Remote Sensing of Environment最新文献

{"title":"Estimating global transpiration from TROPOMI SIF with angular normalization and separation for sunlit and shaded leaves","authors":"Chen Zheng , Shaoqiang Wang , Jing M. Chen , Jingfeng Xiao , Jinghua Chen , Zhaoying Zhang , Giovanni Forzieri","doi":"10.1016/j.rse.2024.114586","DOIUrl":"10.1016/j.rse.2024.114586","url":null,"abstract":"<div><div>Gross primary productivity (GPP) is more accurately estimated by total canopy solar-induced chlorophyll fluorescence (<span><math><msub><mi>SIF</mi><mi>total</mi></msub></math></span>) compared to raw sensor observed SIF signals (<span><math><msub><mi>SIF</mi><mi>obs</mi></msub></math></span>). The use of two-leaf strategy, which distinguishes between SIF from sunlit (<span><math><msub><mi>SIF</mi><mi>sunlit</mi></msub></math></span>) and shaded (<span><math><msub><mi>SIF</mi><mi>shaded</mi></msub></math></span>) leaves, further improves GPP estimates. However, the two-leaf strategy, along with SIF corrections for bidirectional effects, has not been applied to transpiration (T) estimation. In this study, we used the angular normalization method to correct the bidirectional effects and separate <span><math><msub><mi>SIF</mi><mi>sunlit</mi></msub></math></span> and <span><math><msub><mi>SIF</mi><mi>shaded</mi></msub></math></span>. Then we developed <span><math><msub><mi>SIF</mi><mi>sunlit</mi></msub></math></span> and <span><math><msub><mi>SIF</mi><mi>shaded</mi></msub></math></span> driven semi-mechanistic and hybrid models, comparing their T estimates with those from a <span><math><msub><mi>SIF</mi><mi>obs</mi></msub></math></span> driven semi-mechanistic model at both site and global scales. All three types of SIF-driven T models integrate canopy conductance (<span><math><msub><mi>g</mi><mi>c</mi></msub></math></span>) with the Penman-Monteith model, differing in how <span><math><msub><mi>g</mi><mi>c</mi></msub></math></span> is derived: from a <span><math><msub><mi>SIF</mi><mi>obs</mi></msub></math></span> driven semi-mechanistic equation, a <span><math><msub><mi>SIF</mi><mi>sunlit</mi></msub></math></span> and <span><math><msub><mi>SIF</mi><mi>shaded</mi></msub></math></span> driven semi-mechanistic equation, and a <span><math><msub><mi>SIF</mi><mi>sunlit</mi></msub></math></span> and <span><math><msub><mi>SIF</mi><mi>shaded</mi></msub></math></span> driven machine learning model. When evaluated against partitioned T using the underlying water use efficiency method at 72 eddy covariance sites and two global T remote sensing products, a consistent pattern emerged: <span><math><msub><mi>SIF</mi><mi>sunlit</mi></msub></math></span> and <span><math><msub><mi>SIF</mi><mi>shaded</mi></msub></math></span> driven hybrid model > <span><math><msub><mi>SIF</mi><mi>sunlit</mi></msub></math></span> and <span><math><msub><mi>SIF</mi><mi>shaded</mi></msub></math></span> driven semi-mechanistic model > <span><math><msub><mi>SIF</mi><mi>obs</mi></msub></math></span> driven semi-mechanistic model. The <span><math><msub><mi>SIF</mi><mi>sunlit</mi></msub></math></span> and <span><math><msub><mi>SIF</mi><mi>shaded</mi></msub></math></span> driven hybrid model demonstrated a notable proficiency under high vapor pressure deficit and low soil water content conditions. The <span><math><msub><mi>SIF</mi><mi>obs</mi></msub></math></span> driven semi-mec","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"319 ","pages":"Article 114586"},"PeriodicalIF":11.1,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143077122","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":"Individual tree crown delineation in high resolution aerial RGB imagery using StarDist-based model","authors":"Fei Tong,&nbsp;Yun Zhang","doi":"10.1016/j.rse.2025.114618","DOIUrl":"10.1016/j.rse.2025.114618","url":null,"abstract":"<div><div>The availability of high spatial resolution remote sensing imagery has facilitated forestry attribute estimation at the individual tree level. However, producing accurate tree crown delineations for practical applications remains challenging, particularly in mixed forests with overlapping tree crowns. In this study, we propose an individual tree crown delineation method leveraging the StarDist model to improve the delineation accuracy in mixed forests. The StarDist model captures tree crown shapes uniquely through star-convex polygons, which are predicted by the U-Net architecture. The final tree crowns are determined by applying non-maximum suppression (NMS) to all identified star-convex polygons. Performance evaluation on two mixed forest areas reveals a delineation accuracy exceeding 92%, notably outperforming the widely used deep learning model MASK R-CNN by over 6%. In terms of tree crown areas estimation, the <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> for both testing areas is higher than 0.85 for both testing areas. Moreover, the evaluations on precision, recall, and F1-score demonstrate that the proposed model can generate tree crowns fitting well with the true crowns. This study marks the first utilization of the StarDist model for tree crown delineation in mixed forests. Our findings demonstrate the effectiveness of the StarDist model for accurately delineating individual tree crowns, thereby advancing the field of forestry research.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"319 ","pages":"Article 114618"},"PeriodicalIF":11.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143071556","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cloud heights retrieval from passive satellite measurements using lapse rate information","authors":"Weiyuan Zhang ,&nbsp;Jiming Li ,&nbsp;Jiayi Li ,&nbsp;Sihang Xu ,&nbsp;Lijie Zhang ,&nbsp;Yang Wang ,&nbsp;Jianping Huang","doi":"10.1016/j.rse.2025.114622","DOIUrl":"10.1016/j.rse.2025.114622","url":null,"abstract":"<div><div>Cloud top and base height (CTH and CBH) are essential in understanding the role of clouds on the weather and climate systems and improving radiation and precipitation simulations. However, inferring accurate cloud heights from passive satellite observations remains more challenging, especially for CBH. This study developed an effective and convenient method for estimating cloud heights for different cloud types on a global scale. The method is based on the mean lapse rate from surface to cloud top (<em>Γ</em>_<em>ct</em>), the lapse rate within (<em>Γ</em>_<em>cb1</em>) and below cloud (<em>Γ</em>_<em>cb2</em>), which are calculated from collocated active and passive satellite observations. The CTH and CBH can be easily derived based on cloud top temperature (CTT), surface temperature (ST), surface height (SH), <em>Γ</em>_<em>ct</em>, <em>Γ</em>_<em>cb1</em> and <em>Γ</em>_<em>cb2</em>. The lapse rate method was applied to polar-orbiting and geostationary passive satellites and the performances were evaluated using cloud heights measurements from CloudSat and CALIPSO satellite. Overall, our retrieval results can achieve high accuracy and stability in estimating both CTH and CBH. For example, our CTH results have significantly improved the retrieval accuracy, with mean bias error (MBE) is 0 km and R is 0.96, and the absolute bias error (MAE) and root mean square error (RMSE) are reduced from 1.12 km and 1.72 km to 0.85 km and 1.33 km, respectively, compared with the MODIS CTH product. Our CBH retrieval results based on MODIS CTT and ST also agree well with CloudSat and CALIPSO observations, the R is 0.91 and the MAE, MBE and RMSE are 0.73 km, 0 km and 1.26 km, respectively. The cloud geometric thickness derived from the cloud heights retrieval results also agrees well with the active satellite observations (MAE = 0.97 km, MBE = 0 km, RMSE = 1.44 km and <em>R</em> = 0.91). In addition, the good performance of cloud heights retrieval during night and for geostationary satellites can further illustrate the excellent accuracy and strong applicability of the lapse rate method. Specifically, compared with SatCORPS Himawari-8 product, the MAE and RMSE of CTH (CBH) are reduced by 41.5 % (44.2 %) and 39.4 % (36.6 %), respectively. These statistical results confirm that our method has comparable performance to other algorithms (e.g., machine learning and other empirical methods), in the meantime, exhibiting the advantages of simplicity and less input parameters. In addition, the lapse rate method can also be employed to provide a supplemental criterion on determining cloud layers from radiosonde data.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"319 ","pages":"Article 114622"},"PeriodicalIF":11.1,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143071584","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":"Risk identification of mangroves facing Spartina alterniflora invasion using data-driven approaches with UAV and machine learning models","authors":"Zhiyi Kan ,&nbsp;Bin Chen ,&nbsp;Weiwei Yu ,&nbsp;Shunyang Chen ,&nbsp;Guangcheng Chen","doi":"10.1016/j.rse.2025.114613","DOIUrl":"10.1016/j.rse.2025.114613","url":null,"abstract":"<div><div>The rapid and large-scale invasion of <em>Spartina alterniflora</em> has led to extensive biodiversity loss and the degradation of essential ecosystem services, particularly in mangroves. Recent studies have shown that the landscape pattern of mangroves is a key indicator of whether <em>Spartina alterniflora</em> can successfully invade. Therefore, assessing the risk of <em>S. alterniflora</em> invasion from the perspective of mangrove landscape patterns is crucial. This study utilizes drone imagery to extract the spatial distribution of mangroves and <em>S. alterniflora</em> in the Jiulong Estuary. Using an interpretable machine learning model, the relationship between mangrove landscape patterns and <em>S. alterniflora</em> coverage was studied at three plot sizes (10 m, 20 m, 30 m). The results show that: (1) The interpretable machine learning model constructed based on UAV (Unmanned Aerial Vehicle) big data can assess <em>S. alterniflora</em> coverage through mangrove landscape patterns. The validation set achieved good results with Adjusted R² (&gt;0.974), MAE (Mean Absolute Error) (&lt;4.8), and RMSE (Root Mean Square Error) (&lt;7) across the three plot sizes, though the SMAPE (Symmetric Mean Absolute Percentage Error) indicator was relatively poor (&gt;20 %). (2) At all three plot sizes, mangrove coverage, mangrove tree height, average patch perimeter, patch density, landscape shape index, and edge density significantly impacted the suppression of <em>S. alterniflora</em> coverage by mangroves. However, the relative importance of these indicators changes with increasing spatial granularity. (3) The SHAP (SHapley Additive exPlanations) Value of the model revealed the nonlinear response of <em>S. alterniflora</em> coverage to changes in mangrove coverage: mangrove coverage inhibits <em>S. alterniflora</em> growth between 41 %–64 %, and completely suppresses it above 65 %. This study effectively assesses the extent of mangrove invasion by <em>S. alterniflora</em> and provides clear spatial evidence of the threshold for mangrove coverage to suppress <em>S. alterniflora</em> growth. It emphasizes that rationally regulating mangrove distribution through landscape indices will more effectively resist the invasion of <em>S. alterniflora.</em></div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"319 ","pages":"Article 114613"},"PeriodicalIF":11.1,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143071583","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":"Cross-scalar analysis of multisensor land surface phenology","authors":"Xiaojie Gao ,&nbsp;Sophia Stonebrook ,&nbsp;Tristan Green ,&nbsp;Minkyu Moon ,&nbsp;Mark A. Friedl","doi":"10.1016/j.rse.2025.114624","DOIUrl":"10.1016/j.rse.2025.114624","url":null,"abstract":"<div><div>Land surface phenology (LSP) metrics derived from remote sensing are widely used to monitor vegetation phenology over large areas and to characterize how the growing seasons of terrestrial ecosystems are responding to climate change. Until recently, however, most LSP studies relied on coarse spatial resolution sensors, which makes assigning direct linkages between LSP metrics and ecological processes and properties challenging due to scale mismatches and because substantial variation in phenology and ecological properties are often present at sub-pixel scale in coarse resolution LSP metrics. In this study, we leverage publicly available LSP data products with three orders of magnitude difference in spatial resolution derived from Moderate Resolution Imaging Spectroradiometer (MODIS, 500 m), Landsat and Sentinel-2 (HLS, 30 m), and PlanetScope (3 m) imagery to examine and characterize the nature, magnitude, and sources of the agreement and disagreement in LSP metrics across spatial scales. Our results provide three key conclusions: (1) LSP metrics from three sensors showed consistently high cross-scalar agreement across sites (r² = 0.70–0.97), suggesting that they all effectively capture geographic variation in LSP; (2) within-site cross-scalar agreement between LSP metrics was systematically lower relative to agreement across sites, but mean absolute differences were consistent across and within sites (generally &lt;14 days for day of year-based metrics, with a few exceptions); and (3) local-scale composition and heterogeneity in land cover is a key factor that controls cross-scalar agreement in LSP metrics. In particular, we found that site-level heterogeneity in land cover (measured via entropy) and the proportion of evergreen versus deciduous land cover types explain up to half of site-to-site variance in local-scale cross-scalar agreement in LSP metrics. Results from this study support the internal consistency and quality of the three LSP data products examined, and more generally, provide guidance regarding the choice of spatial resolution for different applications and land cover conditions, and yield new insights related to how LSP observations scale across different sensors and spatial resolutions.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"319 ","pages":"Article 114624"},"PeriodicalIF":11.1,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143071557","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 novel strategy of full-waveform light detection and ranging bathymetry based on spatial gram angle difference field conversion and deep-learning network architecture","authors":"Yifu Chen ,&nbsp;Lin Wu ,&nbsp;Yue Qian ,&nbsp;Yuan Le ,&nbsp;Yi Yang ,&nbsp;Dongfang Zhang ,&nbsp;Liqin Zhou ,&nbsp;Haichao Guo ,&nbsp;Lizhe Wang","doi":"10.1016/j.rse.2025.114615","DOIUrl":"10.1016/j.rse.2025.114615","url":null,"abstract":"<div><div>The airborne light detection and ranging (LiDAR) bathymetry (ALB) system is a promising and effective approach for surveying nearshore bathymetry and underwater terrain. Deep-learning techniques have been developed to reduce waveform superposition in shallow and deep areas. These methods avoid the complex transmission process of laser pulses in the water column and the intricate determination of various parameters and thresholds in traditional ALB methods. However, studies on ALB bathymetry using deep-learning techniques remain insufficient. To improve the accuracy and reliability of nearshore bathymetry, this study proposes deep-learning bathymetry fusing waveform features and spatial-angular field features (DBWSF). This method utilizes the waveform curvature to construct an energy curve, enhancing the waveform's features. Additionally, it employs a Gram angle difference field to convert the temporal waveform into a two-dimensional Gram angle difference field image, increasing the dimensions and quantity of waveform features. Finally, this method constructs a dual-path neural network with an attention mechanism to extract the water surface and bottom waveform signals precisely to achieve nearshore bathymetry. In comparison to sample data, DBWSF exhibited high bathymetric accuracy across three study areas (Ganquan Island, Lingyang Reef, and Dong Island), achieving a root mean squared error of 0.21 m. The R² in these regions were 99.6 %, 95.1 %, and 99.8 %, respectively. In the above three study areas, compared with the bathymetric results obtained using the waveform decomposition method, DBWSF was more accurate, with improvements in RMSE of 0.33, 0.27, and 0.04 m. Compared with multilayer perceptron (MLP), the corresponding accurate improvements in RMSE with DBWSF were 0.47, 0.40, and 0.25 m. Compared with other two methods, the R² value for DBWSF in the three study areas exceeded 95 % and reached a highest value of 99.8 %. The results demonstrated the bathymetric capability, reliability, and transferability of DBWSF for determining nearshore bathymetry in different water environments. The novel LiDAR bathymetric deep-learning technique can effectively and intelligently produce precise nearshore bathymetry and seafloor topography maps.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"318 ","pages":"Article 114615"},"PeriodicalIF":11.1,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143044440","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 multi-sensor approach allows confident mapping of forest canopy fuel load and canopy bulk density to assess wildfire risk at the European scale","authors":"Elena Aragoneses ,&nbsp;Mariano García ,&nbsp;Hao Tang ,&nbsp;Emilio Chuvieco","doi":"10.1016/j.rse.2024.114578","DOIUrl":"10.1016/j.rse.2024.114578","url":null,"abstract":"<div><div>With the increasing influence of climate and socio-economic changes, crown fires are becoming the main concern of fire managers and civil protection authorities in Europe. Evaluating and mitigating the negative impacts of these fires requires better tools to identify high-risk areas. Prevention and management strategies for crown fires require accurate and cost-effective tools that can parameterise fuel properties. Here, we use a multi-sensor approach integrating satellite Light Detection and Ranging (LiDAR) observations from the Global Ecosystems Dynamics Investigation (GEDI) sensor, with other remote sensing imagery and biophysical variables to provide spatially-explicit estimates of two key descriptors of crown fire behaviour – canopy fuel load (CFL) and canopy bulk density (CBD) – over the entire European territory at 1 km² grid resolution.</div><div>GEDI L1B and L2A level footprints were used to estimate Leaf Area Density, from which CFL and CBD were subsequently derived. The approach was assessed by applying it to regions of the United States, where bioclimatic conditions are similar to those in Europe, and for which LANDFIRE CBD maps are available (CBD <em>r</em> = 0.6–0.86 and RMSE = 33.1–59.6 %). We then extrapolated the estimates to European areas not covered by GEDI using machine learning models with multispectral (Landsat 8) and radar (Phased Array L-band Synthetic Aperture Radar sensor – PALSAR) imagery, and biophysical variables (CFL <em>r</em> = 0.85 and RMSE = 12.98 %; CBD <em>r</em> = 0.75 and RMSE = 21 %). Pixel-level uncertainty for the spatial extrapolation was also estimated.</div><div>The new wall-to-wall maps of crown fuel properties (<span><span>https://doi.org/10.21950/Z6BWQG</span><svg><path></path></svg></span>) provide new insights into the potential for fire risk prevention in Europe, which together with climate and socio-economic models, would greatly improve the prioritisation of management areas and the targeting of mitigation measures in strategic areas to reduce wildfire risk.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"318 ","pages":"Article 114578"},"PeriodicalIF":11.1,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143050397","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated retrieval of sea-ice salinity, density, and thickness using polarimetric GNSS-R","authors":"Joan Francesc Munoz-Martin,&nbsp;Nereida Rodriguez-Alvarez,&nbsp;Xavier Bosch-Lluis,&nbsp;Kamal Oudrhiri","doi":"10.1016/j.rse.2025.114617","DOIUrl":"10.1016/j.rse.2025.114617","url":null,"abstract":"<div><div>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.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"318 ","pages":"Article 114617"},"PeriodicalIF":11.1,"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}

{"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 ,&nbsp;Shangbo Liu ,&nbsp;Zhihui Wang ,&nbsp;Xun Zhao ,&nbsp;Jianbo Qi ,&nbsp;Yelu Zeng ,&nbsp;Dong Li ,&nbsp;Pengfei Guo ,&nbsp;Zhexiu Yu ,&nbsp;Simei Lin ,&nbsp;Shouyang Liu ,&nbsp;Huaguo Huang","doi":"10.1016/j.rse.2025.114616","DOIUrl":"10.1016/j.rse.2025.114616","url":null,"abstract":"<div><div>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.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"318 ","pages":"Article 114616"},"PeriodicalIF":11.1,"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}

{"title":"Enhanced sea ice classification for ICESat-2 using combined unsupervised and supervised machine learning","authors":"Wenxuan Liu ,&nbsp;Michel Tsamados ,&nbsp;Alek Petty ,&nbsp;Taoyong Jin ,&nbsp;Weibin Chen ,&nbsp;Julienne Stroeve","doi":"10.1016/j.rse.2025.114607","DOIUrl":"10.1016/j.rse.2025.114607","url":null,"abstract":"<div><div>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.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"318 ","pages":"Article 114607"},"PeriodicalIF":11.1,"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}