Pub Date : 2025-03-05DOI: 10.1016/j.envsoft.2025.106421
Nima Zafarmomen , Vidya Samadi
This paper seeks to answer the question “can Large Language Models (LLMs) effectively reason about adverse weather conditions?”. To address this question, we utilized multiple LLMs to harness the US National Weather Service (NWS) flood report data spanning from June 2005 to September 2024. Bidirectional and Auto-Regressive Transformer (BART), Bidirectional Encoder Representations from Transformers (BERT), Large Language Model Meta AI (LLaMA-2), LLaMA-3, and LLaMA-3.1 were employed to categorize data based on predefined labels. The methodology was implemented in Charleston County, South Carolina, USA. Extreme events were unevenly distributed across the training period with the “Cyclonic” category exhibiting significantly fewer instances compared to the “Flood” and “Thunderstorm” categories. Analysis suggests that the LLaMA-3 reached its peak performance at 60% of the dataset size while other LLMs achieved peak performance at approximately 80–100% of the dataset size. This study provided deep insights into the application of LLMs in reasoning adverse weather conditions.
{"title":"Can large language models effectively reason about adverse weather conditions?","authors":"Nima Zafarmomen , Vidya Samadi","doi":"10.1016/j.envsoft.2025.106421","DOIUrl":"10.1016/j.envsoft.2025.106421","url":null,"abstract":"<div><div>This paper seeks to answer the question “can Large Language Models (LLMs) effectively reason about adverse weather conditions?”. To address this question, we utilized multiple LLMs to harness the US National Weather Service (NWS) flood report data spanning from June 2005 to September 2024. Bidirectional and Auto-Regressive Transformer (BART), Bidirectional Encoder Representations from Transformers (BERT), Large Language Model Meta AI (LLaMA-2), LLaMA-3, and LLaMA-3.1 were employed to categorize data based on predefined labels. The methodology was implemented in Charleston County, South Carolina, USA. Extreme events were unevenly distributed across the training period with the “Cyclonic” category exhibiting significantly fewer instances compared to the “Flood” and “Thunderstorm” categories. Analysis suggests that the LLaMA-3 reached its peak performance at 60% of the dataset size while other LLMs achieved peak performance at approximately 80–100% of the dataset size. This study provided deep insights into the application of LLMs in reasoning adverse weather conditions.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"188 ","pages":"Article 106421"},"PeriodicalIF":4.8,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143610213","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 : 2025-03-05DOI: 10.1016/j.envsoft.2025.106424
Ao Li, Ji Li, Zhizhang Shen
Ozone pollution threatens ecosystems and human health, necessitating accurate forecasting for better management and policy implementation. To address this, we developed O3ConvNet, a convolution-based dynamic spatiotemporal deep learning model. It incorporates ModernTCN, a multivariate time series feature module, and a spatial message passing module using a dynamic adjacency matrix with geographic and DTW-based distances. O3ConvNet balances performance and efficiency across datasets with varying station densities and data qualities. In Los Angeles, the mean absolute error ranges from 6.984 μg/m3 to 15.990 μg/m3 for 1-h to 24-h predictions, with R2 values exceeding 0.937. Computational time is reduced by up to 82% compared to the best baseline model. In Wuxi, China, it improves prediction accuracy by 18% and efficiency by 81%. ModernTCN module identifies critical factors for ozone formation, while the dynamic adjacency matrix helps extract spatial dependencies effectively. Overall, this study introduces a robust and generalizable model for regional ozone predictions.
{"title":"An efficient modern convolution-based dynamic spatiotemporal deep learning architecture for ozone prediction","authors":"Ao Li, Ji Li, Zhizhang Shen","doi":"10.1016/j.envsoft.2025.106424","DOIUrl":"10.1016/j.envsoft.2025.106424","url":null,"abstract":"<div><div>Ozone pollution threatens ecosystems and human health, necessitating accurate forecasting for better management and policy implementation. To address this, we developed O3ConvNet, a convolution-based dynamic spatiotemporal deep learning model. It incorporates ModernTCN, a multivariate time series feature module, and a spatial message passing module using a dynamic adjacency matrix with geographic and DTW-based distances. O3ConvNet balances performance and efficiency across datasets with varying station densities and data qualities. In Los Angeles, the mean absolute error ranges from 6.984 μg/m<sup>3</sup> to 15.990 μg/m<sup>3</sup> for 1-h to 24-h predictions, with <em>R</em><sup>2</sup> values exceeding 0.937. Computational time is reduced by up to 82% compared to the best baseline model. In Wuxi, China, it improves prediction accuracy by 18% and efficiency by 81%. ModernTCN module identifies critical factors for ozone formation, while the dynamic adjacency matrix helps extract spatial dependencies effectively. Overall, this study introduces a robust and generalizable model for regional ozone predictions.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"188 ","pages":"Article 106424"},"PeriodicalIF":4.8,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143600157","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 : 2025-03-05DOI: 10.1016/j.envsoft.2025.106412
Junhao Wu , Xi Chen , Jinghan Dong , Nen Tan , Xiaoping Liu , Antonis Chatzipavlis , Philip LH. Yu , Adonis Velegrakis , Yining Wang , Yonggui Huang , Heqin Cheng , Diankai Wang
Dissolved oxygen (DO) is a critical parameter for monitoring water quality. However, most existing deep learning models have overlooked the physical relationship between DO and other parameters during simulation, leading to simulated values that deviate from the actual physical laws. Moreover, the inherent opacity of deep learning models restricts their applicability. Here, we propose the prior knowledge-constrained bidirectional long short-term memory network (PKBiLSTM) model to simulate DO levels in the Dianchi Lake basin. Our results show that the PKBiLSTM model achieves an average Kling-Gupta efficiency coefficient (KGE) of 0.926, which represents a 3.35% and 2.38% increase compared to the gated recurrent unit (GRU) and categorical boosting (CatBoost) models, respectively. The experiments reveal that pH has the greatest effect on DO concentration within the range of 6.5–10. Furthermore, the primary factors affecting DO exhibit seasonal differences. The findings underscore the potential of our method to enhance the scientific management of watersheds.
{"title":"Dissolved oxygen prediction in the Dianchi River basin with explainable artificial intelligence based on physical prior knowledge","authors":"Junhao Wu , Xi Chen , Jinghan Dong , Nen Tan , Xiaoping Liu , Antonis Chatzipavlis , Philip LH. Yu , Adonis Velegrakis , Yining Wang , Yonggui Huang , Heqin Cheng , Diankai Wang","doi":"10.1016/j.envsoft.2025.106412","DOIUrl":"10.1016/j.envsoft.2025.106412","url":null,"abstract":"<div><div>Dissolved oxygen (DO) is a critical parameter for monitoring water quality. However, most existing deep learning models have overlooked the physical relationship between DO and other parameters during simulation, leading to simulated values that deviate from the actual physical laws. Moreover, the inherent opacity of deep learning models restricts their applicability. Here, we propose the prior knowledge-constrained bidirectional long short-term memory network (PKBiLSTM) model to simulate DO levels in the Dianchi Lake basin. Our results show that the PKBiLSTM model achieves an average Kling-Gupta efficiency coefficient (KGE) of 0.926, which represents a 3.35% and 2.38% increase compared to the gated recurrent unit (GRU) and categorical boosting (CatBoost) models, respectively. The experiments reveal that pH has the greatest effect on DO concentration within the range of 6.5–10. Furthermore, the primary factors affecting DO exhibit seasonal differences. The findings underscore the potential of our method to enhance the scientific management of watersheds.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"188 ","pages":"Article 106412"},"PeriodicalIF":4.8,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143577046","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}
This study applies global sensitivity analysis (GSA) to the Iowa Food-Water-Energy system, focusing on nitrogen export into the Mississippi River. A binning method combined with simulation decomposition (SimDec) quantifies and visualizes the influence of crucial aggregate input variables — manure nitrogen (MN), commercial nitrogen (CN), grain nitrogen (GN), and fixation nitrogen (FN) — on nitrogen surplus (NS) at the county level. Unlike traditional Sobol’ indices, the binning method captures dependent variables. In addition, the SimDec procedure provides a detailed visual representation of how these dependencies and interactions drive the nitrogen variability. MN is identified as the most influential factor, followed by CN, with FN and GN having less impact. The study also performs GSA on the low-level input variables, enhancing the overall interpretability of the sensitivity analysis. This approach offers actionable insights for improving nitrogen management practices and contributes to GSA literature by showcasing the analysis of aggregate variables.
{"title":"Simulation decomposition analysis of the Iowa food-water-energy system","authors":"Taeho Jeong , Mariia Kozlova , Leifur Thor Leifsson , Julian Scott Yeomans","doi":"10.1016/j.envsoft.2025.106415","DOIUrl":"10.1016/j.envsoft.2025.106415","url":null,"abstract":"<div><div>This study applies global sensitivity analysis (GSA) to the Iowa Food-Water-Energy system, focusing on nitrogen export into the Mississippi River. A binning method combined with <em>simulation decomposition</em> (SimDec) quantifies and visualizes the influence of crucial aggregate input variables — manure nitrogen (MN), commercial nitrogen (CN), grain nitrogen (GN), and fixation nitrogen (FN) — on nitrogen surplus (NS) at the county level. Unlike traditional Sobol’ indices, the binning method captures dependent variables. In addition, the SimDec procedure provides a detailed visual representation of how these dependencies and interactions drive the nitrogen variability. MN is identified as the most influential factor, followed by CN, with FN and GN having less impact. The study also performs GSA on the low-level input variables, enhancing the overall interpretability of the sensitivity analysis. This approach offers actionable insights for improving nitrogen management practices and contributes to GSA literature by showcasing the analysis of aggregate variables.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"188 ","pages":"Article 106415"},"PeriodicalIF":4.8,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143577047","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 : 2025-03-04DOI: 10.1016/j.envsoft.2025.106418
Bekir Z. Demiray , Yusuf Sermet , Enes Yildirim , Ibrahim Demir
The number and devastating impacts of natural disasters have grown significantly worldwide, and floods are one of the most dangerous and frequent natural disasters. Recent studies emphasize the importance of public awareness in disaster preparedness and response activities. FloodGame is designed as a web-based interactive serious game geared towards educating K-12 and college students and raising public awareness on flood prevention and mitigation strategies so that they are more informed about the implications of future floods. A web-based interactive gaming environment with rich 3D visuals and models is developed that allows users to experiment with different flood mitigation strategies for a real-world location of their choice. This immersive, repeatable, and engaging experience will allow students and the public to comprehend the consequences of individual mitigation measures, build a conceptual understanding of the benefits of mitigation actions, and examine how floods may occur in their communities.
{"title":"FloodGame: An interactive 3D serious game on flood mitigation for disaster awareness and education","authors":"Bekir Z. Demiray , Yusuf Sermet , Enes Yildirim , Ibrahim Demir","doi":"10.1016/j.envsoft.2025.106418","DOIUrl":"10.1016/j.envsoft.2025.106418","url":null,"abstract":"<div><div>The number and devastating impacts of natural disasters have grown significantly worldwide, and floods are one of the most dangerous and frequent natural disasters. Recent studies emphasize the importance of public awareness in disaster preparedness and response activities. FloodGame is designed as a web-based interactive serious game geared towards educating K-12 and college students and raising public awareness on flood prevention and mitigation strategies so that they are more informed about the implications of future floods. A web-based interactive gaming environment with rich 3D visuals and models is developed that allows users to experiment with different flood mitigation strategies for a real-world location of their choice. This immersive, repeatable, and engaging experience will allow students and the public to comprehend the consequences of individual mitigation measures, build a conceptual understanding of the benefits of mitigation actions, and examine how floods may occur in their communities.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"188 ","pages":"Article 106418"},"PeriodicalIF":4.8,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143629600","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 : 2025-03-04DOI: 10.1016/j.envsoft.2025.106416
Chuanda Wang , Wenjiao Duan , Shuiyuan Cheng , Xiaosong Hou , Junfeng Zhang , Yu Wang , Hanyu Zhang , Kai Wang , Rui Liu
A novel algorithm for generating multi-component, high-resolution emission inventories for air quality models (AQMs) was developed, enhancing the 0.01° spatial allocation scheme for points, lines, and surfaces based on land use, population density, and road networks. It incorporated localized chemical species allocation based on recent Volatile Organic Compounds (VOCs) and Particulate Matter (PM) composition, along with time allocation scheme utilizing monthly and hourly non-uniformity coefficients. The algorithm consists of four modules for grid calculation, refined inventory calculation, coarse grid interpolation, and model-ready inventory production. It achieved highly integrated and one-step workflow from raw parameter settings to direct model-ready emission files for simulations with CMAQ and CAMx. The Beijing-Tianjin-Hebei (BTH) region was used as example to introduce the production process and research characteristics of refined emission inventory. The results demonstrated that this algorithm effectively captures the spatiotemporal distribution and dynamics of atmospheric pollutants, offering significant support for emission and simulation research.
{"title":"Gridemis V2.0: A highly integrated algorithm scheme for high-resolution and multi-component allocation of emission inventories used in air quality models","authors":"Chuanda Wang , Wenjiao Duan , Shuiyuan Cheng , Xiaosong Hou , Junfeng Zhang , Yu Wang , Hanyu Zhang , Kai Wang , Rui Liu","doi":"10.1016/j.envsoft.2025.106416","DOIUrl":"10.1016/j.envsoft.2025.106416","url":null,"abstract":"<div><div>A novel algorithm for generating multi-component, high-resolution emission inventories for air quality models (AQMs) was developed, enhancing the 0.01° spatial allocation scheme for points, lines, and surfaces based on land use, population density, and road networks. It incorporated localized chemical species allocation based on recent Volatile Organic Compounds (VOCs) and Particulate Matter (PM) composition, along with time allocation scheme utilizing monthly and hourly non-uniformity coefficients. The algorithm consists of four modules for grid calculation, refined inventory calculation, coarse grid interpolation, and model-ready inventory production. It achieved highly integrated and one-step workflow from raw parameter settings to direct model-ready emission files for simulations with CMAQ and CAMx. The Beijing-Tianjin-Hebei (BTH) region was used as example to introduce the production process and research characteristics of refined emission inventory. The results demonstrated that this algorithm effectively captures the spatiotemporal distribution and dynamics of atmospheric pollutants, offering significant support for emission and simulation research.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"188 ","pages":"Article 106416"},"PeriodicalIF":4.8,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143643038","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 : 2025-03-04DOI: 10.1016/j.envsoft.2025.106417
Arman Haddadchi , Calvin W. Rose
This paper introduces a new high temporal resolution suspended sediment routing model that integrates fine sediment deposition and re-entrainment processes of individual size fractions with suspended sediment transport throughout the river network. This multi-size fraction model provides unique insights into the effects of sediment size classes on key sediment attributes, including suspended sediment size distribution and total concentration, sediment load, and changes in riverbed deposited fine sediment throughout the river network. Data from three high-frequency flow and sediment monitoring sites on the Manawatū River, New Zealand, were used to test the model. After calibrating the model using data from one flood event, it showed good agreement between observed and modelled fine sediment concentration and event load for five subsequent test events. When coupled with catchment erosion predictions, this new sediment routing model should provide valuable information on sediment sources and connectivity within catchments for informing catchment management.
{"title":"An advection-dispersion model for routing suspended sediment down the river network","authors":"Arman Haddadchi , Calvin W. Rose","doi":"10.1016/j.envsoft.2025.106417","DOIUrl":"10.1016/j.envsoft.2025.106417","url":null,"abstract":"<div><div>This paper introduces a new high temporal resolution suspended sediment routing model that integrates fine sediment deposition and re-entrainment processes of individual size fractions with suspended sediment transport throughout the river network. This multi-size fraction model provides unique insights into the effects of sediment size classes on key sediment attributes, including suspended sediment size distribution and total concentration, sediment load, and changes in riverbed deposited fine sediment throughout the river network. Data from three high-frequency flow and sediment monitoring sites on the Manawatū River, New Zealand, were used to test the model. After calibrating the model using data from one flood event, it showed good agreement between observed and modelled fine sediment concentration and event load for five subsequent test events. When coupled with catchment erosion predictions, this new sediment routing model should provide valuable information on sediment sources and connectivity within catchments for informing catchment management.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"188 ","pages":"Article 106417"},"PeriodicalIF":4.8,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143577050","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 : 2025-03-03DOI: 10.1016/j.envsoft.2025.106400
Bo Zhang , Hongsheng Qin , Yuqi Zhang , Maozhen Li , Dongming Qin , Xiaoyang Guo , Meizi Li , Chang Guo
Air pollution problem seriously affects the ecological environment and human health. More accurate predictions over a longer time span would enhance the effectiveness of early warning and prevention measures. Although existing methods have made progress in short sequence prediction, the predictions on long sequences remain challenges due to information loss. In this paper, we propose a spatial–temporal graph-based long sequence air pollutant prediction model. The proposed model first downsamples the time series into different granularities to capture the temporal features. Then, we use the vector production method to construct a spatial–temporal graph for each granularity which combines spatial information with temporal information. The unique spatial–temporal relationships of each city under different time granularities can be extracted by graph attention network (GAT). This approach helps model to capture dependencies in the time series comprehensively, thereby improving the accuracy of long sequence prediction. Based on the scenario and air pollution datasets imported from the detection station in Shanghai, extensive experiments show that the proposed model outperforms existing approaches on MSE and MAE.
{"title":"Multi-granularity PM2.5 concentration long sequence prediction model combined with spatial–temporal graph","authors":"Bo Zhang , Hongsheng Qin , Yuqi Zhang , Maozhen Li , Dongming Qin , Xiaoyang Guo , Meizi Li , Chang Guo","doi":"10.1016/j.envsoft.2025.106400","DOIUrl":"10.1016/j.envsoft.2025.106400","url":null,"abstract":"<div><div>Air pollution problem seriously affects the ecological environment and human health. More accurate predictions over a longer time span would enhance the effectiveness of early warning and prevention measures. Although existing methods have made progress in short sequence prediction, the predictions on long sequences remain challenges due to information loss. In this paper, we propose a spatial–temporal graph-based long sequence air pollutant prediction model. The proposed model first downsamples the time series into different granularities to capture the temporal features. Then, we use the vector production method to construct a spatial–temporal graph for each granularity which combines spatial information with temporal information. The unique spatial–temporal relationships of each city under different time granularities can be extracted by graph attention network (GAT). This approach helps model to capture dependencies in the time series comprehensively, thereby improving the accuracy of long sequence prediction. Based on the scenario and air pollution datasets imported from the detection station in Shanghai, extensive experiments show that the proposed model outperforms existing approaches on MSE and MAE.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"188 ","pages":"Article 106400"},"PeriodicalIF":4.8,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143562364","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}
PlotToSat offers a practical and time efficient way to the challenge of extracting time-series from multiple Earth Observation (EO) datasets at numerous plots spread across a landscape. This opens up new opportunities to understand and model various ecosystems. Regarding forest ecology, plot networks play a vital role in monitoring and understanding the dynamics of forest ecosystems. These networks often contain thousands of plots arranged systematically to represent an ecosystem. Combining field data collected at plots with EO time-series will allow us to better understand phenology and ecosystem composition, structure and distribution. Linking plot networks with EO data without PlotToSat is time consuming and computational expensive because plots are small and spread out, requiring data from multiple satellite tiles. PlotToSat processed a full year of multi-tile Sentinel-1 and Sentinel-2 data (estimated 18.3TB) at 15,962 plots from the fourth Spanish Forest Inventory in less than 24 h. PlotToSat, implemented using the Python API of Google Earth Engine, offers a new and unique workflow that is innovative due to its efficient, scalable and adaptable implementation. It supports Sentinel-1 and Sentinel-2 data, but its flexible design eases integration of additional EO datasets. New environmental modelling is expected to emerge facilitating EO time-series analyses and investigating interactive effects of environmental drivers.
{"title":"PlotToSat: A tool for generating time-series signatures from Sentinel-1 and Sentinel-2 at field-based plots for machine learning applications","authors":"Milto Miltiadou , Stuart Grieve , Paloma Ruiz-Benito , Julen Astigarraga , Verónica Cruz-Alonso , Julián Tijerín Triviño , Emily R. Lines","doi":"10.1016/j.envsoft.2025.106395","DOIUrl":"10.1016/j.envsoft.2025.106395","url":null,"abstract":"<div><div>PlotToSat offers a practical and time efficient way to the challenge of extracting time-series from multiple Earth Observation (EO) datasets at numerous plots spread across a landscape. This opens up new opportunities to understand and model various ecosystems. Regarding forest ecology, plot networks play a vital role in monitoring and understanding the dynamics of forest ecosystems. These networks often contain thousands of plots arranged systematically to represent an ecosystem. Combining field data collected at plots with EO time-series will allow us to better understand phenology and ecosystem composition, structure and distribution. Linking plot networks with EO data without PlotToSat is time consuming and computational expensive because plots are small and spread out, requiring data from multiple satellite tiles. PlotToSat processed a full year of multi-tile Sentinel-1 and Sentinel-2 data (estimated 18.3TB) at 15,962 plots from the fourth Spanish Forest Inventory in less than 24 h. PlotToSat, implemented using the Python API of Google Earth Engine, offers a new and unique workflow that is innovative due to its efficient, scalable and adaptable implementation. It supports Sentinel-1 and Sentinel-2 data, but its flexible design eases integration of additional EO datasets. New environmental modelling is expected to emerge facilitating EO time-series analyses and investigating interactive effects of environmental drivers.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"188 ","pages":"Article 106395"},"PeriodicalIF":4.8,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592243","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 : 2025-03-01DOI: 10.1016/j.envsoft.2025.106397
Elise Mills , Graeme F. Clark , Matthew J. Simpson , Mark Baird , Matthew P. Adams
Sigmoid growth models are often used to study population dynamics. The size of a population at equilibrium commonly depends explicitly on the availability of resources, such as an energy or nutrient source, which is not explicit in standard sigmoid growth models. A simple generalised extension of sigmoid growth models is introduced that can explicitly account for this resource-dependence, demonstrated by three examples of this family of models of increasing mathematical complexity. Each model is calibrated and compared to observed data for algae under sea-ice in Antarctic coastal waters. It was found that through careful construction, models satisfying the proposed framework can estimate key properties of a sea-ice break-out controlled tipping point for the algae, which cannot be estimated using standard sigmoid growth models. The proposed broader family of energy-dependent sigmoid growth models likely has usage in many population growth contexts where resources limit population size.
{"title":"A generalised sigmoid population growth model with energy dependence: Application to quantify the tipping point for Antarctic shallow seabed algae","authors":"Elise Mills , Graeme F. Clark , Matthew J. Simpson , Mark Baird , Matthew P. Adams","doi":"10.1016/j.envsoft.2025.106397","DOIUrl":"10.1016/j.envsoft.2025.106397","url":null,"abstract":"<div><div>Sigmoid growth models are often used to study population dynamics. The size of a population at equilibrium commonly depends explicitly on the availability of resources, such as an energy or nutrient source, which is not explicit in standard sigmoid growth models. A simple generalised extension of sigmoid growth models is introduced that can explicitly account for this resource-dependence, demonstrated by three examples of this family of models of increasing mathematical complexity. Each model is calibrated and compared to observed data for algae under sea-ice in Antarctic coastal waters. It was found that through careful construction, models satisfying the proposed framework can estimate key properties of a sea-ice break-out controlled tipping point for the algae, which cannot be estimated using standard sigmoid growth models. The proposed broader family of energy-dependent sigmoid growth models likely has usage in many population growth contexts where resources limit population size.</div></div>","PeriodicalId":310,"journal":{"name":"Environmental Modelling & Software","volume":"188 ","pages":"Article 106397"},"PeriodicalIF":4.8,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143577049","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号