Bulusu Subrahmanyam, V. S. N. Murty, Sarah B. Hall, Corinne B. Trott
We used NASA’s high-resolution (1/48° or 2.3 km, hourly) Estimating the Circulation and Climate of the Ocean (ECCO) estimates of salinity at a 1 m depth from November 2011 to October 2012 to detect semi-diurnal and diurnal internal tides (ITs) in the Andaman Sea and determine their characteristics in three 2° × 2° boxes off the Myanmar coast (box A), central Andaman Sea (box B), and off the Thailand coast (box C). We also used observed salinity and temperature data for the above period at the BD12-moored buoy in the central Andaman Sea. ECCO salinity data were bandpass-filtered with 11–14 h and 22–26 h periods. Large variations in filtered ECCO salinity (~0.1 psu) in the boxes corresponded with near-surface imprints of propagating ITs. Observed data from the box B domain reveals strong salinity stratification (halocline) in the upper 40 m. Our analyses reveal that the shallow halocline affects the signatures of propagating semi-diurnal ITs reaching the surface, but diurnal ITs propagating in the halocline reach up to the surface and bring variability in ECCO salinity. In box A, the semi-diurnal IT characteristics are higher speeds (0.96 m/s) with larger wavelengths (45 km), that are closer to theoretical mode 2 estimates, but the diurnal ITs propagating in the box A domain, with a possible source over the shelf of Gulf of Martaban, attain lower values (0.45 m/s, 38 km). In box B, the propagation speed is lower (higher) for semi-diurnal (diurnal) ITs. Estimates for box C are closer to those for box A.
我们利用美国国家航空航天局（NASA）2011 年 11 月至 2012 年 10 月的高分辨率（1/48°或 2.3 千米，每小时）海洋环流和气候估算（ECCO）1 米深度的盐度估算数据，探测安达曼海的半日和昼夜内潮（ITs），并确定其在缅甸沿岸（方框 A）、安达曼海中部（方框 B）和泰国沿岸（方框 C）三个 2°×2° 方框内的特征。我们还使用了安达曼海中部 BD12 系泊浮标在上述期间的盐度和温度观测数据。ECCO 盐度数据经过带通滤波，周期分别为 11-14 小时和 22-26 小时。滤波后的 ECCO 盐度在方框内有较大变化（约 0.1 psu），与传播的 ITs 的近海面印迹相吻合。我们的分析表明，浅层卤化线影响了到达海面的半日流 IT 的传播特征，但在卤化线中传播的日流 IT 可以到达海面并带来 ECCO 盐度的变化。在方框 A 中，半昼夜 IT 的特征是传播速度较快（0.96 米/秒），波长较大（45 千米），更接近模式 2 的理论估计值，但在方框 A 域传播的昼夜 IT 值较低 （0.45 米/秒，38 千米），其来源可能在马塔班湾大陆架上。在方框 B 中，半日（昼）IT 传播速度较低（较高）。C 框的估计值与 A 框的估计值较为接近。
{"title":"Identification of Internal Tides in ECCO Estimates of Sea Surface Salinity in the Andaman Sea","authors":"Bulusu Subrahmanyam, V. S. N. Murty, Sarah B. Hall, Corinne B. Trott","doi":"10.3390/rs16183408","DOIUrl":"https://doi.org/10.3390/rs16183408","url":null,"abstract":"We used NASA’s high-resolution (1/48° or 2.3 km, hourly) Estimating the Circulation and Climate of the Ocean (ECCO) estimates of salinity at a 1 m depth from November 2011 to October 2012 to detect semi-diurnal and diurnal internal tides (ITs) in the Andaman Sea and determine their characteristics in three 2° × 2° boxes off the Myanmar coast (box A), central Andaman Sea (box B), and off the Thailand coast (box C). We also used observed salinity and temperature data for the above period at the BD12-moored buoy in the central Andaman Sea. ECCO salinity data were bandpass-filtered with 11–14 h and 22–26 h periods. Large variations in filtered ECCO salinity (~0.1 psu) in the boxes corresponded with near-surface imprints of propagating ITs. Observed data from the box B domain reveals strong salinity stratification (halocline) in the upper 40 m. Our analyses reveal that the shallow halocline affects the signatures of propagating semi-diurnal ITs reaching the surface, but diurnal ITs propagating in the halocline reach up to the surface and bring variability in ECCO salinity. In box A, the semi-diurnal IT characteristics are higher speeds (0.96 m/s) with larger wavelengths (45 km), that are closer to theoretical mode 2 estimates, but the diurnal ITs propagating in the box A domain, with a possible source over the shelf of Gulf of Martaban, attain lower values (0.45 m/s, 38 km). In box B, the propagation speed is lower (higher) for semi-diurnal (diurnal) ITs. Estimates for box C are closer to those for box A.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"734 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142200456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lien Rodríguez-López, Lisandra Bravo Alvarez, Iongel Duran-Llacer, David E. Ruíz-Guirola, Samuel Montejo-Sánchez, Rebeca Martínez-Retureta, Ernesto López-Morales, Luc Bourrel, Frédéric Frappart, Roberto Urrutia
This study examines the dynamics of limnological parameters of a South American lake located in southern Chile with the objective of predicting chlorophyll-a levels, which are a key indicator of algal biomass and water quality, by integrating combined remote sensing and machine learning techniques. Employing four advanced machine learning models (recurrent neural network (RNNs), long short-term memory (LSTM), recurrent gate unit (GRU), and temporal convolutional network (TCNs)), the research focuses on the estimation of chlorophyll-a concentrations at three sampling stations within Lake Ranco. The data span from 1987 to 2020 and are used in three different cases: using only in situ data (Case 1), using in situ and meteorological data (Case 2), using in situ, and meteorological and satellite data from Landsat and Sentinel missions (Case 3). In all cases, each machine learning model shows robust performance, with promising results in predicting chlorophyll-a concentrations. Among these models, LSTM stands out as the most effective, with the best metrics in the estimation, the best performance was Case 1, with R2 = 0.89, an RSME of 0.32 µg/L, an MAE 1.25 µg/L and an MSE 0.25 (µg/L)2, consistently outperforming the others according to the static metrics used for validation. This finding underscores the effectiveness of LSTM in capturing the complex temporal relationships inherent in the dataset. However, increasing the dataset in Case 3 shows a better performance of TCNs (R2 = 0.96; MSE = 0.33 (µg/L)2; RMSE = 0.13 µg/L; and MAE = 0.06 µg/L). The successful application of machine learning algorithms emphasizes their potential to elucidate the dynamics of algal biomass in Lake Ranco, located in the southern region of Chile. These results not only contribute to a deeper understanding of the lake ecosystem but also highlight the utility of advanced computational techniques in environmental research and management.
{"title":"Leveraging Machine Learning and Remote Sensing for Water Quality Analysis in Lake Ranco, Southern Chile","authors":"Lien Rodríguez-López, Lisandra Bravo Alvarez, Iongel Duran-Llacer, David E. Ruíz-Guirola, Samuel Montejo-Sánchez, Rebeca Martínez-Retureta, Ernesto López-Morales, Luc Bourrel, Frédéric Frappart, Roberto Urrutia","doi":"10.3390/rs16183401","DOIUrl":"https://doi.org/10.3390/rs16183401","url":null,"abstract":"This study examines the dynamics of limnological parameters of a South American lake located in southern Chile with the objective of predicting chlorophyll-a levels, which are a key indicator of algal biomass and water quality, by integrating combined remote sensing and machine learning techniques. Employing four advanced machine learning models (recurrent neural network (RNNs), long short-term memory (LSTM), recurrent gate unit (GRU), and temporal convolutional network (TCNs)), the research focuses on the estimation of chlorophyll-a concentrations at three sampling stations within Lake Ranco. The data span from 1987 to 2020 and are used in three different cases: using only in situ data (Case 1), using in situ and meteorological data (Case 2), using in situ, and meteorological and satellite data from Landsat and Sentinel missions (Case 3). In all cases, each machine learning model shows robust performance, with promising results in predicting chlorophyll-a concentrations. Among these models, LSTM stands out as the most effective, with the best metrics in the estimation, the best performance was Case 1, with R2 = 0.89, an RSME of 0.32 µg/L, an MAE 1.25 µg/L and an MSE 0.25 (µg/L)2, consistently outperforming the others according to the static metrics used for validation. This finding underscores the effectiveness of LSTM in capturing the complex temporal relationships inherent in the dataset. However, increasing the dataset in Case 3 shows a better performance of TCNs (R2 = 0.96; MSE = 0.33 (µg/L)2; RMSE = 0.13 µg/L; and MAE = 0.06 µg/L). The successful application of machine learning algorithms emphasizes their potential to elucidate the dynamics of algal biomass in Lake Ranco, located in the southern region of Chile. These results not only contribute to a deeper understanding of the lake ecosystem but also highlight the utility of advanced computational techniques in environmental research and management.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"38 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142200457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xiaoyuan Zhang, Shudong Wang, Kai Liu, Xiankai Huang, Jinlian Shi, Xueke Li
Exploring the dynamic response of land use and ecological vulnerability (EV) to future climate change and human ecological restoration policies is crucial for optimizing regional ecosystem services and formulating sustainable socioeconomic development strategies. This study comprehensively assesses future land use changes and EV in the Yellow River Basin (YRB), a climate-sensitive and ecologically fragile area, by integrating climate change, land management, and ecological protection policies under various scenarios. To achieve this, we developed an EV assessment framework combining a scenario weight matrix, Markov chain, Patch-generating Land Use Simulation model, and exposure–sensitivity–adaptation. We further explored the spatiotemporal variations of EV and their potential socioeconomic impacts at the watershed scale. Our results show significant geospatial variations in future EV under the three scenarios, with the northern region of the upstream area being the most severely affected. Under the ecological conservation management scenario and historical trend scenario, the ecological environment of the basin improves, with a decrease in very high vulnerability areas by 4.45% and 3.08%, respectively, due to the protection and restoration of ecological land. Conversely, under the urban development and construction scenario, intensified climate change and increased land use artificialization exacerbate EV, with medium and high vulnerability areas increasing by 1.86% and 7.78%, respectively. The population in high and very high vulnerability areas is projected to constitute 32.75–33.68% and 34.59–39.21% of the YRB’s total population in 2040 and 2060, respectively, and may continue to grow. Overall, our scenario analysis effectively demonstrates the positive impact of ecological protection on reducing EV and the negative impact of urban expansion and economic development on increasing EV. Our work offers new insights into land resource allocation and the development of ecological restoration policies.
探索土地利用和生态脆弱性（EV）对未来气候变化和人类生态恢复政策的动态响应，对于优化区域生态系统服务和制定可持续的社会经济发展战略至关重要。黄河流域是一个气候敏感、生态脆弱的地区，本研究通过整合各种情景下的气候变化、土地管理和生态保护政策，全面评估了黄河流域未来的土地利用变化和生态脆弱性。为此，我们开发了一个结合情景权重矩阵、马尔科夫链、斑块生成土地利用模拟模型和暴露-敏感-适应的 EV 评估框架。我们进一步探讨了流域尺度上的 EV 时空变化及其潜在的社会经济影响。我们的研究结果表明，在三种情景下，未来电动汽车的时空变化非常明显，其中上游北部地区受到的影响最为严重。在生态保护管理情景和历史趋势情景下，由于生态用地的保护和恢复，流域生态环境有所改善，极高脆弱性区域分别减少了 4.45% 和 3.08%。相反，在城市发展和建设情景下，气候变化加剧，土地利用人工化程度提高，加剧了环境脆弱程度，中度和高度脆弱地区分别增加了 1.86% 和 7.78%。预计到 2040 年和 2060 年，高脆弱区和极高脆弱区的人口将分别占长三角地区总人口的 32.75%-33.68% 和 34.59-39.21%，并可能继续增长。总体而言，我们的情景分析有效地证明了生态保护对减少电动汽车的积极影响，以及城市扩张和经济发展对增加电动汽车的消极影响。我们的工作为土地资源分配和生态恢复政策的制定提供了新的见解。
{"title":"Projecting Response of Ecological Vulnerability to Future Climate Change and Human Policies in the Yellow River Basin, China","authors":"Xiaoyuan Zhang, Shudong Wang, Kai Liu, Xiankai Huang, Jinlian Shi, Xueke Li","doi":"10.3390/rs16183410","DOIUrl":"https://doi.org/10.3390/rs16183410","url":null,"abstract":"Exploring the dynamic response of land use and ecological vulnerability (EV) to future climate change and human ecological restoration policies is crucial for optimizing regional ecosystem services and formulating sustainable socioeconomic development strategies. This study comprehensively assesses future land use changes and EV in the Yellow River Basin (YRB), a climate-sensitive and ecologically fragile area, by integrating climate change, land management, and ecological protection policies under various scenarios. To achieve this, we developed an EV assessment framework combining a scenario weight matrix, Markov chain, Patch-generating Land Use Simulation model, and exposure–sensitivity–adaptation. We further explored the spatiotemporal variations of EV and their potential socioeconomic impacts at the watershed scale. Our results show significant geospatial variations in future EV under the three scenarios, with the northern region of the upstream area being the most severely affected. Under the ecological conservation management scenario and historical trend scenario, the ecological environment of the basin improves, with a decrease in very high vulnerability areas by 4.45% and 3.08%, respectively, due to the protection and restoration of ecological land. Conversely, under the urban development and construction scenario, intensified climate change and increased land use artificialization exacerbate EV, with medium and high vulnerability areas increasing by 1.86% and 7.78%, respectively. The population in high and very high vulnerability areas is projected to constitute 32.75–33.68% and 34.59–39.21% of the YRB’s total population in 2040 and 2060, respectively, and may continue to grow. Overall, our scenario analysis effectively demonstrates the positive impact of ecological protection on reducing EV and the negative impact of urban expansion and economic development on increasing EV. Our work offers new insights into land resource allocation and the development of ecological restoration policies.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"7 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142200463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The atmosphere is a complex nonlinear system, with the information of its temperature, water vapor, pressure, and cloud being crucial aspects of remote-sensing data analysis. There exist intricate interactions among these internal components, such as convection, radiation, and humidity exchange. Atmospheric phenomena span multiple spatial and temporal scales, from small-scale thunderstorms to large-scale events like El Niño. The dynamic interactions across different scales, along with external disturbances to the atmospheric system, such as variations in solar radiation and Earth surface conditions, contribute to the chaotic nature of the atmosphere, making long-term predictions challenging. Grasping the intrinsic chaotic dynamics is essential for advancing atmospheric analysis, which holds profound implications for enhancing meteorological forecasts, mitigating disaster risks, and safeguarding ecological systems. To validate the chaotic nature of the atmosphere, this paper reviewed the definitions and main features of chaotic systems, elucidated the method of phase space reconstruction centered on Takens’ theorem, and categorized the qualitative and quantitative methods for determining the chaotic nature of time series data. Among quantitative methods, the Wolf method is used to calculate the Largest Lyapunov Exponents, while the G–P method is used to calculate the correlation dimensions. A new method named Improved Saturated Correlation Dimension method was proposed to address the subjectivity and noise sensitivity inherent in the traditional G–P method. Subsequently, the Largest Lyapunov Exponents and saturated correlation dimensions were utilized to conduct a quantitative analysis of FY-4A and Himawari-8 remote-sensing infrared observation data, and ERA5 reanalysis data. For both short-term remote-sensing data and long-term reanalysis data, the results showed that more than 99.91% of the regional points have corresponding sequences with positive Largest Lyapunov exponents and all the regional points have correlation dimensions that tended to saturate at values greater than 1 with increasing embedding dimensions, thereby proving that the atmospheric system exhibits chaotic properties on both short and long temporal scales, with extreme sensitivity to initial conditions. This conclusion provided a theoretical foundation for the short-term prediction of atmospheric infrared radiation field variables and the detection of weak, time-sensitive signals in complex atmospheric environments.
{"title":"Research on Multiscale Atmospheric Chaos Based on Infrared Remote-Sensing and Reanalysis Data","authors":"Zhong Wang, Shengli Sun, Wenjun Xu, Rui Chen, Yijun Ma, Gaorui Liu","doi":"10.3390/rs16183376","DOIUrl":"https://doi.org/10.3390/rs16183376","url":null,"abstract":"The atmosphere is a complex nonlinear system, with the information of its temperature, water vapor, pressure, and cloud being crucial aspects of remote-sensing data analysis. There exist intricate interactions among these internal components, such as convection, radiation, and humidity exchange. Atmospheric phenomena span multiple spatial and temporal scales, from small-scale thunderstorms to large-scale events like El Niño. The dynamic interactions across different scales, along with external disturbances to the atmospheric system, such as variations in solar radiation and Earth surface conditions, contribute to the chaotic nature of the atmosphere, making long-term predictions challenging. Grasping the intrinsic chaotic dynamics is essential for advancing atmospheric analysis, which holds profound implications for enhancing meteorological forecasts, mitigating disaster risks, and safeguarding ecological systems. To validate the chaotic nature of the atmosphere, this paper reviewed the definitions and main features of chaotic systems, elucidated the method of phase space reconstruction centered on Takens’ theorem, and categorized the qualitative and quantitative methods for determining the chaotic nature of time series data. Among quantitative methods, the Wolf method is used to calculate the Largest Lyapunov Exponents, while the G–P method is used to calculate the correlation dimensions. A new method named Improved Saturated Correlation Dimension method was proposed to address the subjectivity and noise sensitivity inherent in the traditional G–P method. Subsequently, the Largest Lyapunov Exponents and saturated correlation dimensions were utilized to conduct a quantitative analysis of FY-4A and Himawari-8 remote-sensing infrared observation data, and ERA5 reanalysis data. For both short-term remote-sensing data and long-term reanalysis data, the results showed that more than 99.91% of the regional points have corresponding sequences with positive Largest Lyapunov exponents and all the regional points have correlation dimensions that tended to saturate at values greater than 1 with increasing embedding dimensions, thereby proving that the atmospheric system exhibits chaotic properties on both short and long temporal scales, with extreme sensitivity to initial conditions. This conclusion provided a theoretical foundation for the short-term prediction of atmospheric infrared radiation field variables and the detection of weak, time-sensitive signals in complex atmospheric environments.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"165 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142200445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Angelly de Jesus Pugliese Viloria, Andrea Folini, Daniela Carrion, Maria Antonia Brovelli
With the increase in climate-change-related hazardous events alongside population concentration in urban centres, it is important to provide resilient cities with tools for understanding and eventually preparing for such events. Machine learning (ML) and deep learning (DL) techniques have increasingly been employed to model susceptibility of hazardous events. This study consists of a systematic review of the ML/DL techniques applied to model the susceptibility of air pollution, urban heat islands, floods, and landslides, with the aim of providing a comprehensive source of reference both for techniques and modelling approaches. A total of 1454 articles published between 2020 and 2023 were systematically selected from the Scopus and Web of Science search engines based on search queries and selection criteria. ML/DL techniques were extracted from the selected articles and categorised using ad hoc classification. Consequently, a general approach for modelling the susceptibility of hazardous events was consolidated, covering the data preprocessing, feature selection, modelling, model interpretation, and susceptibility map validation, along with examples of related global/continental data. The most frequently employed techniques across various hazards include random forest, artificial neural networks, and support vector machines. This review also provides, per hazard, the definition, data requirements, and insights into the ML/DL techniques used, including examples of both state-of-the-art and novel modelling approaches.
随着气候变化相关危险事件的增加以及城市中心人口的集中，为具有抗灾能力的城市提供了解并最终准备应对此类事件的工具非常重要。人们越来越多地采用机器学习（ML）和深度学习（DL）技术来模拟危险事件的易感性。本研究对应用于空气污染、城市热岛、洪水和山体滑坡易发性建模的 ML/DL 技术进行了系统回顾，旨在为技术和建模方法提供全面的参考来源。根据搜索查询和选择标准，从 Scopus 和 Web of Science 搜索引擎中系统地选取了 2020 至 2023 年间发表的 1454 篇文章。从所选文章中提取了 ML/DL 技术，并使用特别分类法进行了分类。因此，整合了危险事件易感性建模的一般方法，包括数据预处理、特征选择、建模、模型解释和易感性地图验证，以及相关的全球/大陆数据示例。在各种灾害中最常用的技术包括随机森林、人工神经网络和支持向量机。本综述还提供了每种灾害的定义、数据要求以及对所用 ML/DL 技术的见解，包括最新建模方法和新型建模方法的示例。
{"title":"Hazard Susceptibility Mapping with Machine and Deep Learning: A Literature Review","authors":"Angelly de Jesus Pugliese Viloria, Andrea Folini, Daniela Carrion, Maria Antonia Brovelli","doi":"10.3390/rs16183374","DOIUrl":"https://doi.org/10.3390/rs16183374","url":null,"abstract":"With the increase in climate-change-related hazardous events alongside population concentration in urban centres, it is important to provide resilient cities with tools for understanding and eventually preparing for such events. Machine learning (ML) and deep learning (DL) techniques have increasingly been employed to model susceptibility of hazardous events. This study consists of a systematic review of the ML/DL techniques applied to model the susceptibility of air pollution, urban heat islands, floods, and landslides, with the aim of providing a comprehensive source of reference both for techniques and modelling approaches. A total of 1454 articles published between 2020 and 2023 were systematically selected from the Scopus and Web of Science search engines based on search queries and selection criteria. ML/DL techniques were extracted from the selected articles and categorised using ad hoc classification. Consequently, a general approach for modelling the susceptibility of hazardous events was consolidated, covering the data preprocessing, feature selection, modelling, model interpretation, and susceptibility map validation, along with examples of related global/continental data. The most frequently employed techniques across various hazards include random forest, artificial neural networks, and support vector machines. This review also provides, per hazard, the definition, data requirements, and insights into the ML/DL techniques used, including examples of both state-of-the-art and novel modelling approaches.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"13 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142200480","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Human settlement areas significantly impact the environment, leading to changes in both natural and built environments. Comprehensive information on human settlements, particularly in urban areas, is crucial for effective sustainable development planning. However, urban land use investigations are often limited to two-dimensional building footprint maps, neglecting the three-dimensional aspect of building structures. This paper addresses this issue to contribute to Sustainable Development Goal 11, which focuses on making human settlements inclusive, safe, and sustainable. In this study, Sentinel-1 data are used as the primary source to estimate building heights. One challenge addressed is the issue of multiple backscattering in Sentinel-1’s signal, particularly in densely populated areas with high-rise buildings. To mitigate this, firstly, Sentinel-1 data from different directions, orbit paths, and polarizations are utilized. Combining ascending and descending orbits significantly improves estimation accuracy, and incorporating a higher number of paths provides additional information. However, Sentinel-1 data alone are not sufficiently rich at a global scale across different orbits and polarizations. Secondly, to enhance the accuracy further, Sentinel-1 data are corrected using nighttime light data as additional information, which shows promising results in addressing multiple backscattering issues. Finally, a deep learning model is trained to generate building height maps using these features, achieving a mean absolute error of around 2 m and a mean square error of approximately 13. The generalizability of this method is demonstrated in several cities with diverse built-up structures, including London, Berlin, and others. Finally, a building height map of Iran is generated and evaluated against surveyed buildings, showcasing its large-scale mapping capability.
{"title":"Mapping Building Heights at Large Scales Using Sentinel-1 Radar Imagery and Nighttime Light Data","authors":"Mohammad Kakooei, Yasser Baleghi","doi":"10.3390/rs16183371","DOIUrl":"https://doi.org/10.3390/rs16183371","url":null,"abstract":"Human settlement areas significantly impact the environment, leading to changes in both natural and built environments. Comprehensive information on human settlements, particularly in urban areas, is crucial for effective sustainable development planning. However, urban land use investigations are often limited to two-dimensional building footprint maps, neglecting the three-dimensional aspect of building structures. This paper addresses this issue to contribute to Sustainable Development Goal 11, which focuses on making human settlements inclusive, safe, and sustainable. In this study, Sentinel-1 data are used as the primary source to estimate building heights. One challenge addressed is the issue of multiple backscattering in Sentinel-1’s signal, particularly in densely populated areas with high-rise buildings. To mitigate this, firstly, Sentinel-1 data from different directions, orbit paths, and polarizations are utilized. Combining ascending and descending orbits significantly improves estimation accuracy, and incorporating a higher number of paths provides additional information. However, Sentinel-1 data alone are not sufficiently rich at a global scale across different orbits and polarizations. Secondly, to enhance the accuracy further, Sentinel-1 data are corrected using nighttime light data as additional information, which shows promising results in addressing multiple backscattering issues. Finally, a deep learning model is trained to generate building height maps using these features, achieving a mean absolute error of around 2 m and a mean square error of approximately 13. The generalizability of this method is demonstrated in several cities with diverse built-up structures, including London, Berlin, and others. Finally, a building height map of Iran is generated and evaluated against surveyed buildings, showcasing its large-scale mapping capability.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"46 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142200473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zongqing Cao, Bing Liu, Jianchao Yang, Ke Tan, Zheng Dai, Xingyu Lu, Hong Gu
Interrupted and multi-source track segment association (TSA) are two key challenges in target trajectory research within radar data processing. Traditional methods often rely on simplistic assumptions about target motion and statistical techniques for track association, leading to problems such as unrealistic assumptions, susceptibility to noise, and suboptimal performance limits. This study proposes a unified framework to address the challenges of associating interrupted and multi-source track segments by measuring trajectory similarity. We present TSA-cTFER, a novel network utilizing contrastive learning and TransFormer Encoder to accurately assess trajectory similarity through learned Representations by computing distances between high-dimensional feature vectors. Additionally, we tackle dynamic association scenarios with a two-stage online algorithm designed to manage tracks that appear or disappear at any time. This algorithm categorizes track pairs into easy and hard groups, employing tailored association strategies to achieve precise and robust associations in dynamic environments. Experimental results on real-world datasets demonstrate that our proposed TSA-cTFER network with the two-stage online algorithm outperforms existing methods, achieving 94.59% accuracy in interrupted track segment association tasks and 94.83% in multi-source track segment association tasks.
{"title":"Contrastive Transformer Network for Track Segment Association with Two-Stage Online Method","authors":"Zongqing Cao, Bing Liu, Jianchao Yang, Ke Tan, Zheng Dai, Xingyu Lu, Hong Gu","doi":"10.3390/rs16183380","DOIUrl":"https://doi.org/10.3390/rs16183380","url":null,"abstract":"Interrupted and multi-source track segment association (TSA) are two key challenges in target trajectory research within radar data processing. Traditional methods often rely on simplistic assumptions about target motion and statistical techniques for track association, leading to problems such as unrealistic assumptions, susceptibility to noise, and suboptimal performance limits. This study proposes a unified framework to address the challenges of associating interrupted and multi-source track segments by measuring trajectory similarity. We present TSA-cTFER, a novel network utilizing contrastive learning and TransFormer Encoder to accurately assess trajectory similarity through learned Representations by computing distances between high-dimensional feature vectors. Additionally, we tackle dynamic association scenarios with a two-stage online algorithm designed to manage tracks that appear or disappear at any time. This algorithm categorizes track pairs into easy and hard groups, employing tailored association strategies to achieve precise and robust associations in dynamic environments. Experimental results on real-world datasets demonstrate that our proposed TSA-cTFER network with the two-stage online algorithm outperforms existing methods, achieving 94.59% accuracy in interrupted track segment association tasks and 94.83% in multi-source track segment association tasks.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"6 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142200478","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Visible and infrared image fusion is a strategy that effectively extracts and fuses information from different sources. However, most existing methods largely neglect the issue of lighting imbalance, which makes the same fusion models inapplicable to different scenes. Several methods obtain low-level features from visible and infrared images at an early stage of input or shallow feature extraction. However, these methods do not explore how low-level features provide a foundation for recognizing and utilizing the complementarity and common information between the two types of images. As a result, the complementarity and common information between the images is not fully analyzed and discussed. To address these issues, we propose a Self-Attention Progressive Network for the fusion of infrared and visible images in this paper. Firstly, we construct a Lighting-Aware Sub-Network to analyze lighting distribution, and introduce intensity loss to measure the probability of scene illumination. This approach enhances the model’s adaptability to lighting conditions. Secondly, we introduce self-attention learning to design a multi-state joint feature extraction module (MSJFEM) that fully utilizes the contextual information among input keys. It guides the learning of a dynamic attention matrix to strengthen the capacity for visual representation. Finally, we design a Difference-Aware Propagation Module (DAPM) to extract and integrate edge details from the source images while supplementing differential information. The experiments across three benchmark datasets reveal that the proposed approach exhibits satisfactory performance compared to existing methods.
{"title":"Self-Attention Progressive Network for Infrared and Visible Image Fusion","authors":"Shuying Li, Muyi Han, Yuemei Qin, Qiang Li","doi":"10.3390/rs16183370","DOIUrl":"https://doi.org/10.3390/rs16183370","url":null,"abstract":"Visible and infrared image fusion is a strategy that effectively extracts and fuses information from different sources. However, most existing methods largely neglect the issue of lighting imbalance, which makes the same fusion models inapplicable to different scenes. Several methods obtain low-level features from visible and infrared images at an early stage of input or shallow feature extraction. However, these methods do not explore how low-level features provide a foundation for recognizing and utilizing the complementarity and common information between the two types of images. As a result, the complementarity and common information between the images is not fully analyzed and discussed. To address these issues, we propose a Self-Attention Progressive Network for the fusion of infrared and visible images in this paper. Firstly, we construct a Lighting-Aware Sub-Network to analyze lighting distribution, and introduce intensity loss to measure the probability of scene illumination. This approach enhances the model’s adaptability to lighting conditions. Secondly, we introduce self-attention learning to design a multi-state joint feature extraction module (MSJFEM) that fully utilizes the contextual information among input keys. It guides the learning of a dynamic attention matrix to strengthen the capacity for visual representation. Finally, we design a Difference-Aware Propagation Module (DAPM) to extract and integrate edge details from the source images while supplementing differential information. The experiments across three benchmark datasets reveal that the proposed approach exhibits satisfactory performance compared to existing methods.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"400 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142200469","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jing Ning, Yunjun Yao, Joshua B. Fisher, Yufu Li, Xiaotong Zhang, Bo Jiang, Jia Xu, Ruiyang Yu, Lu Liu, Xueyi Zhang, Zijing Xie, Jiahui Fan, Luna Zhang
As a major agricultural hazard, drought frequently occurs due to a reduction in precipitation resulting in a continuously propagating soil moisture (SM) deficit. Assessment of the high spatial-resolution SM-derived drought index is crucial for monitoring agricultural drought. In this study, we generated a downscaled random forest SM dataset (RF-SM) and calculated the soil water deficit index (RF-SM-SWDI) at 30 m for agricultural drought monitoring. The results showed that the RF-SM dataset exhibited better consistency with in situ SM observations in the detection of extremes than did the SM products, including SMAP, SMOS, NCA-LDAS, and ESA CCI, for different land cover types in the U.S. and yielded a satisfactory performance, with the lowest root mean square error (RMSE, below 0.055 m3/m3) and the highest coefficient of determination (R2, above 0.8) for most observation networks, based on the number of sites. A vegetation health index (VHI), derived from a Landsat 8 optical remote sensing dataset, was also generated for comparison. The results illustrated that the RF-SM-SWDI and VHI exhibited high correlations (R ≥ 0.5) at approximately 70% of the stations. Furthermore, we mapped spatiotemporal drought monitoring indices in California. The RF-SM-SWDI provided drought conditions with more detailed spatial information than did the short-term drought blend (STDB) released by the U.S. Drought Monitor, which demonstrated the expected response of seasonal drought trends, while differences from the VHI were observed mainly in forest areas. Therefore, downscaled SM and SWDI, with a spatial resolution of 30 m, are promising for monitoring agricultural field drought within different contexts, and additional reliable factors could be incorporated to better guide agricultural management practices.
作为一种主要的农业灾害，干旱的发生往往是由于降水量减少导致土壤水分（SM）持续不足。评估高空间分辨率土壤水分衍生干旱指数对于监测农业干旱至关重要。在本研究中，我们生成了降尺度随机森林土壤水分数据集（RF-SM），并计算了 30 米处的土壤水分亏缺指数（RF-SM-SWDI），用于农业干旱监测。结果表明，对于美国不同的土地覆被类型，RF-SM 数据集在探测极端事件方面与原地 SM 观测数据的一致性要好于 SMAP、SMOS、NCA-LDAS 和 ESA CCI 等 SM 产品，并且性能令人满意，在大多数观测网络中，根据站点数量，RF-SM 数据集的均方根误差（RMSE，低于 0.055 m3/m3）最小，判定系数（R2，高于 0.8）最高。此外，还通过 Landsat 8 光学遥感数据集生成了植被健康指数（VHI），以进行比较。结果表明，RF-SM-SWDI 和 VHI 在约 70% 的站点表现出高度相关性（R ≥ 0.5）。此外，我们还绘制了加利福尼亚州的时空干旱监测指数图。与美国干旱监测机构发布的短期干旱混合指数（STDB）相比，RF-SM-SWDI 提供了更详细的干旱状况空间信息，显示了季节性干旱趋势的预期响应，而与 VHI 的差异主要出现在森林地区。因此，空间分辨率为 30 米的降尺度 SM 和 SWDI 有望在不同环境下监测农田干旱，并可纳入更多可靠因素，以更好地指导农业管理实践。
{"title":"Soil Moisture-Derived SWDI at 30 m Based on Multiple Satellite Datasets for Agricultural Drought Monitoring","authors":"Jing Ning, Yunjun Yao, Joshua B. Fisher, Yufu Li, Xiaotong Zhang, Bo Jiang, Jia Xu, Ruiyang Yu, Lu Liu, Xueyi Zhang, Zijing Xie, Jiahui Fan, Luna Zhang","doi":"10.3390/rs16183372","DOIUrl":"https://doi.org/10.3390/rs16183372","url":null,"abstract":"As a major agricultural hazard, drought frequently occurs due to a reduction in precipitation resulting in a continuously propagating soil moisture (SM) deficit. Assessment of the high spatial-resolution SM-derived drought index is crucial for monitoring agricultural drought. In this study, we generated a downscaled random forest SM dataset (RF-SM) and calculated the soil water deficit index (RF-SM-SWDI) at 30 m for agricultural drought monitoring. The results showed that the RF-SM dataset exhibited better consistency with in situ SM observations in the detection of extremes than did the SM products, including SMAP, SMOS, NCA-LDAS, and ESA CCI, for different land cover types in the U.S. and yielded a satisfactory performance, with the lowest root mean square error (RMSE, below 0.055 m3/m3) and the highest coefficient of determination (R2, above 0.8) for most observation networks, based on the number of sites. A vegetation health index (VHI), derived from a Landsat 8 optical remote sensing dataset, was also generated for comparison. The results illustrated that the RF-SM-SWDI and VHI exhibited high correlations (R ≥ 0.5) at approximately 70% of the stations. Furthermore, we mapped spatiotemporal drought monitoring indices in California. The RF-SM-SWDI provided drought conditions with more detailed spatial information than did the short-term drought blend (STDB) released by the U.S. Drought Monitor, which demonstrated the expected response of seasonal drought trends, while differences from the VHI were observed mainly in forest areas. Therefore, downscaled SM and SWDI, with a spatial resolution of 30 m, are promising for monitoring agricultural field drought within different contexts, and additional reliable factors could be incorporated to better guide agricultural management practices.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"52 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142200482","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jianye Yuan, Haofei Wang, Minghao Li, Xiaohan Wang, Weiwei Song, Song Li, Wei Gong
Fire detection is crucial due to the exorbitant annual toll on both human lives and the economy resulting from fire-related incidents. To enhance forest fire detection in complex environments, we propose a new algorithm called FD-Net for various environments. Firstly, to improve detection performance, we introduce a Fire Attention (FA) mechanism that utilizes the position information from feature maps. Secondly, to prevent geometric distortion during image cropping, we propose a Three-Scale Pooling (TSP) module. Lastly, we fine-tune the YOLOv5 network and incorporate a new Fire Fusion (FF) module to enhance the network’s precision in identifying fire targets. Through qualitative and quantitative comparisons, we found that FD-Net outperforms current state-of-the-art algorithms in performance on both fire and fire-and-smoke datasets. This further demonstrates FD-Net’s effectiveness for application in fire detection.
{"title":"FD-Net: A Single-Stage Fire Detection Framework for Remote Sensing in Complex Environments","authors":"Jianye Yuan, Haofei Wang, Minghao Li, Xiaohan Wang, Weiwei Song, Song Li, Wei Gong","doi":"10.3390/rs16183382","DOIUrl":"https://doi.org/10.3390/rs16183382","url":null,"abstract":"Fire detection is crucial due to the exorbitant annual toll on both human lives and the economy resulting from fire-related incidents. To enhance forest fire detection in complex environments, we propose a new algorithm called FD-Net for various environments. Firstly, to improve detection performance, we introduce a Fire Attention (FA) mechanism that utilizes the position information from feature maps. Secondly, to prevent geometric distortion during image cropping, we propose a Three-Scale Pooling (TSP) module. Lastly, we fine-tune the YOLOv5 network and incorporate a new Fire Fusion (FF) module to enhance the network’s precision in identifying fire targets. Through qualitative and quantitative comparisons, we found that FD-Net outperforms current state-of-the-art algorithms in performance on both fire and fire-and-smoke datasets. This further demonstrates FD-Net’s effectiveness for application in fire detection.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"50 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142200499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：