<?xml version="1.0" encoding="utf-8"?>
    <rss version="2.0">
      <channel xmlns:content="http://purl.org/rss/1.0/modules/content/">
        <title>Frontiers in Remote Sensing | New and Recent Articles</title>
        <link>https://www.frontiersin.org/journals/remote-sensing</link>
        <description>RSS Feed for Frontiers in Remote Sensing | New and Recent Articles</description>
        <language>en-us</language>
        <generator>Frontiers Feed Generator,version:1</generator>
        <pubDate>2026-04-29T22:49:36.597+00:00</pubDate>
        <ttl>60</ttl>
        <item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1815646</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1815646</link>
        <title><![CDATA[The impacts of climate variability and agricultural expansion on ecosystem functions in Xinjiang drylands]]></title>
        <pubdate>2026-04-29T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Dechao Zhai</author><author>Yuhao Jiang</author><author>Zhifang Wang</author><author>Zhicheng Huang</author><author>Naijie Peng</author><author>Qunchao He</author><author>Wenjie Fan</author>
        <description><![CDATA[Climate variability and agricultural expansion are fundamentally reshaping ecosystem functions (EFs) in drylands. Xinjiang Region, a vast and typical dryland area in China, also faces dual pressures of natural and anthropogenic disturbances. Numerous local studies in Xinjiang have quantified individual EFs using biophysical models. However, research remains limited on the region-wide spatial distribution of long-term EF transitions, and on how nature-human interactions shape EF transitions across all of Xinjiang. In this paper, these gaps are addressed by using an EF classification framework to assess dominant EF patterns and reveal the ecosystem responses across Xinjiang, China from 2000 to 2024. A suite of multi-year EF maps with high accuracies (0.864–0.882) were produced using a two-level classification scheme with a random forest model, and Landsat imagery. EF maps showed that the water retention (WR) regions are concentrated in mountainous areas, the sandstorm prevention (SP) regions primarily occur in oasis–desert transition zones, and the agricultural production (AP) regions are dispersed in the desert (DS) periphery. Three major transition trajectories were identified: two climate-driven pathways, namely, greening (“DS–SP–WR”) and browning (“WR–SP–DS”) induced mainly by the water changes (R = 0.594 and −0.553 with annual precipitation, respectively), and a policy-driven cropland expansion pathway (“DS–AP”) resulting from aggressive agricultural expansion, particularly cotton cultivation (R = 0.831). This “DS–AP” functional transition in Xinjiang may potentially introduce multiple ecological challenges, including biodiversity loss and increased water demand. These findings reveal how climate variability and agricultural intensification reshape dryland landscapes, with implications for sustainability across water-limited ecosystems globally. Our results underscore the urgent need for adaptive management strategies that balance agricultural development with ecosystem resilience in Earth’s expanding drylands.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1799994</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1799994</link>
        <title><![CDATA[A probabilistic vision-language framework for habitat mapping within biodiversity projects using remote sensing data, ecological text and prior maps]]></title>
        <pubdate>2026-04-29T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Brianna Pickstone</author><author>Karen Anderson</author><author>Richard Delahay</author><author>Sareh Rowlands</author>
        <description><![CDATA[Accurate and time-sensitive spatial ecological information is essential for biodiversity policies and environmental planning, yet existing remote sensing (RS) classification workflows may struggle to integrate ecological semantics and prior domain knowledge, limiting their interpretability and performance for statutory habitat assessments such as Biodiversity Net Gain (BNG). In this study we explore the potential of Vision-Language Models (VLM) for UK Habitat (UKHab) classification, the standard framework for BNG assessments. Through integrating textual descriptors with spectral and structural RS data from Sentinel-2 and UK National lidar datasets, we address two methodological hypotheses: (1) how does the level of ecological detail in textual habitat descriptions influence VLM classification performance, and (2) can probabilistic priors derived from existing land cover products improve reclassification to UKHab categories within a VLM framework? Five model configurations were evaluated ranging from a vision-only approach to four text models, representing increasing semantic detail. Our results show that incorporating carefully designed text consistently improves classification performance, particularly for habitats with limited training data that typically underperform in conventional classification schemes. These habitats often correspond to higher biodiversity value and greater relevance for BNG delivery. For example, Wetland showed an improvement of 19.36% and Heathland and Shrub improved by 9.7% relative to the vision-only approach. The highest performing model, Text Level 3, included core ecological details and achieved the highest macro-averaged scores in precision (0.93), recall (0.87), F1 score (0.90), and overall accuracy (98.1%), representing an improvement of 5%, 3%, 4%, and 0.4%, respectively, compared to the vision-only approach. This study also tested whether the inclusion of probabilistic priors derived from an existing (but differently classified) land cover map (Living England map) improved VLM performance in classifying habitats to UKHab. While improvements were observed for certain classes, such as Heathland and Shrub (+4.54% in overall accuracy), priors produced inconsistent performance including a 16.13% decrease in Wetland accuracy due to errors within the prior map. Overall, this study demonstrates that VLMs, when paired with carefully designed textual inputs provide a promising tool for ecological mapping, offering advantages in low-data scenarios and supporting biodiversity assessments and conservation decision making.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1807693</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1807693</link>
        <title><![CDATA[Enhancing flood inundation mapping technique using Landsat-based remote sensing and 2D hydrodynamic modeling: a case study of the Bihar flood caused by the Koshi embankment breach]]></title>
        <pubdate>2026-04-29T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Aashutosh Aryal</author><author>Rajiv Sinha</author><author>Nawa Raj Pradhan</author><author>Venkataraman Lakshmi</author>
        <description><![CDATA[Floods have become more unpredictable and erratic due to the influence of extreme hydroclimatic events. Therefore, obtaining near-real-time, accurate flood inundation maps for such events is essential for effective flood emergency response, which can be achieved by leveraging remotely sensed data. This study integrated high-resolution remote sensing data to enhance flood inundation mapping in a data-scarce South Asian basin, the Koshi River basin. The study considered the 2008 Bihar flood event, caused by an embankment breach on the Koshi River at Kusaha (12 km upstream of the Kosi barrage) in Nepal, as a case study. An important feature of this event was that it occurred at a relatively low peak discharge of ∼4078 m3/s, far lower than the design capacity of ∼26,901 m3/s for the barrage downstream. The 1500-m wide breach in the embankment resulted in a 15–20 km wide and 150 km long sheet of water, creating an inundation zone of 2722 km2. This study used Landsat satellite surface reflectance data to map flood inundation using the commonly used water index known as the Modified Normalized Difference Water Index (MNDWI) to detect open surface water features. Further, the Normalized Difference Vegetation Index (NDVI), permanent water bodies, and Height Above the Nearest Drainage (HAND) datasets are used to mask the MNDWI-based flood inundation map, thereby improving its accuracy. In addition, different thresholding values are applied to the final masked MNDWI map to obtain more accurate maps. Furthermore, two-dimensional (2D) hydrodynamic modeling, using observed flow data near the breaching location, was set up to simulate the flooding event, provide a near-best estimate of the flood-inundated areas, and validate the accuracy of the remotely sensed inundation map using different binary classification evaluation metrics. The total flooded area within the 2D flow area defined in the model using the MNDWI threshold value of 0.2 is about 1,036 km2, whereas, for the threshold value of 0.3, the total flooded area is about 846 km2. The total flooded area obtained from the 2D modeling was about 1,162 km2. The overall accuracy of the flood inundation map using the 0.2 MNDWI threshold was 79%, whereas the accuracy with the 0.3 MNDWI threshold increased slightly to 81%. However, the precision and recall values for both thresholds were below 50%, as the number of correctly predicted flooded areas was much lower than that of correctly predicted non-flooded areas.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1812755</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1812755</link>
        <title><![CDATA[S3ViT: self-supervised spectral vision transformer framework for hyperspectral unmixing]]></title>
        <pubdate>2026-04-22T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Dario Scilla</author><author>Victor Angulo</author><author>Kasper Johansen</author><author>Naif Alsalem</author><author>Wolfgang Heidrich</author><author>Matthew F. McCabe</author>
        <description><![CDATA[Hyperspectral unmixing aims to decompose each pixel in a hyperspectral image into a set of constituent endmembers and their corresponding abundances. Recent deep learning based approaches have demonstrated strong performance in capturing both spectral and spatial features. However, obtaining reliable per-pixel abundance ground truth in real hyperspectral scenes is generally infeasible, which motivates unsupervised and self-supervised unmixing strategies. In this work, we propose S3ViT, a self-supervised Spectral Vision Transformer designed for pixel-wise hyperspectral unmixing. The transformer captures spectral and spatial dependencies by applying self-attention over the full sequence of pixel tokens (1×1) augmented with learnable positional embeddings. It operates without ground-truth annotations by generating pseudo labels through an unsupervised process: first, Singular Value Decomposition (SVD) is used to estimate the number of endmembers based on a thresholded singular value spectrum; then, k-means clustering provides cluster-derived priors that are used to form a contextual token and initialize spectral prototypes, without being treated as true abundance supervision. To guide training, two initialization tokens, one from Vertex Component Analysis (VCA) and one from the k-means cluster-derived priors, are embedded alongside patch tokens into the transformer. The model learns to estimate abundance maps while enforcing the Abundance Non-negativity and Sum-to-One constraints through ReLU and Softmax layers. Endmember spectra are later estimated from pixels with high predicted abundances. We evaluated S3ViT on the Samson, Jasper Ridge and Washington DC Mall benchmark datasets and compared it to state-of-the-art geometrical and deep learning methods. Our model achieves superior or comparable performance in both RMSE and SAD metrics, with up to 31% improvement in SAD and 25% in RMSE. These results indicate that a compact pixel-token ViT, guided by weak spectral priors and optimized via reconstruction losses, can achieve competitive unmixing performance on standard benchmarks.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1737953</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1737953</link>
        <title><![CDATA[Recording seagrass growth in Mosquito Lagoon, Florida after Hurricanes Ian and Nicole]]></title>
        <pubdate>2026-04-21T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Stephanie A. Insalaco</author><author>Hannah V. Herrero</author><author>Hailey F. Vickich</author><author>Dominic B. Mashak</author>
        <description><![CDATA[Hurricanes Ian and Nicole hit Mosquito Lagoon, Florida in the Fall of 2022 and since then, the ecosystem has greatly shifted. Prior to these storm events, seagrass in Mosquito Lagoon was almost non-existent due to poor ecosystem conditions but made a rapid recovery in 2023. To study this change, a Random Forest Classification was implemented using Harmonized Landsat Sentinel imagery semi-monthly from September 2022 to January 2024. A model was created for each date in the period, and it was evident that while seagrass was still in a significant decline until March 2023, it came back to pre-collapse levels in summer 2023 and beyond. This recovery could be linked with the hurricane events as they would have redistributed seagrass fragments throughout the lagoon to promote previously dormant growth and altered water conditions, but more research is necessary to determine why exactly seagrass has recovered to this extent. Seagrass density varied throughout each date, with the lowest observed being 0% density and peak being 20.32%, as well as the accuracy of each model (average 84%) depending on the amount of seagrass present. This method was successful in identifying seagrass in limited quantities, and this study raises awareness that constant seagrass monitoring is needed in Mosquito Lagoon to promote conservation of the ecosystem.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1768049</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1768049</link>
        <title><![CDATA[Long-term retrieval (1998-2023) of colored dissolved organic matter absorption coefficient and dissolved organic carbon in the Mackenzie River-Beaufort Sea region using the Copernicus GlobColour product]]></title>
        <pubdate>2026-04-17T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>María Sánchez-Urrea</author><author>Martí Galí</author><author>Marta Umbert</author><author>Carolina Gabarró</author><author>Eva De Andrés</author><author>Rafael Gonçalves-Araujo</author>
        <description><![CDATA[In the regime-shifting Arctic, organic carbon export from river watersheds is expected to rise due to changes in hydrological regimes and permafrost thawing, affecting coastal and shelf biogeochemistry. Ocean color remote sensing enables monitoring inaccessible areas like the Beaufort Sea, improving our knowledge of coastal dynamics and land-to-ocean transport of Chromophoric Dissolved Organic Matter (CDOM) and Dissolved Organic Carbon (DOC) during the ice-free season. While multi-sensor datasets provide consistent monitoring over extended periods, the viability of using merged L3 remote sensing reflectance (Rrs) to derive carbon-based regional products remains unexplored. This study uses the Copernicus-GlobColour multi-sensor merged Rrs to generate a long-term dataset of CDOM absorption at 443 nm, aCDOM(443), and DOC concentrations in the Mackenzie River–Beaufort Sea region. We employ a modified GIOP (Generalized Inherent Optical Properties) inversion algorithm to retrieve aCDOM(443); then, derive DOC using a region-specific aCDOM(443)-DOC relationship. Validation against in situ observations shows performance comparable to previous studies, with a MdAPD (r2) of 56% (0.55) for aCDOM(443) and 29% (0.70) for DOC. This novel dataset allows for a detailed analysis of riverine plume dynamics, interannual variability, and trends. Average summer plume extent ranges from 43,068 to 92,388 km2 and is significantly correlated with annual discharge (r = 0.6, p< 0.001). When compared to independent in situ DOC estimates (1999-2017) at the river mouth, our satellite product shows consistent variability patterns. Contrary to initial expectations, a significant decline of −0.017 m−1yr−1 and −3.40 μM yr−1 is observed over a 26-year period for aCDOM(443) and DOC, respectively, at the Mackenzie delta outflow, resulting from a slight increase (1998–2015) followed by a decrease (2016–2023). These results suggest that trends in DOC fluxes need to be assessed using longer time series while considering significant uncertainties in both retrieval accuracy and variable spatiotemporal coverage over multidecadal periods.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1812294</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1812294</link>
        <title><![CDATA[Assessment of seasonal variations in teal carbon of the palustrine wetland in the Grassland Biome of South Africa using remote sensing]]></title>
        <pubdate>2026-04-14T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Sisipho Ngebe</author><author>Laven Naidoo</author><author>Heidi van Deventer</author><author>Philemon Tsele</author><author>Mcebisi Qabaqaba</author>
        <description><![CDATA[Quantifying carbon stocks from the above-ground biomass (AGB) of wetland vegetation across seasons is crucial for assessing ecosystem resilience to anthropogenic and climate pressures. This study aimed to assess differences between summer and winter in aboveground carbon (AGC) of palustrine wetland vegetation using Sentinel-1 and Sentinel-2 data. The Random Forest (RF) and Support Vector Regression (SVR) were implemented with variable importance selection to develop an optimal model from remote sensing derived modelling scenario combinations. Modelling scenarios included field measured Leaf Area index and different combinations of (i) Sentinel-2 derived variables namely vegetation indices (VIs) and reflectance bands, and (ii) Sentinel-1 grey-level co-occurrence matrices, backscatter band ratios, and backscatter channels. Results indicated significant seasonal variation (p < 0.05) with higher total teal carbon in summer (155.1 g C/m2) than winter (115.8 g C/m2). Large macrophytes particularly Phragmites australis stored the highest carbon (93.04 g C/m2 in summer; 78.37 g C/m2 in winter). Sentinel-1-derived models outperformed Sentinel-2-based models for both seasons, achieving R2 of 0.7–0.8, RMSE of 39.9–69.6 g·m-2, and relative RMSE of 17.3%–21.3%. RF consistently performed better than SVR. Thus, seasonal monitoring of teal carbon provide valuable insights of wetlands vegetation contribution in carbon accounting and sequestration.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1770048</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1770048</link>
        <title><![CDATA[Satellite-guided, AI-enhanced framework for assessing UHI-driven cooling and heating loads]]></title>
        <pubdate>2026-04-13T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Seyed Mahmood Hamze-Ziabari</author><author>Michael Lehning</author>
        <description><![CDATA[Quantifying how urban heat islands (UHIs) influence regional electricity consumption remains challenging because station-based indicators and prescribed heating/cooling degree-day (HDD/CDD) thresholds often fail to capture intra-urban thermal heterogeneity and nonlinear demand responses during extremes. This study addresses these limitations by combining machine learning with MODIS thermal remote sensing to reconstruct gap-free daily land surface temperature (LST) fields across Local Climate Zones (LCZs) in the Swiss cantons of Vaud and Geneva (2015–2024) and by deriving demand-relevant temperature thresholds directly from observations. MODIS Terra and Aqua LST are integrated with in-situ meteorological observations, sub-pixel LCZ fractions, and urban morphological predictors to represent within-pixel thermal variability associated with urban form. Multiple machine-learning regressors are benchmarked for LST reconstruction, with XGBoost consistently achieving the highest performance across cantons, satellite overpasses, and independent field stations. Reconstructed LST fields are then linked to decade-long electricity demand records using an interpretable piecewise-linear hinge model, which infers empirical heating and cooling transition temperatures (∼19.5 °C and ∼30.3 °C) from observed demand–LST relationships rather than prescribing them. These thresholds enable LST-based degree-day metrics for mapping exceedances and estimating LCZ-resolved demand sensitivities. Heating sensitivity (demand–temperature slope) is relatively uniform across LCZs, whereas cooling sensitivity varies strongly with urban form—being highest in compact and open mid-rise LCZs and lower in sparsely built and industrial zones. Cooling sensitivity is pronounced on working days but negligible on weekends/holidays, indicating a predominantly non-residential cooling signal, consistent with limited residential air-conditioning adoption in this temperate region. Overall, the framework provides a transferable, spatially explicit basis for quantifying LCZ-specific links between surface-UHI exposure and electricity demand and for identifying urban forms most likely to amplify peak-load risk under intensifying heat extremes.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1779561</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1779561</link>
        <title><![CDATA[Development of an in-season nitrogen application dose estimation algorithm for cotton using multispectral imaging-based nitrogen adequacy index]]></title>
        <pubdate>2026-04-10T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>R. Raja</author><author>D. Kanjana</author><author>P. Nalayini</author><author>K. Rameash</author><author>G. Tamil Amutham</author><author>T. Arumuganathan</author><author>D. Blaise</author><author>Y. G. Prasad</author>
        <description><![CDATA[Managing within-field variability in cotton fields for precision nitrogen (N) management is difficult. The development of multispectral sensors and image data analytics offers a solution to addressing the issue. During the Kharif (monsoon) season of 2021 and 2022, field experiments were conducted at the ICAR-Central Institute for Cotton Research, Regional Station, Coimbatore, Tamil Nadu, involving seven N levels (25%, 50%, 75%, 100%, 125%, 150%, and 200% of the prescribed N-dose (80 kg ha−1)), along with a control (N0). Unmanned aerial systems-based multispectral crop imaging data were collected to develop an algorithm for calculating the N-dose for in-season variable-rate application. The multispectral images were processed for different crop growth stages, and treatment-wise mean normalized difference vegetation index (NDVI) and normalized difference red edge (NDRE) values were determined. The relationship between the mean NDRE index and estimated leaf N content exhibited statistically significant regression at 70 days and 95 days after emergence (DAE) during Kharif 2021 and 75 DAE and 90 DAE during Kharif 2022, respectively. Spatial nitrogen adequacy index (NAI) maps were created using the 95th percentile NDRE values (0.37 for Kharif 2021 and 0.27 for Kharif 2022). The N rate that maximized cotton yield was used to parameterize the NAI-based N-sufficiency response curve. The quadratic model (N rate (kg ha−1) = −191.25 (NAI)2 −22.32(NAI) + 138.5) produced the best fits, with a coefficient of determination (R2) value of 0.84. The developed algorithm can be used in preparing N zone maps for variable-rate N application and has the potential to optimize cotton’s in-season N requirement.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1679383</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1679383</link>
        <title><![CDATA[A multi-metric assessment of publicly available earth observation datasets reveals discrepancies in forest cover estimates in Paraguay, Zambia and Zimbabwe: higher resolution is not always a good indicator for accuracy]]></title>
        <pubdate>2026-04-07T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Natasha Gapare</author><author>Gopika Suresh</author><author>Dominik Sperlich</author>
        <description><![CDATA[Accurate forest monitoring is vital for climate mitigation, carbon credit schemes, and ecosystem management, particularly in tropical and subtropical regions facing rapid deforestation. Global Forest Watch (GFW) is widely used, yet has high commission errors; 45% in Sub-Saharan Africa, 17% in Latin America and 4% omission errors. This risks misleading policies and carbon accounting. To address this, we present a reproducible Google Earth Engine-Python workflow. It compares GFW with ESA WorldCover, Dynamic World, and Global-4 class PALSAR-2 across Paraguay, Zambia, and Zimbabwe. To ensure consistency, all products were standardised to forest/non-forest maps for pixel-based accuracy evaluation. Validation used 4.77 m Planet-NICFI mosaics, which provide high overall and F1 accuracies (87%–90%) and frequent temporal coverage. Their ability to capture seasonal clearing, and rapid regrowth offered a stronger reference. Visual interpretation of 500 random points further enhanced reliability over automated classification. These 500 points were overlaid on the forest/non-forest maps of each dataset to assess their agreement with the Planet-NICFI benchmark. Through this comparison, we derived multi-metric assessment results covering; overall accuracy, kappa coefficient, F1 scores, RMSE, and AUC. Dynamic World achieved the best performance in Paraguay, while GFW and Global-4 class PALSAR-2 performed better in Zambia and Zimbabwe. Importantly, the finer 10 m resolution of Dynamic World did not guarantee higher accuracy, underlining the need for region-specific assessments. Forest area calculations exposed further inconsistencies. In Paraguay, other datasets differ from GFW by approximately 3%–35%. In Zambia, deviations reach up to −47%. Zimbabwe shows the greatest divergence, with other datasets reporting 23%–70% less forest area than GFW. Comparisons with FAO statistics revealed additional discrepancies of −84% to +38%. These findings demonstrate that no single dataset can be assumed reliable across regions. Our framework provides a transparent, transferable approach that helps practitioners and policymakers select the most appropriate EO data for forest monitoring, carbon accounting, and environmental decision-making.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1704067</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1704067</link>
        <title><![CDATA[A novel building metric for PlanetScope© data optimization]]></title>
        <pubdate>2026-04-01T00:00:00Z</pubdate>
        <category>Methods</category>
        <author>Mona Morsy</author><author>Silas Michaelides</author><author>Peter Dietrich</author>
        <description><![CDATA[PlanetScope© data are being widely used by the scientific community in addressing important environmental issues, including cropland and tree cover loss, burned area mapping post-wildfire, imagery for river water masking, and snow-covered area mapping. However, building indices’ applications on PlanetScope data for use in separating blocks and to identify the difference between altered/unaltered constructed areas are not available, primarily due to the absence of the Short-Wave Infrared (SWIR) band. In this research, three types of remote sensing data series are employed, namely, PlanetScope images, Sentinel-2A data, and Google Earth Pro data, to separate the blocks and to capture the differences between damaged and undamaged blocks in the Khan Younis town of the Gaza Strip, utilizing a novel building metric purposely developed to be used with PlanetScope data. Additionally, three change detection methodologies are utilized to assess the efficacy of the new metric, namely, image differencing, Principal Component Analysis (PCA), and metrics, including the one proposed in this paper. The proposed metric exhibited high performance with the PlanetScope data compared to Sentinel-2A data. The results of the three methodologies indicate a strong correlation, with an R-value of 0.793,263 and a P-value of 0.0002456. The average sizes of the affected areas, derived from PlanetScope data from October 2023 to April 2024, are 0, 0.05, 2.4, 2.9, 5.2, 12.8, and 13.1 km2. The Sentinel data from October to January shows that the average sizes of the devastated regions are 0, 0.9, 2.1, and 7 km2, while the rest of the data series was concealed and unavailable by Google Earth Engine (GEE). The accuracy assessment test conducted with the Google Earth Pro scenes to evaluate the strength of the metric showed a considerably strong percentage of compatibility with PlanetScope data. The new proposed building metric yields results that are very close to the other techniques adopted, with slight variations but clearer visual outputs.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1765013</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1765013</link>
        <title><![CDATA[Benchmarking machine learning classifiers for urban mapping in arid environments: a google earth engine analysis of Riyadh’s expansion (1990–2025)]]></title>
        <pubdate>2026-03-24T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Amal Abdelsattar</author>
        <description><![CDATA[Monitoring urban expansion in arid regions is complicated by the spectral similarity between impervious surfaces and bare soil. Although machine learning classifiers on platforms like Google Earth Engine (GEE) offer effective solutions, their performance in these environments has not been systematically benchmarked. This study addresses this gap by comparing five supervised ML algorithms—Random Forest (RF), Support Vector Machine (SVM), Gradient Boosted Trees (GBT), Classification and Regression Tree (CART), and k-Nearest Neighbor (KNN)—for binary urban and non-urban mapping. We applied this analysis to Riyadh, Saudi Arabia, using a 35-year Landsat time series from 1990 to 2025. Annual, radiometrically consistent median composites were generated in GEE, with the 2025 composite based on imagery from January to September 2025. A custom ten-band feature stack, including indices such as the Bare Soil Index (BSI), was used for classification. The Random Forest model (RF-100) achieved the highest accuracy (Overall Accuracy = 0.977, Kappa = 0.954) and was selected for final land-use and land-cover mapping. Validation with 600 independent samples per epoch and comparison to the ESA WorldCover 2020 dataset confirmed the robustness of the results. The analysis found a 293% increase in Riyadh’s built-up area, from 416 km2 in 1990 to 1,219 km2 in 2025, with a notable slowdown in growth after 2010. Variable importance analysis showed that the Bare Soil Index (BSI) and the Normalized Difference Built-up Index (NDBI) were the most significant features for class separation, offering key methodological insights for arid urban remote sensing. This work provides a transferable methodological framework for classifier and feature selection in arid environments and a high-accuracy spatiotemporal dataset establishing a baseline for assessing sustainable urban development under Saudi Vision 2030.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1759371</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1759371</link>
        <title><![CDATA[Estimating flavonoids using radiative transfer model inversion on imaging spectroscopy data]]></title>
        <pubdate>2026-03-20T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Bongokuhle S’phesihle Sibiya</author><author>Moses Azong Cho</author><author>Onisimo Mutanga</author><author>John Odindi</author><author>Cecilia Masemola</author><author>Johannes Van Staden</author>
        <description><![CDATA[Assessing flavonoid content is crucial for understanding plant responses to biotic and abiotic stresses. However, pigments such as chlorophylls, carotenoids, and anthocyanins, absorb light between 410 and 430 nm, thus makes it difficult to quantify flavonoid content in this region. To improve the sensitivity of flavonoids have been developed, but they are largely empirical and lack transferability across environments. Therefore, this study compares the effectiveness of the PROSAIL-D radiative transfer model (RTM) and an empirical approach in mapping flavonoids using AVIRIS data in the Fynbos biome through indices. Results show that PROSAIL-D consistently outperformed the empirical method, with the spectral index FI417,693 achieving an R2 of 0.63 and RMSE of 0.24 (mg catechin (CAE)/g). The empirical approach with the same index yielded an R2 of 0.46 and RMSE of 5.11 mg CAE/g. Overall, the RTM model demonstrated superior accuracy, with RMSE values ranging from 0.24 to 1.70, compared to 5.11 to 6.28 for empirical methods. These findings highlight the potential of integrating AVIRIS data with PROSAIL for non-destructive, remote sensing-based assessment of flavonoids, offering a scalable approach for monitoring plant health and stress responses at the canopy level.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1786848</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1786848</link>
        <title><![CDATA[Study of the changes in the Curonian Lagoon shoreline strip (1995–2024), Lithuania]]></title>
        <pubdate>2026-03-19T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Sérgio Lousada</author><author>Dainora Jankauskienė</author><author>Giedrė Ivavičiūtė</author><author>Lina Kuklienė</author><author>Indrius Kuklys</author><author>Birutė Ruzgienė</author><author>Vivita Pukite</author>
        <description><![CDATA[Coastal lagoons are among the most vulnerable aquatic environments to climate change and human pressures; therefore, studying the Curonian Lagoon is crucial to support evidence-based management and improve understanding of shoreline responses to hydro-meteorological forcing and local land-use pressures. This research investigates the evolution of the Curonian Lagoon shoreline near Preila (Neringa municipality) from 1995 to 2024 using a multi-temporal orthophotographic series (ORT10LT) complemented by a very high-resolution UAV orthomosaic produced in 2024. Shoreline position was consistently delineated and compared across eight observation periods to quantify section-based displacement and hotspot area changes. The analysis reveals a spatially organized pattern, with a persistent accumulation-prone stretch in the mid-profile (250–350 m) and a persistent erosion hotspot toward the latter shoreline (400–566 m). The maximum shoreline retreat reached 32.80 m (425 m section, 2024 relative to 1995–1999), while the maximum shoreline advance reached 18.22 m (275 m section, 2021–2023). Area-based hotspot metrics indicate erosion losses up to 654 m2 (2018–2020) and accumulation gains up to 395 m2 (2009–2010) relative to the baseline. These results provide a reproducible, decision-oriented shoreline-change characterization that supports targeted monitoring and management of this culturally and environmentally significant lagoon margin.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1710758</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1710758</link>
        <title><![CDATA[Case 2 Regional Coast Colour: a neural network-based framework for atmospheric correction and in-water retrievals across multiple ocean colour satellite sensors]]></title>
        <pubdate>2026-03-18T00:00:00Z</pubdate>
        <category>Methods</category>
        <author>Dagmar Müller</author><author>Martin Hieronymi</author><author>Ana B. Ruescas</author><author>Marco Peters</author><author>Rüdiger Röttgers</author><author>Marcel König</author><author>Carole Lebreton</author><author>Kerstin Stelzer</author><author>Carsten Brockmann</author><author>Roland Doerffer</author>
        <description><![CDATA[Since the 1990s, Doerffer and Schiller have been developing physics-based neural network algorithms for analyzing ocean colour in satellite imagery of optically complex coastal waters. At its core, the approach uses neural networks to solve the inversions in various aspects of solar radiative transfer in both the atmosphere and water, including atmospheric correction, towards the estimation of inherent optical properties (IOPs) of the water constituents. Empirical bio-optical models are then applied to derive constituent concentrations from these IOPs. Over the years, this algorithm has evolved significantly and is now widely recognized as Case-2 Regional CoastColour (C2RCC), a trusted tool within the ocean colour research community. Originally designed for the MERIS sensor aboard ENVISAT, C2RCC is now the operational ground segment processor for generating Case-2 (complex) water products from Sentinel-3 OLCI data and from Sentinel-2 MSI data in the Copernicus Marine High Resolution Ocean Colour Service. Adaptations of the algorithm have also been developed for other satellite missions, including SeaWiFS, MODIS, VIIRS, Landsat OLI, and Sentinel-2 MSI. The C2RCC processor is freely accessible through the Sentinel Application Platform (SNAP). This article provides an overview of the background and evolution of the C2RCC algorithm, presenting validation results at coastal sites and in land waters alongside user performance evaluations analyzing the influence of system vicarious calibration gains. It highlights cases where the algorithm delivers reliable results as well as its limitations and areas for future improvement. In its current iteration for Sentinel-3 OLCI, C2RCC performs effectively, particularly in moderately absorbing or scattering Case-2 waters.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1711426</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1711426</link>
        <title><![CDATA[Spectral assessment of nutrient limitation in the savanna landscape: selection of spectral indices towards Sentinel-2 upscaling]]></title>
        <pubdate>2026-03-17T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Nasiphi Ngcoliso</author><author>Abel Ramoelo</author><author>Philemon Tsele</author><author>Mcebisi Qabaqaba</author><author>Siyamthanda Gxokwe</author>
        <description><![CDATA[Nutrient limitations can significantly impact the ecosystem services provided by the savanna biome, potentially leading to degradation and reduced grazing capacity if not detected in time. A key indicator of growth-limiting nutrients is the Nitrogen to Phosphorus (N:P) ratio. However, grass foliar phosphorus content has rarely been studied in African savannas, especially using remote sensing approaches. As a result, there is limited information on the spatial distribution of nutrient limitations in these ecosystems. This study aimed to develop a Sentinel-2-based machine learning regression model to predict and map the distribution of the N:P ratio in the northern region of Kruger National Park (KNP), South Africa, which is dominated by the savanna rangeland biome. Fieldwork was conducted between 15 March and 30 April 2008 to collect grass samples and spectral data using an Analytical Spectral Device (ASD). The hyperspectral field data were then resampled to match the multispectral configuration of Sentinel-2 imagery. A Random Forest Regression (RFR) technique was applied to the simulated Sentinel-2 datasets to develop predictive models of the N:P ratio. Model accuracy was evaluated using the Root Mean Square Error (RMSE) Relative Root Mean Square Error (RRMSE), Percent Bias (PBIAS), and the coefficient of determination (R2). The results showed that vegetation indices (VIs), particularly the Normalized Difference Red Edge (NDRE) derived from Sentinel-2 bands B8 and B5, was optimal for estimating N:P ratio. This index explained over 80% of the N:P variability, with the lowest PBIAS of 0.02%. The best-performing model was used to map nutrient limitations across the study area using Sentinel-2 imagery. The spatial analysis indicated consistent nitrogen limitation and co-limitation across the investigated regions, with no evidence of phosphorus limitation. The high-accuracy models demonstrate the effectiveness of Sentinel-2 imagery for estimating nutrient limitations in heterogeneous savanna landscapes. This study offers a cost-effective, scalable tool for decision-makers involved in the management, sustainability, and restoration of the savanna biome. Future research should consider incorporating textural and environmental variables to enhance model performance and understanding of nutrient dynamics.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1751006</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1751006</link>
        <title><![CDATA[A fully satellite-driven workflow for hydrodynamic modeling in data-scarce coastal systems: integrating ICESat-2, Sentinel-2, SWOT and reanalysis models]]></title>
        <pubdate>2026-03-11T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Ali Reza Payandeh</author><author>Marc Simard</author><author>Daniel Jensen</author><author>Anthony Daniel Campbell</author><author>Heidi van Deventer</author><author>Alexandra Christensen</author>
        <description><![CDATA[Hydrodynamic models in coastal and estuarine systems are typically constrained by sparse bathymetry, boundary, and validation data, especially in regions where field campaigns are costly or impractical. Here we develop and test a fully satellite-driven framework for hydrodynamic modeling in South Africa’s Langebaan Lagoon without using any local in situ measurements. Bathymetry is derived by training multispectral Sentinel-2 reflectance against ICESat-2 ATL24 photon-derived depths using an XGBoost model optimized with Bayesian search. The final satellite derived bathymetry reproduces independent ATL24 points with RMSE = 0.45 m and R2 = 0.97. This bathymetry was used in a depth-averaged Delft3D Flexible Mesh model driven at the open boundary by TPXO tidal harmonics and by ERA5 winds. We validate modeled water surface elevation against 16 SWOT low-rate (250 m, unsmoothed) passes in 2023. SWOT–model comparisons yield an overall RMSE of 0.11 m and R2 = 0.61, with typical point differences <0.10 m (∼5% of the 2 m tidal range), and showed consistent spatial gradients in water level from the offshore boundary, through Saldanha Bay, and into the lagoon. At the offshore boundary, TPXO and SWOT sea surface heights agree closely (R2 = 0.86). A ∼26 min phase lag, determined using a lag-correlation analysis, reduces the TPXO–SWOT RMSE from 0.18 m to 0.11 m, indicating that phase differences explain some of the mismatch, with remaining differences likely linked to non-tidal signals. Our results demonstrate that combining passive optical, photon-counting LiDAR, radar interferometry, and global tidal/atmospheric models enables robust, transferrable hydrodynamic modeling in data-scarce coastal systems, offering a cost-effective pathway for monitoring.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1659305</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1659305</link>
        <title><![CDATA[Mapping small-sized logging disturbances in tropical forests using Sentinel-1 time series and an extensive ground truth dataset]]></title>
        <pubdate>2026-03-11T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Audrey Mercier</author><author>Julie Betbeder</author><author>Frédéric Mortier</author><author>Nicolas Barbier</author><author>Pierre Ploton</author><author>Guillaume Cornu</author><author>Pierre Couteron</author>
        <description><![CDATA[Deforestation and forest degradation are the main threats to biodiversity and carbon stocks in tropical forests. Advances in optical and SAR satellite sensors have enabled the development of real-time monitoring of deforestation on a global scale. SAR is particularly appealing in tropical areas due to its insensitivity to cloud cover. However, the automatic detection of small disturbed areas (such as individual tree felling gaps) remains a major challenge. Thanks to a unique dataset consisting of 23,759 locations of individual tree felling gaps and multi-date drone lidar acquisitions, we evaluated the potential of Sentinel-1 dense time series for monitoring small-sized forest disturbances substantially smaller than 0.1 ha on both FSC-certified and artisanal logging sites in the Congo Basin. We designed a new method for forest monitoring using the fused-lasso technique optimized to detect abrupt changes of at least 0.02 ha in Sentinel-1 time series using the fused-lasso technique (Fused-Lasso Change Detection, FLCD). We assessed our new method along with the Cumulative Sum (CuSum) that also proved promising for detecting small impacts, referring for the first time to precise disturbance dates over large areas. Both approaches reached similar rates of confirmed felling gaps that were similarly increasing with gap size, and similar rates of unconfirmed detected gaps. The FLCD method estimates the dates of tree felling more accurately in FSC-certified areas (−2 days difference for FLCD and −19 for CuSum on average). The effective resolution of the S-1 images limits detection for the smallest gaps, yet the approach can help detect and monitor degradation fronts. Fused lasso regression is relevant for modeling the temporal trajectories of the radar signal, which will allow taking advantage of both the increasing availability of UAV-borne data and the lengthening of the S-1 image series.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1804992</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1804992</link>
        <title><![CDATA[Editorial: Multibeam echosounder backscatter: advances and applications]]></title>
        <pubdate>2026-03-05T00:00:00Z</pubdate>
        <category>Editorial</category>
        <author>Craig J. Brown</author><author>Marc Roche</author><author>Vanessa Lucieer</author><author>Alexandre C. G. Schimel</author>
        <description></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/frsen.2026.1774149</guid>
        <link>https://www.frontiersin.org/articles/10.3389/frsen.2026.1774149</link>
        <title><![CDATA[Low-cost mobile laser scanning for urban tree assessment: accuracy evaluation and application potential]]></title>
        <pubdate>2026-03-04T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Jozef Výbošťok</author><author>Juliána Chudá</author><author>Daniel Tomčík</author><author>Michal Skladan</author><author>Arunima Singh</author><author>Roman Kadlečík</author><author>František Chudý</author><author>Daniel Kükenbrink</author><author>Martin Mokroš</author><author>Janusz Bedkowski</author>
        <description><![CDATA[Recent advances in mobile laser scanning (MLS) have enabled rapid three-dimensional data acquisition for urban tree monitoring, providing an alternative to traditional terrestrial laser scanning (TLS) and photogrammetric approaches. However, the high cost of commercial handheld mobile laser scanning (HMLS) systems limits their routine use in urban green-space inventories. This study evaluates the performance of a low-cost wearable MLS prototype based on a Livox MID-360 sensor and compares it with two commercial HMLS systems (Stonex X120GO and Stonex X200GO) for urban tree assessment. The analysis was conducted in an urban park environment and included 80 individual trees. Tree detection rate (TDR), diameter at breast height (DBH), tree height (TH), crown projection, and point cloud quality were evaluated using commonly applied processing workflows (RayCloud with ITSMe, FSCT, and 3DFin). Using the best-performing workflow, the prototype achieved a DBH RMSE of 2.47 cm and a TH RMSE of 0.43 m, compared to 1.25–2.08 cm (DBH RMSE) and 0.31–0.40 m (TH RMSE) for the commercial systems. Mean cross-section quality metrics further supported data reliability, with Cross Section Quality Index (CSQI) values of 0.78 for the prototype and up to 0.83 for the high-end system, and corresponding Standard Deviation of Radial Distances (SDRD) values of 0.034 m and 0.018 m, respectively. Despite lower point density and increased noise, the low-cost wearable MLS prototype provided comparable TDR, DBH, and TH estimates. Differences in processing time were mainly driven by the selected workflow rather than by the scanning device. Overall, the results demonstrate that low-cost wearable MLS systems can deliver reliable urban tree metrics when combined with suitable processing methods, offering a cost-effective alternative for urban tree inventories and operational monitoring.]]></description>
      </item>
      </channel>
    </rss>