Remote Sensing of Environment最新文献

{"title":"Tracking 35-year dynamics of retrogressive thaw slumps across permafrost regions of the Tibetan Plateau","authors":"Guoqing Yang ,&nbsp;Haijun Qiu ,&nbsp;Ninglian Wang ,&nbsp;Dongdong Yang ,&nbsp;Ya Liu","doi":"10.1016/j.rse.2025.114786","DOIUrl":"10.1016/j.rse.2025.114786","url":null,"abstract":"<div><div>Permafrost degradation on the Tibetan Plateau (TP) has triggered widespread retrogressive thaw slumps (RTSs), affecting hydrology, carbon sequestration and infrastructure stability. To date, there is still a lack of long-term monitoring of RTSs across the TP, the thaw dynamics and comprehensive driving factors remain unclear. Here, using time-series Landsat imagery and change detection algorithm, we identified RTSs on permafrost regions of the TP from 1986 to 2020. Existing RTSs inventories and high-resolution historical imagery were employed to verify the identified results, the temporal validation of RTSs disturbance pixels demonstrated a high accuracy. In the study area, a total of 3537 RTSs were identified, covering a total area of 5997 ha, representing a 26-fold increase since 1986, and 69.2 % of RTSs formed since 2010. Most RTSs are located on gentle slope (4–12°) at elevations between 4500 m and 5300 m, with a tendency to form in alpine grassland and alpine meadow. Annual variations in RTSs area exhibited a significant positive correlation with minimum air temperature, mean land surface temperature, and annual thawing index, while it showing a significant negative correlation with the decrease in downward shortwave radiation. Spatially, RTSs were more common in areas with higher soil water content and shallower active layer. Landsat imagery captured the vast majority of RTSs on the TP and revealed interannual disturbance details, but the 30 m resolution remains inadequate for delineating the refined boundaries of some micro-scale (&lt; 0.18 ha) RTSs. Detected RTSs disturbances on the TP will aid in hazard management and carbon feedback assessments, and our findings provide novel insights into the impacts of climate change and permafrost environments on RTSs formation.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"325 ","pages":"Article 114786"},"PeriodicalIF":11.1,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143886102","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}

{"title":"A vision foundation model-based method for large-scale forest disturbance mapping using time series Sentinel-1 SAR data","authors":"Yuping Tian ,&nbsp;Feng Zhao ,&nbsp;Ran Meng ,&nbsp;Rui Sun ,&nbsp;Yuan Zhang ,&nbsp;Yanyan Shen ,&nbsp;Bin Wang ,&nbsp;Jie Liu ,&nbsp;Mingze Li","doi":"10.1016/j.rse.2025.114775","DOIUrl":"10.1016/j.rse.2025.114775","url":null,"abstract":"<div><div>Accurate and timely forest disturbance mapping at large-scale is crucial for ecosystem protection and management. Sentinel-1 SAR data, with its all-weather capability and fine spatial-temporal resolutions, offers unique advantages for timely mapping of forest disturbance. Although deep learning models have been used for this purpose, they still struggle to fully exploit Sentinel-1 data's potential due to challenges in extracting multi-scale features and capturing context in complex landscape patterns. The advent of vision foundation models like the Segment Anything Model (SAM) offer new possibilities to improve large-scale forest disturbance mapping with Sentinel-1 data. However, trained with natural images, SAM had difficulty processing speckle noises in SAR and recognizing intricate forest disturbance patterns; additionally, challenges remained in developing model transfer strategies and improving model efficiency for large-scale applications. To address these issues, we propose SAMSR, a SAM-based framework adapted for forest disturbance mapping using time series Sentinel-1 data by adding CNN branches and cross-branch attention modules to enhance fine spatial details. Firstly, we evaluated SAMSR's performances against SAM, U-Net and DeepLabv3+ at four global disturbance hotspots with large variations in environmental conditions (Rondônia in Brazil, Guangxi in China, California in USA, and Hainan in China). Then, we test the model's transferability using fine-tuned transfer learning to identify the best transfer strategy and explore the potential of active learning to further enhance model efficiency. The results indicated that the SAMSR (IoU 0.50–0.78; F1 0.67–0.88) outperformed SAM, U-Net and DeepLabv3+ for about 23.08 %, 3.23 % and 1.59 % in IoU, respectively, while the multi-region model achieved optimal transfer performance (IoU 0.60–0.77; F1 0.75–0.87). Moreover, applying active learning can meet the saturation accuracy of traditional method with about 20 %–70 % less training samples, significantly reducing costs for fine-tuning the foundational model. This study thus provides a novel and efficient framework for large-scale forest disturbance monitoring at fine spatial-temporal resolutions, which could be critical for forest protection and ecological studies.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"325 ","pages":"Article 114775"},"PeriodicalIF":11.1,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143882489","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}

{"title":"Characterizing dynamics of built-up height in China from 2005 to 2020 based on GEDI, Landsat, and PALSAR data","authors":"Peimin Chen ,&nbsp;Huabing Huang ,&nbsp;Peng Qin ,&nbsp;Xiangjiang Liu ,&nbsp;Zhenbang Wu ,&nbsp;Feng Zhao ,&nbsp;Chong Liu ,&nbsp;Jie Wang ,&nbsp;Zhan Li ,&nbsp;Xiao Cheng ,&nbsp;Peng Gong","doi":"10.1016/j.rse.2025.114776","DOIUrl":"10.1016/j.rse.2025.114776","url":null,"abstract":"<div><div>The unprecedented urbanization in China has driven rapid urban and rural development in recent decades. While existing studies have extensively focused on horizontal urban expansion, research on vertical urban expansion patterns remains limited. To address this gap, this study proposed a Multi-Temporal Built-up Height estimation Network (MTBH-Net) to estimate 30-m China Multi-Temporal Built-up Height (CMTBH-30) by integrating Global Ecosystem Dynamics Investigation (GEDI), Landsat, and PALSAR data. Specifically, we introduced sample migration to generate reference built-up height data and applied the Continuous Change Detection and Classification (CCDC) disturbance feature to reduce inconsistency in unchanged built-up areas. Validation using the GEDI test set demonstrated that CMTBH-30 achieved RMSEs of 5.10 m, 5.53 m, 6.16 m, and 6.21 m for 2005, 2010, 2015, and 2020. Further validation with field-collected data yielded an RMSE of 4.54 m. Additionally, CMTBH-30 is consistent with the 3D-GIoBFP dataset, achieving RMSEs ranging from 5.34 m to 8.95 m across ten cities. Our findings reveal an increase in average built-up heights in China from 10.28 m in 2005 to 10.92 m in 2020, reflecting an upward trend in urban development. Additionally, the standard deviation of built-up heights rises from 5.16 m in 2005 to 7.71 m in 2020, indicating increased height variation nationwide. Regional analysis from 2005 to 2020 highlights notable vertical growth in newly expanded built-up areas in Macau (+14.4 m), Hong Kong (+12.3 m), and Guangdong (+12.3 m), while Qinghai (+3.8 m) and Chongqing (+3.0 m) also experienced significant growth in stable built-up areas. Heilongjiang, Jilin, Hebei, and Taiwan exhibited minimal growth. The CMTBH-30 dataset effectively captures fine-grained built-up heights, addressing the gap in multi-temporal built-up height estimation. This study provides a new dimension for urban research and is valuable for a multitude of applications such as urban planning, disaster management, and sustainable development. The CMTBH-30 dataset is available at <span><span>https://data-starcloud.pcl.ac.cn/iearthdata/</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"325 ","pages":"Article 114776"},"PeriodicalIF":11.1,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873200","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}

{"title":"ESTIMET: Enhanced and Spatial-Temporal Improvement of MODIS EvapoTranspiration algorithm for all sky conditions in tropical biomes","authors":"Cinthia M.A. Claudino ,&nbsp;Guillaume F. Bertrand ,&nbsp;Rodolfo L.B. Nóbrega ,&nbsp;Cristiano das N. Almeida ,&nbsp;Ana Cláudia V. Gusmão ,&nbsp;Suzana M.G.L. Montenegro ,&nbsp;Bernardo B. Silva ,&nbsp;Eduardo G. Patriota ,&nbsp;Filipe C. Lemos ,&nbsp;Jaqueline V. Coutinho ,&nbsp;José Welton Gonçalo de Sousa ,&nbsp;João M. Andrade ,&nbsp;Davi C.D. Melo ,&nbsp;Diogo Francisco B. Rodrigues ,&nbsp;Leidjane M. Oliveira ,&nbsp;Yunqing Xuan ,&nbsp;Magna S.B. Moura ,&nbsp;Abelardo A.A. Montenegro ,&nbsp;Luca Brocca ,&nbsp;Chiara Corbari ,&nbsp;Victor Hugo R. Coelho","doi":"10.1016/j.rse.2025.114771","DOIUrl":"10.1016/j.rse.2025.114771","url":null,"abstract":"<div><div>We developed an ET model, namely the Enhanced and Spatial-Temporal Improvement of MODIS EvapoTranspiration (ESTIMET), for local-to-regional ET monitoring and applications in the tropics, based on the original MOD16 evapotranspiration (ET) algorithm. The main distinguishing features of ESTIMET are providing a near-real-time product with increased spatial (from 500 to 250 m) and temporal (from 8-day to daily) resolutions, minimising gaps in cloud cover and adjusting specific tropical characteristics of diverse vegetation and microclimate types. We compared the results of ESTIMET with the MOD16A2GF, PML_V2, and GLEAM 4.1a ET products, using eddy covariance (EC) data from 14 sites in Brazil, as well as the water balance-based annual ET in 25 Brazilian catchments. Overall, the ESTIMET estimates captured the daily seasonal variations of the EC data, especially in the Caatinga, Pantanal, and Cerrado biomes, with concordance correlation coefficients (ρc) ranging from 0.45 to 0.80 at eight sites located in these three biomes. The comparisons of the 8-day cumulative ET show that the ESTIMET algorithm exhibits a mean ρc of 0.63, greater than that of MOD16A2GF (ρc = 0.58), GLEAM 4.1a (ρc = 0.47), and PML_V2 (ρc = 0.45). Similarly, for the catchment water balance, ESTIMET exhibits a better representation of annual ET than other ET products in the three major South American biomes, i.e. the Amazon, Atlantic Forest, and Cerrado, which cover over 85 % of the Brazilian territory. Thus, ESTIMET improves remote sensing-based ET estimates in tropical biomes, operating at a finer spatiotemporal scale and latency (i.e. monthly) under all sky conditions.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"325 ","pages":"Article 114771"},"PeriodicalIF":11.1,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873201","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}

{"title":"A physically based differentiable radiative transfer model (DRTM) for land surface optical and biochemical parameters retrieval","authors":"Lisai Cao ,&nbsp;Zhijun Zhen ,&nbsp;Shengbo Chen ,&nbsp;Tiangang Yin","doi":"10.1016/j.rse.2025.114764","DOIUrl":"10.1016/j.rse.2025.114764","url":null,"abstract":"<div><div>The differential path tracing method and automatic differentiation can effectively calculate the derivatives of the loss function, enabling the estimation of surface properties such as reflectivity and transmissivity from sensor images. However, their full potential has not been completely explored in remote sensing. We developed a differentiable radiative transfer model (DRTM) to efficiently simulate and retrieve leaf optical properties, leaf biochemical components, and sensor observation angles from passive remote sensing imagery. The modeling accuracy is verified using various three-dimensional (3D) heterogeneous landscapes, including natural vegetation-covered and artificial urban landscapes. The forward modeling part of DRTM has proved to be faster and more efficient in computer resource usage. In addition, DRTM demonstrated a much more effective adaptation of deep learning than the traditional look-up table method, to better resolve the most challenging inversions from canopy level to foliar level in vegetation remote sensing. In this context, DRTM can potentially address various inverse challenges in remote sensing within a unified framework.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"325 ","pages":"Article 114764"},"PeriodicalIF":11.1,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143872463","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}

{"title":"Detection of multi-year ex-fast ice in the Weddell Sea, Antarctica, using ICESat-2 satellite altimeter data","authors":"Younghyun Koo ,&nbsp;Hongjie Xie ,&nbsp;Walter N. Meier ,&nbsp;Stephen F. Ackley ,&nbsp;Nathan T. Kurtz","doi":"10.1016/j.rse.2025.114750","DOIUrl":"10.1016/j.rse.2025.114750","url":null,"abstract":"<div><div>Landfast ice, sea ice fastened to coastal land or ice shelves, generally undergoes distinctive thermodynamic growth and less active dynamic deformation due to its prolonged attachment to the land, resulting in a thicker and smoother surface compared to drifting pack ice. In 2019, large landfast ice floes were detached from the Ronne Ice Shelf, and the broken pieces started to drift into the Weddell Sea. This study employs a random forest (RF) machine learning model to detect these ex-fast ice floes using six key variables from the ICESat-2 ATL10 sea ice freeboard product: freeboard, Gaussian width of photon height distribution, standard deviation of freeboard, floe length, modal freeboard, and sea ice concentration. The RF model achieves an overall accuracy of 99 % in detecting ex-fast ice, effectively capturing the drift, freeboard distribution, and size distribution of ex-fast ice floes across the western Weddell Sea in 2019. Among six variables, freeboard, standard deviation of freeboard, and Gaussian width of photon height distribution contribute over 94 % to the model performance. Furthermore, the detection of ex-fast ice improves the quantification of sea ice topographical features derived from ICESat-2, including modal freeboard, ridge fraction, and surface roughness. This study highlights the effectiveness of discriminating heterogeneous ex-fast ice from typical pack ice to enhance sea ice measurements using ICESat-2 satellite altimeter data.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"325 ","pages":"Article 114750"},"PeriodicalIF":11.1,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143864848","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}

{"title":"An integrating pre-temperature description method for generating all-weather land surface temperature via passive microwave and thermal infrared remote sensing","authors":"Weizhen Ji ,&nbsp;Yunhao Chen ,&nbsp;Xiaohui Li ,&nbsp;Kangning Li ,&nbsp;Haiping Xia ,&nbsp;Ji Zhou ,&nbsp;Han Gao","doi":"10.1016/j.rse.2025.114767","DOIUrl":"10.1016/j.rse.2025.114767","url":null,"abstract":"<div><div>Integrating passive microwave (PMW) and thermal infrared (TIR) remote sensing to generate all-weather land surface temperature (LST) is essential for effective land thermal monitoring. Previous studies have attempted to adapt TIR-interactive kernel-driven downscaling techniques into the PMW-TIR integration process. However, large-scale spans often introduce significant uncertainties in the generated LST, potentially leading to spatial streaks. To address these challenges, it is critical to introduce a reliable temperature representation at the target resolution to generate accurate all-weather LST. In this study, we propose an integrated pre-temperature description model (ITDM) comprising three modules. The first module is a machine learning-based bias correction-driven generation module (BCDM), which generates relatively precise LST, particularly during the daytime, though it may smooth some spatial textures in certain regions. The second module, a spatial detail-aware generation module (SDAM), utilizes an annual temperature cycle model-based LST as a temperature description, ensuring spatial consistency in the generated LST. The third module integrates the two previous modules, addressing their differences to optimize the final output. Validation results based on MODIS LST indicate that the proposed method achieves a daytime root mean squared error (RMSE) of 3.20 K and a standard deviation of bias (STD) of 3.08 K. For nighttime, the RMSE and STD are 2.24 K and 2.15 K, respectively. Additionally, ten in-situ measurements reveal an average RMSE of 3.90 K in the daytime and 3.34 K in the nighttime. Comparative results with two other advanced methods based on MODIS LST and in-situ LST show that the proposed approach reduces RMSE by 0.04–0.91 K and mitigates streaking phenomena more effectively. The study also discusses feature importance, module performance, and the extendibility of the method. The proposed model significantly contributes to the generation of high-quality all-weather LST.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"324 ","pages":"Article 114767"},"PeriodicalIF":11.1,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143860475","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}

{"title":"The relationship between the ratio of far-red to red leaf SIF and leaf chlorophyll content: Theoretical derivation and experimental validation","authors":"Runfei Zhang ,&nbsp;Peiqi Yang ,&nbsp;Shan Xu ,&nbsp;Long Li ,&nbsp;Tingrui Guo ,&nbsp;Dalei Han ,&nbsp;Jing Liu","doi":"10.1016/j.rse.2025.114762","DOIUrl":"10.1016/j.rse.2025.114762","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Leaf chlorophyll content (LCC) is an important indicator of photosynthetic capacity. Sun-induced chlorophyll fluorescence (SIF) is an optical signal emitted from the leaf interior, providing a unique technique for accurately estimating LCC. The far-red to red ratio of chlorophyll fluorescence (&lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt;) has been used to empirically estimate LCC in some previous studies. While these studies support the use of the &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt; for LCC estimation, its theoretical underpinning remains less well-defined and its effectiveness across a wider range of scenarios remains unclear. In this study, we established the relationship between the &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt; and LCC using the light use efficiency (LUE)-based SIF model and spectral invariant radiative transfer theory. Firstly, the LUE-based SIF model demonstrates that the change in the leaf &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt; is controlled by the ratio of the fluorescence escape fraction (i.e., &lt;em&gt;f&lt;/em&gt;&lt;sub&gt;&lt;em&gt;esc&lt;/em&gt;&lt;/sub&gt; from the photosystem to the leaf surface) at the corresponding bands. Secondly, a &lt;em&gt;f&lt;/em&gt;&lt;sub&gt;&lt;em&gt;esc&lt;/em&gt;&lt;/sub&gt; modeling approach is presented using the spectral invariant theory and thus the &lt;em&gt;f&lt;/em&gt;&lt;sub&gt;&lt;em&gt;esc&lt;/em&gt;&lt;/sub&gt; ratio is linked to LCC. Theoretical analysis shows that the &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt; has a strong correlation with LCC, which explains over 90 % of the variation in &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt;. Both experimental measurements and model simulations from a radiative transfer model Fluspect were used to validate the relationship between LCC and three &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt; (i.e., &lt;span&gt;&lt;math&gt;&lt;msubsup&gt;&lt;mi&gt;F&lt;/mi&gt;&lt;mtext&gt;ratio&lt;/mtext&gt;&lt;mo&gt;↑&lt;/mo&gt;&lt;/msubsup&gt;&lt;/math&gt;&lt;/span&gt;, &lt;span&gt;&lt;math&gt;&lt;msubsup&gt;&lt;mi&gt;F&lt;/mi&gt;&lt;mtext&gt;ratio&lt;/mtext&gt;&lt;mo&gt;↓&lt;/mo&gt;&lt;/msubsup&gt;&lt;/math&gt;&lt;/span&gt; and &lt;span&gt;&lt;math&gt;&lt;msubsup&gt;&lt;mi&gt;F&lt;/mi&gt;&lt;mtext&gt;ratio&lt;/mtext&gt;&lt;mi&gt;tot&lt;/mi&gt;&lt;/msubsup&gt;&lt;/math&gt;&lt;/span&gt;), which were derived from the upward and downward SIF of leaves, as well as the total SIF observed from both sides. The Fluspect simulations were used to assess the sensitivity of the &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt;-LCC relationship to the leaf structure. Two types of experimental measurements, including the field measurements of three crops and the laboratory measurements of 20 tundra plants, were employed to examine the species dependence of the &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt;-LCC relationship. The performance of &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt; for LCC estimation was evaluated and compared with spectral indices and the PROSPECT model using the experimental measurements and leave-one-out cross-validation (LOOCV) approach. Both the Fluspect simulations and the experimental measurements indicate that the &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt; is strongly correlated with LCC for a wide range of leaf scenarios. The &lt;em&gt;F&lt;/em&gt;&lt;sub&gt;ratio&lt;/sub&gt;-LCC relationship remains relatively stable across different leaf structures and plant species, since the relationship is almost consistent. The LOOCV of experimental measurem","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"324 ","pages":"Article 114762"},"PeriodicalIF":11.1,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143855876","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}

{"title":"Machine learning-based generation of high-resolution 3D full-coverage aerosol distribution data over China using multisource data","authors":"Wenze Li ,&nbsp;Wenchao Han ,&nbsp;Jiachen Meng ,&nbsp;Zipeng Dong ,&nbsp;Jun Xu ,&nbsp;Qimeng Wang ,&nbsp;Lulu Yuan ,&nbsp;Han Wang ,&nbsp;Zhongzhi Zhang ,&nbsp;Miaomiao Cheng","doi":"10.1016/j.rse.2025.114772","DOIUrl":"10.1016/j.rse.2025.114772","url":null,"abstract":"<div><div>Aerosol pollution significantly influences the interaction between solar radiation and the earth's atmosphere and seriously threatens human health. Numerous studies have applied machine learning models such as Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) to estimate aerosol-related parameters, including aerosol optical depth and particulate matter concentrations (e.g., PM_2.5). However, current aerosol products primarily provide horizontal or spatially discontinuous vertical data, lacking comprehensive three-dimensional (3D) coverage. To address this gap, we developed the XGBoost-LightGBM-Wavelet (XLW) model, integrating XGBoost, LightGBM, and wavelet transforms to merge multisource data. This approach, for the first time, produced high-resolution, three-dimensional, full-coverage aerosol distribution data for China in 2015. The model outputs a dataset of aerosol spatial distribution with a horizontal resolution of 0.05°, and 167 layers within 10 km in the vertical direction. The XLW model demonstrates excellent predictive ability, effectively filling gaps in aerosol distribution. It enhances signal continuity and strengthens lower-layer signals, closely matching ground LiDAR observations and providing a more accurate representation compared to the Cloud-Aerosol LiDAR and Infrared Pathfinder Satellite Observation (CALIPSO) data. The dataset accurately reveals the 3D distribution of aerosols, which is meaningful for a comprehensive study of aerosol distribution at different altitudes in various regions. At 300 m height above ground level, the most polluted regions are the North China Plain and the Yangtze River Delta region, with an average aerosol extinction coefficient (AEC) of 0.34 and 0.40 km⁻¹, respectively. As the height increases to 1 km, the average AEC notably decreases to 0.23 and 0.24 km⁻¹ in the North China Plain and the Yangtze River Delta. By 3 km, aerosol distribution becomes sparse over most regions of China. For the vertical variations of aerosol distributions in typical cities, in the North China Plain and Yangtze River Delta, aerosol concentrations consistently decrease from the near-surface to 4 km. However, in the Pearl River Delta, aerosol concentrations decrease consistently from 0 to 2 km, with relatively stable between 2 and 3 km. Above 4 km, aerosol concentrations are nearly negligible in all typical cities. The XLW model can accurately produce a high-resolution, 3D, full-coverage aerosol spatial distribution dataset, which is vital for conducting thorough studies on aerosol transport, aerosol radiative effects, and climate change.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"324 ","pages":"Article 114772"},"PeriodicalIF":11.1,"publicationDate":"2025-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143851386","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}

{"title":"GDCM: Generalized data completion model for satellite observations","authors":"Haoyu Wang ,&nbsp;Yinfei Zhou ,&nbsp;Xiaofeng Li","doi":"10.1016/j.rse.2025.114760","DOIUrl":"10.1016/j.rse.2025.114760","url":null,"abstract":"<div><div>Ocean remote sensing data is crucial in understanding the global climate system. Due to satellite orbital coverage gaps and cloud cover, satellite ocean remote sensing products have significant data gaps. This paper introduces a Generalized Data Completion Model (GDCM) based on deep learning to reconstruct gap-free and cloud-free key oceanic variables such as sea surface temperature (SST), wind speed, water vapor, cloud liquid water, and precipitation rate derived from polar-orbiting satellite sensors including Advanced Microwave Scanning Radiometer 2 (AMSR2), the Special Sensor Microwave Imager (SSMI), and the Moderate Resolution Imaging Spectroradiometer (MODIS). Utilizing Convolutional Neural Networks (CNNs) and attention mechanisms, the GDCM model effectively leverages spatio-temporal information within remote sensing data to fill in missing regions accurately. We use reanalysis data to simulate various data missing scenarios during model training for model development. We tested the model with the US East Coast region's global-coverage AMSR2/SSMI and local-coverage MODIS datasets. The experiments demonstrate that the GDCM model successfully and precisely completes the data for different satellites and types of missing data. To enable the model to capture enough data for the dynamical change patterns, we used seven consecutive days of observation data as inputs to improve the model's data-completion ability, significantly enhancing the handling of MODIS SST missing data due to cloud cover. When the input data's duration increased from one day to seven days, the model's R² value improved from 0.062 to 0.93, and the Root Mean Square Difference (RMSD) decreased from 6.58 to 0.92. Besides the model framework design, we implemented the incremental learning training strategy to enhance the model's data completion capability for different types of missing data, especially for SST data from AMSR2 satellites. The model's completed SST data R² value improved from 0.93 to 0.99, and the RMSD decreased from 2.64 °C to 0.50 °C. The Mean Absolute Difference (MAD) of water vapor data decreased from 0.88 kg/m² to 0.76 kg/m², and the RMSD decreased from 1.39 kg/m² to 1.27 kg/m². This study provides a generalized new solution to the problem of missing ocean data at different resolutions, contributing to a more comprehensive and supporting ocean science research and related applications.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"324 ","pages":"Article 114760"},"PeriodicalIF":11.1,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143838424","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}