalIF":4.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143020245","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}
Pub Date : 2025-02-01DOI: 10.1016/j.envsoft.2024.106307
Mohamed M. Fathi , Zihan Liu , Anjali M. Fernandes , Michael T. Hren , Dennis O. Terry , C. Nataraj , Virginia Smith
Computational hydrodynamic models support river science and management. However, current physics-based models face computational challenges; they require extensive processing time for large-scale two-dimensional flood simulations. Despite the success of Deep Learning (DL) applications in generating inundation maps, accurate prediction of unsteady flood hydrodynamic maps remains challenging. This paper compares traditional approaches to a novel DL approach, which integrates convolutional neural networks with long short-term memory, to deliver precise, rapid, and continuous simulation of the spatiotemporal dynamics of river floods. This is the first DL framework able to generate essential hydrodynamic variables: water depth, velocity magnitude, and flow direction maps. Water depth and velocity magnitude predictions across the testing dataset are robust, with average RMSE of 0.14 m and 0.02 m/s, respectively. The DL predictions are 415 times faster compared to traditional computational approaches, representing a paradigm shift in hydrodynamics modeling that advances long-term flood simulations and resilient river management.
{"title":"Spatiotemporal flood depth and velocity dynamics using a convolutional neural network within a sequential Deep-Learning framework","authors":"Mohamed M. Fathi , Zihan Liu , Anjali M. Fernandes , Michael T. Hren , Dennis O. Terry , C. Nataraj , Virginia Smith","doi":"10.1016/j.envsoft.2024.106307","DOIUrl":"10.1016/j.envsoft.2024.106307","url":null,"abstract":"<div><div>Computational hydrodynamic models support river science and management. However, current physics-based models face computational challenges; they require extensive processing time for large-scale two-dimensional flood simulations. Despite the success of Deep Learning (DL) applications in generating inundation maps, accurate prediction of unsteady flood hydrodynamic maps remains challenging. This paper compares traditional approaches to a novel DL approach, which integrates convolutional neural networks with long short-term memory, to deliver precise, rapid, and continuous simulation of the spatiotemporal dynamics of river floods. This is the first DL framework able to generate essential hydrodynamic variables: water depth, velocity magnitude, and flow direction maps. Water depth and velocity magnitude predictions across the testing dataset are robust, with average RMSE of 0.14 m and 0.02 m/s, respectively. The DL predictions are 415 times faster compared to traditional computational approaches, representing a paradigm shift in hydrodynamics modeling that advances long-term flood simulations and resilient river management.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"185 ","pages":"Article 106307"},"PeriodicalIF":4.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142905661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.envsoft.2024.106266
Daniel Luna , Ranran Chen , Ahmed Sheba , Ryan Young , Yao Liang , Xu Liang
Effective data and model integration is crucial for exploring scientific questions in hydrology and other geosciences. The increasing heterogeneity and complexity of data and models pose integration challenges. CyberWater addresses these with an open-data and open-modeling framework. Featuring GUI-based workflows, it includes Data Agents for accessing diverse online data sources and a Generic Model Agent Toolkit for seamless, code-free model integration. This study introduces the Static Parameter Agent suite, a novel toolkit designed to streamline the creation and organization of parameter files required for various models. The toolkit enables users to efficiently and automatically generate files on demand, minimizing the time-consuming and error-prone manual preparation of complex parameter files. It further logs all changes to parameter values across each model simulation, ensuring a reproducible end-to-end process. It connects seamlessly with Geographic Information System (GIS) engines like GRASS GIS and has been tested on models including VIC, DHSVM, and CASA-CNP.
{"title":"Facilitating open data and open model integration with generic parameter input file generators in the CyberWater framework","authors":"Daniel Luna , Ranran Chen , Ahmed Sheba , Ryan Young , Yao Liang , Xu Liang","doi":"10.1016/j.envsoft.2024.106266","DOIUrl":"10.1016/j.envsoft.2024.106266","url":null,"abstract":"<div><div>Effective data and model integration is crucial for exploring scientific questions in hydrology and other geosciences. The increasing heterogeneity and complexity of data and models pose integration challenges. CyberWater addresses these with an open-data and open-modeling framework. Featuring GUI-based workflows, it includes Data Agents for accessing diverse online data sources and a Generic Model Agent Toolkit for seamless, code-free model integration. This study introduces the Static Parameter Agent suite, a novel toolkit designed to streamline the creation and organization of parameter files required for various models. The toolkit enables users to efficiently and automatically generate files on demand, minimizing the time-consuming and error-prone manual preparation of complex parameter files. It further logs all changes to parameter values across each model simulation, ensuring a reproducible end-to-end process. It connects seamlessly with Geographic Information System (GIS) engines like GRASS GIS and has been tested on models including VIC, DHSVM, and CASA-CNP.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"185 ","pages":"Article 106266"},"PeriodicalIF":4.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143128006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The commonly used random sampling method in machine learning-based flood susceptibility studies has two major issues: a default invalid assumption of spatial homogeneity and an inadequate number of non-flood samples. To address these issues, this study proposed a novel sample-enhancement framework to improve the quality of training samples on both flood and non-flood sides. Three one-way enhancements (two flood and one non-flood) and two joint enhancements were designed. The enhancements were evaluated against random sampling using four mainstream machine learning algorithms (ANN, RF, SVM, and XGBoost) across two heterogeneous urban regions in Guangzhou, China. The highest performances are achieved by the joint enhancements, which are followed by one-way enhancements and random sampling (no enhancement). Another important conclusion is that one-way enhancements exhibit divergent yet complementary effects. Flood enhancements primarily affect susceptibility distribution (mean value and standard deviation), while non-flood enhancements mainly influence binary classification performance (AUC).
{"title":"A novel sample-enhancement framework for machine learning-based urban flood susceptibility assessment","authors":"Huabing Huang, Changpeng Wang, Zhiwen Tao, Jiayin Zhan","doi":"10.1016/j.envsoft.2024.106314","DOIUrl":"10.1016/j.envsoft.2024.106314","url":null,"abstract":"<div><div>The commonly used random sampling method in machine learning-based flood susceptibility studies has two major issues: a default invalid assumption of spatial homogeneity and an inadequate number of non-flood samples. To address these issues, this study proposed a novel sample-enhancement framework to improve the quality of training samples on both flood and non-flood sides. Three one-way enhancements (two flood and one non-flood) and two joint enhancements were designed. The enhancements were evaluated against random sampling using four mainstream machine learning algorithms (ANN, RF, SVM, and XGBoost) across two heterogeneous urban regions in Guangzhou, China. The highest performances are achieved by the joint enhancements, which are followed by one-way enhancements and random sampling (no enhancement). Another important conclusion is that one-way enhancements exhibit divergent yet complementary effects. Flood enhancements primarily affect susceptibility distribution (mean value and standard deviation), while non-flood enhancements mainly influence binary classification performance (AUC).</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"185 ","pages":"Article 106314"},"PeriodicalIF":4.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142905678","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}
Pub Date : 2025-02-01DOI: 10.1016/j.envsoft.2024.106309
Prince Agyemang , Ebenezer M. Kwofie , Jamie I. Baum , Dongyi Wang , Emmanuel A. Kwofie
To generate evidence to address food system challenges, we developed an adaptable framework for multimodel assessment of the convergence effect of health and environmental drivers in food systems. We achieved this goal by developing a modeling framework that facilitates testing and applying four deep-learning algorithms using a case study of the United States's food system. Among the models tested, the bidirectional and single-layer long short-term memory models outperformed the others with αE(2.75) and αH(3.51) when predicting environmental drivers and health drivers, respectively. All the models tested performed better at predicting environmental than health drivers. The best-performing model for each dimension was deployed into the Food System Rapid Overview Assessment through Scenarios (FS-ROAS) tool. As we approach the endpoint of the transformative 2030 agenda, FS-ROAS can be a timely toolkit that enables stakeholders to explore diverse intervention scenarios in the context of short-medium and long-term goals for future food systems and generate evidence to guide future actions.
{"title":"Environmental-Health Convergence: A deep learning-oriented decision support system for catalyzing sustainable healthy food systems","authors":"Prince Agyemang , Ebenezer M. Kwofie , Jamie I. Baum , Dongyi Wang , Emmanuel A. Kwofie","doi":"10.1016/j.envsoft.2024.106309","DOIUrl":"10.1016/j.envsoft.2024.106309","url":null,"abstract":"<div><div>To generate evidence to address food system challenges, we developed an adaptable framework for multimodel assessment of the convergence effect of health and environmental drivers in food systems. We achieved this goal by developing a modeling framework that facilitates testing and applying four deep-learning algorithms using a case study of the United States's food system. Among the models tested, the bidirectional and single-layer long short-term memory models outperformed the others with α<sub><em>E</em></sub>(2.75) and α<sub><em>H</em></sub>(3.51) when predicting environmental drivers and health drivers, respectively. All the models tested performed better at predicting environmental than health drivers. The best-performing model for each dimension was deployed into the Food System Rapid Overview Assessment through Scenarios (FS-ROAS) tool. As we approach the endpoint of the transformative 2030 agenda, FS-ROAS can be a timely toolkit that enables stakeholders to explore diverse intervention scenarios in the context of short-medium and long-term goals for future food systems and generate evidence to guide future actions.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"185 ","pages":"Article 106309"},"PeriodicalIF":4.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142905650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}