Pub Date : 2025-01-26DOI: 10.1016/j.rse.2025.114617
Joan Francesc Munoz-Martin, Nereida Rodriguez-Alvarez, Xavier Bosch-Lluis, Kamal Oudrhiri
This study presents a novel methodology for estimating sea-ice thickness (SIT) using polarimetric Global Navigation Satellite System – Reflectometry (GNSS-R). Building on previous work that demonstrated the capability of GNSS-R to measure thin sea ice, this research extends the application to thicker and multi-year sea ice using data from the Soil Moisture Active Passive (SMAP) mission. The study employs three key datasets: polarimetric GNSS-R data from SMAP, sea-ice thickness data from CryoSat-2 and SMOS, and ice temperature data from ERA5. A detailed model correlating the GNSS-R reflectivity to SIT and incorporating the impact of sea-ice salinity is developed. Results show high correlation coefficients between the GNSS-R derived parameters and the CryoSat-2/SMOS SIT data, indicating the method's robustness. The study concludes that full polarimetric GNSS-R can be useful to estimate sea ice salinity and density, critical to improve SIT models for its use in GNSS-R, other radar, and microwave radiometry instruments.
{"title":"Integrated retrieval of sea-ice salinity, density, and thickness using polarimetric GNSS-R","authors":"Joan Francesc Munoz-Martin, Nereida Rodriguez-Alvarez, Xavier Bosch-Lluis, Kamal Oudrhiri","doi":"10.1016/j.rse.2025.114617","DOIUrl":"https://doi.org/10.1016/j.rse.2025.114617","url":null,"abstract":"This study presents a novel methodology for estimating sea-ice thickness (SIT) using polarimetric Global Navigation Satellite System – Reflectometry (GNSS-R). Building on previous work that demonstrated the capability of GNSS-R to measure thin sea ice, this research extends the application to thicker and multi-year sea ice using data from the Soil Moisture Active Passive (SMAP) mission. The study employs three key datasets: polarimetric GNSS-R data from SMAP, sea-ice thickness data from CryoSat-2 and SMOS, and ice temperature data from ERA5. A detailed model correlating the GNSS-R reflectivity to SIT and incorporating the impact of sea-ice salinity is developed. Results show high correlation coefficients between the GNSS-R derived parameters and the CryoSat-2/SMOS SIT data, indicating the method's robustness. The study concludes that full polarimetric GNSS-R can be useful to estimate sea ice salinity and density, critical to improve SIT models for its use in GNSS-R, other radar, and microwave radiometry instruments.","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"31 1","pages":""},"PeriodicalIF":13.5,"publicationDate":"2025-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143035067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Foliage area volume density (FAVD) and leaf chlorophyll content (LCC) are two key traits closely linked to the structure and physiological status of trees. However, their physically-based retrieval at the individual tree level has remained challenging due to the complex interactions of scattering and absorption within the irregularly shaped tree crowns, as well as multiple scattering among neighboring trees, particularly in the near-infrared (NIR) spectrum. In this study, we proposed a tree-specific retrieval strategy that leverages unmanned aerial vehicle (UAV) imagery and corresponding photogrammetric point clouds to establish a tree-specific spatial adjacency constraint within the three-dimensional (3D) RTM-based inversion procedure for each individual tree. Unlike previous approaches that relied exclusively on pixel-level information from the region of interest, the proposed method fully accounted for the multiple scattering from adjacent trees and explicitly incorporates the irregularity of tree crown shapes. In the RTM-based prediction of the spectral reflectance of a focal tree (i.e., the target tree), the structures of adjacent trees were integrated alongside the focal tree, thereby forming a spatial adjacency constraint. This ensures that the scattering regime of the focal tree in the simulated scenario aligns with that of the actual scenario. The proposed method was assessed using both real UAV data and synthetic datasets. The results showed that tree-level retrieval under the adjacency constraint was highly consistent with reference (RRMSE of less than 0.22), whereas retrieval without the adjacency constraint exhibited substantial mis-estimation, particularly for FAVD (RRMSE of up to 0.44). Although the multiple scattering from adjacent trees was primarily influenced by the illumination geometry and tree canopy cover (TCC), sensitivity analysis of the sun zenith angle (SZA) and TCC revealed that retrieval accuracy slightly improved with a decreasing SZA and an increasing TCC. This improvement can be attributed to the enhanced treatment of multiple scattering under these conditions. These findings underscore the effectiveness of the tree-specific retrieval strategy for accurately estimating plant functional traits across forest stands. Moreover, they suggest the potential for monitoring functional diversity and long-term ecosystem process at the forest landscape scale through the use of functional traits.
{"title":"Seeing into individual trees: Tree-specific retrieval of tree-level traits using 3D radiative transfer model and spatial adjacency constraint from UAV multispectral imagery","authors":"Linyuan Li, Shangbo Liu, Zhihui Wang, Xun Zhao, Jianbo Qi, Yelu Zeng, Dong Li, Pengfei Guo, Zhexiu Yu, Simei Lin, Shouyang Liu, Huaguo Huang","doi":"10.1016/j.rse.2025.114616","DOIUrl":"https://doi.org/10.1016/j.rse.2025.114616","url":null,"abstract":"Foliage area volume density (FAVD) and leaf chlorophyll content (LCC) are two key traits closely linked to the structure and physiological status of trees. However, their physically-based retrieval at the individual tree level has remained challenging due to the complex interactions of scattering and absorption within the irregularly shaped tree crowns, as well as multiple scattering among neighboring trees, particularly in the near-infrared (NIR) spectrum. In this study, we proposed a tree-specific retrieval strategy that leverages unmanned aerial vehicle (UAV) imagery and corresponding photogrammetric point clouds to establish a tree-specific spatial adjacency constraint within the three-dimensional (3D) RTM-based inversion procedure for each individual tree. Unlike previous approaches that relied exclusively on pixel-level information from the region of interest, the proposed method fully accounted for the multiple scattering from adjacent trees and explicitly incorporates the irregularity of tree crown shapes. In the RTM-based prediction of the spectral reflectance of a focal tree (i.e., the target tree), the structures of adjacent trees were integrated alongside the focal tree, thereby forming a spatial adjacency constraint. This ensures that the scattering regime of the focal tree in the simulated scenario aligns with that of the actual scenario. The proposed method was assessed using both real UAV data and synthetic datasets. The results showed that tree-level retrieval under the adjacency constraint was highly consistent with reference (RRMSE of less than 0.22), whereas retrieval without the adjacency constraint exhibited substantial mis-estimation, particularly for FAVD (RRMSE of up to 0.44). Although the multiple scattering from adjacent trees was primarily influenced by the illumination geometry and tree canopy cover (TCC), sensitivity analysis of the sun zenith angle (SZA) and TCC revealed that retrieval accuracy slightly improved with a decreasing SZA and an increasing TCC. This improvement can be attributed to the enhanced treatment of multiple scattering under these conditions. These findings underscore the effectiveness of the tree-specific retrieval strategy for accurately estimating plant functional traits across forest stands. Moreover, they suggest the potential for monitoring functional diversity and long-term ecosystem process at the forest landscape scale through the use of functional traits.","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"48 1","pages":""},"PeriodicalIF":13.5,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143031103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ICESat-2 provides the potential for high-resolution and accurate measurements of the sea ice state. However, the current ATL07 sea ice height and type product relies on a threshold method for surface type classification, which introduces uncertainties in lead detection, especially in summer. In addition, it only categorizes into sea ice and lead types, excluding gray ice and the dark lead category has been shown to misclassify leads in cloudy conditions. To address these issues, we seek to improve the surface type classification by combining unsupervised and supervised machine learning methods and leveraging coincident imagery obtained from Sentinel-2. First, we use an unsupervised Gaussian Mixture Model (GMM) with four statistical parameters—photon rate, background rate, width of distribution, and height—to group ATL07 segments into 80 clusters. These clusters are then assigned specific surface types—sea ice, gray ice, or lead—based on coincident Sentinel-2 imagery. In the second step, we train a supervised K-nearest neighbor (KNN) classification model using the labeled segments from the GMM as training data. We conduct Leave One Group Out cross-validation of our model using coincident Sentinel-2 images as the ground truth, analyzing 717,009 strong beam and 702,843 weak beam ATL07 segments. The results demonstrate an improvement in lead detection, with precision values reaching approximately 98.6 % for strong beams and 97.5 % for weak beams and recall values of 91.8 % for strong beams and 90.3 % for weak beams. Our approach is applied to both Antarctic and Arctic sea ice, and is extended to include a new gray ice category, which agrees reasonably well with the coincident Sentinel-2 images. Our new sea ice and lead classification approach shows great promise for improving sea surface height and sea ice freeboard retrievals from ICESat-2 and highlights the significant value of coincident satellite imagery for classification training and validation.
{"title":"Enhanced sea ice classification for ICESat-2 using combined unsupervised and supervised machine learning","authors":"Wenxuan Liu, Michel Tsamados, Alek Petty, Taoyong Jin, Weibin Chen, Julienne Stroeve","doi":"10.1016/j.rse.2025.114607","DOIUrl":"https://doi.org/10.1016/j.rse.2025.114607","url":null,"abstract":"ICESat-2 provides the potential for high-resolution and accurate measurements of the sea ice state. However, the current ATL07 sea ice height and type product relies on a threshold method for surface type classification, which introduces uncertainties in lead detection, especially in summer. In addition, it only categorizes into sea ice and lead types, excluding gray ice and the dark lead category has been shown to misclassify leads in cloudy conditions. To address these issues, we seek to improve the surface type classification by combining unsupervised and supervised machine learning methods and leveraging coincident imagery obtained from Sentinel-2. First, we use an unsupervised Gaussian Mixture Model (GMM) with four statistical parameters—photon rate, background rate, width of distribution, and height—to group ATL07 segments into 80 clusters. These clusters are then assigned specific surface types—sea ice, gray ice, or lead—based on coincident Sentinel-2 imagery. In the second step, we train a supervised K-nearest neighbor (KNN) classification model using the labeled segments from the GMM as training data. We conduct Leave One Group Out cross-validation of our model using coincident Sentinel-2 images as the ground truth, analyzing 717,009 strong beam and 702,843 weak beam ATL07 segments. The results demonstrate an improvement in lead detection, with precision values reaching approximately 98.6 % for strong beams and 97.5 % for weak beams and recall values of 91.8 % for strong beams and 90.3 % for weak beams. Our approach is applied to both Antarctic and Arctic sea ice, and is extended to include a new gray ice category, which agrees reasonably well with the coincident Sentinel-2 images. Our new sea ice and lead classification approach shows great promise for improving sea surface height and sea ice freeboard retrievals from ICESat-2 and highlights the significant value of coincident satellite imagery for classification training and validation.","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"35 1","pages":""},"PeriodicalIF":13.5,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143031327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-24DOI: 10.1016/j.rse.2025.114614
Jinchen He, Shuhang Zhang, Wei Feng, Xiaodong Cui, Min Zhong
Coastal bathymetry is of great significance to the development and protection of islands and reefs. Traditional ship-based sonar bathymetry and airborne LiDAR (Light Detection and Ranging) bathymetry make it difficult to efficiently map the water depth of remote islands and reefs. Notably, the photon-counting LiDAR on board ICESat-2 (Ice, Cloud, and Land Elevation Satellite-2) has the capability of shallow water bathymetry. However, the photon data acquired by this instrument is contaminated with substantial noise of varying density. In this study, a sliding window-based coastal bathymetric method (SWCBM-Ph) is proposed for photon data with variable density. Experiments are carried out on six island coasts as an example, and the results show that the method is effective in extracting underwater terrain photons for bathymetry, with RMSE (Root Mean Square Error) of 0.60 m and 0.53 m on low- and high-density photon datasets separately within water depth of 30 m. Compared with existing bathymetry methods, the SWCBM-Ph is less affected by noise signals, and adapts to variations in photon density, including diversities between different datasets and within the same dataset. Therefore, the proposed method helps to improve the stability of spaceborne photon bathymetry for complex situations in coastal waters.
{"title":"A sliding window-based coastal bathymetric method for ICESat-2 photon-counting LiDAR data with variable photon density","authors":"Jinchen He, Shuhang Zhang, Wei Feng, Xiaodong Cui, Min Zhong","doi":"10.1016/j.rse.2025.114614","DOIUrl":"https://doi.org/10.1016/j.rse.2025.114614","url":null,"abstract":"Coastal bathymetry is of great significance to the development and protection of islands and reefs. Traditional ship-based sonar bathymetry and airborne LiDAR (Light Detection and Ranging) bathymetry make it difficult to efficiently map the water depth of remote islands and reefs. Notably, the photon-counting LiDAR on board ICESat-2 (Ice, Cloud, and Land Elevation Satellite-2) has the capability of shallow water bathymetry. However, the photon data acquired by this instrument is contaminated with substantial noise of varying density. In this study, a sliding window-based coastal bathymetric method (SWCBM-Ph) is proposed for photon data with variable density. Experiments are carried out on six island coasts as an example, and the results show that the method is effective in extracting underwater terrain photons for bathymetry, with RMSE (Root Mean Square Error) of 0.60 m and 0.53 m on low- and high-density photon datasets separately within water depth of 30 m. Compared with existing bathymetry methods, the SWCBM-Ph is less affected by noise signals, and adapts to variations in photon density, including diversities between different datasets and within the same dataset. Therefore, the proposed method helps to improve the stability of spaceborne photon bathymetry for complex situations in coastal waters.","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"15 1","pages":""},"PeriodicalIF":13.5,"publicationDate":"2025-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143031326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-23DOI: 10.1016/j.rse.2025.114608
Zihan Liu, Wenfeng Zhan, Yanlan Wu, Jiufeng Li, Huilin Du, Long Li, Shasha Wang, Chunli Wang
Land surface temperature (LST) acquired from polar orbiters serves as a critical dataset for investigating daily clear-sky climatology of surface urban heat islands (SUHIs). However, these orbiters only capture instantaneous LSTs at specific overpass times within a diurnal cycle, thus limiting the analysis of temporally sensitive SUHI metrics such as daily mean, maximum, and minimum SUHI intensity (SUHII). Consequently, the representativeness of daily clear-sky SUHI climatology derived from these instantaneous LSTs remains unclear, especially across global cities. Here we employ Aqua & Terra MODIS LST data alongside a well-established diurnal temperature cycle (DTC) model to assess such representativeness of daily clear-sky SUHI climatology, primarily based on the SUHII biases estimated from DTC-derived diurnally continuous and satellite-based temporally discrete LSTs across global cities. This approach discloses the fidelity of satellite-derived instantaneous LSTs to in representing daily clear-sky SUHI climatology. We further dissect the drivers of SUHII biases using the LightGBM model and SHAP algorithm. Our results reveal substantial underestimation of daily mean and maximum SUHIIs alongside overestimation of minimum SUHII when compared to the estimates directly based on instantaneous LSTs. The annual global mean SUHII biases for daily mean, maximum, and minimum conditions are 0.21 ± 0.13 K, 0.51 ± 0.18 K, and − 0.43 ± 0.17 K, respectively. We observe substantial seasonal and geographic variability in SUHII biases, with greater SUHII biases during winter, in snow climates, and across Europe and Oceania when compared to other seasons, climates, and continents. Notably, background climate is the principal contributor (34 %) to variation in SUHII bias, followed by surface properties (28 %), urban metrics (20 %), and human activity (18 %). Our findings emphasize the importance and show the feasibility of correcting SUHII biases in daily clear-sky SUHI climatology derived from instantaneous LSTs from polar orbiters across global cities.
极地轨道器获取的地表温度（LST）是研究地表城市热岛晴空气候的重要数据。然而，这些轨道器只捕获日周期内特定立交桥时间的瞬时lst，因此限制了对日平均、最大和最小SUHI强度（SUHII）等时间敏感的SUHI指标的分析。因此，从这些瞬时lst得出的每日晴空SUHI气候学的代表性仍然不清楚，特别是在全球城市中。这里我们使用Aqua &；Terra MODIS LST数据与一个完善的日温度循环（DTC）模型一起评估每日晴空SUHI气候学的代表性，主要基于DTC衍生的全球城市日连续和基于卫星的时间离散LST估计的SUHI偏差。这种方法揭示了卫星获取的瞬时地表温度对代表每日晴空SUHI气候的保真度。我们使用LightGBM模型和SHAP算法进一步剖析了SUHII偏差的驱动因素。我们的研究结果显示，与直接基于瞬时lst的估计相比，日平均和最大SUHII被严重低估，而最小SUHII被高估。日平均、最大值和最小值条件下的全球年平均SUHII偏差分别为0.21±0.13 K、0.51±0.18 K和- 0.43±0.17 K。我们观察到SUHII偏差存在明显的季节和地理差异，与其他季节、气候和大陆相比，冬季、雪气候、欧洲和大洋洲的SUHII偏差更大。值得注意的是，背景气候是SUHII偏差变化的主要贡献者（34%），其次是地表性质（28%）、城市指标（20%）和人类活动（18%）。我们的研究结果强调了更正SUHII偏差的重要性，并表明了在全球城市极地轨道器瞬时lst的每日晴空SUHI气候学中纠正SUHII偏差的可行性。
{"title":"Assessment of instantaneous sampling on quantifying satellite-derived surface urban heat islands: Biases and driving factors","authors":"Zihan Liu, Wenfeng Zhan, Yanlan Wu, Jiufeng Li, Huilin Du, Long Li, Shasha Wang, Chunli Wang","doi":"10.1016/j.rse.2025.114608","DOIUrl":"https://doi.org/10.1016/j.rse.2025.114608","url":null,"abstract":"Land surface temperature (LST) acquired from polar orbiters serves as a critical dataset for investigating daily clear-sky climatology of surface urban heat islands (SUHIs). However, these orbiters only capture instantaneous LSTs at specific overpass times within a diurnal cycle, thus limiting the analysis of temporally sensitive SUHI metrics such as daily mean, maximum, and minimum SUHI intensity (SUHII). Consequently, the representativeness of daily clear-sky SUHI climatology derived from these instantaneous LSTs remains unclear, especially across global cities. Here we employ Aqua & Terra MODIS LST data alongside a well-established diurnal temperature cycle (DTC) model to assess such representativeness of daily clear-sky SUHI climatology, primarily based on the SUHII biases estimated from DTC-derived diurnally continuous and satellite-based temporally discrete LSTs across global cities. This approach discloses the fidelity of satellite-derived instantaneous LSTs to in representing daily clear-sky SUHI climatology. We further dissect the drivers of SUHII biases using the LightGBM model and SHAP algorithm. Our results reveal substantial underestimation of daily mean and maximum SUHIIs alongside overestimation of minimum SUHII when compared to the estimates directly based on instantaneous LSTs. The annual global mean SUHII biases for daily mean, maximum, and minimum conditions are 0.21 ± 0.13 K, 0.51 ± 0.18 K, and − 0.43 ± 0.17 K, respectively. We observe substantial seasonal and geographic variability in SUHII biases, with greater SUHII biases during winter, in snow climates, and across Europe and Oceania when compared to other seasons, climates, and continents. Notably, background climate is the principal contributor (34 %) to variation in SUHII bias, followed by surface properties (28 %), urban metrics (20 %), and human activity (18 %). Our findings emphasize the importance and show the feasibility of correcting SUHII biases in daily clear-sky SUHI climatology derived from instantaneous LSTs from polar orbiters across global cities.","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"27 1","pages":""},"PeriodicalIF":13.5,"publicationDate":"2025-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143020957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-23DOI: 10.1016/j.rse.2025.114604
Raquel N. Buendía, Sajad Tabibi, Olivier Francis
Sea Level Anomaly (SLA) measurements are essential for understanding oceanic dynamics, climate variability, and climate change impacts. While satellite-based radar altimetry missions are the primary source of such measurements, their spatiotemporal resolution may sometimes be insufficient. This study explores the potential of Global Navigation Satellite Systems-Reflectometry (GNSS-R) as an additional approach for SLA retrieval. It exploits L-Band coherent carrier phase measurements collected by radio occultation receivers in Low Earth Orbit (LEO), known as Grazing angle GNSS-R (GG-R). We compare these GNSS-R measurements with those from traditional radar altimetry, in- cluding Sentinel-3 A/3B, Saral, and Cryosat-2. Our analysis of SLA data spanning from May 2019 to October 2021 reveals an average Root Mean Square Error (RMSE) of ∼47 cm among nearly 10,000 samples. We find that measure- ments derived from both techniques often complement each other when they meet recommended quality standards. Enhancing GG-R estimates could serve as a valuable complement to existing radar altimetry missions, which alone may not provide sufficient data. Furthermore, a comparison exclusively focused on GG-R events has been made to ensure consistency in the Spire GG-R retrievals, resulting in a 25 cm RMSE. Additionally, we conducted an assess- ment to evaluate the coherency and coverage of GG-R measurements. Approximately 24 % of the tracks are coherent, primarily located in the polar regions and calm waters.
{"title":"Exploring grazing angle GNSS-R for precision altimetry: A comparative study","authors":"Raquel N. Buendía, Sajad Tabibi, Olivier Francis","doi":"10.1016/j.rse.2025.114604","DOIUrl":"https://doi.org/10.1016/j.rse.2025.114604","url":null,"abstract":"Sea Level Anomaly (SLA) measurements are essential for understanding oceanic dynamics, climate variability, and climate change impacts. While satellite-based radar altimetry missions are the primary source of such measurements, their spatiotemporal resolution may sometimes be insufficient. This study explores the potential of Global Navigation Satellite Systems-Reflectometry (GNSS-R) as an additional approach for SLA retrieval. It exploits L-Band coherent carrier phase measurements collected by radio occultation receivers in Low Earth Orbit (LEO), known as Grazing angle GNSS-R (GG-R). We compare these GNSS-R measurements with those from traditional radar altimetry, in- cluding Sentinel-3 A/3B, Saral, and Cryosat-2. Our analysis of SLA data spanning from May 2019 to October 2021 reveals an average Root Mean Square Error (RMSE) of ∼47 cm among nearly 10,000 samples. We find that measure- ments derived from both techniques often complement each other when they meet recommended quality standards. Enhancing GG-R estimates could serve as a valuable complement to existing radar altimetry missions, which alone may not provide sufficient data. Furthermore, a comparison exclusively focused on GG-R events has been made to ensure consistency in the Spire GG-R retrievals, resulting in a 25 cm RMSE. Additionally, we conducted an assess- ment to evaluate the coherency and coverage of GG-R measurements. Approximately 24 % of the tracks are coherent, primarily located in the polar regions and calm waters.","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"1 1","pages":""},"PeriodicalIF":13.5,"publicationDate":"2025-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143020959","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Land surface temperature (LST) is a crucial parameter of the surface-atmosphere system, driving the water and heat exchange between the surface and the atmosphere. However, existing LST retrieval methods are highly sensitive to input errors. This study proposed a robust framework for retrieving LST, termed SW-NN, which integrates the physics-based Split-Window (SW) algorithm with a data-driven Neural Network (NN). The framework comprises of two main components: (1) a NN model that estimates SW coefficients as functions of key parameters such as brightness temperature (BT), water vapor content (WVC), land surface emissivity (LSE), and viewing zenith angle (VZA); and (2) a SW model that applies these coefficients to compute LST based on physical principles. By embedding the SW algorithm into the NN's loss function, this integrated design ensures that physical relationships guide the training process. The training data for the framework were generated by simulating satellite BT under a broad range of atmospheric and surface conditions using a radiative transfer model. To address the challenge of input errors, the proposed framework incorporates Gaussian noise into the training data to simulate realistic uncertainties in BT, WVC, and LSE. Specifically, noise with standard deviations of 0.05 K, 10 % of the WVC value, and 0.01 was added to BT, WVC, and LSE, respectively. Simulation analysis on an independent test set demonstrates that the proposed framework achieves a theoretical root-mean-square error (RMSE) of 0.60 K under the noise strategy, outperforming standalone NN and SW models. Sensitivity analysis, conducted using the same noise strategy applied during training, indicates that input errors affect LST retrieval by approximately 0.20 K, significantly enhancing the model's generalization and robustness. The proposed framework was also applied to MODIS data to retrieve LST, which was directly validated against global measurements from fifteen sites. Additionally, the proposed framework was compared with the NN method, the generalized split-window (GSW) method (MOD11 LST), and the Temperature Emissivity Separation (TES) method (MOD21 LST). The results showed that the proposed framework achieved an RMSE of 1.99 K, outperforming the NN method (RMSE = 2.08 K) and the GSW method (RMSE = 2.52 K), and performing comparably to the TES method (RMSE = 2.03 K). Further analysis in arid areas, where LSE accuracy is relatively lower, showed that the proposed framework improved the RMSE to 1.94 K compared to MOD11 LST, which had an RMSE of 3.02 K, utilizing the same LSE inputs. The proposed framework leverages the SW model's mechanism and the NN model's nonlinear fitting capability. It also demonstrates high robustness against input error, particularly LSE error. In summary, the proposed framework achieves robust and accurate LST retrieval, offering interpretability and a significant improvement over existing methods designed for sensors with two thermal infra
{"title":"A robust framework for accurate land surface temperature retrieval: Integrating split-window into knowledge-guided machine learning approach","authors":"Yuanliang Cheng, Hua Wu, Zhao-Liang Li, Frank-M. Göttsche, Xingxing Zhang, Xiujuan Li, Huanyu Zhang, Yitao Li","doi":"10.1016/j.rse.2025.114609","DOIUrl":"https://doi.org/10.1016/j.rse.2025.114609","url":null,"abstract":"Land surface temperature (LST) is a crucial parameter of the surface-atmosphere system, driving the water and heat exchange between the surface and the atmosphere. However, existing LST retrieval methods are highly sensitive to input errors. This study proposed a robust framework for retrieving LST, termed SW-NN, which integrates the physics-based Split-Window (SW) algorithm with a data-driven Neural Network (NN). The framework comprises of two main components: (1) a NN model that estimates SW coefficients as functions of key parameters such as brightness temperature (BT), water vapor content (WVC), land surface emissivity (LSE), and viewing zenith angle (VZA); and (2) a SW model that applies these coefficients to compute LST based on physical principles. By embedding the SW algorithm into the NN's loss function, this integrated design ensures that physical relationships guide the training process. The training data for the framework were generated by simulating satellite BT under a broad range of atmospheric and surface conditions using a radiative transfer model. To address the challenge of input errors, the proposed framework incorporates Gaussian noise into the training data to simulate realistic uncertainties in BT, WVC, and LSE. Specifically, noise with standard deviations of 0.05 K, 10 % of the WVC value, and 0.01 was added to BT, WVC, and LSE, respectively. Simulation analysis on an independent test set demonstrates that the proposed framework achieves a theoretical root-mean-square error (RMSE) of 0.60 K under the noise strategy, outperforming standalone NN and SW models. Sensitivity analysis, conducted using the same noise strategy applied during training, indicates that input errors affect LST retrieval by approximately 0.20 K, significantly enhancing the model's generalization and robustness. The proposed framework was also applied to MODIS data to retrieve LST, which was directly validated against global measurements from fifteen sites. Additionally, the proposed framework was compared with the NN method, the generalized split-window (GSW) method (MOD11 LST), and the Temperature Emissivity Separation (TES) method (MOD21 LST). The results showed that the proposed framework achieved an RMSE of 1.99 K, outperforming the NN method (RMSE = 2.08 K) and the GSW method (RMSE = 2.52 K), and performing comparably to the TES method (RMSE = 2.03 K). Further analysis in arid areas, where LSE accuracy is relatively lower, showed that the proposed framework improved the RMSE to 1.94 K compared to MOD11 LST, which had an RMSE of 3.02 K, utilizing the same LSE inputs. The proposed framework leverages the SW model's mechanism and the NN model's nonlinear fitting capability. It also demonstrates high robustness against input error, particularly LSE error. In summary, the proposed framework achieves robust and accurate LST retrieval, offering interpretability and a significant improvement over existing methods designed for sensors with two thermal infra","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"22 1","pages":""},"PeriodicalIF":13.5,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142991982","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-22DOI: 10.1016/j.rse.2025.114597
Yang Wang, Y. Jade Morton, J. Toby Minear, Alexa Putnam, Alex Conrad, Penina Axelrad, R. Steven Nerem, April Warnock, Christopher Ruf, Daniel Medeiros Moreira, Matthieu Talpe
River slope, a crucial parameter in hydrological modeling, has historically been difficult to measure continuously on a regional or global scale. Satellite altimetry missions often have long revisit times, such as 10 to 20 days for the Surface Water and Ocean Topography (SWOT) mission. In this paper, a novel approach is presented utilizing spaceborne GNSS Reflectometry (GNSS-R) to measure river slopes with high accuracy and potentially short revisit times. Our Earth is enveloped in radio signals from over 100 GNSS satellites. These signals can be coherently reflected from river surfaces and detected by low Earth orbit (LEO) satellites with sufficient energy to estimate carrier phase. The carrier phase measurement captures water surface height variations, which can be extracted through modeling of the reflection signal propagation geometry and space environment effects to estimate river slopes. This study processes both the raw intermediate frequency (IF) data obtained by NASA’s Cyclone GNSS (CYGNSS) microsatellites and the grazing-angle GNSS-R data generated by Spire Global nanosatellites to demonstrate the feasibility and performance of the GNSS-R based river slope retrieval. This paper focuses on selected river sections with width greater than 500 meters. Detailed methodologies and error analyses are presented, indicating total uncertainty of approximately 0.38 cm/km plus ionospheric TEC model error for CYGNSS and 0.69 cm/km for Spire (with dual-frequency ionospheric correction) over an ideal 5-km river section at 30° elevation angle. The retrieval results are validated in areas with nearby flat water surfaces (such as lakes or wide and slow river sections) and against in situ gauge measurements and satellite altimetry, consistently demonstrating the high accuracy and reliability of spaceborne GNSS-R for measuring river slopes.
{"title":"Measuring river slope using spaceborne GNSS reflectometry: Methodology and first performance assessment","authors":"Yang Wang, Y. Jade Morton, J. Toby Minear, Alexa Putnam, Alex Conrad, Penina Axelrad, R. Steven Nerem, April Warnock, Christopher Ruf, Daniel Medeiros Moreira, Matthieu Talpe","doi":"10.1016/j.rse.2025.114597","DOIUrl":"https://doi.org/10.1016/j.rse.2025.114597","url":null,"abstract":"River slope, a crucial parameter in hydrological modeling, has historically been difficult to measure continuously on a regional or global scale. Satellite altimetry missions often have long revisit times, such as 10 to 20 days for the Surface Water and Ocean Topography (SWOT) mission. In this paper, a novel approach is presented utilizing spaceborne GNSS Reflectometry (GNSS-R) to measure river slopes with high accuracy and potentially short revisit times. Our Earth is enveloped in radio signals from over 100 GNSS satellites. These signals can be coherently reflected from river surfaces and detected by low Earth orbit (LEO) satellites with sufficient energy to estimate carrier phase. The carrier phase measurement captures water surface height variations, which can be extracted through modeling of the reflection signal propagation geometry and space environment effects to estimate river slopes. This study processes both the raw intermediate frequency (IF) data obtained by NASA’s Cyclone GNSS (CYGNSS) microsatellites and the grazing-angle GNSS-R data generated by Spire Global nanosatellites to demonstrate the feasibility and performance of the GNSS-R based river slope retrieval. This paper focuses on selected river sections with width greater than <span><span style=\"\"></span><span data-mathml='<math xmlns=\"http://www.w3.org/1998/Math/MathML\"><mo is=\"true\">&#x223C;</mo></math>' role=\"presentation\" style=\"font-size: 90%; display: inline-block; position: relative;\" tabindex=\"0\"><svg aria-hidden=\"true\" focusable=\"false\" height=\"1.163ex\" role=\"img\" style=\"vertical-align: 0.307ex; margin-bottom: -0.427ex;\" viewbox=\"0 -449.1 778.5 500.8\" width=\"1.808ex\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><g fill=\"currentColor\" stroke=\"currentColor\" stroke-width=\"0\" transform=\"matrix(1 0 0 -1 0 0)\"><g is=\"true\"><use xlink:href=\"#MJMAIN-223C\"></use></g></g></svg><span role=\"presentation\"><math xmlns=\"http://www.w3.org/1998/Math/MathML\"><mo is=\"true\">∼</mo></math></span></span><script type=\"math/mml\"><math><mo is=\"true\">∼</mo></math></script></span>500 meters. Detailed methodologies and error analyses are presented, indicating total uncertainty of approximately 0.38 cm/km plus ionospheric TEC model error for CYGNSS and 0.69 cm/km for Spire (with dual-frequency ionospheric correction) over an ideal 5-km river section at 30° elevation angle. The retrieval results are validated in areas with nearby flat water surfaces (such as lakes or wide and slow river sections) and against in situ gauge measurements and satellite altimetry, consistently demonstrating the high accuracy and reliability of spaceborne GNSS-R for measuring river slopes.","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"37 1","pages":""},"PeriodicalIF":13.5,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142991981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-22DOI: 10.1016/j.rse.2025.114611
Zheng Du, Bao Zhang, Yibin Yao, Qingzhi Zhao, Liang Zhang
Various techniques have been developed to monitor water vapor because of its important role in weather forecasting and climate change studies. However, high-resolution, spatially continuous water vapor data remain scarce due to the sparsity of ground stations, coarse observational resolution, unavailability of remote sensing data during cloudy conditions, and systematic biases among different techniques. In this study we developed the Global Navigation Satellite System (GNSS) aided algorithms to retrieve Precipitable Water Vapor (PWV) from near-infrared (NIR), thermal infrared (TIR), and microwave (MW) observations from the Medium Resolution Spectral Imager II (MERSI-II) and the Microwave Radiation Imager (MWRI) onboard the Fengyun-3D satellite. We also proposed an improved iterative tropospheric decomposition algorithm to fuse the multiband PWV data, yielding the NIR + TIR PWV (0.01°), the MW PWV (0.25°), and the fused PWV (0.001°) for Australia. Validation against the GNSS PWV shows that the NIR + TIR PWV has a Root Mean Square Error (RMSE) of 1.45 mm and a bias of 0.07 mm, implying a 34 % improvement over the operational NIR products in terms of RMSE. The MW PWV shows RMSE and bias of 1.86 mm and 0.05 mm. The fused PWV integrates the advantages of different datasets, further enhancing the accuracy by 15 % for the NIR + TIR PWV and 21 % for the MW PWV. This study made the first attempt to retrieve PWV from three-band observations and delivers high-quality PWV products, which fills the data gap for high-resolution, spatially continuous PWV information.
由于水汽在天气预报和气候变化研究中的重要作用，人们开发了各种技术来监测水汽。然而，由于地面站稀疏、观测分辨率粗糙、多云条件下遥感数据不可用以及不同技术之间的系统偏差，高分辨率、空间连续的水汽数据仍然稀缺。在这项研究中，我们开发了全球导航卫星系统（GNSS）辅助算法，从中分辨率光谱成像仪II （MERSI-II）和微波辐射成像仪（MWRI）上的近红外（NIR）、热红外（TIR）和微波（MW）观测数据中检索可降水量（PWV）。我们还提出了一种改进的迭代对流层分解算法来融合多波段PWV数据，得到澳大利亚的NIR + TIR PWV（0.01°）、MW PWV（0.25°）和融合的PWV（0.001°）。对GNSS PWV的验证表明，近红外+ TIR PWV的均方根误差（RMSE）为1.45 mm，偏差为0.07 mm，这意味着在RMSE方面比实际近红外产品提高了34%。MW PWV的均方根误差和偏置分别为1.86 mm和0.05 mm。融合的PWV集成了不同数据集的优势，进一步提高了NIR + TIR PWV的精度15%，MW PWV的精度21%。该研究首次尝试从三波段观测中提取PWV，并提供了高质量的PWV产品，填补了高分辨率、空间连续PWV信息的数据空白。
{"title":"Integrating near-infrared, thermal infrared, and microwave satellite observations to retrieve high-resolution precipitable water vapor","authors":"Zheng Du, Bao Zhang, Yibin Yao, Qingzhi Zhao, Liang Zhang","doi":"10.1016/j.rse.2025.114611","DOIUrl":"https://doi.org/10.1016/j.rse.2025.114611","url":null,"abstract":"Various techniques have been developed to monitor water vapor because of its important role in weather forecasting and climate change studies. However, high-resolution, spatially continuous water vapor data remain scarce due to the sparsity of ground stations, coarse observational resolution, unavailability of remote sensing data during cloudy conditions, and systematic biases among different techniques. In this study we developed the Global Navigation Satellite System (GNSS) aided algorithms to retrieve Precipitable Water Vapor (PWV) from near-infrared (NIR), thermal infrared (TIR), and microwave (MW) observations from the Medium Resolution Spectral Imager II (MERSI-II) and the Microwave Radiation Imager (MWRI) onboard the Fengyun-3D satellite. We also proposed an improved iterative tropospheric decomposition algorithm to fuse the multiband PWV data, yielding the NIR + TIR PWV (0.01°), the MW PWV (0.25°), and the fused PWV (0.001°) for Australia. Validation against the GNSS PWV shows that the NIR + TIR PWV has a Root Mean Square Error (RMSE) of 1.45 mm and a bias of 0.07 mm, implying a 34 % improvement over the operational NIR products in terms of RMSE. The MW PWV shows RMSE and bias of 1.86 mm and 0.05 mm. The fused PWV integrates the advantages of different datasets, further enhancing the accuracy by 15 % for the NIR + TIR PWV and 21 % for the MW PWV. This study made the first attempt to retrieve PWV from three-band observations and delivers high-quality PWV products, which fills the data gap for high-resolution, spatially continuous PWV information.","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"58 1","pages":""},"PeriodicalIF":13.5,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142991980","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
River discharge plays an indispensable role in maintaining the stability of the hydrosphere system and eco-environment. Previous methods that utilize satellite imagery to estimate discharge over poorly gauged basins are generally tailored for large rivers and heavily reliant on ground-based measurements. Consequently, uncertainties often escalate when these methods are applied to medium-sized rivers. Based on Landsat 5 Thematic Mapper (TM) and unmanned aerial vehicle (UAV) images, this study proposed a framework for estimating the discharge of large and medium rivers with limited ground observations. It comprises (1) a modified C/M method, which considers the spatial heterogeneity of rivers using single-site observation data, and (2) a newly developed method for estimating river bathymetry with zero discharge measurements (RIBA-zero). Results show that, utilizing the modified C/M method, rivers wider than three times the satellite resolution (i.e., 90 m) exhibit a relative root mean square error (rRMSE) of 0.23 in the velocity estimation. Narrower rivers display a slight increase in the rRMSE (0.41), which is still within an encouraging range. For both types of river widths, the accuracy of flow velocity estimation is higher during high-flow periods compared with the low-flow counterparts. In terms of the flow area estimation, the RIBA-zero method is much more suited for parabola-shaped cross-sections (rRMSE = 0.22) and flood seasons (rRMSE = 0.35). Additionally, when replacing 30-m Landsat 5 TM with 10 m-resolution Sentinel-2 imageries, the approaches make a significant improvement in velocity estimation for rivers narrower than 90 m across all periods, exhibiting great potential to estimate discharge in medium rivers with finer resolution satellite imageries. The framework requires a few ground observations for discharge estimates with the Nash–Sutcliffe efficiency coefficient (NSE) reaching ∼0.9, thereby greatly facilitating hydrology-related studies with profound implications for sustainable water resources management worldwide.
{"title":"Combining Landsat 5 TM and UAV images to estimate river discharge with limited ground-based flow velocity and water level observations","authors":"Maomao Li, Changsen Zhao, Qi Huang, Tianli Pan, Hervé Yesou, Françoise Nerry, Zhao-Liang Li","doi":"10.1016/j.rse.2025.114610","DOIUrl":"https://doi.org/10.1016/j.rse.2025.114610","url":null,"abstract":"River discharge plays an indispensable role in maintaining the stability of the hydrosphere system and eco-environment. Previous methods that utilize satellite imagery to estimate discharge over poorly gauged basins are generally tailored for large rivers and heavily reliant on ground-based measurements. Consequently, uncertainties often escalate when these methods are applied to medium-sized rivers. Based on Landsat 5 Thematic Mapper (TM) and unmanned aerial vehicle (UAV) images, this study proposed a framework for estimating the discharge of large and medium rivers with limited ground observations. It comprises (1) a modified C/M method, which considers the spatial heterogeneity of rivers using single-site observation data, and (2) a newly developed method for estimating river bathymetry with zero discharge measurements (RIBA-zero). Results show that, utilizing the modified <em>C</em>/<em>M</em> method, rivers wider than three times the satellite resolution (i.e., 90 m) exhibit a relative root mean square error (rRMSE) of 0.23 in the velocity estimation. Narrower rivers display a slight increase in the rRMSE (0.41), which is still within an encouraging range. For both types of river widths, the accuracy of flow velocity estimation is higher during high-flow periods compared with the low-flow counterparts. In terms of the flow area estimation, the RIBA-zero method is much more suited for parabola-shaped cross-sections (rRMSE = 0.22) and flood seasons (rRMSE = 0.35). Additionally, when replacing 30-m Landsat 5 TM with 10 m-resolution Sentinel-2 imageries, the approaches make a significant improvement in velocity estimation for rivers narrower than 90 m across all periods, exhibiting great potential to estimate discharge in medium rivers with finer resolution satellite imageries. The framework requires a few ground observations for discharge estimates with the Nash–Sutcliffe efficiency coefficient (NSE) reaching ∼0.9, thereby greatly facilitating hydrology-related studies with profound implications for sustainable water resources management worldwide.","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"62 1","pages":""},"PeriodicalIF":13.5,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142991983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：