Pub Date : 2025-12-12DOI: 10.1016/j.rsase.2025.101829
Kishore Bhamidipati , M. Kaur , Tarandeep Singh Walia , D. Garg , Mohammed Amoon , Ekasnh Bhardwaj , Robertas Damaševičius
Pansharpening plays an important role in improving the spatial resolution of multispectral images while preserving their spectral information. It enables more detailed and accurate analysis in various applications, such as remote sensing and environmental monitoring. Recent advances in deep learning-based pansharpening models have resulted in substantial improvements in performance. However, these models still suffer from the balancing of spectral accuracy and spatial detail, which can lead to artifacts, quality degradation, and overfitting problems. To overcome these limitations, an efficient pansharpening model is proposed. Initially, a dual transformer block is designed which integrates Swin and DeiT transformers to improve both local and global feature extraction. These features are then processed through a proposed U-shaped encoder–decoder network. This network utilizes the dual transformer block in both encoding and decoding stages. Finally, a customized multi-aspect pansharpening loss (MAPL) is introduced to preserve spectral fidelity, enhance spatial resolution, and improve perceptual quality. Extensive experimental results demonstrate that the proposed model significantly outperforms competitive models on various performance metrics. Thus, compared to competitive models, the proposed model shows significant improvements in preserving fine spatial details and maintaining spectral accuracy.
{"title":"Hybrid dual-transformer pansharpening network for enhanced spatial-spectral fidelity","authors":"Kishore Bhamidipati , M. Kaur , Tarandeep Singh Walia , D. Garg , Mohammed Amoon , Ekasnh Bhardwaj , Robertas Damaševičius","doi":"10.1016/j.rsase.2025.101829","DOIUrl":"10.1016/j.rsase.2025.101829","url":null,"abstract":"<div><div>Pansharpening plays an important role in improving the spatial resolution of multispectral images while preserving their spectral information. It enables more detailed and accurate analysis in various applications, such as remote sensing and environmental monitoring. Recent advances in deep learning-based pansharpening models have resulted in substantial improvements in performance. However, these models still suffer from the balancing of spectral accuracy and spatial detail, which can lead to artifacts, quality degradation, and overfitting problems. To overcome these limitations, an efficient pansharpening model is proposed. Initially, a dual transformer block is designed which integrates Swin and DeiT transformers to improve both local and global feature extraction. These features are then processed through a proposed U-shaped encoder–decoder network. This network utilizes the dual transformer block in both encoding and decoding stages. Finally, a customized multi-aspect pansharpening loss (MAPL) is introduced to preserve spectral fidelity, enhance spatial resolution, and improve perceptual quality. Extensive experimental results demonstrate that the proposed model significantly outperforms competitive models on various performance metrics. Thus, compared to competitive models, the proposed model shows significant improvements in preserving fine spatial details and maintaining spectral accuracy.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"41 ","pages":"Article 101829"},"PeriodicalIF":4.5,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791459","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-12DOI: 10.1016/j.rsase.2025.101823
Mohammed Q. Alkhatib , Ali Jamali
Over the past decade, hyperspectral image (HSI) classification has drawn considerable interest due to HSIs’ ability to effectively distinguish terrestrial objects by capturing detailed, continuous spectral information. The strong performance of recent deep learning techniques in tasks like image classification and semantic segmentation has led to their growing use in HSI classification, due to their ability to capture complex spatial and spectral features more effectively than traditional methods. This paper presents MixerCA, a novel lightweight model for HSI classification that leverages depthwise convolution and a self-attention mechanism. MixerCA integrates depth-wise convolutions, token and channel mixing, and coordinate attention into a unified structure to decouple spatial and channel interactions, maintain consistent resolution throughout the network, and directly process HSI patches. Extensive experiments on four hyperspectral benchmark datasets reveal MixerCA’s clear advantages over several competing algorithms, including 2D-CNN, 3D-CNN, Tri-CNN, HybridSN, ViT, and Swin Transformer. The source code is publicly available at https://github.com/mqalkhatib/MixerCA.
{"title":"MixerCA: An efficient and accurate model for high-performance hyperspectral image classification","authors":"Mohammed Q. Alkhatib , Ali Jamali","doi":"10.1016/j.rsase.2025.101823","DOIUrl":"10.1016/j.rsase.2025.101823","url":null,"abstract":"<div><div>Over the past decade, hyperspectral image (HSI) classification has drawn considerable interest due to HSIs’ ability to effectively distinguish terrestrial objects by capturing detailed, continuous spectral information. The strong performance of recent deep learning techniques in tasks like image classification and semantic segmentation has led to their growing use in HSI classification, due to their ability to capture complex spatial and spectral features more effectively than traditional methods. This paper presents MixerCA, a novel lightweight model for HSI classification that leverages depthwise convolution and a self-attention mechanism. MixerCA integrates depth-wise convolutions, token and channel mixing, and coordinate attention into a unified structure to decouple spatial and channel interactions, maintain consistent resolution throughout the network, and directly process HSI patches. Extensive experiments on four hyperspectral benchmark datasets reveal MixerCA’s clear advantages over several competing algorithms, including 2D-CNN, 3D-CNN, Tri-CNN, HybridSN, ViT, and Swin Transformer. The source code is publicly available at <span><span>https://github.com/mqalkhatib/MixerCA</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"41 ","pages":"Article 101823"},"PeriodicalIF":4.5,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790942","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-12DOI: 10.1016/j.rsase.2025.101838
Axel Belemtougri , Roland Yonaba , Claire I. Michailovsky , Tibor Stigter , Lawani Adjadi Mounirou , Pieter van der Zaag
This study presents a new geospatial framework for detecting and mapping ephemeral sand rivers (ESRs) across the West African Sahel, focusing on Burkina Faso, Mali, and Niger, where food security challenges are acute. ESRs, which remain dry most of the year, act as vital subsurface water reservoirs in arid environments. During the wet season, infiltrated streamflow is stored within sandy beds, minimizing evaporative losses and providing shallow groundwater with potential to support domestic, livestock, and agricultural needs during dry periods. The methodology integrates hydrological analyses, remote sensing, and machine learning. A high-resolution drainage network was derived from the 90 m MERIT DEM, based on national reference river networks and satellite-derived information. A Random Forest model predicted river flow intermittency and identified ephemeral rivers (flowing 1–4 months annually, catchment area ≥1000 km2), around which 500 m buffer zones were delineated for analysis. Two composite thresholds (CTs) combining NDESI–NDVI spectral indices achieved moderate accuracy: CT1 (42 %) and CT2 (72 %), with CT2 serving as a first-order tool for sandy riverbed detection. A multi-temporal supervised land use/land cover classification achieved high accuracy (92 %) and F1 scores >0.86, outperforming the spectral thresholds. Using vegetation presence as a proxy for shallow groundwater, 19 % of ESRs (402 km) were identified as areas of potentially accessible water storage near settlements representing about 3 million people (4.8 % of the population) across the three countries. These findings highlight the importance of ESRs for sustainable water management and climate-resilient livelihoods in the Sahel.
本研究提出了一个新的地理空间框架，用于探测和绘制整个西非萨赫勒地区的短暂沙河（ESRs），重点关注粮食安全挑战严峻的布基纳法索、马里和尼日尔。esr在一年中大部分时间保持干燥，在干旱环境中充当重要的地下水库。在雨季，渗透的水流被储存在沙质河床中，最大限度地减少蒸发损失，并提供浅层地下水，有可能在干旱时期支持家庭、牲畜和农业需求。该方法集成了水文分析、遥感和机器学习。基于国家参考河网和卫星衍生信息，从90 m MERIT DEM中获得了一个高分辨率的排水网络。随机森林模型预测了河流的间歇性，并确定了短暂河流（每年流动1-4个月，集水区面积≥1000 km2），并在其周围划定了500 m缓冲区进行分析。结合NDESI-NDVI光谱指数的两个复合阈值（ct）达到了中等精度：CT1（42%）和CT2(72%)，其中CT2作为砂质河床检测的一级工具。一个多时间监督的土地利用/土地覆盖分类获得了很高的准确性（92%），F1得分>；0.86，优于光谱阈值。使用植被存在作为浅层地下水的代表，19%的esr（402公里）被确定为潜在的可达储水区域，靠近三个国家约300万人口（占人口的4.8%）的定居点。这些发现突出了可持续水资源管理和气候适应型生计在萨赫勒地区的重要性。
{"title":"Where rivers sleep: mapping ephemeral sand rivers in the West African Sahel","authors":"Axel Belemtougri , Roland Yonaba , Claire I. Michailovsky , Tibor Stigter , Lawani Adjadi Mounirou , Pieter van der Zaag","doi":"10.1016/j.rsase.2025.101838","DOIUrl":"10.1016/j.rsase.2025.101838","url":null,"abstract":"<div><div>This study presents a new geospatial framework for detecting and mapping ephemeral sand rivers (ESRs) across the West African Sahel, focusing on Burkina Faso, Mali, and Niger, where food security challenges are acute. ESRs, which remain dry most of the year, act as vital subsurface water reservoirs in arid environments. During the wet season, infiltrated streamflow is stored within sandy beds, minimizing evaporative losses and providing shallow groundwater with potential to support domestic, livestock, and agricultural needs during dry periods. The methodology integrates hydrological analyses, remote sensing, and machine learning. A high-resolution drainage network was derived from the 90 m MERIT DEM, based on national reference river networks and satellite-derived information. A Random Forest model predicted river flow intermittency and identified ephemeral rivers (flowing 1–4 months annually, catchment area ≥1000 km<sup>2</sup>), around which 500 m buffer zones were delineated for analysis. Two composite thresholds (CTs) combining NDESI–NDVI spectral indices achieved moderate accuracy: CT1 (42 %) and CT2 (72 %), with CT2 serving as a first-order tool for sandy riverbed detection. A multi-temporal supervised land use/land cover classification achieved high accuracy (92 %) and F1 scores >0.86, outperforming the spectral thresholds. Using vegetation presence as a proxy for shallow groundwater, 19 % of ESRs (402 km) were identified as areas of potentially accessible water storage near settlements representing about 3 million people (4.8 % of the population) across the three countries. These findings highlight the importance of ESRs for sustainable water management and climate-resilient livelihoods in the Sahel.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"41 ","pages":"Article 101838"},"PeriodicalIF":4.5,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-12DOI: 10.1016/j.rsase.2025.101835
Aamir Ali , Guanhua Zhou , Yumin Tan , Franz Pablo Antezana Lopez
Biochemical oxygen demand over five days (BOD5) is a cornerstone indicator of organic pollution, yet its retrieval from remote sensing is hindered by its non-optically active nature. We present an explainable ensemble-learning framework that predicts BOD5 in Hong Kong's marine waters by fusing multi-year (2019–2023) Sentinel-2 imagery with cyclic temporal features and four physicochemical and climatic proxies—chlorophyll-a (Chl-a), salinity, suspended solids (SS) and temperature. Initially, each proxy is estimated and subsequently utilized for BOD5 prediction using CatBoost, LightGBM, XGBoost and Random Forest. XGBoost best captures Chl-a (r = 0.81) and temperature (r = 0.99), whereas CatBoost excels for salinity (r = 0.93), SS (r = 0.85) and ultimately BOD5 (r = 0.88). SHapley Additive exPlanations reveal the dominant predictors and spatio-temporal mapping across four representative dates shows persistently elevated Chl-a, SS and BOD5 and depressed salinity in eutrophic Deep Bay zone. This transparent, high-accuracy framework can guide Environmental Protection Department in prioritizing field sampling and streamlining pollution mitigation.
{"title":"Biochemical oxygen demand estimation using explainable ensemble learning methods","authors":"Aamir Ali , Guanhua Zhou , Yumin Tan , Franz Pablo Antezana Lopez","doi":"10.1016/j.rsase.2025.101835","DOIUrl":"10.1016/j.rsase.2025.101835","url":null,"abstract":"<div><div>Biochemical oxygen demand over five days (BOD<sub>5</sub>) is a cornerstone indicator of organic pollution, yet its retrieval from remote sensing is hindered by its non-optically active nature. We present an explainable ensemble-learning framework that predicts BOD<sub>5</sub> in Hong Kong's marine waters by fusing multi-year (2019–2023) Sentinel-2 imagery with cyclic temporal features and four physicochemical and climatic proxies—chlorophyll-a (Chl-a), salinity, suspended solids (SS) and temperature. Initially, each proxy is estimated and subsequently utilized for BOD<sub>5</sub> prediction using CatBoost, LightGBM, XGBoost and Random Forest. XGBoost best captures Chl-a (r = 0.81) and temperature (r = 0.99), whereas CatBoost excels for salinity (r = 0.93), SS (r = 0.85) and ultimately BOD<sub>5</sub> (r = 0.88). SHapley Additive exPlanations reveal the dominant predictors and spatio-temporal mapping across four representative dates shows persistently elevated Chl-a, SS and BOD<sub>5</sub> and depressed salinity in eutrophic Deep Bay zone. This transparent, high-accuracy framework can guide Environmental Protection Department in prioritizing field sampling and streamlining pollution mitigation.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"41 ","pages":"Article 101835"},"PeriodicalIF":4.5,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145738345","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-12DOI: 10.1016/j.rsase.2025.101826
Mohammad Hossein Kazeminezhad
Accurate wind speed estimation is essential for applications in ocean engineering, renewable energy, and climate modeling. Although the ERA5 reanalysis dataset provides high-resolution global wind data, its accuracy varies across different regions and wind regimes. This study presents a comprehensive assessment and calibration of ERA5 wind speed data in the Northwestern Indian Ocean, using satellite observations from multiple scatterometers and altimeters over a 32-year period (1992–2023). The evaluation reveals systematic biases in ERA5, including a general underestimation of high wind speeds and localized discrepancies near coastlines, particularly in the Persian Gulf, Gulf of Oman, and equatorial regions. To enhance the accuracy of wind data, three calibration techniques including Linear Regression (LR), Quantile Mapping (QM), and Artificial Neural Networks (ANN), were applied and compared. The results demonstrate that while LR reduces the normalized mean bias, it offers only limited improvements in the Scatter Index (SI), decreasing it from 20.24 % to 19.07 %. QM improves the alignment of high wind percentiles and reduces the SI for the 99th percentile from 28.69 % to 26.47 %, but it does not significantly enhance the overall dataset. ANN, on the other hand, offers the most effective correction, reducing the overall SI to 17.93 % and the SI at the 99th percentile to 24.76 %. A dedicated assessment under tropical cyclone (TC) conditions further confirms the robustness of the ANN calibration, showing substantial improvements in the representation of extreme wind speed (with the SI reduced from 36.53 % to 28.77 % at the 99th percentile). Despite persistent residual biases at the highest wind speeds, the ANN approach significantly enhances agreement with satellite observations, outperforming raw ERA5 in all evaluated metrics. These findings highlight the potential of machine learning techniques to improve reanalysis datasets in regions where accurate wind representation is critical for climate resilience, offshore safety, and renewable energy planning.
{"title":"Neural network–enhanced calibration of ERA5 wind speeds in the northwestern Indian Ocean using 32 years of satellite observations","authors":"Mohammad Hossein Kazeminezhad","doi":"10.1016/j.rsase.2025.101826","DOIUrl":"10.1016/j.rsase.2025.101826","url":null,"abstract":"<div><div>Accurate wind speed estimation is essential for applications in ocean engineering, renewable energy, and climate modeling. Although the ERA5 reanalysis dataset provides high-resolution global wind data, its accuracy varies across different regions and wind regimes. This study presents a comprehensive assessment and calibration of ERA5 wind speed data in the Northwestern Indian Ocean, using satellite observations from multiple scatterometers and altimeters over a 32-year period (1992–2023). The evaluation reveals systematic biases in ERA5, including a general underestimation of high wind speeds and localized discrepancies near coastlines, particularly in the Persian Gulf, Gulf of Oman, and equatorial regions. To enhance the accuracy of wind data, three calibration techniques including Linear Regression (LR), Quantile Mapping (QM), and Artificial Neural Networks (ANN), were applied and compared. The results demonstrate that while LR reduces the normalized mean bias, it offers only limited improvements in the Scatter Index (SI), decreasing it from 20.24 % to 19.07 %. QM improves the alignment of high wind percentiles and reduces the SI for the 99th percentile from 28.69 % to 26.47 %, but it does not significantly enhance the overall dataset. ANN, on the other hand, offers the most effective correction, reducing the overall SI to 17.93 % and the SI at the 99th percentile to 24.76 %. A dedicated assessment under tropical cyclone (TC) conditions further confirms the robustness of the ANN calibration, showing substantial improvements in the representation of extreme wind speed (with the SI reduced from 36.53 % to 28.77 % at the 99th percentile). Despite persistent residual biases at the highest wind speeds, the ANN approach significantly enhances agreement with satellite observations, outperforming raw ERA5 in all evaluated metrics. These findings highlight the potential of machine learning techniques to improve reanalysis datasets in regions where accurate wind representation is critical for climate resilience, offshore safety, and renewable energy planning.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"41 ","pages":"Article 101826"},"PeriodicalIF":4.5,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145738346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-12DOI: 10.1016/j.rsase.2025.101827
Mingyu Ouyang , Bowei Zeng , Guoru Huang
Rapid and precise waterlogging depth measurements in the context of urban floods are key in guiding the management of such flooding events. Traditional urban flooding monitoring methods are labor-intensive, expensive, and ineffective for comprehensive and timely monitoring. To overcome these limitations, we propose a method to detect the waterlogging depth on urban roads. In particular, the method integrates deep-learning and ellipse detection algorithms for the detection and segmentation of wheels from various vehicle types using Cascade Mask R-CNN. These detected wheels serve as reference objects for the waterlogging depth calculations. The geometric information on the submerged wheels is then obtained using the ellipse and minimum area rectangle detection algorithms. These parameters are subsequently employed to calculate the waterlogging depth on roads. The model was validated on a representative surveillance video site located in Dongying City, China. The model achieves an average bounding box precision and segmentation precision of over 97 % on the validation dataset. Following this, 246 validation samples were compared with manually measured depth. The absolute errors of all samples are below 0.1 m. The proposed method can facilitate the advancement of related studies and offer technical assistance in areas of urban waterlogging monitoring.
{"title":"A deep learning method for identifying waterlogging depth on urban roadways from surveillance camera images","authors":"Mingyu Ouyang , Bowei Zeng , Guoru Huang","doi":"10.1016/j.rsase.2025.101827","DOIUrl":"10.1016/j.rsase.2025.101827","url":null,"abstract":"<div><div>Rapid and precise waterlogging depth measurements in the context of urban floods are key in guiding the management of such flooding events. Traditional urban flooding monitoring methods are labor-intensive, expensive, and ineffective for comprehensive and timely monitoring. To overcome these limitations, we propose a method to detect the waterlogging depth on urban roads. In particular, the method integrates deep-learning and ellipse detection algorithms for the detection and segmentation of wheels from various vehicle types using Cascade Mask R-CNN. These detected wheels serve as reference objects for the waterlogging depth calculations. The geometric information on the submerged wheels is then obtained using the ellipse and minimum area rectangle detection algorithms. These parameters are subsequently employed to calculate the waterlogging depth on roads. The model was validated on a representative surveillance video site located in Dongying City, China. The model achieves an average bounding box precision and segmentation precision of over 97 % on the validation dataset. Following this, 246 validation samples were compared with manually measured depth. The absolute errors of all samples are below 0.1 m. The proposed method can facilitate the advancement of related studies and offer technical assistance in areas of urban waterlogging monitoring.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"41 ","pages":"Article 101827"},"PeriodicalIF":4.5,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145738456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-12DOI: 10.1016/j.rsase.2025.101828
Sumantra Chatterjee , Bala Ram Sapkota , Gurjinder S. Baath , K. Colton Flynn , Douglas R. Smith
Crop height is a key biophysical parameter closely linked to plant growth, biomass, and yield. With the advancement of remote sensing technologies, unmanned aerial vehicles (UAV) have emerged as a promising tool for estimating crop height using structure from motion (SfM) point clouds. However, accurately mapping the digital terrain model (DTM) beneath dense canopies remains a major challenge, as existing methods struggle to capture the soil surface effectively under full vegetation cover. This study aimed to develop and evaluate efficient workflows for generating DTMs from UAV-derived point clouds for in-season crop height estimation, with a focus on eliminating the need for pre-season UAV flights. The experiment was conducted during the 2023 maize growing season in Temple, TX, and compared three workflows: (i) UAV-BCH, which used UAV-derived bare soil surfaces as the DTM; (ii) SentCH, which used Sentinel-1A-derived surfaces from the dormant season; and (iii) UAV-PCH, a novel approach that applied a low-pass filter to select the lowest 1 % of elevation points within a moving window, followed by fitting a 2.5D regression surface to generate the DTM. Results showed that UAV-PCH consistently outperformed the other methods across two fields with varying elevation patterns, achieving higher accuracy (R2 ≈ 0.89) compared to UAV-BCH (R2 ≈ 0.66) and SentCH (R2 ≈ 0.69). UAV-PDTM effectively minimized temporal inconsistencies and vertical misalignments commonly associated with multi-date UAV acquisitions. This approach offers a scalable and efficient solution for crop height estimation using a single UAV flight. Future research should explore its applicability across diverse crop types and canopy structures to enhance its utility in precision agriculture.
{"title":"Advanced workflows for UAV-based crop height estimation using structure from motion (SfM) point clouds","authors":"Sumantra Chatterjee , Bala Ram Sapkota , Gurjinder S. Baath , K. Colton Flynn , Douglas R. Smith","doi":"10.1016/j.rsase.2025.101828","DOIUrl":"10.1016/j.rsase.2025.101828","url":null,"abstract":"<div><div>Crop height is a key biophysical parameter closely linked to plant growth, biomass, and yield. With the advancement of remote sensing technologies, unmanned aerial vehicles (UAV) have emerged as a promising tool for estimating crop height using structure from motion (SfM) point clouds. However, accurately mapping the digital terrain model (DTM) beneath dense canopies remains a major challenge, as existing methods struggle to capture the soil surface effectively under full vegetation cover. This study aimed to develop and evaluate efficient workflows for generating DTMs from UAV-derived point clouds for in-season crop height estimation, with a focus on eliminating the need for pre-season UAV flights. The experiment was conducted during the 2023 maize growing season in Temple, TX, and compared three workflows: (i) <em>UAV-B</em><sub><em>CH</em></sub>, which used UAV-derived bare soil surfaces as the DTM; (ii) <em>Sent</em><sub><em>CH</em></sub>, which used Sentinel-1A-derived surfaces from the dormant season; and (iii) <em>UAV-P</em><sub><em>CH</em></sub>, a novel approach that applied a low-pass filter to select the lowest 1 % of elevation points within a moving window, followed by fitting a 2.5D regression surface to generate the DTM. Results showed that <em>UAV-P</em><sub><em>CH</em></sub> consistently outperformed the other methods across two fields with varying elevation patterns, achieving higher accuracy (R<sup>2</sup> ≈ 0.89) compared to <em>UAV-B</em><sub><em>CH</em></sub> (R<sup>2</sup> ≈ 0.66) and <em>Sent</em><sub><em>CH</em></sub> (R<sup>2</sup> ≈ 0.69). <em>UAV-P</em><sub><em>DTM</em></sub> effectively minimized temporal inconsistencies and vertical misalignments commonly associated with multi-date UAV acquisitions. This approach offers a scalable and efficient solution for crop height estimation using a single UAV flight. Future research should explore its applicability across diverse crop types and canopy structures to enhance its utility in precision agriculture.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"41 ","pages":"Article 101828"},"PeriodicalIF":4.5,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Anthropogenic carbon dioxide (CO2) emission from fossil fuel combustion has been the key driving force for the global warming in the last century. Previous studies have demonstrated the potential of satellites to quantify CO2 emissions from large discrete power plants, which constitute roughly half of the global CO2 emissions from fossil fuels. Most of these studies relied on OCO-2 or OCO-3 data, which have narrow-swath imaging coverage and kilometer-level spatial resolution. As a result, only a small fraction of global power plants could be effectively assessed. In this study, we identified and quantified a larger number of power plants CO2 emission plumes from space using two hyperspectral imagers: Earth Mineral Dust Source Investigation (EMIT) and PRecursore IperSpettrale della Missione Applicativa (PRISMA). We applied the Scene Specific Iterative Matched Filter (SSIMF) algorithm to quantify emissions from 165 power plants across 23 countries on five continents and rectified the systematic underestimation by the algorithm using high-quality hourly emission data reported by the Environmental Protection Agency (EPA) in the U.S. The bias-corrected emission fluxes from hyperspectral imagers were consistent with bottom-up inventories. In addition, we show that hyperspectral imagers can distinguish nearby sources as close as 2 km, which is challenging for coarse resolution OCO. Our study highlights the potential of hyperspectral imagers in the global monitoring of CO2 emissions from individual power plants.
在上个世纪，化石燃料燃烧产生的人为二氧化碳（CO2）排放一直是全球变暖的主要驱动力。以前的研究已经证明，卫星有可能量化大型分散发电厂的二氧化碳排放量，这些发电厂约占全球化石燃料二氧化碳排放量的一半。这些研究大多依赖于OCO-2或OCO-3数据，这些数据具有窄条成像覆盖范围和公里级空间分辨率。因此，全球只有一小部分电厂能够得到有效评估。在这项研究中，我们使用两个高光谱成像仪：地球矿物尘埃源调查（EMIT）和precursoiperspettrale della Missione Applicativa （PRISMA）识别和量化了来自太空的大量发电厂二氧化碳排放羽流。我们应用场景特定迭代匹配滤波（SSIMF）算法对五大洲23个国家的165个发电厂的排放量进行了量化，并利用美国环境保护署（EPA）报告的高质量每小时排放数据纠正了该算法的系统性低估。高光谱成像仪的偏差校正排放通量与自下而上的清单一致。此外，我们表明，高光谱成像仪可以分辨近至2公里的附近源，这对粗分辨率OCO来说是一个挑战。我们的研究强调了高光谱成像仪在全球监测单个发电厂二氧化碳排放方面的潜力。
{"title":"Quantifying thermal power plants CO2 emissions globally from space using hyperspectral imagers","authors":"Menglin Lei , Yuzhong Zhang , Xuyang Huang , Shutao Zhao , Shuai Zhang","doi":"10.1016/j.rsase.2025.101822","DOIUrl":"10.1016/j.rsase.2025.101822","url":null,"abstract":"<div><div>Anthropogenic carbon dioxide (CO<sub>2</sub>) emission from fossil fuel combustion has been the key driving force for the global warming in the last century. Previous studies have demonstrated the potential of satellites to quantify CO<sub>2</sub> emissions from large discrete power plants, which constitute roughly half of the global CO<sub>2</sub> emissions from fossil fuels. Most of these studies relied on OCO-2 or OCO-3 data, which have narrow-swath imaging coverage and kilometer-level spatial resolution. As a result, only a small fraction of global power plants could be effectively assessed. In this study, we identified and quantified a larger number of power plants CO<sub>2</sub> emission plumes from space using two hyperspectral imagers: Earth Mineral Dust Source Investigation (EMIT) and PRecursore IperSpettrale della Missione Applicativa (PRISMA). We applied the Scene Specific Iterative Matched Filter (SSIMF) algorithm to quantify emissions from 165 power plants across 23 countries on five continents and rectified the systematic underestimation by the algorithm using high-quality hourly emission data reported by the Environmental Protection Agency (EPA) in the U.S. The bias-corrected emission fluxes from hyperspectral imagers were consistent with bottom-up inventories. In addition, we show that hyperspectral imagers can distinguish nearby sources as close as 2 km, which is challenging for coarse resolution OCO. Our study highlights the potential of hyperspectral imagers in the global monitoring of CO<sub>2</sub> emissions from individual power plants.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"41 ","pages":"Article 101822"},"PeriodicalIF":4.5,"publicationDate":"2025-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-09DOI: 10.1016/j.rsase.2025.101816
Lokanath S, Govindaraju, Rakesh C J, Kishor Kumar A
Land use and land cover (LULC) changes induce unrivalled environmental changes at various spatial and temporal scales, affecting land surface processes in present catchment and it has experienced detrimental LULC changes over the past three decades, characterized by mismanagement of land and water resources. Hence, the current investigation aimed to analyze the spatial–temporal dynamics of LULC changes from 1990 to 2023 and to predict future trend scenario (1990–2053) for the Vanivilasa Sagara reservoir catchment, Karnataka state in India. The analysis was employed using integrated remote sensing and multitemporal geospatial data by utilizing Landsat – 5, Landsat – 7, IRS LISS – III, and Sentinel – 2A data, with aligning spatial resolutions, uncertainties, along with cumulative statistical model, and ground survey. First, a hybrid–image classification technique was employed to create LULC maps spanning 33 years (1990–2023) across seven reference periods: 1990, 1995, 2000, 2007, 2015, 2020, and 2023, and validated by accuracy assessment using meticulous field data. Subsequently, the magnitude, extent, trajectories, change rate, overall gains, losses, and net LULC changes were derived using statistical evaluation. The concerted results have indicated significant LULC alterations, including transforming forests, scrublands, and grasslands into agricultural and built-up regions such as human settlements, industries, and quarries/mining; cropping pattern conversions; marked rise in wasteland ecosystems such as erosional gullied land, salt-affected land, marshes, and land without scrubs. Moreover, unforeseen thrive in surface water bodies has been found. Key driving parameters and implications of LULC dynamics that hamper the environment of the study area were comprehensively studied and deciphered. Further, future LULC trend analysis obtained by Visual Basics Analysis (VBA) module, serves as an early warning system for resource degradation in catchments amidst climate change is addressed by this study, which contributes to regional planning between researchers, farmers, and government authorities at every level to achieve harmony.
{"title":"Spatio-temporal analysis of land use and land cover (LULC) dynamics: Trends, drivers and implications in semi-arid Vanivilasa sagara reservoir catchment, India","authors":"Lokanath S, Govindaraju, Rakesh C J, Kishor Kumar A","doi":"10.1016/j.rsase.2025.101816","DOIUrl":"10.1016/j.rsase.2025.101816","url":null,"abstract":"<div><div>Land use and land cover (LULC) changes induce unrivalled environmental changes at various spatial and temporal scales, affecting land surface processes in present catchment and it has experienced detrimental LULC changes over the past three decades, characterized by mismanagement of land and water resources. Hence, the current investigation aimed to analyze the spatial–temporal dynamics of LULC changes from 1990 to 2023 and to predict future trend scenario (1990–2053) for the Vanivilasa Sagara reservoir catchment, Karnataka state in India. The analysis was employed using integrated remote sensing and multitemporal geospatial data by utilizing Landsat – 5, Landsat – 7, IRS LISS – III, and Sentinel – 2A data, with aligning spatial resolutions, uncertainties, along with cumulative statistical model, and ground survey. First, a hybrid–image classification technique was employed to create LULC maps spanning 33 years (1990–2023) across seven reference periods: 1990, 1995, 2000, 2007, 2015, 2020, and 2023, and validated by accuracy assessment using meticulous field data. Subsequently, the magnitude, extent, trajectories, change rate, overall gains, losses, and net LULC changes were derived using statistical evaluation. The concerted results have indicated significant LULC alterations, including transforming forests, scrublands, and grasslands into agricultural and built-up regions such as human settlements, industries, and quarries/mining; cropping pattern conversions; marked rise in wasteland ecosystems such as erosional gullied land, salt-affected land, marshes, and land without scrubs. Moreover, unforeseen thrive in surface water bodies has been found. Key driving parameters and implications of LULC dynamics that hamper the environment of the study area were comprehensively studied and deciphered. Further, future LULC trend analysis obtained by Visual Basics Analysis (VBA) module, serves as an early warning system for resource degradation in catchments amidst climate change is addressed by this study, which contributes to regional planning between researchers, farmers, and government authorities at every level to achieve harmony.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"41 ","pages":"Article 101816"},"PeriodicalIF":4.5,"publicationDate":"2025-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790943","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号