Remote Sensing of Environment最新文献

{"title":"GROUNDED EO: Data-driven Sentinel-2 LAI and FAPAR retrieval using Gaussian processes trained with extensive fiducial reference measurements","authors":"Luke A. Brown ,&nbsp;Richard Fernandes ,&nbsp;Jochem Verrelst ,&nbsp;Harry Morris ,&nbsp;Najib Djamai ,&nbsp;Pablo Reyes-Muñoz ,&nbsp;Dávid D.Kovács ,&nbsp;Courtney Meier","doi":"10.1016/j.rse.2025.114797","DOIUrl":"10.1016/j.rse.2025.114797","url":null,"abstract":"<div><div>Due to their importance in monitoring and modelling Earth's climate, the Global Climate Observing System (GCOS) designates leaf area index (LAI) and the fraction of absorbed photosynthetically active radiation (FAPAR) as essential climate variables (ECVs). The Simplified Level 2 Biophysical Processor (SL2P) has proven particularly popular for decametric (i.e. 10 m to 100 m) retrieval of these ECVs. Comprehensive validation has shown that due to simplifying assumptions in the underlying radiative transfer models (RTMs), biases persist in SL2P retrievals. To avoid RTM assumptions altogether, an empirical data-driven approach might be considered. Yet, such a strategy has historically been prevented by the limited quantity and quality of available in situ reference measurements, as well as the large number of training samples traditionally required by machine learning regression algorithms. New opportunities are now offered by recently established continental-scale environmental monitoring networks, advances in automated data processing and uncertainty evaluation, and machine learning regression algorithms that require many fewer training samples. The Ground Reference Observations Underlying Novel Decametric Vegetation Data Products from Earth Observation (GROUNDED EO) project was initiated to take advantage of these opportunities. We describe the empirical data-driven LAI and FAPAR retrieval approach adopted within the project, involving i) generation of a database containing over 16,000 fiducial reference measurements covering 81 National Ecological Observatory Network (NEON), Integrated Carbon Observation System (ICOS), and Terrestrial Ecosystem Research Network (TERN) sites between 2013 and 2022, ii) development of an empirical data-driven algorithm for Sentinel-2 LAI and FAPAR retrieval based on Gaussian processes, and iii) evaluation of GROUNDED EO retrievals through intercomparison with the current state-of-the-art in decametric retrieval (i.e. SL2P, and a modified version of SL2P developed by the Canada Centre for Remote Sensing – SL2P-CCRS), as well as validation against unseen fiducial reference measurements. In the majority of cases (and despite not making use of ancillary data such as land cover), the empirical data-driven GROUNDED EO retrievals were subject to reduced bias than those from SL2P and SL2P-CCRS, as well as increased fulfilment of user requirements (i.e. 74% of LAI and 69% of FAPAR retrievals overall). Consequently, the approach has potential to reduce uncertainty in key inputs for climate monitoring and modelling, agricultural and forest management, and biodiversity assessment.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"326 ","pages":"Article 114797"},"PeriodicalIF":11.1,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144067364","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":"Dynamic landslide susceptibility mapping over last three decades to uncover variations in landslide causation in subtropical urban mountainous areas","authors":"Peifeng Ma ,&nbsp;Li Chen ,&nbsp;Chang Yu ,&nbsp;Qing Zhu ,&nbsp;Yulin Ding ,&nbsp;Zherong Wu ,&nbsp;Hongsheng Li ,&nbsp;Changyao Tian ,&nbsp;Xuanmei Fan","doi":"10.1016/j.rse.2025.114800","DOIUrl":"10.1016/j.rse.2025.114800","url":null,"abstract":"<div><div>Landslide susceptibility assessment (LSA) plays a vital role in disaster prevention and mitigation. Recently, numerous data-driven LSA approaches have emerged. Nonetheless, most of them neglected the rapid oscillations within the landslide-prone environment, primarily due to significant changes in external triggers such as rainfall, which would render landslides susceptible to varying causations over time. Thus, conducting dynamic landslide susceptibility mapping (D-LSM) and revealing the underlying trends in landslide causes, become increasingly important for effective landslide hazard assessment. This study decomposed the entire D-LSM task into yearly LSA subtasks, and innovatively meta-learned intermediate representations that can be well-generalized and fine-tuned in a fast-adaptation manner. Then, to interpret the model predictions and characterize the variations in landslide causation, Shapley Additive exPlanations (SHAP) was utilized for feature permutation year by year. In addition, MT-InSAR techniques were applied to enhance and validate the D-LSM results. The study area was Lantau Island, Hong Kong, where the yearly LSA was executed from 1992 to 2019. The performance comparison results show that the proposed method outperformed the other approaches with regard to accuracy (3 %–7 %), precision (2 %–9 %), recall (3 %–5 %), and F1-score (2 %–7 %), even when adopting a fast adaptation strategy using only 5 samples and 5 gradient descent updates. This validates the applicability of meta-learning for identifying commonalities across multi-temporal LSA tasks. The overall model interpretation results indicate that slope and extreme rainfall were the primary contributors to landslide occurrences in Hong Kong. The feature permutation results over the 30 years reveal a variation in landslide causation, particularly a dramatic shift in the ranking of some contributing factors under extreme weather conditions. Remarkably, the importance of AERD (Annual Extreme Rainfall Days), a factor indicating extreme rainfall intensity, was deeply affected by global climate change and the government's Landslide Prevention and Mitigation Programme (LPMitP).</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"326 ","pages":"Article 114800"},"PeriodicalIF":11.1,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144066427","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":"Monitoring Spartina Alterniflora removal dynamics across coastal China using time series Sentinel-1 imagery","authors":"Yukui Min ,&nbsp;Yinghai Ke ,&nbsp;Zhaojun Zhuo ,&nbsp;Weichun Qi ,&nbsp;Jinyuan Li ,&nbsp;Peng Li ,&nbsp;Nana Zhao","doi":"10.1016/j.rse.2025.114813","DOIUrl":"10.1016/j.rse.2025.114813","url":null,"abstract":"<div><div>Invasions by <em>Spartina</em> species have posed serious threats to coastal ecosystems worldwide. Since the introduction of <em>Spartina alterniflora</em> (<em>S. alterniflora</em>) in China in 1979, it has expanded across 68,000 ha of coastal wetlands by 2020. In 2022, the Chinese government issued the “Special Action Plan for the Prevention and Control of <em>Spartina alterniflora</em> (2022–2025)”, aiming for nationwide eradication by 2025. As local and provincial removal efforts progress, timely information on removal status and timing is crucial for tracking the project's progress, assessing the effectiveness of control measures, and facilitating in-depth research on <em>S. alterniflora</em> re-establishment mechanism. Frequent cloud cover hinders the application of optical satellite imagery in timely monitoring of <em>S. alterniflora</em> removal in coastal areas. The all-weather Sentinel-1 SAR sensor overcomes this limitation, offering frequent acquisitions suitable for accurate mapping of <em>S. alterniflora</em> removal. In this study, we present the SAR Time Series Change and change Time Detection (STS-CTD) model, a deep learning framework designed to detect <em>S. alterniflora</em> removal events using Sentinel-1 time series imagery, providing information on where and when <em>S. alterniflora</em> was removed. The model integrates Transformer encoder, multi-kernel Conv1D decoder, and Band Dropout training strategy to learn the abrupt changes in time series backscatters and radar indices caused by plant removal. We applied the model within the boundaries of national-scale <em>S. alterniflora</em> map and generated the first <em>S. alterniflora</em> removal maps across China's coastline during 2021 to 2023. Our key findings include: (1) The resultant <em>S. alterniflora</em> removal map achieved an overall accuracy (OA) of 97.95 % and an F1-score of 98.39 % for removal identification, and the removal timing estimation exhibited a Mean Absolute Error (MAE) of 6.77 days and a Root Mean Square Error (RMSE) of 14.68 days; (2) the multi-kernel Conv1D and Band Dropout strategy considerably improved the model performance compared to models using only Transformer encoder and conventional Dropout; (3) the STS-CTD model outperformed state-of-the-art time series analysis models, including Bi-LSTM, Bi-GRU, TCN, and InceptionTime; (4) the model demonstrated strong temporal transferability, showing promise for application in future years; (5) it effectively mitigated noise from SAR imaging and tidal inundation, though continuous inundation and incomplete removal reduced accuracy in certain areas. The STS-CTD model and the resulting national-scale maps offer an operational solution for assessing invasive species management in coastal wetlands.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"326 ","pages":"Article 114813"},"PeriodicalIF":11.1,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144066428","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":"Satellite canopy water content from Sentinel-2, Landsat-8 and MODIS: Principle, algorithm and assessment","authors":"Hongliang Ma ,&nbsp;Marie Weiss ,&nbsp;Daria Malik ,&nbsp;Beatrice Berthelot ,&nbsp;Marta Yebra ,&nbsp;Rachael H. Nolan ,&nbsp;Arnaud Mialon ,&nbsp;Jiangyuan Zeng ,&nbsp;Xingwen Quan ,&nbsp;Håkan Torbern Tagesson ,&nbsp;Albert Olioso ,&nbsp;Frederic Baret","doi":"10.1016/j.rse.2025.114801","DOIUrl":"10.1016/j.rse.2025.114801","url":null,"abstract":"<div><div>In spite of the efforts made for canopy water content (CWC) mapping in the community, including vegetation water proxy from microwave-based vegetation optical depth (VOD) as well as optical-based indices, there is still no operational CWC product from optical satellites up to now. To fill this gap, this study proposes a unified algorithm for CWC mapping at both decametric and coarse spatial resolution from several widely used optical satellites. Based on machine learning trained on radiative transfer model simulations, we comprehensively parameterized the distribution of the canopy and vegetation input variables (i.e., leaf traits and soil background) of the PROSAIL model, by relying on the largest open integrated global plant and soil spectral databases. We investigated the impact of diverse band combinations as well as the inclusion of optical indices for CWC estimation using RTMs. The performances of this algorithm were first evaluated at decametric resolution based on ground measurements distributed over five ground campaigns corresponding to diverse climate and biome types. The retrieved CWC from Sentinel-2 and Landsat-8 exhibits satisfactory performance, with coefficient R of 0.81 and RMSE of 0.046 g/cm². We then evaluated CWC at 500 m resolution from MODIS by comparing it with Landsat-8 and Sentinel-2 aggregated values over a globally distributed selection of LANDVAL sites, representative of the existing biome types combined with a range of precipitation, soil moisture and vegetation density conditions. The MODIS CWC global maps show reasonable seasonal and spatial patterns compared to multi-frequencies microwave-based VOD, and improvements compared to the conventionally and extensively used optical indices such as NDWI. The CWC product developed in this study is expected to provide new insights for global or regional vegetation water variations monitoring from optical satellites, with the strength of high spatial resolution compared to the microwave passive VOD (i.e., 20-500 m vs 22.5 km). These two products could be further combined for more accurate global vegetation water and biomass mapping in the future to improve our understanding of carbon uptake and hydrological applications.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"326 ","pages":"Article 114801"},"PeriodicalIF":11.1,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144066429","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":"Improved satellite-scale land surface temperature components retrieval with hotspot effect correction and temperature difference constraints","authors":"Yifan Lu ,&nbsp;Zunjian Bian ,&nbsp;Jean-Louis Roujean ,&nbsp;Hua Li ,&nbsp;Frank M. Göttsche ,&nbsp;Yajun Huang ,&nbsp;Tengyuan Fan ,&nbsp;Biao Cao ,&nbsp;Yongming Du ,&nbsp;Qing Xiao","doi":"10.1016/j.rse.2025.114794","DOIUrl":"10.1016/j.rse.2025.114794","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Land surface temperature (LST) plays an important role in Earth energy balance and water/carbon cycle processes and is recognized as an Essential Climate Variable (ECV) and an Essential Agricultural Variable (EAV). LST products that are issued from satellite observations mostly depict landscape-scale temperature due to their generally large footprint. This means that a pixel-based temperature integrates over various components, whereas temperature individual components are better suited for the purpose of evapotranspiration estimation, crop growth assessment, drought monitoring, etc. Thus, disentangling soil and vegetation temperatures is a real matter of concern. Moreover, most satellite-based LSTs are contaminated by directional effects due to the inherent anisotropy properties of most terrestrial targets. The characteristics of directional effects are closely linked to the properties of the target and controlled by the view and solar geometry. A singular angular signature is obtained in the hotspot geometry, i.e., when the sun, the satellite and the target are aligned. The hotspot phenomenon highlights the temperature differences between sunlit and shaded areas. However, due to the lack of adequate multi-angle observations and inaccurate portrayal or neglect of solar influence, the hotspot effect is often overlooked and has become a barrier for better inversion results at satellite scale. Therefore, hotspot effect needs to be better characterized, which here is achieved with a three-component model that distinguishes vegetation, sunlit and shaded soil temperature components and accounts for vegetation structure. Our work combines thermal infrared (TIR) observations from the Sea and Land Surface Temperature Radiometer (SLSTR) onboard the LEO (Low Earth Orbit) Sentinel-3, and two sensors onboard GEO (geostationary) satellites, i.e. the Advanced Himawari Imager (AHI) and Spinning Enhanced Visible and Infrared Imager (SEVIRI). Based on inversion with a Bayesian method and prior information associated with component temperature differences as constrained, the findings include: 1) Satellite observations throughout East Asia around noon indicate that for every 10 degrees change in angular distance from the sun, LST will on average vary by 0.6 K; 2) As a better constraint, the hotspot effect can benefit from multi-angle TIR observations to improve the retrieval of LST components, thereby reducing the root mean squared error (RMSE) from approximately 3.5 K, 5.8 K, and 4.1 K to 2.8 K, 3.5 K, and 3.1 K, at DM, EVO and KAL sites, respectively; 3) Based on a dataset simulated with a three-dimensional radiative transfer model, a significant inversion error may result if the hotspot is ignored for an angular distance between the viewing and solar directions that is smaller than 30&lt;span&gt;&lt;math&gt;&lt;msup&gt;&lt;mrow&gt;&lt;/mrow&gt;&lt;mo&gt;°&lt;/mo&gt;&lt;/msup&gt;&lt;/math&gt;&lt;/span&gt;. Overall, considering the hotspot effect has the potential to reduce inversion noise and to separate the tem","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"326 ","pages":"Article 114794"},"PeriodicalIF":11.1,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144066430","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":"Enhancing terrestrial net primary productivity estimation with EXP-CASA: A novel light use efficiency model approach","authors":"Guanzhou Chen ,&nbsp;Kaiqi Zhang ,&nbsp;Xiaodong Zhang ,&nbsp;Hong Xie ,&nbsp;Haobo Yang ,&nbsp;Xiaoliang Tan ,&nbsp;Tong Wang ,&nbsp;Yule Ma ,&nbsp;Qing Wang ,&nbsp;Jinzhou Cao ,&nbsp;Weihong Cui","doi":"10.1016/j.rse.2025.114790","DOIUrl":"10.1016/j.rse.2025.114790","url":null,"abstract":"&lt;div&gt;&lt;div&gt;The Light Use Efficiency (LUE) model, epitomized by the Carnegie-Ames-Stanford Approach (CASA) model, is extensively applied in the quantitative estimation and analysis of vegetation Net Primary Productivity (NPP). However, the classic CASA model is marked by significant complexity: the estimation of environmental stress, in particular, necessitates multi-source observation data and model parameters, adding to the complexity and uncertainty of the model’s operation. Additionally, the saturation effect of the Normalized Difference Vegetation Index (NDVI), a key variable in the CASA model, weakens the accuracy of CASA’s NPP predictions in densely vegetated areas. To address these limitations, this study introduces the Exponential-CASA (EXP-CASA) model. The EXP-CASA model effectively improves the CASA model with RMSE decreasing by 37.5% by using novel functions for estimating the fraction of absorbed photosynthetically active radiation (FPAR) and environmental stress, utilizing long-term observational data from FLUXNET and MODIS surface reflectance data. In a comparative analysis of NPP estimation accuracy, EXP-CASA (&lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msup&gt;&lt;mrow&gt;&lt;mi&gt;R&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mn&gt;2&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;0&lt;/mn&gt;&lt;mo&gt;.&lt;/mo&gt;&lt;mn&gt;68&lt;/mn&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;R&lt;/mi&gt;&lt;mi&gt;M&lt;/mi&gt;&lt;mi&gt;S&lt;/mi&gt;&lt;mi&gt;E&lt;/mi&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;1&lt;/mn&gt;&lt;mo&gt;.&lt;/mo&gt;&lt;mn&gt;1&lt;/mn&gt;&lt;mspace&gt;&lt;/mspace&gt;&lt;mi&gt;gC&lt;/mi&gt;&lt;mi&gt;⋅&lt;/mi&gt;&lt;msup&gt;&lt;mrow&gt;&lt;mi&gt;m&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mo&gt;−&lt;/mo&gt;&lt;mn&gt;2&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;mi&gt;⋅&lt;/mi&gt;&lt;msup&gt;&lt;mrow&gt;&lt;mi&gt;d&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mo&gt;−&lt;/mo&gt;&lt;mn&gt;1&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;) performs better than the NPP product from GLASS (&lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msup&gt;&lt;mrow&gt;&lt;mi&gt;R&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mn&gt;2&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;0&lt;/mn&gt;&lt;mo&gt;.&lt;/mo&gt;&lt;mn&gt;61&lt;/mn&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;R&lt;/mi&gt;&lt;mi&gt;M&lt;/mi&gt;&lt;mi&gt;S&lt;/mi&gt;&lt;mi&gt;E&lt;/mi&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;1&lt;/mn&gt;&lt;mo&gt;.&lt;/mo&gt;&lt;mn&gt;2&lt;/mn&gt;&lt;mspace&gt;&lt;/mspace&gt;&lt;mi&gt;gC&lt;/mi&gt;&lt;mi&gt;⋅&lt;/mi&gt;&lt;msup&gt;&lt;mrow&gt;&lt;mi&gt;m&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mo&gt;−&lt;/mo&gt;&lt;mn&gt;2&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;mi&gt;⋅&lt;/mi&gt;&lt;msup&gt;&lt;mrow&gt;&lt;mi&gt;d&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mo&gt;−&lt;/mo&gt;&lt;mn&gt;1&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;). Additionally, this research assesses the EXP-CASA model’s adaptability to various vegetation indices, evaluates the sensitivity and stability of its parameters over time, and compares its accuracy against other leading NPP estimation products across different seasons, latitudinal zones, ecological types, and temporal sequences. The findings reveal that the EXP-CASA model exhibits strong adaptability to diverse vegetation indices and stability of model parameters over time series. Importantly, EXP-CASA displays superior sensitivity to NPP anomalies at flux sites and more accurately simulates short-term NPP fluctuations than GLASS-NPP and captures periodic trends. By introducing a novel estimation approach that optimizes model construction, the EXP-CASA model remarkably improves the accuracy of NPP estimations, paving the way for global-scale, consistent, and continuous assessment of vegetation NPP. I","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"326 ","pages":"Article 114790"},"PeriodicalIF":11.1,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143946707","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":"Mapping recreational marine traffic from Sentinel-2 imagery using YOLO object detection models","authors":"Janne Mäyrä ,&nbsp;Elina A. Virtanen ,&nbsp;Ari-Pekka Jokinen ,&nbsp;Joni Koskikala ,&nbsp;Sakari Väkevä ,&nbsp;Jenni Attila","doi":"10.1016/j.rse.2025.114791","DOIUrl":"10.1016/j.rse.2025.114791","url":null,"abstract":"<div><div>Identifying where maritime activities take place, and quantifying their potential impact on marine biodiversity, is important for the sustainable management of marine areas, spatial planning and marine conservation. Detection and monitoring of small vessels, such as pleasure crafts, has been challenging due to limited data availability with adequate temporal and spatial resolution. Here, we develop an analysis framework to detect and quantify small vessels, using openly available Sentinel-2 optical imagery, and the YOLO-family object detection models. We also show how existing spatial datasets can be used to improve the quality of the model output.</div><div>We chose five distinct marine areas along the Finnish coast in the northern Baltic Sea and manually annotated 8,768 vessels for training and validating object detection models. The top-performing model, based on F1-score, achieved a test set F1-score of 0.863 and an mAP50 of 0.888. This model was then used to quantify pleasure crafts in the southern part of Finland from May to September 2022, using all available Sentinel-2 images with sufficient image quality. Detected recreational traffic was primarily concentrated on boating lanes and close to marinas, with weekend traffic averaging approximately twice the volume of weekday traffic.</div><div>The developed approach can be used to identify areas with high boat traffic, thus supporting sustainable spatial planning, ecological impact avoidance, and serving as an indicator of recreational popularity of marine areas. Additionally, it provides a basis for assessing the potential impacts of pleasure crafts on marine biodiversity. By leveraging openly available satellite images with frequent revisits, the approach offers broad geographical coverage and high spatial accuracy, making it scalable for estimating boat traffic across large areas, even globally.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"326 ","pages":"Article 114791"},"PeriodicalIF":11.1,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143946702","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 knowledge-augmented deep fusion method for estimating near-surface air temperature","authors":"Fengrui Chen ,&nbsp;Xi Li ,&nbsp;Yiguo Wang","doi":"10.1016/j.rse.2025.114819","DOIUrl":"10.1016/j.rse.2025.114819","url":null,"abstract":"<div><div>Near-surface air temperature (Ta) is a critical meteorological variable, and obtaining its precise spatiotemporal distribution is essential for numerous scientific domains beyond meteorology and hydrology. Despite the promising advancements in Ta mapping using machine learning, these models often suffer from inadequate generalization capabilities due to their heavy reliance on data. A critical limitation is that their “free” learning style fails to deeply uncover the intricate spatiotemporal patterns of Ta. Addressing this problem, we propose a novel knowledge-augmented deep fusion method (KADF), designed to enhance the accuracy of Ta mapping through integrating prior knowledge. KADF integrates three categories of prior knowledge concerning Ta: spatial autocorrelation, temporal autocorrelation, and temporal heterogeneity in the relationship between Ta and predictive variables. This tailored strategy enables the model to more efficiently explore the intricate spatiotemporal relationships between grounded Ta observations and satellite-derived auxiliary variables, culminating in accurate Ta estimates through the deep fusion of these datasets. The efficacy of KADF was thoroughly evaluated over Chinese mainland. The validation results show that KADF accurately mapped the spatiotemporal distribution of daily Ta, with root mean square error (RMSE) values of 1.0 °C for mean Ta (T_mean), 1.22 °C for maximum Ta (T_max), and 1.33 °C for minimum Ta (T_min). Moreover, the integration of prior knowledge regarding Ta significantly enhanced the generalizability of the data-driven mapping model. Compared to the state-of-the-art machine learning-based estimation method, KADF reduced the mean absolute error (MAE) values by 23–31 % and RMSEs by 24–29 %. Furthermore, this method considerably improved the ability to capture spatial and temporal variations in Ta across various environmental conditions. Finally, a 1 km daily Ta dataset for the time frame spanning from 2010 to 2018 was produced. Overall, KADF holds great promise for accurately estimating Ta and can be easily adapted to other regions. The source code of KADF has been made publicly available at <span><span>https://github.com/Henu-frch/KADF</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"326 ","pages":"Article 114819"},"PeriodicalIF":11.1,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942423","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":"Across-scale thermal infrared anisotropy in forests: Insights from a multi-angular laboratory-based approach","authors":"Jennifer Susan Adams ,&nbsp;Alexander Damm ,&nbsp;Mike Werfeli ,&nbsp;Julian Gröbner ,&nbsp;Kathrin Naegeli","doi":"10.1016/j.rse.2025.114766","DOIUrl":"10.1016/j.rse.2025.114766","url":null,"abstract":"&lt;div&gt;&lt;div&gt;The Land Surface Temperature (LST) is well suited to monitor biosphere–atmosphere interactions in forests, as it depends on water availability and atmospheric/meteorological conditions above and below the canopy. Satellite-based LST has proven integral in observing evapotranspiration, estimating surface heat fluxes and characterising vegetation properties. Since the radiative regime of forests is complex, driven by canopy structure, components radiation properties and their arrangement, forest radiative temperatures are subject to strong angular effects. However, this depends on the scale of observation, where scattering mechanisms from canopy – to satellite – scales influence anisotropy with varying orders of magnitude. Given the heterogeneous and complex nature of forests, multi-angular data collection is particularly difficult, necessitating instrumentation distant enough from the canopy to obtain significant canopy brightness temperature and concurrent observations to exclude turbulence/atmospheric effects. Accordingly, current research and understanding on forest anisotropy at varying scales (from local validation level to satellite footprint) remain insufficient to provide practical solutions for addressing angular effects for upcoming thermal satellite sensors and associated validation schemes. This study presents a novel method founded in the optical remote sensing domain to explore the use of microcanopies that represent forests at different scales in the footprint of a multi-angular goniometer observing system. Both Geometric Optical (GO) and volumetric scattering dominated canopies are constructed to simulate impacts of anisotropy in heterogeneous and homogeneous canopies, and observed using a thermal infrared radiometer. Results show that heterogeneous canopies dominated by GO scattering are subject to much higher magnitudes of anisotropy, reaching maximum temperature differences of 3 °C off-nadir. Magnitudes of anisotropy are higher in sparse forests, where the gap fraction and crown arrangement (inducing sunlit/shaded portions of soil and vegetation) drive larger off-nadir differences. In dense forests, anisotropy is driven by viewing the maximum portion of sunlit vegetation (hotspot), where the soil is mostly obscured. Canopy structural metrics such as the fractional cover and gap fraction were found to have significant correlation with off-nadir differences. In more homogeneous canopies, anisotropy reaches a lower magnitude with temperature differences up to 1 °C, driven largely by volumetric scattering and components radiation properties. Optimal placement of instrumentation at the canopy-scale (more heterogeneous behaviour due to proximity to the canopy and small pixel size) used to validate satellite observations (more homogeneous behaviour due to larger pixel size) was found to be in cases of viewing maximum sunlit vegetation, for dense canopies. Given upcoming high spatial resolution sensors and associated validatio","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"326 ","pages":"Article 114766"},"PeriodicalIF":11.1,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942314","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 impact of map accuracy on area estimation with remotely sensed data within the stratified random sampling design","authors":"Sergii Skakun","doi":"10.1016/j.rse.2025.114805","DOIUrl":"10.1016/j.rse.2025.114805","url":null,"abstract":"<div><div>One of the core applications of satellite-based classification maps is area estimation. Regardless of the algorithms used, maps will always contain errors stemming from imperfect input and training/calibration data, incomplete data coverage, and spectral and/or temporal confusion between land cover and land use classes. Because of omission and commission errors, the <em>pixel-counting area estimator</em> will be a biased estimator for area estimation. Therefore, the remote sensing research and application communities have developed a framework and recommended practices to address this problem. One such approach is a stratified random sampling design, in which classification maps could be used for stratification in the sampling design, and areas are estimated from the sample data, which represent reference data or reference class labels. As such, the quality of the map, i.e., producer's (PA) and user's accuracy (UA), will not affect the bias of the estimator, as the bias depends on the sampling design and the choice of estimator. However, map quality will impact the efficiency of stratification: a more accurate map will require a smaller sample size to reach the target variance of the estimate, or it will yield improved precision if the sample size is fixed. This study aims to provide a quantitative assessment of the impact of map accuracies on area estimation within the stratified random sampling design. The relative bias of the pixel-counting estimator is expressed using class-specific PA and UA, and shown to be <span><math><mfrac><mi>PA</mi><mi>UA</mi></mfrac><mo>−</mo><mn>1</mn></math></span>. Furthermore, for the case of binary classification, elements of the confusion matrix, as well as the sample size, variance of the area estimator, and relative efficiency of stratification (the ratio of the products of variance and sample size for the two sampling approaches) are expressed using PA and UA. Numerical simulations demonstrate how relative efficiency depends on area estimation objectives, target area proportion, and the map's performance metrics (PA and UA). Such dependence is nonlinear, and the impact of those parameters varies. For example, when the target class is minor or rare (i.e., its true proportion is &lt;&lt;0.5), the impact of PA outweighs that of UA. As the target area proportion increases, the impact of accuracies converges, and UA has a greater impact on efficiency than PA. There are multiple values in the PA/UA space, though constrained, to reach the same objectives, e.g., in terms of relative efficiency, sample size, and target variance. Overall, this study offers map producers a criterion that can be used to benchmark algorithm performance for map generation when area estimation is the primary objective of the classification maps.</div></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":"326 ","pages":"Article 114805"},"PeriodicalIF":11.1,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942422","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}