Pub Date : 2023-12-21DOI: 10.3389/frsen.2023.1307976
T. Harmel
Above-water radiometry (AWR) methods have been developed to provide “ground-truth” (or fiducial) measurements for calibration and validation of the water color satellite missions. AWR is also an important tool for environmental survey from dedicated field missions. Under clear sky, the critical step of AWR is to retrieve the water-leaving radiance from radiometric measurements of the upward radiance that also includes the reflection of the direct sunlight and diffuse skylight reflected by the wind ruffled water surface toward the sensor. In order to correct for the surface reflection, sky radiance measurements are performed and converted into surface radiance through a factor often called “sea surface reflectance factor” or “effective Fresnel reflectance coefficient”. Based on theoretical and practical considerations, this factor was renamed surface-to-sky radiance ratio, Rss, to avoid misuse of the term reflectance as often encountered in the literature. Vector radiative transfer computations were performed over the spectral range 350–1,000 nm to provide angular values of Rss for a comprehensive set of aerosol loads and types (including maritime, continental desert and polluted models) and water surface roughness expressed in wave slope variances or in equivalent Cox-Munk wind speeds, for practical use. After separating direct and diffuse light components, it was shown that the spectral shape and amplitude of Rss are very sensitive to aerosol load and type even for extremely low values of the aerosol optical thickness. Uncertainty attached to Rss was computed based on propagation of errors made in aerosol and surface roughness parameters demonstrating the need to adapt the viewing geometry according to the Sun elevation and to associate concurrent aerosol measurements for optimal AWR protocols.
{"title":"Apparent surface-to-sky radiance ratio of natural waters including polarization and aerosol effects: implications for above-water radiometry","authors":"T. Harmel","doi":"10.3389/frsen.2023.1307976","DOIUrl":"https://doi.org/10.3389/frsen.2023.1307976","url":null,"abstract":"Above-water radiometry (AWR) methods have been developed to provide “ground-truth” (or fiducial) measurements for calibration and validation of the water color satellite missions. AWR is also an important tool for environmental survey from dedicated field missions. Under clear sky, the critical step of AWR is to retrieve the water-leaving radiance from radiometric measurements of the upward radiance that also includes the reflection of the direct sunlight and diffuse skylight reflected by the wind ruffled water surface toward the sensor. In order to correct for the surface reflection, sky radiance measurements are performed and converted into surface radiance through a factor often called “sea surface reflectance factor” or “effective Fresnel reflectance coefficient”. Based on theoretical and practical considerations, this factor was renamed surface-to-sky radiance ratio, Rss, to avoid misuse of the term reflectance as often encountered in the literature. Vector radiative transfer computations were performed over the spectral range 350–1,000 nm to provide angular values of Rss for a comprehensive set of aerosol loads and types (including maritime, continental desert and polluted models) and water surface roughness expressed in wave slope variances or in equivalent Cox-Munk wind speeds, for practical use. After separating direct and diffuse light components, it was shown that the spectral shape and amplitude of Rss are very sensitive to aerosol load and type even for extremely low values of the aerosol optical thickness. Uncertainty attached to Rss was computed based on propagation of errors made in aerosol and surface roughness parameters demonstrating the need to adapt the viewing geometry according to the Sun elevation and to associate concurrent aerosol measurements for optimal AWR protocols.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"42 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138948935","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 : 2023-12-12DOI: 10.3389/frsen.2023.1289963
Riad Guehaz, Venkataraman Sivakumar
In this study, space-borne remote sensing (Landsat-8, MODIS) was employed to evaluate the effects of forest fires occurring on 05 July 2021, over Khenchela province, Algeria. Our objective is to understand the severity of damage caused by the fire and its implications for vegetation and land cover. Utilizing the Normalized Difference Vegetation Index (NDVI) from MODIS data and Landsat-8 imagery, we report changes in vegetation health and land cover. To identify areas affected by forest fires and evaluate the severity of damage, the Normalized Burn Ratio (NBR) and Differenced Normalized Burn Ratio (dNBR) were calculated. Analysis showed that −1825.11 ha (1.21%) of the total area experienced severe burns, 3843.54 ha (2.54%) moderate to high severity burns, 3927.97 ha (2.59%) moderate to low severity burns and 9864.45 ha (6.51%) low severity burns. The area covered by vegetation decreased from 2014 to 2021, indicating a negative trend in vegetation cover over the study period.
{"title":"A case study about the forest fire occurred on 05 July 2021 over Khenchela province, Algeria, using space-borne remote sensing","authors":"Riad Guehaz, Venkataraman Sivakumar","doi":"10.3389/frsen.2023.1289963","DOIUrl":"https://doi.org/10.3389/frsen.2023.1289963","url":null,"abstract":"In this study, space-borne remote sensing (Landsat-8, MODIS) was employed to evaluate the effects of forest fires occurring on 05 July 2021, over Khenchela province, Algeria. Our objective is to understand the severity of damage caused by the fire and its implications for vegetation and land cover. Utilizing the Normalized Difference Vegetation Index (NDVI) from MODIS data and Landsat-8 imagery, we report changes in vegetation health and land cover. To identify areas affected by forest fires and evaluate the severity of damage, the Normalized Burn Ratio (NBR) and Differenced Normalized Burn Ratio (dNBR) were calculated. Analysis showed that −1825.11 ha (1.21%) of the total area experienced severe burns, 3843.54 ha (2.54%) moderate to high severity burns, 3927.97 ha (2.59%) moderate to low severity burns and 9864.45 ha (6.51%) low severity burns. The area covered by vegetation decreased from 2014 to 2021, indicating a negative trend in vegetation cover over the study period.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"26 10","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139009615","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 : 2023-12-07DOI: 10.3389/frsen.2023.1303347
J. Mcintire, D. Moyer, A. Angal, Xiaoxiong Xiong
Scientific studies of the Earth’s climate increasingly rely on high-quality satellite observations. The Visible Infrared Imaging Radiometer Suite (VIIRS) is a key sensor onboard a series of satellites [Suomi National Polar-orbiting Partnership (SNPP) and Joint Polar-orbiting Satellite System 1–4 (JPSS-1–JPSS-4)] that generate scientific data from land, ocean, and atmosphere used in these climate models. Providing quality scientific data from space-borne sensors requires the instruments to be well-calibrated. While much of the calibration can be maintained on-orbit, some aspects of the calibration can best be measured prior to launch. One VIIRS parameter that needs to be measured pre-launch is the response versus scan angle (RVS). The RVS measures the relative change in the reflectance of the scanning optics as a function of the angle of incidence. With the RVS, the gain calibration measured on-orbit can be transferred to any scan angle. The JPSS-3 and JPSS-4 instruments have undergone ground testing including the RVS measurements, which is the subject of this work. Results indicate that the measurements are comparable to previous VIIRS builds and are expected to contribute to the generation of high-quality science data once JPSS-3 and JPSS-4 are on-orbit.
{"title":"JPSS-3 / 4 VIIRS response versus scan angle characterization and performance","authors":"J. Mcintire, D. Moyer, A. Angal, Xiaoxiong Xiong","doi":"10.3389/frsen.2023.1303347","DOIUrl":"https://doi.org/10.3389/frsen.2023.1303347","url":null,"abstract":"Scientific studies of the Earth’s climate increasingly rely on high-quality satellite observations. The Visible Infrared Imaging Radiometer Suite (VIIRS) is a key sensor onboard a series of satellites [Suomi National Polar-orbiting Partnership (SNPP) and Joint Polar-orbiting Satellite System 1–4 (JPSS-1–JPSS-4)] that generate scientific data from land, ocean, and atmosphere used in these climate models. Providing quality scientific data from space-borne sensors requires the instruments to be well-calibrated. While much of the calibration can be maintained on-orbit, some aspects of the calibration can best be measured prior to launch. One VIIRS parameter that needs to be measured pre-launch is the response versus scan angle (RVS). The RVS measures the relative change in the reflectance of the scanning optics as a function of the angle of incidence. With the RVS, the gain calibration measured on-orbit can be transferred to any scan angle. The JPSS-3 and JPSS-4 instruments have undergone ground testing including the RVS measurements, which is the subject of this work. Results indicate that the measurements are comparable to previous VIIRS builds and are expected to contribute to the generation of high-quality science data once JPSS-3 and JPSS-4 are on-orbit.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"35 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138593331","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 : 2023-12-07DOI: 10.3389/frsen.2023.1283719
L. P. Gaspar, Marina D. A. Scarpelli, Eliziane G. Oliveira, Rafael Souza Cruz Alves, Arthur Monteiro Gomes, Rafaela Wolf, Rafaela Vitti Ferneda, Silvia Harumi Kamazuka, C. Gussoni, Milton Cezar Ribeiro
The increasing conversion of natural areas for anthropic land use has been a major cause of habitat loss, destabilizing ecosystems and leading to a biodiversity crisis. Passive acoustic sensors open the possibility of remotely sensing fauna on large spatial and temporal scales, improving our understanding of the current state of biodiversity and the effects of human influences. Acoustic indices have been widely used and tested in recent years, with an aim towards understanding the relationship between indices and the acoustic activity of several taxa in different types of environments. However, studies have shown divergent relationships between acoustic indices and the vocal activity of most soniferous taxa. A combination of indices has, in turn, been reported as a promising tool for representing biodiversity in different contexts. We used uni- and bivariate models to test different combinations of 8 common indices in relation to bird assemblage metrics. We recorded twenty-two study sites in Brazil’s Atlantic Forest and three different types of environments in each site (forest, pasture, and swamp). Our results showed that 1) the best acoustic indices for explaining bird richness, abundance, and diversity were Bioacoustic and Acoustic Complexity; 2) the type of environment (forest, pasture, and swamp) influenced the performance of acoustic indices in explaining bird biodiversity, with the highest score model (biggest R2 value) being a combination between Acoustic Diversity and Bioacoustic indices. Our results do support the use of acoustic indices in monitoring the acoustic activity of birds, but combining indices is encouraged since it provided the best results. However, given the divergence we found across environments, we recommend that sets of indices are tested to determine which of them best describe the biodiversity pattern models for a specific habitat. Based on our results, we propose that biodiversity patterns can be predicted through acoustic patterns. However, the level of confidence will depend on the acoustic index used and on focal taxa of interest (i.e., birds, amphibians, insects, and mammals).
{"title":"Predicting bird diversity through acoustic indices within the Atlantic Forest biodiversity hotspot","authors":"L. P. Gaspar, Marina D. A. Scarpelli, Eliziane G. Oliveira, Rafael Souza Cruz Alves, Arthur Monteiro Gomes, Rafaela Wolf, Rafaela Vitti Ferneda, Silvia Harumi Kamazuka, C. Gussoni, Milton Cezar Ribeiro","doi":"10.3389/frsen.2023.1283719","DOIUrl":"https://doi.org/10.3389/frsen.2023.1283719","url":null,"abstract":"The increasing conversion of natural areas for anthropic land use has been a major cause of habitat loss, destabilizing ecosystems and leading to a biodiversity crisis. Passive acoustic sensors open the possibility of remotely sensing fauna on large spatial and temporal scales, improving our understanding of the current state of biodiversity and the effects of human influences. Acoustic indices have been widely used and tested in recent years, with an aim towards understanding the relationship between indices and the acoustic activity of several taxa in different types of environments. However, studies have shown divergent relationships between acoustic indices and the vocal activity of most soniferous taxa. A combination of indices has, in turn, been reported as a promising tool for representing biodiversity in different contexts. We used uni- and bivariate models to test different combinations of 8 common indices in relation to bird assemblage metrics. We recorded twenty-two study sites in Brazil’s Atlantic Forest and three different types of environments in each site (forest, pasture, and swamp). Our results showed that 1) the best acoustic indices for explaining bird richness, abundance, and diversity were Bioacoustic and Acoustic Complexity; 2) the type of environment (forest, pasture, and swamp) influenced the performance of acoustic indices in explaining bird biodiversity, with the highest score model (biggest R2 value) being a combination between Acoustic Diversity and Bioacoustic indices. Our results do support the use of acoustic indices in monitoring the acoustic activity of birds, but combining indices is encouraged since it provided the best results. However, given the divergence we found across environments, we recommend that sets of indices are tested to determine which of them best describe the biodiversity pattern models for a specific habitat. Based on our results, we propose that biodiversity patterns can be predicted through acoustic patterns. However, the level of confidence will depend on the acoustic index used and on focal taxa of interest (i.e., birds, amphibians, insects, and mammals).","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"47 23","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138593964","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 : 2023-12-05DOI: 10.3389/frsen.2023.1148328
D. Marzi, Antonietta Sorriso, Paolo Gamba
This study introduces a methodology for land cover mapping across extensive areas, utilizing multitemporal Sentinel-1 Synthetic Aperture Radar (SAR) data. The objective is to effectively process SAR data to extract spatio-temporal features that encapsulate temporal patterns within various land cover classes. The paper outlines the approach for processing multitemporal SAR data and presents an innovative technique for the selection of training points from an existing Medium Resolution Land Cover (MRLC) map. The methodology was tested across four distinct regions of interest, each spanning 100 × 100 km2, located in Siberia, Italy, Brazil, and Africa. These regions were chosen to evaluate the methodology’s applicability in diverse climate environments. The study reports both qualitative and quantitative results, showcasing the validity of the proposed procedure and the potential of SAR data for land cover mapping. The experimental outcomes demonstrate an average increase of 16% in overall accuracy compared to existing global products. The results suggest that the presented approach holds promise for enhancing land cover mapping accuracy, particularly when applied to extensive areas with varying land cover classes and environmental conditions. The ability to leverage multitemporal SAR data for this purpose opens new possibilities for improving global land cover maps and their applications.
{"title":"Automatic wide area land cover mapping using Sentinel-1 multitemporal data","authors":"D. Marzi, Antonietta Sorriso, Paolo Gamba","doi":"10.3389/frsen.2023.1148328","DOIUrl":"https://doi.org/10.3389/frsen.2023.1148328","url":null,"abstract":"This study introduces a methodology for land cover mapping across extensive areas, utilizing multitemporal Sentinel-1 Synthetic Aperture Radar (SAR) data. The objective is to effectively process SAR data to extract spatio-temporal features that encapsulate temporal patterns within various land cover classes. The paper outlines the approach for processing multitemporal SAR data and presents an innovative technique for the selection of training points from an existing Medium Resolution Land Cover (MRLC) map. The methodology was tested across four distinct regions of interest, each spanning 100 × 100 km2, located in Siberia, Italy, Brazil, and Africa. These regions were chosen to evaluate the methodology’s applicability in diverse climate environments. The study reports both qualitative and quantitative results, showcasing the validity of the proposed procedure and the potential of SAR data for land cover mapping. The experimental outcomes demonstrate an average increase of 16% in overall accuracy compared to existing global products. The results suggest that the presented approach holds promise for enhancing land cover mapping accuracy, particularly when applied to extensive areas with varying land cover classes and environmental conditions. The ability to leverage multitemporal SAR data for this purpose opens new possibilities for improving global land cover maps and their applications.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"19 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138598151","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}
Introduction: Monitoring surface water through the extraction of water bodies from high-resolution remote sensing images is of significant importance. With the advancements in deep learning, deep neural networks have been increasingly applied to high-resolution remote sensing image segmentation. However, conventional convolutional models face challenges in water body extraction, including issues like unclear water boundaries and a high number of training parameters.Methods: In this study, we employed the DeeplabV3+ network for water body extraction in high-resolution remote sensing images. However, the traditional DeeplabV3+ network exhibited limitations in segmentation accuracy for high-resolution remote sensing images and incurred high training costs due to a large number of parameters. To address these issues, we made several improvements to the traditional DeeplabV3+ network: Replaced the backbone network with MobileNetV2. Added a Channel Attention (CA) module to the MobileNetV2 feature extraction network. Introduced an Atrous Spatial Pyramid Pooling (ASPP) module. Implemented Focal loss for balanced loss computation.Results: Our proposed method yielded significant enhancements. It not only improved the segmentation accuracy of water bodies in high-resolution remote sensing images but also effectively reduced the number of network parameters and training time. Experimental results on the Water dataset demonstrated superior performance compared to other networks: Outperformed the U-Net network by 3.06% in terms of mean Intersection over Union (mIoU). Outperformed the MACU-Net network by 1.03%. Outperformed the traditional DeeplabV3+ network by 2.05%. The proposed method surpassed not only the traditional DeeplabV3+ but also U-Net, PSP-Net, and MACU-Net networks.Discussion: These results highlight the effectiveness of our modified DeeplabV3+ network with MobileNetV2 backbone, CA module, ASPP module, and Focal loss for water body extraction in high-resolution remote sensing images. The reduction in training time and parameters makes our approach a promising solution for accurate and efficient water body segmentation in remote sensing applications.
{"title":"Research on water extraction from high resolution remote sensing images based on deep learning","authors":"Peng Wu, Junjie Fu, Xiaomei Yi, Guoying Wang, Lufeng Mo, Brian Tapiwanashe Maponde, Hao Liang, Chunling Tao, Wenying Ge, Tengteng Jiang, Zhen Ren","doi":"10.3389/frsen.2023.1283615","DOIUrl":"https://doi.org/10.3389/frsen.2023.1283615","url":null,"abstract":"Introduction: Monitoring surface water through the extraction of water bodies from high-resolution remote sensing images is of significant importance. With the advancements in deep learning, deep neural networks have been increasingly applied to high-resolution remote sensing image segmentation. However, conventional convolutional models face challenges in water body extraction, including issues like unclear water boundaries and a high number of training parameters.Methods: In this study, we employed the DeeplabV3+ network for water body extraction in high-resolution remote sensing images. However, the traditional DeeplabV3+ network exhibited limitations in segmentation accuracy for high-resolution remote sensing images and incurred high training costs due to a large number of parameters. To address these issues, we made several improvements to the traditional DeeplabV3+ network: Replaced the backbone network with MobileNetV2. Added a Channel Attention (CA) module to the MobileNetV2 feature extraction network. Introduced an Atrous Spatial Pyramid Pooling (ASPP) module. Implemented Focal loss for balanced loss computation.Results: Our proposed method yielded significant enhancements. It not only improved the segmentation accuracy of water bodies in high-resolution remote sensing images but also effectively reduced the number of network parameters and training time. Experimental results on the Water dataset demonstrated superior performance compared to other networks: Outperformed the U-Net network by 3.06% in terms of mean Intersection over Union (mIoU). Outperformed the MACU-Net network by 1.03%. Outperformed the traditional DeeplabV3+ network by 2.05%. The proposed method surpassed not only the traditional DeeplabV3+ but also U-Net, PSP-Net, and MACU-Net networks.Discussion: These results highlight the effectiveness of our modified DeeplabV3+ network with MobileNetV2 backbone, CA module, ASPP module, and Focal loss for water body extraction in high-resolution remote sensing images. The reduction in training time and parameters makes our approach a promising solution for accurate and efficient water body segmentation in remote sensing applications.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"11 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138602101","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 : 2023-11-09DOI: 10.3389/frsen.2023.1249521
Thomas M. Jordan, Stefan G. H. Simis, Nick Selmes, Giulia Sent, Federico Ienna, Victor Martinez-Vicente
Validation of satellite-derived aquatic reflectance involves relating meter-scale in situ observations to satellite pixels with typical spatial resolution ∼ 10–100 m within a temporal “match-up window” of an overpass. Due to sub-pixel variation these discrepancies in measurement scale are a source of uncertainty in the validation result. Additionally, validation protocols and statistics do not normally account for spatial autocorrelation when pairing in situ data from moving platforms with satellite pixels. Here, using high-frequency autonomous mobile radiometers deployed on ships, we characterize the spatial structure of in situ R rs in inland and coastal waters (Lake Balaton, Western English Channel, Tagus Estuary). Using variogram analysis, we partition R rs variability into spatial and intrinsic (non-spatial) components. We then demonstrate the capacity of mobile radiometers to spatially sample in situ R rs within a temporal window broadly representative of satellite validation and provide spatial statistics to aid satellite validation practice. At a length scale typical of a medium resolution sensor (300 m) between 5% and 35% (median values across spectral bands and deployments) of the variation in in situ R rs was due to spatial separation. This result illustrates the extent to which mobile radiometers can reduce validation uncertainty due to spatial discrepancy via sub-pixel sampling. The length scale at which in situ R rs became spatially decorrelated ranged from ∼ 100–1,000 m. This information serves as a guideline for selection of spatially independent in situ R rs when matching with a satellite image, emphasizing the need for either downsampling or using modified statistics when selecting data to validate high resolution sensors (sub 100 m pixel size).
{"title":"Spatial structure of in situ reflectance in coastal and inland waters: implications for satellite validation","authors":"Thomas M. Jordan, Stefan G. H. Simis, Nick Selmes, Giulia Sent, Federico Ienna, Victor Martinez-Vicente","doi":"10.3389/frsen.2023.1249521","DOIUrl":"https://doi.org/10.3389/frsen.2023.1249521","url":null,"abstract":"Validation of satellite-derived aquatic reflectance involves relating meter-scale in situ observations to satellite pixels with typical spatial resolution ∼ 10–100 m within a temporal “match-up window” of an overpass. Due to sub-pixel variation these discrepancies in measurement scale are a source of uncertainty in the validation result. Additionally, validation protocols and statistics do not normally account for spatial autocorrelation when pairing in situ data from moving platforms with satellite pixels. Here, using high-frequency autonomous mobile radiometers deployed on ships, we characterize the spatial structure of in situ R rs in inland and coastal waters (Lake Balaton, Western English Channel, Tagus Estuary). Using variogram analysis, we partition R rs variability into spatial and intrinsic (non-spatial) components. We then demonstrate the capacity of mobile radiometers to spatially sample in situ R rs within a temporal window broadly representative of satellite validation and provide spatial statistics to aid satellite validation practice. At a length scale typical of a medium resolution sensor (300 m) between 5% and 35% (median values across spectral bands and deployments) of the variation in in situ R rs was due to spatial separation. This result illustrates the extent to which mobile radiometers can reduce validation uncertainty due to spatial discrepancy via sub-pixel sampling. The length scale at which in situ R rs became spatially decorrelated ranged from ∼ 100–1,000 m. This information serves as a guideline for selection of spatially independent in situ R rs when matching with a satellite image, emphasizing the need for either downsampling or using modified statistics when selecting data to validate high resolution sensors (sub 100 m pixel size).","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":" 33","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135293474","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 : 2023-11-02DOI: 10.3389/frsen.2023.1240895
Dorothy Menefee, Trey O. Lee, K. Colton Flynn, Jiquan Chen, Michael Abraha, John Baker, Andy Suyker
Introduction: Machine learning methods combined with satellite imagery have the potential to improve estimates of carbon uptake of terrestrial ecosystems, including croplands. Studying carbon uptake patterns across the U.S. using research networks, like the Long-Term Agroecosystem Research (LTAR) network, can allow for the study of broader trends in crop productivity and sustainability. Methods: In this study, gross primary productivity (GPP) estimates from the Moderate Resolution Imaging Spectroradiometer (MODIS) for three LTAR cropland sites were integrated for use in a machine learning modeling effort. They are Kellogg Biological Station (KBS, 2 towers and 20 site-years), Upper Mississippi River Basin (UMRB - Rosemount, 1 tower and 12 site-years), and Platte River High Plains Aquifer (PRHPA, 3 towers and 52 site-years). All sites were planted to maize ( Zea mays L .) and soybean ( Glycine max L .). The MODIS GPP product was initially compared to in-situ measurements from Eddy Covariance (EC) instruments at each site and then to all sites combined. Next, machine learning algorithms were used to create refined GPP estimates using air temperature, precipitation, crop type (maize or soybean), agroecosystem, and the MODIS GPP product as inputs. The AutoML program in the h2o package tested a variety of individual and combined algorithms, including Gradient Boosting Machines (GBM), eXtreme Gradient Boosting Models (XGBoost), and Stacked Ensemble. Results and discussion: The coefficient of determination ( r 2 ) of the raw comparison (MODIS GPP to EC GPP) was 0.38, prior to machine learning model incorporation. The optimal model for simulating GPP across all sites was a Stacked Ensemble type with a validated r 2 value of 0.87, RMSE of 2.62 units, and MAE of 1.59. The machine learning methodology was able to successfully simulate GPP across three agroecosystems and two crops.
导读:结合卫星图像的机器学习方法有可能改善对陆地生态系统(包括农田)碳吸收的估计。利用长期农业生态系统研究(LTAR)网络等研究网络研究美国各地的碳吸收模式，可以研究作物生产力和可持续性的更广泛趋势。方法:在本研究中，对三个LTAR农田的中分辨率成像光谱仪(MODIS)估算的总初级生产力(GPP)进行了整合，用于机器学习建模工作。它们是凯洛格生物站(KBS, 2个塔和20个站点年)，密西西比河上游流域(UMRB -罗斯蒙特，1个塔和12个站点年)和普拉特河高平原含水层(PRHPA, 3个塔和52个站点年)。所有试验点均种植玉米(Zea mays L .)和大豆(Glycine max L .)。MODIS GPP产品首先与每个站点的Eddy Covariance (EC)仪器的现场测量结果进行比较，然后与所有站点的测量结果进行比较。接下来，使用机器学习算法将气温、降水、作物类型(玉米或大豆)、农业生态系统和MODIS GPP产品作为输入，创建精细的GPP估计。h2o包中的AutoML程序测试了各种单独和组合算法，包括梯度增强机(GBM)，极端梯度增强模型(XGBoost)和堆叠集成。结果和讨论:在机器学习模型纳入之前，原始比较(MODIS GPP与EC GPP)的决定系数(r 2)为0.38。所有站点GPP的最优模拟模型为堆叠集成模型，验证r 2值为0.87,RMSE为2.62,MAE为1.59。机器学习方法能够成功地模拟三种农业生态系统和两种作物的GPP。
{"title":"Machine learning algorithms improve MODIS GPP estimates in United States croplands","authors":"Dorothy Menefee, Trey O. Lee, K. Colton Flynn, Jiquan Chen, Michael Abraha, John Baker, Andy Suyker","doi":"10.3389/frsen.2023.1240895","DOIUrl":"https://doi.org/10.3389/frsen.2023.1240895","url":null,"abstract":"Introduction: Machine learning methods combined with satellite imagery have the potential to improve estimates of carbon uptake of terrestrial ecosystems, including croplands. Studying carbon uptake patterns across the U.S. using research networks, like the Long-Term Agroecosystem Research (LTAR) network, can allow for the study of broader trends in crop productivity and sustainability. Methods: In this study, gross primary productivity (GPP) estimates from the Moderate Resolution Imaging Spectroradiometer (MODIS) for three LTAR cropland sites were integrated for use in a machine learning modeling effort. They are Kellogg Biological Station (KBS, 2 towers and 20 site-years), Upper Mississippi River Basin (UMRB - Rosemount, 1 tower and 12 site-years), and Platte River High Plains Aquifer (PRHPA, 3 towers and 52 site-years). All sites were planted to maize ( Zea mays L .) and soybean ( Glycine max L .). The MODIS GPP product was initially compared to in-situ measurements from Eddy Covariance (EC) instruments at each site and then to all sites combined. Next, machine learning algorithms were used to create refined GPP estimates using air temperature, precipitation, crop type (maize or soybean), agroecosystem, and the MODIS GPP product as inputs. The AutoML program in the h2o package tested a variety of individual and combined algorithms, including Gradient Boosting Machines (GBM), eXtreme Gradient Boosting Models (XGBoost), and Stacked Ensemble. Results and discussion: The coefficient of determination ( r 2 ) of the raw comparison (MODIS GPP to EC GPP) was 0.38, prior to machine learning model incorporation. The optimal model for simulating GPP across all sites was a Stacked Ensemble type with a validated r 2 value of 0.87, RMSE of 2.62 units, and MAE of 1.59. The machine learning methodology was able to successfully simulate GPP across three agroecosystems and two crops.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"24 10","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135974245","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 : 2023-10-25DOI: 10.3389/frsen.2023.1233803
Ray Nassar, Cameron G. MacDonald, Bruce Kuwahara, Alexander Fogal, Joshua Issa, Anthony Girmenia, Safwan Khan, Chris E. Sioris
For most CO 2 and CH 4 satellites, only a small percentage (∼10%) of observations yield successful retrievals, with the remaining ∼90% rejected, primarily due to the effects of clouds. Discarding this large fraction of data is an inefficient strategy worth reconsidering due to the costs involved in developing, launching and operating the satellites to make these observations. However, if real-time cloud data are available together with pointing capability, cloud data can guide the instrument pointing in an “intelligent pointing” strategy for cloud avoidance. In this work, multiple intelligent pointing simulations were conducted, demonstrating the significant advantages of this approach for satellites in a highly elliptical orbit (HEO), from which nearly the whole Earth disk can be observed. Multiple factors are shown to contribute to intelligent pointing efficiency such as the size and shape (or aspect ratio) of the field of view (FOV). For the current baseline orbit and Imaging Fourier Transform Spectrometer (IFTS) observing characteristics for the proposed Arctic Observing Mission (AOM), the monthly fraction of cloud-free observations is roughly a factor of 2 (ranging from ∼1.5–2.5) more than obtained with standard pointing (in which cloud information is not used). A similar efficiency is expected in a geostationary orbit (GEO) with an IFTS, however, for a dispersive instrument in HEO or GEO, the gain is more modest. This result is primarily attributed to the ∼1:1 aspect ratio of the IFTS FOV, since it is more efficient for cloud avoidance and scanning irregularly-shaped land masses than the long and narrow slit projection of a typical dispersive spectrometer. These results have implications for the design of future CO 2 or CH 4 monitoring satellites and constellation architectures, as well as other fields of satellite earth observation in which clouds significantly impact observations.
{"title":"Intelligent pointing increases the fraction of cloud-free CO2 and CH4 observations from space","authors":"Ray Nassar, Cameron G. MacDonald, Bruce Kuwahara, Alexander Fogal, Joshua Issa, Anthony Girmenia, Safwan Khan, Chris E. Sioris","doi":"10.3389/frsen.2023.1233803","DOIUrl":"https://doi.org/10.3389/frsen.2023.1233803","url":null,"abstract":"For most CO 2 and CH 4 satellites, only a small percentage (∼10%) of observations yield successful retrievals, with the remaining ∼90% rejected, primarily due to the effects of clouds. Discarding this large fraction of data is an inefficient strategy worth reconsidering due to the costs involved in developing, launching and operating the satellites to make these observations. However, if real-time cloud data are available together with pointing capability, cloud data can guide the instrument pointing in an “intelligent pointing” strategy for cloud avoidance. In this work, multiple intelligent pointing simulations were conducted, demonstrating the significant advantages of this approach for satellites in a highly elliptical orbit (HEO), from which nearly the whole Earth disk can be observed. Multiple factors are shown to contribute to intelligent pointing efficiency such as the size and shape (or aspect ratio) of the field of view (FOV). For the current baseline orbit and Imaging Fourier Transform Spectrometer (IFTS) observing characteristics for the proposed Arctic Observing Mission (AOM), the monthly fraction of cloud-free observations is roughly a factor of 2 (ranging from ∼1.5–2.5) more than obtained with standard pointing (in which cloud information is not used). A similar efficiency is expected in a geostationary orbit (GEO) with an IFTS, however, for a dispersive instrument in HEO or GEO, the gain is more modest. This result is primarily attributed to the ∼1:1 aspect ratio of the IFTS FOV, since it is more efficient for cloud avoidance and scanning irregularly-shaped land masses than the long and narrow slit projection of a typical dispersive spectrometer. These results have implications for the design of future CO 2 or CH 4 monitoring satellites and constellation architectures, as well as other fields of satellite earth observation in which clouds significantly impact observations.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135169247","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}
Soil moisture variability caused by soil erosion, weather extremes, and spatial variations in soil health is a limiting factor for crop growth and productivity. Crop evapotranspiration (ET) is significant for irrigation water management systems. The variability in crop water requirements at various growth stages is a common concern at a global level. In Canada’s Prince Edward Island (PEI), where agriculture is particularly prominent, this concern is predominantly evident. The island’s most prominent business, agriculture, finds it challenging to predict agricultural water needs due to shifting climate extremes, weather patterns, and precipitation patterns. Thus, accurate estimations for irrigation water requirements are essential for water conservation and precision farming. This work used a satellite-based normalized difference vegetation index (NDVI) technique to simulate the crop coefficient (K c ) and crop evapotranspiration (ET c ) for field-scale potato cultivation at various crop growth stages for the growing seasons of 2021 and 2022. The standard FAO Penman–Monteith equation was used to estimate the reference evapotranspiration (ET r ) using weather data from the nearest weather stations. The findings showed a statistically significant ( p < 0.05) positive association between NDVI and tabulated K c values extracted from all three satellites (Landsat 8, Sentinel-2A, and Planet) for the 2021 season. However, the correlation weakened in the subsequent year, particularly for Sentinel-2A and Planet data, while the association with Landsat 8 data became statistically insignificant ( p > 0.05). Sentinel-2A outperformed Landsat 8 and Planet overall. The K c values peaked at the halfway stage, fell before the maturity period, and were at their lowest at the start of the season. A similar pattern was observed for ET c (mm/day), which peaked at midseason and decreased with each developmental stage of the potato crop. Similar trends were observed for ET c (mm/day), which peaked at the mid-stage with mean values of 4.0 (2021) and 3.7 (2022), was the lowest in the initial phase with mean values of 1.8 (2021) and 1.5 (2022), and grew with each developmental stage of the potato crop. The study’s ET maps show how agricultural water use varies throughout a growing season. Farmers in Prince Edward Island may find the applied technique helpful in creating sustainable growth plans at different phases of crop development. Integrating high-resolution imagery with soil health, yield mapping, and crop growth parameters can help develop a decision support system to tailor sustainable management practices to improve profit margins, crop yield, and quality.
{"title":"Mapping crop evapotranspiration with high-resolution imagery and meteorological data: insights into sustainable agriculture in Prince Edward Island","authors":"Fatima Imtiaz, Aitazaz Farooque, Xander Wang, Farhat Abbas, Hassan Afzaal, Travis Esau, Bishnu Acharya, Qamar Zaman","doi":"10.3389/frsen.2023.1274019","DOIUrl":"https://doi.org/10.3389/frsen.2023.1274019","url":null,"abstract":"Soil moisture variability caused by soil erosion, weather extremes, and spatial variations in soil health is a limiting factor for crop growth and productivity. Crop evapotranspiration (ET) is significant for irrigation water management systems. The variability in crop water requirements at various growth stages is a common concern at a global level. In Canada’s Prince Edward Island (PEI), where agriculture is particularly prominent, this concern is predominantly evident. The island’s most prominent business, agriculture, finds it challenging to predict agricultural water needs due to shifting climate extremes, weather patterns, and precipitation patterns. Thus, accurate estimations for irrigation water requirements are essential for water conservation and precision farming. This work used a satellite-based normalized difference vegetation index (NDVI) technique to simulate the crop coefficient (K c ) and crop evapotranspiration (ET c ) for field-scale potato cultivation at various crop growth stages for the growing seasons of 2021 and 2022. The standard FAO Penman–Monteith equation was used to estimate the reference evapotranspiration (ET r ) using weather data from the nearest weather stations. The findings showed a statistically significant ( p &lt; 0.05) positive association between NDVI and tabulated K c values extracted from all three satellites (Landsat 8, Sentinel-2A, and Planet) for the 2021 season. However, the correlation weakened in the subsequent year, particularly for Sentinel-2A and Planet data, while the association with Landsat 8 data became statistically insignificant ( p &gt; 0.05). Sentinel-2A outperformed Landsat 8 and Planet overall. The K c values peaked at the halfway stage, fell before the maturity period, and were at their lowest at the start of the season. A similar pattern was observed for ET c (mm/day), which peaked at midseason and decreased with each developmental stage of the potato crop. Similar trends were observed for ET c (mm/day), which peaked at the mid-stage with mean values of 4.0 (2021) and 3.7 (2022), was the lowest in the initial phase with mean values of 1.8 (2021) and 1.5 (2022), and grew with each developmental stage of the potato crop. The study’s ET maps show how agricultural water use varies throughout a growing season. Farmers in Prince Edward Island may find the applied technique helpful in creating sustainable growth plans at different phases of crop development. Integrating high-resolution imagery with soil health, yield mapping, and crop growth parameters can help develop a decision support system to tailor sustainable management practices to improve profit margins, crop yield, and quality.","PeriodicalId":198378,"journal":{"name":"Frontiers in Remote Sensing","volume":"183 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135888805","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：