Pub Date : 2024-07-26DOI: 10.3389/frsen.2024.1416550
Bertrand Ygorra, Frédéric Frappart, J. Wigneron, T. Catry, Benjamin Pillot, Antoine Pfefer, Jonas Courtalon, S. Riazanoff
Tropical forests are currently under pressure from increasing threats. These threats are mostly related to human activities. Earth observations (EO) are increasingly used for monitoring forest cover, especially synthetic aperture radar (SAR), that is less affected than optical sensors by atmospheric conditions. Since the launch of the Sentinel-1 satellites, numerous methods for forest disturbance monitoring have been developed, including near real-time (NRT) operational algorithms as systems providing early warnings on deforestation. These systems include Radar for Detecting Deforestation (RADD), Global Land Analysis and Discovery (GLAD), Real Time Deforestation Detection System (DETER), and Jica-Jaxa Forest Early Warning System (JJ-FAST). These algorithms provide online disturbance maps and are applied at continental/global scales with a Minimum Mapping Unit (MMU) ranging from 0.1 ha to 6.25 ha. For local operators, these algorithms are hard to customize to meet users’ specific needs. Recently, the Cumulative sum change detection (CuSum) method has been developed for the monitoring of forest disturbances from long time series of Sentinel-1 images. Here, we present the development of a NRT version of CuSum with a MMU of 0.03 ha. The values of the different parameters of this NRT CuSum algorithm were determined to optimize the detection of changes using the F1-score. In the best configuration, 68% precision, 72% recall, 93% accuracy and 0.71 F1-score were obtained.
{"title":"A near-real-time tropical deforestation monitoring algorithm based on the CuSum change detection method","authors":"Bertrand Ygorra, Frédéric Frappart, J. Wigneron, T. Catry, Benjamin Pillot, Antoine Pfefer, Jonas Courtalon, S. Riazanoff","doi":"10.3389/frsen.2024.1416550","DOIUrl":"https://doi.org/10.3389/frsen.2024.1416550","url":null,"abstract":"Tropical forests are currently under pressure from increasing threats. These threats are mostly related to human activities. Earth observations (EO) are increasingly used for monitoring forest cover, especially synthetic aperture radar (SAR), that is less affected than optical sensors by atmospheric conditions. Since the launch of the Sentinel-1 satellites, numerous methods for forest disturbance monitoring have been developed, including near real-time (NRT) operational algorithms as systems providing early warnings on deforestation. These systems include Radar for Detecting Deforestation (RADD), Global Land Analysis and Discovery (GLAD), Real Time Deforestation Detection System (DETER), and Jica-Jaxa Forest Early Warning System (JJ-FAST). These algorithms provide online disturbance maps and are applied at continental/global scales with a Minimum Mapping Unit (MMU) ranging from 0.1 ha to 6.25 ha. For local operators, these algorithms are hard to customize to meet users’ specific needs. Recently, the Cumulative sum change detection (CuSum) method has been developed for the monitoring of forest disturbances from long time series of Sentinel-1 images. Here, we present the development of a NRT version of CuSum with a MMU of 0.03 ha. The values of the different parameters of this NRT CuSum algorithm were determined to optimize the detection of changes using the F1-score. In the best configuration, 68% precision, 72% recall, 93% accuracy and 0.71 F1-score were obtained.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"57 18","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141799228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-24DOI: 10.3389/frsen.2024.1423332
Ave Ansper-Toomsalu, Mirjam Uusõue, Kersti Kangro, Martin Hieronymi, K. Alikas
Optically complex waters present significant challenges for remote sensing due to high concentrations of optically active substances (OASs) and their inherent optical properties (IOPs), as well as the adjacency effect. OASs and IOPs can be derived from atmospheric correction processors’ in-water algorithms applied to data from Sentinel-2 MultiSpectral Instrument (S2 MSI) and Sentinel-3 Ocean and Land Color Instrument (S3 OLCI). This study compared S3 OLCI Level-2 in-water products for Case-2 waters with alternative in-water algorithms derived from ACOLITE, POLYMER, C2RCC, and A4O. Fifty in-water algorithms were evaluated using an extensive match-up dataset from lakes and coastal areas, focusing particularly on small lakes with high colored dissolved organic matter absorption at 442 nm (up to 48 m-1). The Chl a band ratio introduced by Gons et al. (2022) applied to data processed by ACOLITE performed best for S3 OLCI Chl a retrieval (dispersion = 23%, bias = 10%). Gons et al. (2022) band ratio also showed consistent agreement between S3 OLCI and S2 MSI resampled data (intercept of 6.27 and slope of 0.83, close to the 1:1 line); however, lower Chl a values (<20 mg/m3) were overestimated by S2 MSI. When estimating errors associated with proximity to land, S2 MSI Chl a in-water algorithms had higher errors close to the shore (on average 315%) compared to S3 OLCI (on average 150%). Chl a retrieved with POLYMER had the lowest errors close to the shore for both S2 MSI and S3 OLCI data (on average 70%). Total suspended matter (TSM) retrieval with C2RCC performed well for S2 MSI (dispersion 24% and bias −12%). Total absorption was most accurately derived from C2RCC applied to S3 OLCI L1 data (dispersion < 43% and bias < −39%), and it was better estimated than its individual components: phytoplankton, mineral particles, and colored dissolved organic matter absorption. However, none of the colored dissolved organic matter absorption in-water algorithms performed well (dispersion > 59% and bias < −29%).
{"title":"Suitability of different in-water algorithms for eutrophic and absorbing waters applied to Sentinel-2 MSI and Sentinel-3 OLCI data","authors":"Ave Ansper-Toomsalu, Mirjam Uusõue, Kersti Kangro, Martin Hieronymi, K. Alikas","doi":"10.3389/frsen.2024.1423332","DOIUrl":"https://doi.org/10.3389/frsen.2024.1423332","url":null,"abstract":"Optically complex waters present significant challenges for remote sensing due to high concentrations of optically active substances (OASs) and their inherent optical properties (IOPs), as well as the adjacency effect. OASs and IOPs can be derived from atmospheric correction processors’ in-water algorithms applied to data from Sentinel-2 MultiSpectral Instrument (S2 MSI) and Sentinel-3 Ocean and Land Color Instrument (S3 OLCI). This study compared S3 OLCI Level-2 in-water products for Case-2 waters with alternative in-water algorithms derived from ACOLITE, POLYMER, C2RCC, and A4O. Fifty in-water algorithms were evaluated using an extensive match-up dataset from lakes and coastal areas, focusing particularly on small lakes with high colored dissolved organic matter absorption at 442 nm (up to 48 m-1). The Chl a band ratio introduced by Gons et al. (2022) applied to data processed by ACOLITE performed best for S3 OLCI Chl a retrieval (dispersion = 23%, bias = 10%). Gons et al. (2022) band ratio also showed consistent agreement between S3 OLCI and S2 MSI resampled data (intercept of 6.27 and slope of 0.83, close to the 1:1 line); however, lower Chl a values (<20 mg/m3) were overestimated by S2 MSI. When estimating errors associated with proximity to land, S2 MSI Chl a in-water algorithms had higher errors close to the shore (on average 315%) compared to S3 OLCI (on average 150%). Chl a retrieved with POLYMER had the lowest errors close to the shore for both S2 MSI and S3 OLCI data (on average 70%). Total suspended matter (TSM) retrieval with C2RCC performed well for S2 MSI (dispersion 24% and bias −12%). Total absorption was most accurately derived from C2RCC applied to S3 OLCI L1 data (dispersion < 43% and bias < −39%), and it was better estimated than its individual components: phytoplankton, mineral particles, and colored dissolved organic matter absorption. However, none of the colored dissolved organic matter absorption in-water algorithms performed well (dispersion > 59% and bias < −29%).","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"4 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141808700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-24DOI: 10.3389/frsen.2024.1399839
Bing Lin, Matthew Walker Mclinden, Xia Cai, G. Heymsfield, Nikki Privé, S. Harrah, Lihua Li
Sea surface air pressure observations are a significant gap in the current Earth observing systems. This study addresses retrieval algorithm development and the evaluation of the potential impact of instrumental and environmental uncertainties on sea level pressure retrievals for the measurements of O2 differential absorption radar systems operating at three spectrally evenly spaced close-frequency bands (65.5, 67.75, and 70.0 GHz). A simulated northern hemispheric summer case is used to simulate retrieval uncertainties. To avoid high attenuation and a low signal-to-noise ratio, radar measurements from weather conditions with a rain rate ≥1 mm/h are not used in the retrieval. This study finds that a retrieval algorithm combining all three channels, i.e., the 3-channel approach, can effectively mitigate major atmospheric and sea surface influences on sea surface air pressure retrieval. The major uncertainty of sea surface pressure retrieval is due to the standard deviation in radar power returns. Analysis and simulation demonstrate the potential of global sea surface pressure observations with errors of about 1∼2 mb, which is urgently needed for the improvement of numerical weather prediction models. Future work will emphasize instrument development and field experiments. It is anticipated that an O2 differential absorption radar system will be available for meteorological applications in a few years.
{"title":"Sea surface barometry with an O2 differential absorption radar: retrieval algorithm development and simulation","authors":"Bing Lin, Matthew Walker Mclinden, Xia Cai, G. Heymsfield, Nikki Privé, S. Harrah, Lihua Li","doi":"10.3389/frsen.2024.1399839","DOIUrl":"https://doi.org/10.3389/frsen.2024.1399839","url":null,"abstract":"Sea surface air pressure observations are a significant gap in the current Earth observing systems. This study addresses retrieval algorithm development and the evaluation of the potential impact of instrumental and environmental uncertainties on sea level pressure retrievals for the measurements of O2 differential absorption radar systems operating at three spectrally evenly spaced close-frequency bands (65.5, 67.75, and 70.0 GHz). A simulated northern hemispheric summer case is used to simulate retrieval uncertainties. To avoid high attenuation and a low signal-to-noise ratio, radar measurements from weather conditions with a rain rate ≥1 mm/h are not used in the retrieval. This study finds that a retrieval algorithm combining all three channels, i.e., the 3-channel approach, can effectively mitigate major atmospheric and sea surface influences on sea surface air pressure retrieval. The major uncertainty of sea surface pressure retrieval is due to the standard deviation in radar power returns. Analysis and simulation demonstrate the potential of global sea surface pressure observations with errors of about 1∼2 mb, which is urgently needed for the improvement of numerical weather prediction models. Future work will emphasize instrument development and field experiments. It is anticipated that an O2 differential absorption radar system will be available for meteorological applications in a few years.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"41 16","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141809744","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-18DOI: 10.3389/frsen.2024.1383147
Arun M. Saranathan, Mortimer Werther, S. Balasubramanian, Daniel Odermatt, N. Pahlevan
Given the use of machine learning-based tools for monitoring the Water Quality Indicators (WQIs) over lakes and coastal waters, understanding the properties of such models, including the uncertainties inherent in their predictions is essential. This has led to the development of two probabilistic NN-algorithms: Mixture Density Network (MDN) and Bayesian Neural Network via Monte Carlo Dropout (BNN-MCD). These NNs are complex, featuring thousands of trainable parameters and modifiable hyper-parameters, and have been independently trained and tested. The model uncertainty metric captures the uncertainty present in each prediction based on the properties of the model—namely, the model architecture and the training data distribution. We conduct an analysis of MDN and BNN-MCD under near-identical conditions of model architecture, training, and test sets, etc., to retrieve the concentration of chlorophyll-a pigments (Chl a), total suspended solids (TSS), and the absorption by colored dissolved organic matter at 440 nm (acdom (440)). The spectral resolutions considered correspond to the Hyperspectral Imager for the Coastal Ocean (HICO), PRecursore IperSpettrale della Missione Applicativa (PRISMA), Ocean Colour and Land Imager (OLCI), and MultiSpectral Instrument (MSI). The model performances are tested in terms of both predictive residuals and predictive uncertainty metric quality. We also compared the simultaneous WQI retrievals against a single-parameter retrieval framework (for Chla). Ultimately, the models’ real-world applicability was investigated using a MSI satellite-matchup dataset N=3,053) of Chla and TSS. Experiments show that both models exhibit comparable estimation performance. Specifically, the median symmetric accuracy (MdSA) on the test set for the different parameters in both algorithms range from 30% to 60%. The uncertainty estimates, on the other hand, differ strongly. MDN’s uncertainty estimate is ∼50%, encompassing estimation residuals for 75% of test samples, whereas BNN-MCD’s average uncertainty estimate is ∼25%, encompassing the residuals for 50% of samples. Our analysis also revealed that simultaneous estimation results in improvements in both predictive performance and uncertainty metric quality. Interestingly, the trends mentioned above hold across different sensor resolutions, as well as experimental regimes. This disparity calls for additional research to determine whether such trends in model uncertainty are inherent to specific models or can be more broadly generalized across different algorithms and sensor setups.
{"title":"Assessment of advanced neural networks for the dual estimation of water quality indicators and their uncertainties","authors":"Arun M. Saranathan, Mortimer Werther, S. Balasubramanian, Daniel Odermatt, N. Pahlevan","doi":"10.3389/frsen.2024.1383147","DOIUrl":"https://doi.org/10.3389/frsen.2024.1383147","url":null,"abstract":"Given the use of machine learning-based tools for monitoring the Water Quality Indicators (WQIs) over lakes and coastal waters, understanding the properties of such models, including the uncertainties inherent in their predictions is essential. This has led to the development of two probabilistic NN-algorithms: Mixture Density Network (MDN) and Bayesian Neural Network via Monte Carlo Dropout (BNN-MCD). These NNs are complex, featuring thousands of trainable parameters and modifiable hyper-parameters, and have been independently trained and tested. The model uncertainty metric captures the uncertainty present in each prediction based on the properties of the model—namely, the model architecture and the training data distribution. We conduct an analysis of MDN and BNN-MCD under near-identical conditions of model architecture, training, and test sets, etc., to retrieve the concentration of chlorophyll-a pigments (Chl a), total suspended solids (TSS), and the absorption by colored dissolved organic matter at 440 nm (acdom (440)). The spectral resolutions considered correspond to the Hyperspectral Imager for the Coastal Ocean (HICO), PRecursore IperSpettrale della Missione Applicativa (PRISMA), Ocean Colour and Land Imager (OLCI), and MultiSpectral Instrument (MSI). The model performances are tested in terms of both predictive residuals and predictive uncertainty metric quality. We also compared the simultaneous WQI retrievals against a single-parameter retrieval framework (for Chla). Ultimately, the models’ real-world applicability was investigated using a MSI satellite-matchup dataset N=3,053) of Chla and TSS. Experiments show that both models exhibit comparable estimation performance. Specifically, the median symmetric accuracy (MdSA) on the test set for the different parameters in both algorithms range from 30% to 60%. The uncertainty estimates, on the other hand, differ strongly. MDN’s uncertainty estimate is ∼50%, encompassing estimation residuals for 75% of test samples, whereas BNN-MCD’s average uncertainty estimate is ∼25%, encompassing the residuals for 50% of samples. Our analysis also revealed that simultaneous estimation results in improvements in both predictive performance and uncertainty metric quality. Interestingly, the trends mentioned above hold across different sensor resolutions, as well as experimental regimes. This disparity calls for additional research to determine whether such trends in model uncertainty are inherent to specific models or can be more broadly generalized across different algorithms and sensor setups.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":" 44","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141827079","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-17DOI: 10.3389/frsen.2024.1333851
Paul Chamberlain, Robert J. Frouin, Jing Tan, Matt Mazloff, Andrew Barnard, Emmanuel Boss, N. Haëntjens, Cristina Orrico
A novel ocean profiling float system for calibrating and validating satellite-based ocean color observations has been developed and tested. The float-based radiometric sampling system, herein referred to as HyperNav, is complementary to traditional moored in-situ observing systems and provides additional capability due to the relatively small platform size and high radiometric accuracy that allows for opportunistic deployments at locations during seasons and conditions that are best for ocean color observations. The purpose of this study is to optimize the deployment locations of an array of HyperNav systems to support the Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) mission by performing System Vicarious Calibration (SVC) observations. Specifically, we present the development of logistical and scientific criteria for selecting suitable sites for developing an SVC network of profiling-float-based radiometric systems capable of calibrating and validating ocean color observations. As part of the analyses described in this paper, we have synthetically deployed HyperNav at potential US-based and international sites, including: north of Crete island; south-east of Bermuda island; south of Puerto Rico island; southwest of Port Hueneme, CA; west of Monterey, CA; west of Kona, HI; and south-west of Tahiti island. The synthetic analyses identified Kona, Puerto Rico, Crete, and Tahiti as promising SVC sites. All sites considered are suitable for generating a significant number of validation match-ups. Optimally deploying HyperNav systems at these sites during the PACE post-launch SVC campaign is expected to cost-effectively provide a large number of SVC match-ups to fulfill the PACE calibration requirements.
开发并测试了一种新型海洋剖面浮标系统，用于校准和验证卫星海洋颜色观测结果。浮筒辐射采样系统（以下简称 HyperNav）是对传统系泊式现场观测系统的补充，由于平台体积相对较小，辐射精度高，可在最适合海洋颜色观测的季节和条件下择机部署，因此具有更强的能力。本研究的目的是通过执行系统虚拟校准（SVC）观测，优化 HyperNav 系统阵列的部署位置，以支持浮游生物、气溶胶、云层和海洋生态系统（PACE）任务。具体来说，我们介绍了如何制定后勤和科学标准，以选择合适的地点来开发能够校准和验证海洋颜色观测数据的基于辐射测量系统的剖面浮标 SVC 网络。作为本文所述分析的一部分，我们在潜在的美国和国际站点综合部署了 HyperNav，这些站点包括：克里特岛北部、百慕大群岛东南部、波多黎各岛南部、加利福尼亚州胡内姆港西南部、加利福尼亚州蒙特雷西部、夏威夷州科纳西部和塔希提岛西南部。合成分析确定科纳、波多黎各、克里特岛和塔希提岛为有希望的 SVC 地点。所有考虑的地点都适合生成大量验证匹配。在 PACE 发射后 SVC 活动期间，在这些站点优化部署 HyperNav 系统，预计将以具有成本效益的方式提供大量 SVC 匹配，以满足 PACE 校准要求。
{"title":"Selecting HyperNav deployment sites for calibrating and validating PACE ocean color observations","authors":"Paul Chamberlain, Robert J. Frouin, Jing Tan, Matt Mazloff, Andrew Barnard, Emmanuel Boss, N. Haëntjens, Cristina Orrico","doi":"10.3389/frsen.2024.1333851","DOIUrl":"https://doi.org/10.3389/frsen.2024.1333851","url":null,"abstract":"A novel ocean profiling float system for calibrating and validating satellite-based ocean color observations has been developed and tested. The float-based radiometric sampling system, herein referred to as HyperNav, is complementary to traditional moored in-situ observing systems and provides additional capability due to the relatively small platform size and high radiometric accuracy that allows for opportunistic deployments at locations during seasons and conditions that are best for ocean color observations. The purpose of this study is to optimize the deployment locations of an array of HyperNav systems to support the Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) mission by performing System Vicarious Calibration (SVC) observations. Specifically, we present the development of logistical and scientific criteria for selecting suitable sites for developing an SVC network of profiling-float-based radiometric systems capable of calibrating and validating ocean color observations. As part of the analyses described in this paper, we have synthetically deployed HyperNav at potential US-based and international sites, including: north of Crete island; south-east of Bermuda island; south of Puerto Rico island; southwest of Port Hueneme, CA; west of Monterey, CA; west of Kona, HI; and south-west of Tahiti island. The synthetic analyses identified Kona, Puerto Rico, Crete, and Tahiti as promising SVC sites. All sites considered are suitable for generating a significant number of validation match-ups. Optimally deploying HyperNav systems at these sites during the PACE post-launch SVC campaign is expected to cost-effectively provide a large number of SVC match-ups to fulfill the PACE calibration requirements.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":" 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141828137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-16DOI: 10.3389/frsen.2024.1395442
L. Siu, Xubin Zeng, A. Sorooshian, Brian Cairns, R. Ferrare, J. Hair, C. Hostetler, D. Painemal, J. Schlosser
With the ongoing expansion of global observation networks, it is expected that we shall routinely analyze records of geophysical variables such as temperature from multiple collocated instruments. Validating datasets in this situation is not a trivial task because every observing system has its own bias and noise. Triple collocation is a general statistical framework to estimate the error characteristics in three or more observational-based datasets. In a triple colocation analysis, several metrics are routinely reported but traditional multiple-panel plots are not the most effective way to display information. A new formula of error variance is derived for connecting the key terms in the triple collocation theory. A diagram based on this formula is devised to facilitate triple collocation analysis of any data from observations, as illustrated using three aerosol optical depth datasets from the recent Aerosol Cloud meTeorology Interactions oVer the western ATlantic Experiment (ACTIVATE). An observational-based skill score is also derived to evaluate the quality of three datasets by taking into account both error variance and correlation coefficient. Several applications are discussed and sample plotting routines are provided.
{"title":"Summarizing multiple aspects of triple collocation analysis in a single diagram","authors":"L. Siu, Xubin Zeng, A. Sorooshian, Brian Cairns, R. Ferrare, J. Hair, C. Hostetler, D. Painemal, J. Schlosser","doi":"10.3389/frsen.2024.1395442","DOIUrl":"https://doi.org/10.3389/frsen.2024.1395442","url":null,"abstract":"With the ongoing expansion of global observation networks, it is expected that we shall routinely analyze records of geophysical variables such as temperature from multiple collocated instruments. Validating datasets in this situation is not a trivial task because every observing system has its own bias and noise. Triple collocation is a general statistical framework to estimate the error characteristics in three or more observational-based datasets. In a triple colocation analysis, several metrics are routinely reported but traditional multiple-panel plots are not the most effective way to display information. A new formula of error variance is derived for connecting the key terms in the triple collocation theory. A diagram based on this formula is devised to facilitate triple collocation analysis of any data from observations, as illustrated using three aerosol optical depth datasets from the recent Aerosol Cloud meTeorology Interactions oVer the western ATlantic Experiment (ACTIVATE). An observational-based skill score is also derived to evaluate the quality of three datasets by taking into account both error variance and correlation coefficient. Several applications are discussed and sample plotting routines are provided.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"66 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141643010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-16DOI: 10.3389/frsen.2024.1347507
Joel Kuusk, Alexandre Corizzi, D. Doxaran, Kim Duong, Kenneth Flight, Joosep Kivastik, Kaspars Laizāns, E. Leymarie, Silvar Muru, Christophe Penkerc’h, Kevin G. Ruddick
Optical Earth observation satellites provide vast amounts of data on a daily basis. The top-of-atmosphere radiance measured by these satellites is usually converted to bottom-of-atmosphere radiance or reflectance which is then used for deriving numerous higher level products used for monitoring environmental conditions, climate change, stock of natural resources, etc. The increase of available remote sensing data impacts decision-making on both regional and global scales, and demands appropriate quality control and validation procedures. A HYperspectral Pointable System for Terrestrial and Aquatic Radiometry (HYPSTAR®) has been designed to provide automated, in-situ multiangular reflectance measurements of land and water targets. HYPSTAR-SR covers 380–1020 nm spectral range at 3 nm spectral resolution and is used at water sites. For land sites the HYPSTAR-XR variant is used with the spectral range extended to 1680 nm at 10 nm spectral resolution. The spectroradiometer has multiplexed radiance and irradiance entrances, an internal mechanical shutter, and an integrated imaging camera for capturing snapshots of the targets. The spectroradiometer is mounted on a two-axis pointing system with 360° range of free movement in both axes. The system also incorporates a stable light emitting diode as a light source, used for monitoring the stability of the radiometric calibration during the long-term unattended field deployment. Autonomous operation is managed by a host system which handles data acquisition, storage, and transmission to a central WATERHYPERNET or LANDHYPERNET server according to a pre-programmed schedule. The system is remotely accessible over the internet for configuration changes and software updates. The HYPSTAR systems have been deployed at 10 water and 11 land sites for different periods ranging from a few days to a few years. The data are automatically processed at the central servers by the HYPERNETS processor and the derived radiance, irradiance, and reflectance products with associated measurement uncertainties are distributed at the WATERHYPERNET and LANDHYPERNET data portals.
{"title":"HYPSTAR: a hyperspectral pointable system for terrestrial and aquatic radiometry","authors":"Joel Kuusk, Alexandre Corizzi, D. Doxaran, Kim Duong, Kenneth Flight, Joosep Kivastik, Kaspars Laizāns, E. Leymarie, Silvar Muru, Christophe Penkerc’h, Kevin G. Ruddick","doi":"10.3389/frsen.2024.1347507","DOIUrl":"https://doi.org/10.3389/frsen.2024.1347507","url":null,"abstract":"Optical Earth observation satellites provide vast amounts of data on a daily basis. The top-of-atmosphere radiance measured by these satellites is usually converted to bottom-of-atmosphere radiance or reflectance which is then used for deriving numerous higher level products used for monitoring environmental conditions, climate change, stock of natural resources, etc. The increase of available remote sensing data impacts decision-making on both regional and global scales, and demands appropriate quality control and validation procedures. A HYperspectral Pointable System for Terrestrial and Aquatic Radiometry (HYPSTAR®) has been designed to provide automated, in-situ multiangular reflectance measurements of land and water targets. HYPSTAR-SR covers 380–1020 nm spectral range at 3 nm spectral resolution and is used at water sites. For land sites the HYPSTAR-XR variant is used with the spectral range extended to 1680 nm at 10 nm spectral resolution. The spectroradiometer has multiplexed radiance and irradiance entrances, an internal mechanical shutter, and an integrated imaging camera for capturing snapshots of the targets. The spectroradiometer is mounted on a two-axis pointing system with 360° range of free movement in both axes. The system also incorporates a stable light emitting diode as a light source, used for monitoring the stability of the radiometric calibration during the long-term unattended field deployment. Autonomous operation is managed by a host system which handles data acquisition, storage, and transmission to a central WATERHYPERNET or LANDHYPERNET server according to a pre-programmed schedule. The system is remotely accessible over the internet for configuration changes and software updates. The HYPSTAR systems have been deployed at 10 water and 11 land sites for different periods ranging from a few days to a few years. The data are automatically processed at the central servers by the HYPERNETS processor and the derived radiance, irradiance, and reflectance products with associated measurement uncertainties are distributed at the WATERHYPERNET and LANDHYPERNET data portals.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"11 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141644065","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-12DOI: 10.3389/frsen.2024.1404461
Alexander Kostinski, A. Marshak, T. Várnai
The Deep Space Climate Observatory (DSCOVR) spacecraft drifts about the Lagrangian point ≈ 1.4 − 1.6 × 106 km from Earth, where its Earth Polychromatic Imaging Camera (EPIC) observes the entire sunlit face of Earth every 1–2 h. In an attempt to detect “signals,” i.e., longer-term changes and semi-permanent features such as the ever-present ocean glitter, while suppressing geographic “noise,” in this study, we introduce temporally and conditionally averaged reflectance images, performed on a fixed grid of pixels and uniquely suited to the DSCOVR/EPIC observational circumstances. The resulting images (maps), averaged in time over months and conditioned on surface/cover type such as land, ocean, or clouds, show seasonal dependence literally at a glance, e.g., by an apparent extent of polar caps. Clear ocean-only aggregate maps feature central patches of ocean glitter, linking directly to surface roughness and, thereby, global winds. When combined with clouds, these blue planet “moving average” maps also serve as diagnostic tools for cloud retrieval algorithms. Land-only images convey the prominence of Earth’s deserts and the variable opacity of the atmosphere at different wavelengths. Insights into climate science and diagnostic and educational tools are likely to emerge from such average reflectance maps.
深空气候观测站（DSCOVR）航天器在距离地球 ≈ 1.4 - 1.6 × 106 km 的拉格朗日点附近漂移，其地球多色成像相机（EPIC）每隔 1-2 h 对地球的整个日照面进行观测、为了探测 "信号"，即较长期的变化和半永久性特征（如无处不在的海洋闪光），同时抑制地理 "噪声"，在本研究中，我们引入了时间和条件平均反射率图像，在固定的像素网格上进行，非常适合 DSCOVR/EPIC 的观测环境。由此产生的图像（地图）按月进行时间平均，并以陆地、海洋或云层等表面/覆盖类型为条件，一目了然地显示出季节相关性，如极地帽的明显范围。清晰的纯海洋集合图显示了海洋闪烁的中心斑块，与地表粗糙度直接相关，进而与全球风向相关。当与云层结合时，这些蓝色星球 "移动平均 "图还可作为云层检索算法的诊断工具。纯陆地图像可显示地球沙漠的显著位置，以及不同波长下大气的不透明度。这种平均反射率地图很可能会为气候科学以及诊断和教育工具提供启示。
{"title":"Deep space observations of conditionally averaged global reflectance patterns","authors":"Alexander Kostinski, A. Marshak, T. Várnai","doi":"10.3389/frsen.2024.1404461","DOIUrl":"https://doi.org/10.3389/frsen.2024.1404461","url":null,"abstract":"The Deep Space Climate Observatory (DSCOVR) spacecraft drifts about the Lagrangian point ≈ 1.4 − 1.6 × 106 km from Earth, where its Earth Polychromatic Imaging Camera (EPIC) observes the entire sunlit face of Earth every 1–2 h. In an attempt to detect “signals,” i.e., longer-term changes and semi-permanent features such as the ever-present ocean glitter, while suppressing geographic “noise,” in this study, we introduce temporally and conditionally averaged reflectance images, performed on a fixed grid of pixels and uniquely suited to the DSCOVR/EPIC observational circumstances. The resulting images (maps), averaged in time over months and conditioned on surface/cover type such as land, ocean, or clouds, show seasonal dependence literally at a glance, e.g., by an apparent extent of polar caps. Clear ocean-only aggregate maps feature central patches of ocean glitter, linking directly to surface roughness and, thereby, global winds. When combined with clouds, these blue planet “moving average” maps also serve as diagnostic tools for cloud retrieval algorithms. Land-only images convey the prominence of Earth’s deserts and the variable opacity of the atmosphere at different wavelengths. Insights into climate science and diagnostic and educational tools are likely to emerge from such average reflectance maps.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"7 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141834333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-10DOI: 10.3389/frsen.2024.1401653
Weibin Chen, M. Tsamados, R. Willatt, So Takao, D. Brockley, Claude De Rijke-Thomas, Alistair Francis, Thomas Johnson, Jack Landy, Isobel R. Lawrence, Sanggyun Lee, Dorsa Nasrollahi Shirazi, Wenxuan Liu, Connor Nelson, Julienne Stroeve, Len Hirata, M. Deisenroth
The Sentinel-3A and Sentinel-3B satellites, launched in February 2016 and April 2018 respectively, build on the legacy of CryoSat-2 by providing high-resolution Ku-band radar altimetry data over the polar regions up to 81° North. The combination of synthetic aperture radar (SAR) mode altimetry (SRAL instrument) from Sentinel-3A and Sentinel-3B, and the Ocean and Land Colour Instrument (OLCI) imaging spectrometer, results in the creation of the first satellite platform that offers coincident optical imagery and SAR radar altimetry. We utilise this synergy between altimetry and imagery to demonstrate a novel application of deep learning to distinguish sea ice from leads in spring. We use SRAL classified leads as training input for pan-Arctic lead detection from OLCI imagery. This surface classification is an important step for estimating sea ice thickness and to predict future sea ice changes in the Arctic and Antarctic regions. We propose the use of Vision Transformers (ViT), an approach adapting the popular deep learning algorithm Transformer, for this task. Their effectiveness, in terms of both quantitative metric including accuracy and qualitative metric including model roll-out, on several entire OLCI images is demonstrated and we show improved skill compared to previous machine learning and empirical approaches. We show the potential for this method to provide lead fraction retrievals at improved accuracy and spatial resolution for sunlit periods before melt onset.
{"title":"Co-located OLCI optical imagery and SAR altimetry from Sentinel-3 for enhanced Arctic spring sea ice surface classification","authors":"Weibin Chen, M. Tsamados, R. Willatt, So Takao, D. Brockley, Claude De Rijke-Thomas, Alistair Francis, Thomas Johnson, Jack Landy, Isobel R. Lawrence, Sanggyun Lee, Dorsa Nasrollahi Shirazi, Wenxuan Liu, Connor Nelson, Julienne Stroeve, Len Hirata, M. Deisenroth","doi":"10.3389/frsen.2024.1401653","DOIUrl":"https://doi.org/10.3389/frsen.2024.1401653","url":null,"abstract":"The Sentinel-3A and Sentinel-3B satellites, launched in February 2016 and April 2018 respectively, build on the legacy of CryoSat-2 by providing high-resolution Ku-band radar altimetry data over the polar regions up to 81° North. The combination of synthetic aperture radar (SAR) mode altimetry (SRAL instrument) from Sentinel-3A and Sentinel-3B, and the Ocean and Land Colour Instrument (OLCI) imaging spectrometer, results in the creation of the first satellite platform that offers coincident optical imagery and SAR radar altimetry. We utilise this synergy between altimetry and imagery to demonstrate a novel application of deep learning to distinguish sea ice from leads in spring. We use SRAL classified leads as training input for pan-Arctic lead detection from OLCI imagery. This surface classification is an important step for estimating sea ice thickness and to predict future sea ice changes in the Arctic and Antarctic regions. We propose the use of Vision Transformers (ViT), an approach adapting the popular deep learning algorithm Transformer, for this task. Their effectiveness, in terms of both quantitative metric including accuracy and qualitative metric including model roll-out, on several entire OLCI images is demonstrated and we show improved skill compared to previous machine learning and empirical approaches. We show the potential for this method to provide lead fraction retrievals at improved accuracy and spatial resolution for sunlit periods before melt onset.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"17 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141661470","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-08DOI: 10.3389/frsen.2024.1404877
R. Barton-Grimley, A. Nehrir
Measurements of water vapor are important for understanding the hydrological cycle, the thermodynamic structure of the lower troposphere, and broader atmospheric circulation. Subsequently, many scientific communities have emphasized a need for high-accuracy and spatial resolution profiles of water vapor within and above the planetary boundary layer (PBL). Advancements in lidar technologies at the NASA Langley Research Center are ongoing to enable the first space-based water vapor differential absorption lidar (DIAL) that can provide high-accuracy and vertical resolution retrievals of moisture in the PBL and through the mid-troposphere. The performance of this space-based DIAL is assessed here for sensitivity throughout the troposphere and globally with representative canonical cases of water vapor and aerosol loading. The specific humidity retrieval sensitivity to systematic and random errors is assessed, and measurement resolutions and capabilities are provided. We show that tunable operation along the side of the 823-nm absorption line allows for the optimization of the lower-tropospheric water vapor retrievals across different meteorological regimes and latitudes and provides the operational flexibility needed to dynamically optimize random errors for different scientific applications. The analysis presented here suggests that baseline and threshold systematic error requirements of <1.5% and <2.5%, respectively, are achievable. Random error is shown to dominate the retrieval, with errors on the order of 5% within the PBL being achievable with 300-m vertical 50-km horizontal resolutions over open ocean and on the order of 10%–15% over high-albedo surfaces. The flexibility of the DIAL method to trade retrieval precision for spatial resolution is shown, highlighting its strengths over passive techniques to tailor retrievals to different scientific applications. Combined, the total error budget demonstrated here indicates a high impact for space-based DIAL, with technologies being advanced for space missions within the next 5–10 years.
{"title":"Sensitivity analysis of space-based water vapor differential absorption lidar at 823 nm","authors":"R. Barton-Grimley, A. Nehrir","doi":"10.3389/frsen.2024.1404877","DOIUrl":"https://doi.org/10.3389/frsen.2024.1404877","url":null,"abstract":"Measurements of water vapor are important for understanding the hydrological cycle, the thermodynamic structure of the lower troposphere, and broader atmospheric circulation. Subsequently, many scientific communities have emphasized a need for high-accuracy and spatial resolution profiles of water vapor within and above the planetary boundary layer (PBL). Advancements in lidar technologies at the NASA Langley Research Center are ongoing to enable the first space-based water vapor differential absorption lidar (DIAL) that can provide high-accuracy and vertical resolution retrievals of moisture in the PBL and through the mid-troposphere. The performance of this space-based DIAL is assessed here for sensitivity throughout the troposphere and globally with representative canonical cases of water vapor and aerosol loading. The specific humidity retrieval sensitivity to systematic and random errors is assessed, and measurement resolutions and capabilities are provided. We show that tunable operation along the side of the 823-nm absorption line allows for the optimization of the lower-tropospheric water vapor retrievals across different meteorological regimes and latitudes and provides the operational flexibility needed to dynamically optimize random errors for different scientific applications. The analysis presented here suggests that baseline and threshold systematic error requirements of <1.5% and <2.5%, respectively, are achievable. Random error is shown to dominate the retrieval, with errors on the order of 5% within the PBL being achievable with 300-m vertical 50-km horizontal resolutions over open ocean and on the order of 10%–15% over high-albedo surfaces. The flexibility of the DIAL method to trade retrieval precision for spatial resolution is shown, highlighting its strengths over passive techniques to tailor retrievals to different scientific applications. Combined, the total error budget demonstrated here indicates a high impact for space-based DIAL, with technologies being advanced for space missions within the next 5–10 years.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":" 841","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141669216","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：