Remote Sensing of Environment最新文献_第4页

A three-stage framework for stand-level automated stem volume estimation in temperate forests using Mobile laser scanning 基于移动激光扫描的温带森林林分水平茎体积自动估算的三阶段框架

IF 11.4 1区地球科学 Q1 ENVIRONMENTAL SCIENCES

Remote Sensing of Environment

Pub Date : 2026-01-21 DOI: 10.1016/j.rse.2026.115246

Jinyuan Shao , Dennis Heejoon Choi , Jidong Liu , Xiangxi Tian , Bina Thapa , Seunghyeon Lee , Ayman Habib , Songlin Fei

Accurate stem-level volume estimation at large scale is highly desired in temperate natural forests due to their economic and ecological significance. Mobile Laser Scanning (MLS) systems (e.g., handheld or backpack) offer the ability to efficiently capture high-density point clouds over large areas, creating opportunities for automated, large-scale individual stem volume estimation. However, effective algorithms that can automatically and accurately analyze the dense and complex MLS point clouds of temperate natural forests are lacking. To address this issue, we propose a novel three-stage method to automatically detect, segment, and reconstruct individual stems, enabling direct volume estimations from MLS point clouds of temperate natural forests. First, a deep learning model is employed to separate understory vegetation from overstory trees, reducing point cloud complexity. Next, we introduce a Bidirectional Section Growing (BSG) method for individual stem detection and segmentation, specifically for the segmentation of merchantable logs and multi-stem scenarios, using a novel Least Squares with Similarity Optimization (LeSSO) algorithm. Finally, the Sector Median Points (SMP) method is developed to reconstruct stem shapes for precise volume estimation. Our method is evaluated on four datasets collected in temperate natural forests across the U.S. and Europe. Experimental results demonstrate its superior performance compared to state-of-the-art algorithms, achieving 89.2% Intersection over Union (IoU) for understory removal, 99.4% F-score for stem detection, 91.5% IoU for stem segmentation, and reconstruction accuracy with a Point-to-Mesh distance of 0.0004 m² and a Chamfer distance of 0.05 m. Moreover, we record 42 stem locations in the field for one of the U.S. datasets and conduct destructive measurements of section-wise diameters for each of them to serve as independent reference data to evaluate stem detection and volume estimation. Our method is able to detect all 42 trees from the point cloud, and reconstructed stem models yield the most accurate section-wise diameter estimates with a Root Mean Square Error (RMSE) of 2.27 cm and R² of 0.96, and best volume estimation with RMSE of 0.18 m³ and R² of 0.97. Our method paves the way for automated and accurate estimation of merchantable stem volume from MLS point clouds collected in complex temperate natural forests.

由于其经济和生态意义，在温带天然林中迫切需要大尺度准确的茎位体积估算。移动激光扫描（MLS）系统（例如，手持式或背包式）提供了有效捕获大面积高密度点云的能力，为自动化、大规模的单个茎体积估计创造了机会。然而，目前还缺乏能够自动准确分析温带天然林密集复杂MLS点云的有效算法。为了解决这一问题，我们提出了一种新的三阶段方法来自动检测、分割和重建单个茎干，从而实现了从温带天然林的MLS点云中直接估计体积。首先，采用深度学习模型分离林下植被和林下乔木，降低点云复杂度；接下来，我们引入了一种双向截面生长（BSG）方法，用于单个茎的检测和分割，特别是用于可销售日志和多茎场景的分割，该方法使用了一种新颖的最小二乘相似度优化（LeSSO）算法。最后，提出了扇形中值点（SMP）方法来重建茎的形状，以获得精确的体积估计。我们的方法在美国和欧洲温带天然林收集的四个数据集上进行了评估。实验结果表明，与现有算法相比，该算法具有更优异的性能，林下植被去除的交叉比对（Intersection over Union, IoU）达到89.2%，茎干检测的f值达到99.4%，茎干分割的IoU达到91.5%，点对网格距离为0.0004 m2，切角距离为0.05 m的重建精度。此外，我们在美国的一个数据集中记录了42个阀杆位置，并对每个阀杆的截面直径进行了破坏性测量，作为评估阀杆检测和体积估计的独立参考数据。我们的方法能够从点云中检测到所有42棵树，重建的树干模型产生了最准确的截面直径估计，均方根误差（RMSE）为2.27 cm， R2为0.96，最佳体积估计RMSE为0.18 m3， R2为0.97。我们的方法为从复杂温带天然林中收集的MLS点云中自动准确估计可销售茎体积铺平了道路。

{"title":"A three-stage framework for stand-level automated stem volume estimation in temperate forests using Mobile laser scanning","authors":"Jinyuan Shao , Dennis Heejoon Choi , Jidong Liu , Xiangxi Tian , Bina Thapa , Seunghyeon Lee , Ayman Habib , Songlin Fei","doi":"10.1016/j.rse.2026.115246","DOIUrl":"10.1016/j.rse.2026.115246","url":null,"abstract":"<div><div>Accurate stem-level volume estimation at large scale is highly desired in temperate natural forests due to their economic and ecological significance. Mobile Laser Scanning (MLS) systems (e.g., handheld or backpack) offer the ability to efficiently capture high-density point clouds over large areas, creating opportunities for automated, large-scale individual stem volume estimation. However, effective algorithms that can automatically and accurately analyze the dense and complex MLS point clouds of temperate natural forests are lacking. To address this issue, we propose a novel three-stage method to automatically detect, segment, and reconstruct individual stems, enabling direct volume estimations from MLS point clouds of temperate natural forests. First, a deep learning model is employed to separate understory vegetation from overstory trees, reducing point cloud complexity. Next, we introduce a Bidirectional Section Growing (BSG) method for individual stem detection and segmentation, specifically for the segmentation of merchantable logs and multi-stem scenarios, using a novel Least Squares with Similarity Optimization (LeSSO) algorithm. Finally, the Sector Median Points (SMP) method is developed to reconstruct stem shapes for precise volume estimation. Our method is evaluated on four datasets collected in temperate natural forests across the U.S. and Europe. Experimental results demonstrate its superior performance compared to state-of-the-art algorithms, achieving 89.2% Intersection over Union (IoU) for understory removal, 99.4% F-score for stem detection, 91.5% IoU for stem segmentation, and reconstruction accuracy with a Point-to-Mesh distance of 0.0004 m<sup>2</sup> and a Chamfer distance of 0.05 m. Moreover, we record 42 stem locations in the field for one of the U.S. datasets and conduct destructive measurements of section-wise diameters for each of them to serve as independent reference data to evaluate stem detection and volume estimation. Our method is able to detect all 42 trees from the point cloud, and reconstructed stem models yield the most accurate section-wise diameter estimates with a Root Mean Square Error (RMSE) of 2.27 cm and R<sup>2</sup> of 0.96, and best volume estimation with RMSE of 0.18 m<sup>3</sup> and R<sup>2</sup> of 0.97. Our method paves the way for automated and accurate estimation of merchantable stem volume from MLS point clouds collected in complex temperate natural forests.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"335 ","pages":"Article 115246"},"PeriodicalIF":11.4,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146015015","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}

引用次数: 0

Global analysis of nitrogen dioxide and formaldehyde column densities from the Pandora global network: Variability and implications for satellite validation 来自潘多拉全球网络的二氧化氮和甲醛柱密度的全球分析：变异性及其对卫星验证的影响

IF 11.4 1区地球科学 Q1 ENVIRONMENTAL SCIENCES

Remote Sensing of Environment

Pub Date : 2026-01-21 DOI: 10.1016/j.rse.2026.115249

Jong-Uk Park , Subin Lim , Thomas F. Hanisco , Nader Abuhassan , Bryan K. Place , Apoorva Pandey , Alexander Cede , Martin Tiefengraber , Manuel Gebetsberger , Jinsoo Park , Jinsoo Choi , James H. Crawford , Chang-Keun Song , Sang-Woo Kim

This study harnesses quality-assured global Pandora observations (2019–2023) from the Pandonia Global Network (PGN) to investigate diurnal and seasonal variations of NO₂ and HCHO—key proxies for tropospheric O₃—and to evaluate TROPOMI satellite observations. NO₂ vertical column densities (VCDs) at Polluted Urban stations peak in winter and gradually increase throughout the day, but show a decrease in afternoons in summer due to photochemical loss. Conversely, Rural/Background stations exhibit summer maxima with monotonic daytime increases across seasons, driven by stratospheric NO₂ variability. HCHO VCDs are higher in summer at most sites, with a morning increase followed by elevated concentrations throughout the afternoon. The spatial representativeness mismatch between satellite and Pandora observations results in negative biases in TROPOMI NO₂ VCDs at Polluted Urban stations and a valley station, while overestimations are found at high-altitude stations. Considerable random uncertainties in TROPOMI HCHO VCDs lead to low correlations (r² = 0.08–0.11) and high random errors (0.27–0.33 DU) across environments. Averaging collocated data points prior to intercomparison effectively reduces random biases, whereas increasing the spatial collocation range introduces biases due to spatial averaging effects. Tropospheric HCHO-to-NO₂ ratios (FNR_trop) retrieved from Pandora observations indicate that Polluted Urban (0.82 ± 0.08) and Rural/Background (1.64 ± 0.07) stations are generally under VOC-limited and NO_x-limited O₃ production regimes, respectively, while summertime increases in FNR_trop put Polluted Urban stations in a transitional range, yielding higher O₃ production efficiencies. TROPOMI-derived FNR_trop shows good agreement with Pandora in Polluted Urban stations (ΔFNR_median = 0.18), whereas random error increases in rural areas with lower tropospheric NO₂.

本研究利用潘多尼亚全球网络（PGN） 2019-2023年有质量保证的潘多拉全球观测数据，研究对流层臭氧的关键代用物NO2和hcho的日变化和季节变化，并评估TROPOMI卫星观测结果。城市污染站NO2垂直柱密度（vcd）在冬季达到峰值，白天逐渐升高，但在夏季下午由于光化学损失而下降。相反，在平流层NO2变率的驱动下，农村/背景站呈现夏季最大值，各季节日间单调增加。在大多数地点，HCHO vcd在夏季较高，上午浓度增加，随后整个下午浓度升高。卫星和Pandora观测数据的空间代表性不匹配导致城市污染站和山谷站的TROPOMI NO2 vcd出现负偏倚，而高海拔站则出现高估。TROPOMI HCHO vcd中相当大的随机不确定性导致不同环境下的低相关性（r2 = 0.08-0.11）和高随机误差（0.27-0.33 DU）。在相互比较之前对并置数据点进行平均可以有效地减少随机偏差，而增加空间搭配范围则会由于空间平均效应而引入偏差。从Pandora观测数据获取的对流层hho - no2比值（FNRtrop）表明，受污染的城市站（0.82±0.08）和农村/背景站（1.64±0.07）通常分别处于voc限制和nox限制的O3生产状态，而夏季FNRtrop的增加使受污染的城市站处于过渡范围，产生更高的O3生产效率。在受污染的城市站点，tropomi衍生的FNRtrop与Pandora具有良好的一致性（ΔFNRmedian = 0.18），而在对流层NO2较低的农村地区，随机误差增加。

{"title":"Global analysis of nitrogen dioxide and formaldehyde column densities from the Pandora global network: Variability and implications for satellite validation","authors":"Jong-Uk Park , Subin Lim , Thomas F. Hanisco , Nader Abuhassan , Bryan K. Place , Apoorva Pandey , Alexander Cede , Martin Tiefengraber , Manuel Gebetsberger , Jinsoo Park , Jinsoo Choi , James H. Crawford , Chang-Keun Song , Sang-Woo Kim","doi":"10.1016/j.rse.2026.115249","DOIUrl":"10.1016/j.rse.2026.115249","url":null,"abstract":"<div><div>This study harnesses quality-assured global Pandora observations (2019–2023) from the Pandonia Global Network (PGN) to investigate diurnal and seasonal variations of NO<sub>2</sub> and HCHO—key proxies for tropospheric O<sub>3</sub>—and to evaluate TROPOMI satellite observations. NO<sub>2</sub> vertical column densities (VCDs) at Polluted Urban stations peak in winter and gradually increase throughout the day, but show a decrease in afternoons in summer due to photochemical loss. Conversely, Rural/Background stations exhibit summer maxima with monotonic daytime increases across seasons, driven by stratospheric NO<sub>2</sub> variability. HCHO VCDs are higher in summer at most sites, with a morning increase followed by elevated concentrations throughout the afternoon. The spatial representativeness mismatch between satellite and Pandora observations results in negative biases in TROPOMI NO<sub>2</sub> VCDs at Polluted Urban stations and a valley station, while overestimations are found at high-altitude stations. Considerable random uncertainties in TROPOMI HCHO VCDs lead to low correlations (r<sup>2</sup> = 0.08–0.11) and high random errors (0.27–0.33 DU) across environments. Averaging collocated data points prior to intercomparison effectively reduces random biases, whereas increasing the spatial collocation range introduces biases due to spatial averaging effects. Tropospheric HCHO-to-NO<sub>2</sub> ratios (FNR<sub>trop</sub>) retrieved from Pandora observations indicate that Polluted Urban (0.82 ± 0.08) and Rural/Background (1.64 ± 0.07) stations are generally under VOC-limited and NO<sub>x</sub>-limited O<sub>3</sub> production regimes, respectively, while summertime increases in FNR<sub>trop</sub> put Polluted Urban stations in a transitional range, yielding higher O<sub>3</sub> production efficiencies. TROPOMI-derived FNR<sub>trop</sub> shows good agreement with Pandora in Polluted Urban stations (ΔFNR<sub>median</sub> = 0.18), whereas random error increases in rural areas with lower tropospheric NO<sub>2</sub>.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"335 ","pages":"Article 115249"},"PeriodicalIF":11.4,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146015018","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}

引用次数: 0

Automated rice mapping under diverse cropping patterns and establishment methods by integrating phenological knowledge and synergy of optical and SAR imagery 综合物候知识和光学与SAR影像协同作用的不同种植模式下水稻自动制图及建立方法

IF 11.4 1区地球科学 Q1 ENVIRONMENTAL SCIENCES

Remote Sensing of Environment

Pub Date : 2026-01-21 DOI: 10.1016/j.rse.2026.115255

Xingrong Li , Gaoxiang Yang , Meng He , Yuan Xiong , Leilei Liu , Hifza Mariam , Iftikhar Ali , Chongya Jiang , Xia Yao , Yan Zhu , Weixing Cao , Tao Cheng

Accurate and timely information on rice cultivation areas and cropping intensity is essential for precision crop management, food security and environmental sustainability. However, the generation of high-quality rice products is often hindered by the lack of ground truth samples, particularly in the regions with complex cropping patterns. While most rice mapping methods rely on the use of remotely sensed flooding signal from the transplanting period to distinguish rice from other crops, they face significant challenges from the increasingly prevalent direct-seeded rice for which the flooding signal at the beginning of season is weak due to the unique management measures. To address these issues and achieve the comprehensive extraction of rice under diverse cropping patterns and establishment methods, this study proposed the phenological knowledge-guided automatic rice mapping approach using optical and synthetic aperture radar (SAR) data (PHAROS). This method first determined the cropping intensity of rice and its concurrent crops using harmonized multi-resource Normalized Difference Vegetation Index (NDVI) time series and phenological knowledge. Subsequently, the Double Phase SAR Index (DPSI) was constructed to extract candidate rice samples by synthesizing the early and peak growth phases retrieved from NDVI time series and the corresponding SAR polarization features. Consequently, training data were generated automatically and fed into a machine learning classifier for rice mapping. The effectiveness of the PHAROS was evaluated over the Middle and Lower Reaches of the Yangtze River (MLRYR) of China and four other major rice production regions in Asia. Furthermore, the earliest timing of early, middle and late rice in the MLRYR was also quantified via the PHAROS and cross-year model transfer.

The results demonstrated that the PHAROS could identify rice of diverse cropping pattern with an overall accuracy (OA) from 0.965 to 0.976 in the MLRYR from 2019 to 2023. The classification maps exhibited well-delineated rice parcels and clear separation between single- and double-cropping rice. The PHAROS also yielded OA values of 0.902–0.979 and similar distribution pattern with the reference rice products in Suihua of China, Sataka of Japan, An Giang of Vietnam and Punjab of Pakistan, which represent diverse climatic conditions and cropping patterns across Asia. Compared to its counterpart methods, PHAROS demonstrated significant improvements by 0.022–0.235 in OA in rice planting area extraction and cropping intensity detection. The early, middle and late rice could be identified as early as at tillering, tillering-jointing and sowing/transplanting stages, respectively. This study reveals the necessity of handling the overlooked weak flooding signal from direct-seeded rice and offers a viable solution for large-scale rice cropping intensity detection and mapping under diverse establishment methods.

准确和及时的水稻种植面积和种植强度信息对于精准作物管理、粮食安全和环境可持续性至关重要。然而，高质量稻米产品的生产往往受到缺乏地面真实样本的阻碍，特别是在种植模式复杂的地区。虽然大多数水稻制图方法依赖于利用移栽期的遥感洪涝信号来区分水稻与其他作物，但它们面临着日益普遍的直接播种水稻的重大挑战，由于独特的管理措施，直接播种水稻的季初洪涝信号较弱。为了解决这些问题，实现不同种植模式和建立方法下水稻的综合提取，本研究提出了物候知识引导下基于光学和合成孔径雷达（SAR）数据的水稻自动制图方法（PHAROS）。该方法首先利用协调的多资源归一化植被指数（NDVI）时间序列和物候知识确定水稻及其同期作物的种植强度；随后，通过综合NDVI时间序列反演的水稻生长早期和高峰阶段及其对应的SAR极化特征，构建双相SAR指数（DPSI）来提取候选水稻样品。因此，训练数据被自动生成，并输入到机器学习分类器中用于水稻制图。在中国长江中下游地区和亚洲其他四个主要水稻产区对PHAROS的有效性进行了评价。此外，通过PHAROS和跨年模型转移，还量化了MLRYR早、中、晚稻的最早时间。结果表明，2019 ~ 2023年，PHAROS在MLRYR中能够识别出不同种植模式的水稻，总体准确率（OA）在0.965 ~ 0.976之间。分类图显示出水稻圈定清晰，单季稻和双季稻区分清晰。在中国绥化、日本萨塔卡、越南安江和巴基斯坦旁遮普省，PHAROS测定的OA值为0.902 ~ 0.979，分布格局与参考稻谷相似，反映了亚洲地区不同的气候条件和种植模式。与同类方法相比，PHAROS在水稻种植面积提取和种植强度检测方面的OA值提高了0.022 ~ 0.235。早、中、晚稻可分别在分蘖期、分蘖拔节期和播种移栽期进行鉴定。本研究揭示了对直播水稻被忽视的弱洪涝信号进行处理的必要性，为多种建立方法下的大规模水稻种植强度检测与制图提供了可行的解决方案。

{"title":"Automated rice mapping under diverse cropping patterns and establishment methods by integrating phenological knowledge and synergy of optical and SAR imagery","authors":"Xingrong Li , Gaoxiang Yang , Meng He , Yuan Xiong , Leilei Liu , Hifza Mariam , Iftikhar Ali , Chongya Jiang , Xia Yao , Yan Zhu , Weixing Cao , Tao Cheng","doi":"10.1016/j.rse.2026.115255","DOIUrl":"10.1016/j.rse.2026.115255","url":null,"abstract":"<div><div>Accurate and timely information on rice cultivation areas and cropping intensity is essential for precision crop management, food security and environmental sustainability. However, the generation of high-quality rice products is often hindered by the lack of ground truth samples, particularly in the regions with complex cropping patterns. While most rice mapping methods rely on the use of remotely sensed flooding signal from the transplanting period to distinguish rice from other crops, they face significant challenges from the increasingly prevalent direct-seeded rice for which the flooding signal at the beginning of season is weak due to the unique management measures. To address these issues and achieve the comprehensive extraction of rice under diverse cropping patterns and establishment methods, this study proposed the phenological knowledge-guided automatic rice mapping approach using optical and synthetic aperture radar (SAR) data (PHAROS). This method first determined the cropping intensity of rice and its concurrent crops using harmonized multi-resource Normalized Difference Vegetation Index (NDVI) time series and phenological knowledge. Subsequently, the Double Phase SAR Index (DPSI) was constructed to extract candidate rice samples by synthesizing the early and peak growth phases retrieved from NDVI time series and the corresponding SAR polarization features. Consequently, training data were generated automatically and fed into a machine learning classifier for rice mapping. The effectiveness of the PHAROS was evaluated over the Middle and Lower Reaches of the Yangtze River (MLRYR) of China and four other major rice production regions in Asia. Furthermore, the earliest timing of early, middle and late rice in the MLRYR was also quantified via the PHAROS and cross-year model transfer.</div><div>The results demonstrated that the PHAROS could identify rice of diverse cropping pattern with an overall accuracy (OA) from 0.965 to 0.976 in the MLRYR from 2019 to 2023. The classification maps exhibited well-delineated rice parcels and clear separation between single- and double-cropping rice. The PHAROS also yielded OA values of 0.902–0.979 and similar distribution pattern with the reference rice products in Suihua of China, Sataka of Japan, An Giang of Vietnam and Punjab of Pakistan, which represent diverse climatic conditions and cropping patterns across Asia. Compared to its counterpart methods, PHAROS demonstrated significant improvements by 0.022–0.235 in OA in rice planting area extraction and cropping intensity detection. The early, middle and late rice could be identified as early as at tillering, tillering-jointing and sowing/transplanting stages, respectively. This study reveals the necessity of handling the overlooked weak flooding signal from direct-seeded rice and offers a viable solution for large-scale rice cropping intensity detection and mapping under diverse establishment methods.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"335 ","pages":"Article 115255"},"PeriodicalIF":11.4,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146015019","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}

引用次数: 0

Global assessment of landslide monitoring applicability with the Harmony mission 基于Harmony任务的滑坡监测适用性全球评估

IF 11.4 1区地球科学 Q1 ENVIRONMENTAL SCIENCES

Remote Sensing of Environment

Pub Date : 2026-01-20 DOI: 10.1016/j.rse.2026.115236

Shaokun Guo , Jie Dong , Mingsheng Liao

The Harmony mission, comprising two passive companion satellites that will operate in a bistatic configuration with Sentinel-1D after its planned launch, is designed to enhance Earth observation through improved diversity of observation geometry. The benefits are substantial, particularly for landslide monitoring, where three-dimensional deformation can be more effectively resolved. However, a quantitative global-scale assessment of its anticipated performance is still lacking. This study addresses this gap through a systematic geometric analysis. We modify the conventional applicability framework to support a bistatic configuration, leveraging multiple mission-related datasets to produce global landslide applicability products for the three stations, including Sentinel-1D and the two Harmony satellites. The products are accessible through a cloud platform. Findings reveal that, under the ascending-descending combined scheme, invalid regions rise by 9% globally, surpassing 25% in certain rugged areas. This is accompanied by notable improvements: the global sensitivity index increases by 0.11 (from 0.61 to 0.72), with low-sensitivity (<0.3) regions declining from 10.4% to 2.1%. Theoretical experiments reveal a northward/southward sensitivity improvement exceeding 0.15, while the direct 3D inversion capability, a key benefit of the additional observation directions, is confirmed under moderate noise. Overall, Harmony provides a robust and effective approach to 3D landslide monitoring, markedly enhancing reliable landslide detection globally.

“和谐”任务由两颗被动卫星组成，在计划发射后将与“哨兵- 1d”一起以双基地配置运行，旨在通过改善观测几何形状的多样性来增强对地观测。这样做的好处是巨大的，特别是在滑坡监测方面，三维变形可以更有效地解决。然而，对其预期表现的定量全球规模评估仍然缺乏。本研究通过系统的几何分析来解决这一差距。我们修改了传统的适用性框架以支持双基地配置，利用多个任务相关数据集为三个站点（包括Sentinel-1D和两颗Harmony卫星）生产全球滑坡适用性产品。产品可通过云平台访问。结果表明，在上升-下降联合方案下，全球无效区域增加了9%，在某些崎岖地区超过25%。这伴随着显著的改善：全球敏感性指数增加了0.11（从0.61到0.72），低敏感性（<0.3）区域从10.4%下降到2.1%。理论实验表明，北/南方向的灵敏度提高超过0.15，而在中等噪声条件下，直接三维反演能力是额外观测方向的一个关键优势。总体而言，Harmony提供了一种强大而有效的三维滑坡监测方法，显著提高了全球滑坡检测的可靠性。

{"title":"Global assessment of landslide monitoring applicability with the Harmony mission","authors":"Shaokun Guo , Jie Dong , Mingsheng Liao","doi":"10.1016/j.rse.2026.115236","DOIUrl":"10.1016/j.rse.2026.115236","url":null,"abstract":"<div><div>The Harmony mission, comprising two passive companion satellites that will operate in a bistatic configuration with Sentinel-1D after its planned launch, is designed to enhance Earth observation through improved diversity of observation geometry. The benefits are substantial, particularly for landslide monitoring, where three-dimensional deformation can be more effectively resolved. However, a quantitative global-scale assessment of its anticipated performance is still lacking. This study addresses this gap through a systematic geometric analysis. We modify the conventional applicability framework to support a bistatic configuration, leveraging multiple mission-related datasets to produce global landslide applicability products for the three stations, including Sentinel-1D and the two Harmony satellites. The products are accessible through a cloud platform. Findings reveal that, under the ascending-descending combined scheme, invalid regions rise by 9% globally, surpassing 25% in certain rugged areas. This is accompanied by notable improvements: the global sensitivity index increases by 0.11 (from 0.61 to 0.72), with low-sensitivity (<0.3) regions declining from 10.4% to 2.1%. Theoretical experiments reveal a northward/southward sensitivity improvement exceeding 0.15, while the direct 3D inversion capability, a key benefit of the additional observation directions, is confirmed under moderate noise. Overall, Harmony provides a robust and effective approach to 3D landslide monitoring, markedly enhancing reliable landslide detection globally.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"335 ","pages":"Article 115236"},"PeriodicalIF":11.4,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146015021","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}

引用次数: 0

Reconstructing all-weather remotely sensed air temperature via a kernel-based temporal filling and bias correction (KTF-BC) framework 基于时间填充和偏差校正（KTF-BC）框架的全天候遥感气温重建

IF 11.4 1区地球科学 Q1 ENVIRONMENTAL SCIENCES

Remote Sensing of Environment

Pub Date : 2026-01-20 DOI: 10.1016/j.rse.2026.115253

Yan Xin , Yongming Xu , Xudong Tong , Meng Ji , Yaping Mo , Yonghong Liu , Shanyou Zhu

Thermal infrared (TIR) remote sensing provides an effective means of mapping near-surface air temperature (T_a) at large scales. However, cloud coverage introduces substantial data gaps, posing a considerable challenge for producing all-weather T_a datasets. This study proposed a two-step framework, termed Kernel-based Temporal Filling and Bias Correction (KTF-BC), to reconstruct all-weather remotely sensed T_a. In the first stage, a kernel-based temporal filling method was developed to estimate the theoretical clear-sky T_a for cloud-covered pixels. In the second stage, a bias correction model was constructed to adjust these theoretical estimates toward the actual T_a under cloudy conditions. The proposed framework was applied across China from 2019 to 2023 to generate spatially complete daily mean T_a at 1-km resolution. Validation against meteorological stations under cloudy conditions demonstrated consistently high accuracy, with R² values up to 0.99, mean absolute errors (MAEs) ranging from 0.91 to 0.95 °C, root mean square errors (RMSEs) ranging from 1.21 to 1.26 °C, and biases close to 0 °C. The method effectively captured fine-scale thermal heterogeneity and demonstrated robust performance across varying cloud conditions and surface environments. This study provides a practical and reliable solution for reconstructing all-weather T_a from satellite observations.

热红外（TIR）遥感提供了大尺度近地表气温（Ta）测绘的有效手段。然而，云层覆盖带来了大量的数据缺口，对生成全天候的气象数据集构成了相当大的挑战。本研究提出了一种基于核的时间填充和偏差校正（KTF-BC）两步框架来重建全天候遥感Ta。在第一阶段，提出了一种基于核的时间填充方法来估计云覆盖像素的理论晴空Ta。在第二阶段，建立了一个偏差校正模型，将这些理论估计调整到多云条件下的实际Ta。该框架于2019年至2023年在中国各地应用，以生成1公里分辨率的空间完整日平均Ta。在多云条件下对气象站的验证显示出一贯的高准确性，R2值高达0.99，平均绝对误差（MAEs）范围为0.91至0.95°C，均方根误差（rmse）范围为1.21至1.26°C，偏差接近0°C。该方法有效地捕获了精细尺度的热非均质性，并在不同的云条件和地表环境中展示了稳健的性能。本研究为利用卫星观测资料重建全天候热场提供了一种实用可靠的解决方案。

{"title":"Reconstructing all-weather remotely sensed air temperature via a kernel-based temporal filling and bias correction (KTF-BC) framework","authors":"Yan Xin , Yongming Xu , Xudong Tong , Meng Ji , Yaping Mo , Yonghong Liu , Shanyou Zhu","doi":"10.1016/j.rse.2026.115253","DOIUrl":"10.1016/j.rse.2026.115253","url":null,"abstract":"<div><div>Thermal infrared (TIR) remote sensing provides an effective means of mapping near-surface air temperature (<em>T</em><sub><em>a</em></sub>) at large scales. However, cloud coverage introduces substantial data gaps, posing a considerable challenge for producing all-weather <em>T</em><sub><em>a</em></sub> datasets. This study proposed a two-step framework, termed Kernel-based Temporal Filling and Bias Correction (KTF-BC), to reconstruct all-weather remotely sensed <em>T</em><sub><em>a</em></sub>. In the first stage, a kernel-based temporal filling method was developed to estimate the theoretical clear-sky <em>T</em><sub><em>a</em></sub> for cloud-covered pixels. In the second stage, a bias correction model was constructed to adjust these theoretical estimates toward the actual <em>T</em><sub><em>a</em></sub> under cloudy conditions. The proposed framework was applied across China from 2019 to 2023 to generate spatially complete daily mean <em>T</em><sub><em>a</em></sub> at 1-km resolution. Validation against meteorological stations under cloudy conditions demonstrated consistently high accuracy, with R<sup>2</sup> values up to 0.99, mean absolute errors (MAEs) ranging from 0.91 to 0.95 °C, root mean square errors (RMSEs) ranging from 1.21 to 1.26 °C, and biases close to 0 °C. The method effectively captured fine-scale thermal heterogeneity and demonstrated robust performance across varying cloud conditions and surface environments. This study provides a practical and reliable solution for reconstructing all-weather <em>T</em><sub><em>a</em></sub> from satellite observations.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"335 ","pages":"Article 115253"},"PeriodicalIF":11.4,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145996477","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}

引用次数: 0

Generating an annual 30 m rice cover product for monsoon Asia (2018–2023) using harmonized Landsat and Sentinel-2 data and the NASA-IBM geospatial foundation model 使用统一的Landsat和Sentinel-2数据以及NASA-IBM地理空间基础模型，为亚洲季风（2018-2023）生成每年30米的水稻覆盖产品

IF 11.4 1区地球科学 Q1 ENVIRONMENTAL SCIENCES

Remote Sensing of Environment

Pub Date : 2026-01-20 DOI: 10.1016/j.rse.2026.115256

Husheng Fang , Shunlin Liang , Wenyuan Li , Yongzhe Chen , Han Ma , Jianglei Xu , Yichuan Ma , Tao He , Feng Tian , Fengjiao Zhang , Hui Liang

Timely and accurate information on rice distribution is crucial for food security and agricultural decision-making. Monsoon Asia hosts the world's largest rice cultivation area and is dominated by smallholder farming systems with fragmented farmlands, making it challenging for accurate rice field mapping based on coarse-resolution satellite imagery. Machine learning methods have been widely used in remote sensing classification tasks due to their high efficiency and accuracy. However, their performance can be limited in scenarios where sufficient training samples are unavailable. Recently, the advent of geospatial foundation models has offered a promising solution with a pre-training strategy on massive unlabeled satellite data, which achieves higher accuracy than traditional deep learning models in downstream tasks. This study proposes a conceptual framework for adapting geospatial foundation models (GFMs) to large-scale agricultural mapping, representing the first application of the NASA–IBM Prithvi model for continental-scale rice monitoring across Monsoon Asia. We first automatically generate rice labels from multiple existing regional rice products, then fine-tune Prithvi using the Harmonized Landsat and Sentinel-2 satellite data with high spatiotemporal resolution. Finally, the 30 m rice distribution product for Monsoon Asia from 2018 to 2023 is generated. Validation shows that the overall accuracy is 84.14% for the entire Monsoon Asia, with accuracies ranging from 83.07% to 90.06% across four climatic zones. Our product exhibits strong consistency with existing scattered local-scale high-resolution products and shows improved accuracy compared to the MODIS-based and SAR-based products covering the Monsoon Asia. This study indicates that geospatial foundation models can play a key role in remote sensing. Combining geospatial foundation models with massive satellite data has enormous potential for future applications. The rice cover product and code for fine tuning are available at www.glass.hku.hk.

及时准确的大米分配信息对粮食安全和农业决策至关重要。季风亚洲拥有世界上最大的水稻种植面积，并且以小农农业系统为主，农田分散，这使得基于粗分辨率卫星图像的精确稻田测绘具有挑战性。机器学习方法以其高效、准确的特点在遥感分类任务中得到了广泛的应用。然而，在没有足够的训练样本的情况下，它们的性能会受到限制。近年来，地理空间基础模型的出现为大规模未标记卫星数据的预训练策略提供了一种有前景的解决方案，该方法在下游任务中比传统深度学习模型具有更高的精度。本研究提出了一个将地理空间基础模型（GFMs）应用于大尺度农业制图的概念框架，代表了NASA-IBM Prithvi模型在季风亚洲大陆尺度水稻监测中的首次应用。我们首先从多个现有的区域大米产品中自动生成大米标签，然后使用高时空分辨率的Harmonized Landsat和Sentinel-2卫星数据对Prithvi进行微调。最后，生成了2018年至2023年季风亚洲的30米大米分布产品。验证结果表明，对整个季风亚洲的总体精度为84.14%，四个气候带的精度范围为83.07% ~ 90.06%。我们的产品与现有的分散的局地尺度高分辨率产品具有很强的一致性，并且与覆盖季风亚洲的基于modis和sar的产品相比，具有更高的精度。研究表明，地理空间基础模型在遥感研究中具有重要作用。将地理空间基础模型与大量卫星数据相结合，在未来的应用中具有巨大的潜力。米盖产品和微调代码可在www.glass.hku.hk上获得。

{"title":"Generating an annual 30 m rice cover product for monsoon Asia (2018–2023) using harmonized Landsat and Sentinel-2 data and the NASA-IBM geospatial foundation model","authors":"Husheng Fang , Shunlin Liang , Wenyuan Li , Yongzhe Chen , Han Ma , Jianglei Xu , Yichuan Ma , Tao He , Feng Tian , Fengjiao Zhang , Hui Liang","doi":"10.1016/j.rse.2026.115256","DOIUrl":"10.1016/j.rse.2026.115256","url":null,"abstract":"<div><div>Timely and accurate information on rice distribution is crucial for food security and agricultural decision-making. Monsoon Asia hosts the world's largest rice cultivation area and is dominated by smallholder farming systems with fragmented farmlands, making it challenging for accurate rice field mapping based on coarse-resolution satellite imagery. Machine learning methods have been widely used in remote sensing classification tasks due to their high efficiency and accuracy. However, their performance can be limited in scenarios where sufficient training samples are unavailable. Recently, the advent of geospatial foundation models has offered a promising solution with a pre-training strategy on massive unlabeled satellite data, which achieves higher accuracy than traditional deep learning models in downstream tasks. This study proposes a conceptual framework for adapting geospatial foundation models (GFMs) to large-scale agricultural mapping, representing the first application of the NASA–IBM Prithvi model for continental-scale rice monitoring across Monsoon Asia. We first automatically generate rice labels from multiple existing regional rice products, then fine-tune Prithvi using the Harmonized Landsat and Sentinel-2 satellite data with high spatiotemporal resolution. Finally, the 30 m rice distribution product for Monsoon Asia from 2018 to 2023 is generated. Validation shows that the overall accuracy is 84.14% for the entire Monsoon Asia, with accuracies ranging from 83.07% to 90.06% across four climatic zones. Our product exhibits strong consistency with existing scattered local-scale high-resolution products and shows improved accuracy compared to the MODIS-based and SAR-based products covering the Monsoon Asia. This study indicates that geospatial foundation models can play a key role in remote sensing. Combining geospatial foundation models with massive satellite data has enormous potential for future applications. The rice cover product and code for fine tuning are available at <span><span>www.glass.hku.hk</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"335 ","pages":"Article 115256"},"PeriodicalIF":11.4,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146015022","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}

引用次数: 0

Characterizing mangrove forest succession in Suriname using GEDI waveform metrics 利用GEDI波形指标表征苏里南红树林演替

IF 11.4 1区地球科学 Q1 ENVIRONMENTAL SCIENCES

Remote Sensing of Environment

Pub Date : 2026-01-19 DOI: 10.1016/j.rse.2026.115244

Jasper Feyen , Verginia Wortel , Kim Calders , John Armston , Frieke Vancoillie

Mangroves are critical coastal ecosystems known for their carbon storage capacity, biodiversity, and role in shoreline stabilization. In Suriname, mangroves develop within a dynamic coastal setting shaped by migrating mudbanks and high sedimentation rates. This study examines how 30 structural metrics derived from the Global Ecosystem Dynamics Investigation (GEDI) vary across gradients of mangrove stand age and seaward distance. Forest stand age and yearly coastline positions were derived from Landsat time series data, enabling the integration of temporal and spatial drivers to uncover patterns of mangrove succession and structural development. Nonlinear growth models, more specifically the Chapman–Richards function, captured early growth and stabilization phases, while Generalized Additive Models (GAMs) provided flexibility to represent more complex structural changes observed in mature and decaying stands. Results show that structural metrics related to forest growth, such as canopy height and aboveground biomass density, increase rapidly during early successional stages but plateau beyond approximately 12 years or 2 km from the coastline. Complexity-oriented metrics, such as Foliage Height Diversity (FHD) and the Waveform Structural Complexity Index (WSCI), continue to evolve, reflecting increased vertical stratification in mature stands. By combining GEDI spaceborne LiDAR with Landsat-derived chronosequences, this study demonstrates how remote sensing can be used to monitor mangrove successional trajectories and structural complexity, including in inaccessible coastal regions. Our findings extend traditional mangrove successional models by quantifying how both temporal (age) and spatial (seaward distance) gradients jointly determine mangrove structure across the Surinamese coastline.

红树林是重要的沿海生态系统，以其碳储存能力、生物多样性和海岸线稳定作用而闻名。在苏里南，红树林生长在由迁移的泥滩和高沉积率形成的动态海岸环境中。本研究考察了全球生态系统动力学调查（GEDI）得出的30个结构指标在红树林林龄和向海距离梯度中的变化。森林林分年龄和年海岸线位置来源于Landsat时间序列数据，能够整合时空驱动因素来揭示红树林演替和结构发展的模式。非线性生长模型，更具体地说，Chapman-Richards函数，捕获了早期生长和稳定阶段，而广义可加模型（GAMs）提供了灵活性，可以表示成熟和腐烂林分中观察到的更复杂的结构变化。结果表明，与森林生长有关的结构指标，如冠层高度和地上生物量密度，在演替早期迅速增加，但在约12年或距海岸线2公里后趋于稳定。以复杂性为导向的指标，如叶片高度多样性（FHD）和波形结构复杂性指数（WSCI），继续演变，反映了成熟林分垂直分层的增加。通过将GEDI星载激光雷达与landsat衍生的时间序列相结合，本研究展示了遥感如何用于监测红树林的演替轨迹和结构复杂性，包括在难以到达的沿海地区。我们的研究结果通过量化时间（年龄）和空间（向海距离）梯度如何共同决定苏里南海岸线上的红树林结构，扩展了传统的红树林演替模型。

{"title":"Characterizing mangrove forest succession in Suriname using GEDI waveform metrics","authors":"Jasper Feyen , Verginia Wortel , Kim Calders , John Armston , Frieke Vancoillie","doi":"10.1016/j.rse.2026.115244","DOIUrl":"10.1016/j.rse.2026.115244","url":null,"abstract":"<div><div>Mangroves are critical coastal ecosystems known for their carbon storage capacity, biodiversity, and role in shoreline stabilization. In Suriname, mangroves develop within a dynamic coastal setting shaped by migrating mudbanks and high sedimentation rates. This study examines how 30 structural metrics derived from the Global Ecosystem Dynamics Investigation (GEDI) vary across gradients of mangrove stand age and seaward distance. Forest stand age and yearly coastline positions were derived from Landsat time series data, enabling the integration of temporal and spatial drivers to uncover patterns of mangrove succession and structural development. Nonlinear growth models, more specifically the Chapman–Richards function, captured early growth and stabilization phases, while Generalized Additive Models (GAMs) provided flexibility to represent more complex structural changes observed in mature and decaying stands. Results show that structural metrics related to forest growth, such as canopy height and aboveground biomass density, increase rapidly during early successional stages but plateau beyond approximately 12 years or 2 km from the coastline. Complexity-oriented metrics, such as Foliage Height Diversity (FHD) and the Waveform Structural Complexity Index (WSCI), continue to evolve, reflecting increased vertical stratification in mature stands. By combining GEDI spaceborne LiDAR with Landsat-derived chronosequences, this study demonstrates how remote sensing can be used to monitor mangrove successional trajectories and structural complexity, including in inaccessible coastal regions. Our findings extend traditional mangrove successional models by quantifying how both temporal (age) and spatial (seaward distance) gradients jointly determine mangrove structure across the Surinamese coastline.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"334 ","pages":"Article 115244"},"PeriodicalIF":11.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146001413","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}

引用次数: 0

Mapping urban built-up types from 2000 to 2022 at 10-m resolution using super-resolution of Landsat spectral-temporal metrics and center-patch classification 利用超分辨率Landsat光谱-时间指标和中心斑块分类，以10 m分辨率绘制2000 - 2022年城市建成区类型

IF 11.4 1区地球科学 Q1 ENVIRONMENTAL SCIENCES

Remote Sensing of Environment

Pub Date : 2026-01-19 DOI: 10.1016/j.rse.2026.115251

Vu-Dong Pham , Franz Schug , David Frantz , Sebastian van der Linden

Detailed information on urban development types, e.g. residential, industrial, or transportation infrastructure, is essential for assessing urban growth rates, population dynamics, and environmental impacts. Earth observation imagery from Landsat and Sentinel-2 provides valuable data for characterizing urban areas and their development over large spatial extents and long temporal scales. However, mapping large areas over multiple decades poses several challenges. Specifically, the coarse resolution of historical Landsat data (30-m) limits the capacity to capture the spatial detail of diverse built-up types. Furthermore, existing mapping techniques—such as pixel-based, scene-based, and semantic segmentation - often face limitations in capturing spatial context, compromise mapping resolution, or rely on hand-crafted training data. To address these challenges, this study proposes a novel workflow comprising two key components: (1) enhancing historical Landsat spatial resolution to 10-m using a generative super-resolution model, with a focus on synthetic images derived from spectral-temporal metrics, and (2) a “center-patch classification” method, wherein patch images serve as input for the central pixel classification. We applied the proposed methodologies to produce tri-annual maps of urban built-up classes—including Residential buildings, Industrial buildings, Open spaces, and Non built-up—across the Baltic Sea region for two decades (2000−2021). Evaluation of the baseline performance demonstrated that the 10-m maps derived from super-resolution Landsat data achieved higher accuracy than existing Sentinel-2-based products. Furthermore, when applied to earlier years within the study period, the super-resolution Landsat data consistently exhibited improved classification accuracy across all urban classes compared to native-resolution Landsat data.

关于城市发展类型的详细信息，例如住宅、工业或交通基础设施，对于评估城市增长率、人口动态和环境影响至关重要。来自Landsat和Sentinel-2的地球观测图像为描述城市地区及其在大空间范围和长时间尺度上的发展提供了宝贵的数据。然而，在几十年的时间里绘制大面积的地图会带来一些挑战。具体来说，历史Landsat数据的粗分辨率（30米）限制了捕捉不同建筑类型空间细节的能力。此外，现有的映射技术（如基于像素的、基于场景的和语义分割）在捕获空间上下文、降低映射分辨率或依赖手工制作的训练数据方面经常面临限制。为了解决这些挑战，本研究提出了一种新的工作流程，其中包括两个关键部分：(1)使用生成式超分辨率模型将历史Landsat空间分辨率提高到10米，重点关注从光谱-时间度量衍生的合成图像；(2)“中心-补丁分类”方法，其中补丁图像作为中心像素分类的输入。我们应用所提出的方法制作了横跨波罗的海地区二十年（2000 - 2021年）的城市建成区（包括住宅建筑、工业建筑、开放空间和非建成区）三年一次的地图。基线性能评估表明，来自超分辨率Landsat数据的10米地图比现有的基于sentinel -2的产品具有更高的精度。此外，当应用于研究期间的早期时，与原始分辨率Landsat数据相比，超分辨率Landsat数据在所有城市类别中始终表现出更高的分类准确性。

{"title":"Mapping urban built-up types from 2000 to 2022 at 10-m resolution using super-resolution of Landsat spectral-temporal metrics and center-patch classification","authors":"Vu-Dong Pham , Franz Schug , David Frantz , Sebastian van der Linden","doi":"10.1016/j.rse.2026.115251","DOIUrl":"10.1016/j.rse.2026.115251","url":null,"abstract":"<div><div>Detailed information on urban development types, e.g. residential, industrial, or transportation infrastructure, is essential for assessing urban growth rates, population dynamics, and environmental impacts. Earth observation imagery from Landsat and Sentinel-2 provides valuable data for characterizing urban areas and their development over large spatial extents and long temporal scales. However, mapping large areas over multiple decades poses several challenges. Specifically, the coarse resolution of historical Landsat data (30-m) limits the capacity to capture the spatial detail of diverse built-up types. Furthermore, existing mapping techniques—such as pixel-based, scene-based, and semantic segmentation - often face limitations in capturing spatial context, compromise mapping resolution, or rely on hand-crafted training data. To address these challenges, this study proposes a novel workflow comprising two key components: (1) enhancing historical Landsat spatial resolution to 10-m using a generative super-resolution model, with a focus on synthetic images derived from spectral-temporal metrics, and (2) a “center-patch classification” method, wherein patch images serve as input for the central pixel classification. We applied the proposed methodologies to produce tri-annual maps of urban built-up classes—including <em>Residential buildings, Industrial buildings, Open spaces,</em> and <em>Non built-up</em>—across the Baltic Sea region for two decades (2000−2021). Evaluation of the baseline performance demonstrated that the 10-m maps derived from super-resolution Landsat data achieved higher accuracy than existing Sentinel-2-based products. Furthermore, when applied to earlier years within the study period, the super-resolution Landsat data consistently exhibited improved classification accuracy across all urban classes compared to native-resolution Landsat data.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"335 ","pages":"Article 115251"},"PeriodicalIF":11.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145996475","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}

引用次数: 0

Spatial-X fusion for multi-source satellite imageries 多源卫星图像的空间- x融合

IF 11.4 1区地球科学 Q1 ENVIRONMENTAL SCIENCES

Remote Sensing of Environment

Pub Date : 2026-01-19 DOI: 10.1016/j.rse.2025.115214

Jiang He , Liupeng Lin , Zhuo Zheng , Qiangqiang Yuan , Jie Li , Liangpei Zhang , Xiao xiang Zhu

Multi-source remote sensing data can highlight different types of information based on user needs, resulting in large volumes of data and significant challenges. Hardware and environmental constraints create mutual dependencies between information types, particularly between spatial data and other types, limiting the development of high-precision applications. Traditional methods are task-specific, leading to many algorithms without a unified solution, which greatly increases the computational and deployment costs of image fusion. In this paper, we summarize four remote sensing fusion tasks, including pan-sharpening, hyperspectral-multispectral fusion, spatio-temporal fusion, and polarimetric SAR fusion. By defining the spectral, temporal, and polarimetric information, as X, we propose the concept of generalized spatial-channel fusion, referred to as Spatial-X fusion. Then, we design an end-to-end network SpaXFus, a generalized spatial-channel fusion framework through a model-driven unfolding approach that exploits spatial-X intrinsic interactions to capture internal dependencies and self-interactions. Comprehensive experimental results demonstrate the superiority of SpaXFus, e.g., SpaXFus can achieve four remote sensing image fusion tasks with superior performance (across all fusion tasks, spectral distortion decreases by 25.48 %, while spatial details improve by 7.5 %) and shows huge improvements across multiple types of downstream applications, including vegetation index generation, fine-grained image classification, change detection, and SAR vegetation extraction.

多源遥感数据可以根据用户需要突出不同类型的信息，导致数据量大，并带来重大挑战。硬件和环境约束在信息类型之间（特别是空间数据和其他类型之间）造成相互依赖，限制了高精度应用程序的开发。传统的图像融合方法是针对任务的，导致许多算法没有统一的解决方案，这大大增加了图像融合的计算和部署成本。本文综述了泛锐化、高光谱-多光谱融合、时空融合和极化SAR融合等四种遥感融合技术。通过将光谱、时间和偏振信息定义为X，我们提出了广义空间信道融合的概念，称为空间-X融合。然后，我们设计了一个端到端网络SpaXFus，这是一个广义的空间通道融合框架，通过模型驱动的展开方法，利用空间- x内在相互作用来捕获内部依赖关系和自相互作用。综合实验结果证明了SpaXFus的优势，SpaXFus可以实现4个遥感图像融合任务，并且在所有融合任务中，光谱失真降低25.48%，空间细节提高7.5%，在植被指数生成、细粒度图像分类、变化检测和SAR植被提取等多种下游应用中表现出巨大的改进。

{"title":"Spatial-X fusion for multi-source satellite imageries","authors":"Jiang He , Liupeng Lin , Zhuo Zheng , Qiangqiang Yuan , Jie Li , Liangpei Zhang , Xiao xiang Zhu","doi":"10.1016/j.rse.2025.115214","DOIUrl":"10.1016/j.rse.2025.115214","url":null,"abstract":"<div><div>Multi-source remote sensing data can highlight different types of information based on user needs, resulting in large volumes of data and significant challenges. Hardware and environmental constraints create mutual dependencies between information types, particularly between spatial data and other types, limiting the development of high-precision applications. Traditional methods are task-specific, leading to many algorithms without a unified solution, which greatly increases the computational and deployment costs of image fusion. In this paper, we summarize four remote sensing fusion tasks, including pan-sharpening, hyperspectral-multispectral fusion, spatio-temporal fusion, and polarimetric SAR fusion. By defining the spectral, temporal, and polarimetric information, as X, we propose the concept of generalized spatial-channel fusion, referred to as Spatial-X fusion. Then, we design an end-to-end network SpaXFus, a generalized spatial-channel fusion framework through a model-driven unfolding approach that exploits spatial-X intrinsic interactions to capture internal dependencies and self-interactions. Comprehensive experimental results demonstrate the superiority of SpaXFus, e.g., SpaXFus can achieve four remote sensing image fusion tasks with superior performance (across all fusion tasks, spectral distortion decreases by 25.48 %, while spatial details improve by 7.5 %) and shows huge improvements across multiple types of downstream applications, including vegetation index generation, fine-grained image classification, change detection, and SAR vegetation extraction.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"334 ","pages":"Article 115214"},"PeriodicalIF":11.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146001412","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}

引用次数: 0

Amplified deviation flood index (ADFI) for fast non-prior flood detection 用于快速非先验洪水检测的放大偏差洪水指数（ADFI）

IF 11.4 1区地球科学 Q1 ENVIRONMENTAL SCIENCES

Remote Sensing of Environment

Pub Date : 2026-01-19 DOI: 10.1016/j.rse.2026.115258

Hui Zhang , Ming Luo , Zhixin Qi , Xing Li , Yongquan Zhao

Climate change causes widespread increases in the frequency, magnitude, and extent of flood events, which pose increasing threats to societal and natural systems and highlight the urgency for timely and accurate flood mapping. However, previous flood mapping methods often require prior knowledge (such as the timing and location) of flood events that is usually incomplete or even unavailable when studying historical floods. Here we propose a new amplified deviation flood index (ADFI) using the time-series anomaly statistics from the Synthetic Aperture Radar (SAR) data for mapping fully flooded areas without relying on prior knowledge of flood events. ADFI is constructed by considering two fundamentals of flood events: a decrease in backscatter intensity when ground objects are fully flooded and an increase in the variance of backscatter intensity owing to infrequently sudden occurrence of flood events, thus enabling a fast non-prior detection of flood events and extents. The performance of ADFI is assessed in four study areas across different climate zones of the globe, and the assessment shows that the overall accuracies of ADFI in all study areas exceed 93%, with precision >95% and recall >94%. Further comparison with two existing flood indices suggests that our proposed ADFI-based mapping method can improve the overall accuracy by 12.11%–3.97%, precision by 12.59%–10.17%, and recall by 54.32%–6.37%. A time-series flood mapping based on ADFI demonstrates that our proposed method enables a non-prior, precise, and fast detection of flood events and allows prompt monitoring of flood disasters. Our proposed approach enhances the efficiency and scalability of flood monitoring, providing a valuable tool for rapid disaster response and the reconstruction of long-term flood histories across diverse environments and climates.

气候变化导致洪水事件的频率、规模和范围普遍增加，对社会和自然系统构成越来越大的威胁，并突出了及时和准确绘制洪水地图的紧迫性。然而，以往的洪水制图方法往往需要洪水事件的先验知识（如时间和位置），而这些知识在研究历史洪水时通常是不完整的，甚至是不可用的。本文提出了一种新的放大偏差洪水指数（ADFI），该指数利用合成孔径雷达（SAR）数据的时间序列异常统计量来绘制全洪水区域，而不依赖于洪水事件的先验知识。ADFI的构建考虑了洪水事件的两个基本原理：地面物体被完全淹没时，后向散射强度会减小；洪水事件不经常突然发生，后向散射强度的方差会增大，从而可以快速地非先验地检测洪水事件和范围。在全球不同气候带的4个研究区对ADFI的性能进行了评估，评估结果表明，ADFI在所有研究区的总体准确度均超过93%，精密度为95%，召回率为94%。进一步与已有的2个洪水指数进行对比，表明基于adfi的制图方法整体精度提高12.11% ~ 3.97%，精密度提高12.59% ~ 10.17%，查全率提高54.32% ~ 6.37%。基于ADFI的时间序列洪水映射表明，我们提出的方法可以实现对洪水事件的非先验、精确和快速检测，并允许对洪水灾害进行及时监测。我们提出的方法提高了洪水监测的效率和可扩展性，为快速响应灾害和重建不同环境和气候下的长期洪水历史提供了有价值的工具。

{"title":"Amplified deviation flood index (ADFI) for fast non-prior flood detection","authors":"Hui Zhang , Ming Luo , Zhixin Qi , Xing Li , Yongquan Zhao","doi":"10.1016/j.rse.2026.115258","DOIUrl":"10.1016/j.rse.2026.115258","url":null,"abstract":"<div><div>Climate change causes widespread increases in the frequency, magnitude, and extent of flood events, which pose increasing threats to societal and natural systems and highlight the urgency for timely and accurate flood mapping. However, previous flood mapping methods often require prior knowledge (such as the timing and location) of flood events that is usually incomplete or even unavailable when studying historical floods. Here we propose a new amplified deviation flood index (ADFI) using the time-series anomaly statistics from the Synthetic Aperture Radar (SAR) data for mapping fully flooded areas without relying on prior knowledge of flood events. ADFI is constructed by considering two fundamentals of flood events: a decrease in backscatter intensity when ground objects are fully flooded and an increase in the variance of backscatter intensity owing to infrequently sudden occurrence of flood events, thus enabling a fast non-prior detection of flood events and extents. The performance of ADFI is assessed in four study areas across different climate zones of the globe, and the assessment shows that the overall accuracies of ADFI in all study areas exceed 93%, with precision >95% and recall >94%. Further comparison with two existing flood indices suggests that our proposed ADFI-based mapping method can improve the overall accuracy by 12.11%–3.97%, precision by 12.59%–10.17%, and recall by 54.32%–6.37%. A time-series flood mapping based on ADFI demonstrates that our proposed method enables a non-prior, precise, and fast detection of flood events and allows prompt monitoring of flood disasters. Our proposed approach enhances the efficiency and scalability of flood monitoring, providing a valuable tool for rapid disaster response and the reconstruction of long-term flood histories across diverse environments and climates.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"335 ","pages":"Article 115258"},"PeriodicalIF":11.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145996476","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}

引用次数: 0