Search Results (3,676)

Accurately identifying dominant forest tree species from remote sensing supports biodiversity assessment, carbon stock estimation, and sustainable forest management. Sentinel-2 satellite time series capture phenological dynamics and canopy spectral variation, providing valuable information for regional tree species classification. However, interannual shifts in phenological timing can reduce the temporal correspondence among repeated annual cycles. This temporal misalignment complicates the consistent representation of multiyear observations. To address this problem, this study proposes the Hierarchical Season-Year Representation Alignment Network (HSRA-Net). Using multiyear Sentinel-2 NDVI time series, HSRA-Net builds robust temporal representations, aligns comparable seasonal phases across years, and adaptively integrates informative temporal features. The model was evaluated on dominant tree species in representative temperate and subtropical forests in China using ground reference observations and four temporal scales: intra-growing, annual, biennial, and triennial. It was compared with standard models and representative recent spatiotemporal methods. Among the four temporal scales, HSRA-Net achieved its highest classification accuracy in the triennial scale in both study areas. HSRA-Net achieved OA of 88.90% and 83.17% in the temperate and subtropical regions, respectively. Compared with the intra-growing season, the triennial scale increased OA by 3.72% and 14.31%, respectively. These findings demonstrate that organizing comparable phenological information across repeated annual cycles can improve dominant tree species classification and highlight the effectiveness of HSRA-Net for multiyear temporal representation. Full article

Floods in monsoon-dominated river basins exhibit high spatio-temporal variability, necessitating high-resolution, multi-sensor approaches for reliable monitoring and impact assessment. In flood-prone agricultural regions, continuous monitoring using optical remote sensing is frequently hindered by dense monsoon cloud cover. This study establishes a comprehensive multi-sensor framework within the Google Earth Engine (GEE) to examine the spatio-temporal dynamics and land surface impacts of the 2025 monsoon floods in Punjab, Pakistan. Flood inundation mapping was executed using a 12-day Sentinel-1 Synthetic Aperture Radar (SAR) time series via a dual-threshold change detection methodology. Concurrently, Sentinel-2 imagery facilitated the derivation of land use/land cover (LULC) changes and vegetation dynamics using a Random Forest classifier, achieving overall accuracy of 93% (pre-flood), 91% (during flood), and 94% (post-flood). These accuracy levels were consistent across all three phases despite spectral confusion between water, saturated soil, and vegetation during peak inundation, indicating consistent classification performance under monsoon conditions. The analysis revealed a distinct bimodal flooding regime, characterized by an early monsoon peak in July–August and a more severe late monsoon peak in August-September. The cumulative maximum flood extent reached 9495.33 km², with peak single-date inundation reaching 5449 km². Mapped cropland declined by 6.7% (8181 km²) during peak flooding, with 3.9% (4796 km²) remaining non-cropland by the end of the observation period; 5892 km² of pre-flood cropland was identified as inundated through spatial intersection. In addition, the Normalized Difference Vegetation Index (NDVI) declined by 28.6%, from 0.28 to 0.20, indicating a substantial reduction in vegetation greenness. Spatial consistency was checked with the United Nations Satellite Centre (UNOSAT) and the Food and Agriculture Organization (FAO), independently collected data showing moderate spatial agreement. The proposed framework is highly scalable for continuous flood monitoring, offering critical insights for disaster management and climate adaptation planning in monsoon regions plagued by data scarcity and persistent cloudiness. The approach is particularly relevant for near-real-time operational monitoring, given its reliance on freely available Sentinel data and cloud-based processing that requires no specialized ground infrastructure. Full article

A comprehensive and publicly available geospatial database of PV installations is crucial for effective policymaking and infrastructure planning towards Thailand’s carbon neutrality and net-zero greenhouse gas emissions plan. This study introduces a national-scale mapping framework for ground-mounted photovoltaic power stations using geospatial artificial intelligence (GeoAI). A pretrained deep learning model is applied to Sentinel-2 satellite imagery to delineate solar photovoltaic sites. The outputs are cross-checked and verified with official statistical data. Geospatial analyses and regression-based comparisons are conducted to validate the infrastructure inventory. The results indicate that all registered solar PV sites were identified, with a total mapped area of 14.55 km² and a capacity of 506.4 megawatts. Furthermore, the research clarifies that without official administrative ownership data, GeoAI cannot be applied more broadly for policy evaluation. Closing these data infrastructure gaps is necessary to support more effective energy planning and monitoring in Thailand. Full article

Multi-temporal satellite data were used to examine changes in aufeis area and spatial configuration in the Chuluut River valley, Mongolia, and to compare seasonal vegetation dynamics among sites with different return frequencies of aufeis. Annual aufeis masks were derived from Landsat imagery for 1986–2025, while vegetation phenology was assessed using Harmonized Landsat–Sentinel-2 (HLS) data for 2016–2025. Mean aufeis area was 11.98 km² and declined significantly by approximately 0.084 km² per year, accompanied by spatial reorganization. The strongest climatic association was found with April–June precipitation in the preceding year. Hydrologically, aufeis acts as a temporary seasonal water store, retaining part of winter discharge and releasing meltwater during spring. Its decline and redistribution may alter the timing and pathways of seasonal water release, reduce delayed moisture inputs to floodplain surfaces, and modify water availability for riparian and meadow ecosystems. In meadows, the start of season (SOS) occurred later under frequent than infrequent aufeis recurrence. Peak and mean summer normalized difference vegetation index (NDVI) values remained high after aufeis melt-out. The highest integrated seasonal NDVI values occurred under intermittent aufeis recurrence. Full article

Mine reclamation is a process in which a mine pit is back-filled with overburden, covered with topsoil, and revegetated. It is essential to monitor barren areas that require revegetation for long-term reclamation monitoring. The study introduces a Modified Barren Index (MBaI) to classify barren areas in a coal mining region. The MbaI is a computationally efficient and effective approach for differentiating barren areas from active mines, overburden dumps, built-up areas and vegetation to support mine reclamation and monitoring. The new proposed index utilizes near- and shortwave infrared to distinguish between bare soil and other surfaces. Using Landsat 8 and Sentinel 2, the study was carried out in the Jharia Coal Field region in India and compared with commonly used indices, i.e., the Biophysical Composition Index, Modified Bare Soil Index and Normalized Difference Bare Soil Index. The comparison between actual reflectance data and laboratory ECOSTRESS data attests that MbaI can effectively differentiate between barren and non-barren areas while other indices struggled. The index was tested in coastal, snow and desert regions for assessing efficiency, with accuracies of 98%, 97% and 91% using Landsat 8 and 94%, 94% and 96% using Sentinel 2 data, respectively. The extracted barren areas using MbaI exhibited lower NDVI and NDMI values compared to areas extracted from other indices, suggesting better efficiency. The study can be useful for achieving faster and accurate classification of barren areas from non-barren areas in mining and non-mining regions in the Indian subcontinent using multi-satellite data. However, wet soil can limit the accuracy of barren area extraction by MbaI, which can become a major limitation with coal mining regions experiencing regular rainfall. Full article

This review synthesizes recent regional applications of satellite, unmanned aerial vehicle (UAV), and hyperspectral/spectroradiometric remote sensing for vegetation cover discrimination, including crop and natural vegetation discrimination, plant disease and pest detection, and yield assessment. The reviewed evidence demonstrates complementary rather than universally superior capabilities among sensing platforms. Satellite observations provide repeated large-area monitoring but remain constrained by spatial resolution, cloud interference, and spectral mixing, whereas UAVs offer very-high-resolution and flexible field-scale observations at the expense of spatial coverage and greater acquisition and processing requirements. Hyperspectral and spectroradiometric approaches provide detailed spectral information for distinguishing subtle vegetation differences, but are limited by data complexity and operational scalability. The quantitative results reported in the reviewed studies illustrate this variability: satellite-based crop discrimination achieved approximately 90% overall accuracy with QuickBird and 81% overall accuracy (κ = 0.74) with Sentinel-2 at a 10 m resolution, while a UAV hyperspectral vegetation classification study achieved 94.5% accuracy. However, these values are not directly comparable because the vegetation targets, sensors, acquisition conditions, and analytical methods differed among studies. Recent evidence also indicates that the phenological timing, spectral band selection, spatial resolution, and representative training data strongly influence the discrimination performance, while the transfer of disease detection models from controlled experiments to operational field conditions remains a major challenge. By integrating evidence from satellite, UAV, and ground-based spectroradiometric approaches, this review provides a comprehensive framework for understanding the complementary capabilities of these technologies for vegetation cover discrimination and highlights their importance for improving vegetation monitoring, precision agriculture, and sustainable ecosystem management. Full article

Remote sensing-based mapping of crop planting structures in irrigation districts plays a vital role in forecasting regional production and optimizing water resource allocation. However, optical satellite imagery is limited by insufficient spatial resolution when applied to fragmented farmland landscapes, while unmanned aerial vehicle (UAV) imagery is limited by spatial coverage and high data processing costs. To address these bottlenecks, this study proposed a cross-scale collaborative extraction framework utilizing the CycleGAN network and transfer learning to map summer harvest crops (winter wheat and rapeseed) in the Baojixia Irrigation District for the year 2023. First, multiple semantic segmentation models—including U-Net, DeepLabv3+, SegFormer, and HRNet—were evaluated on a joint satellite–UAV dataset, with U-Net selected as the optimal backbone. Next, CycleGAN was introduced to perform style translation from the UAV domain to the satellite domain. This step generated high-fidelity, satellite-like images that preserve UAV-derived high-resolution spatial details, which were subsequently used to pre-train the U-Net backbone, significantly reducing the labor of manual annotation. Finally, the model was fine-tuned with real satellite images to achieve precise crop extraction. Results indicated that this framework accelerates model convergence and improves segmentation accuracy. The proposed method achieved an mIoU of 85.09%, an mPA (Recall) of 91.63%, a Precision of 91.97%, an Accuracy of 93.57%, and an F1-Score of 91.80%, outperforming the baseline U-Net model by 2.98%, 2.00%, 1.66%, 1.52%, and 1.83%, respectively. By successfully transferring high-resolution prior knowledge into the satellite feature space, this study provides a cost-effective and highly accurate solution for crop identification in complex agricultural landscapes, breaking the spatial limitations of UAV remote sensing. Full article

Wheat grain protein content (GPC), quantified as the mass proportion of protein relative to the total dry grain weight, is a key indicator for wheat quality evaluation. Based on space–air–ground collaborative observations, this study screened GPC-sensitive spectral parameters and multi-stage and multi-scale remote-sensing GPC estimation models. Results showed that the optimal feature set derived from unmanned aerial vehicle (UAV) imagery consisted of four vegetation indices (VIs) and three optimized texture indices (TIs). The Random Forest-based Soil–Plant Analysis Development (SPAD) model for field plots achieved the highest accuracy at the anthesis stage, with a coefficient of determination (R²) of 0.91 and a root mean square error (RMSE) of 1.77. Cross-year validation gave a correlation coefficient of 0.80 and RMSE of 3.83. Data analysis indicated that GPC of mature wheat had an extremely significant correlation with SPAD values across various growth stages, higher than leaf area index (LAI). We thus established a quantitative GPC estimation framework with UAV-derived SPAD as an intermediate variable, which kept stable R² (0.60) and RMSE (1.1%) except during grain-filling. On this basis, UAV-retrieved GPC data were scaled up to 10 m. Eight optimal parameters were selected from Sentinel-2 satellite data, and the regional-scale GPC estimation model was developed via the partial least squares regression (PLSR) algorithm. Compared with the model established directly using ground measured data without scale conversion, the scale-up model increased R² by 8.9% and reduced RMSE by 40%, improving the model inversion accuracy. The proposed regional-scale GPC remote-sensing estimation model effectively solves the scale mismatch between ground observation and satellite data, and provides technical support for field and regional wheat quality remote-sensing estimation. Full article

Reliable orbit estimation from position observations depends on both nominal propagation and the specification of process and measurement uncertainty. An offline covariance calibration framework is presented for a TLE-driven differentiable SGP4 extended Kalman filter (dSGP4–EKF). dSGP4 supplies the fixed nominal trajectory, while the sensitivities to six TLE initialization elements define a local Cartesian error transition. Error state re-baselining maintains consistent coordinates across catalog changes. Each condition-specific checkpoint contains six positive scalars: three continuous white noise acceleration scales defining the process covariance Q, and three position noise scales defining the measurement covariance R. One-step-ahead innovation negative log-likelihood calibrates these parameters without ground-truth state supervision. The calibrated parameters are frozen during deployment, which requires no neural network inference. Across three Sentinel satellites, three synthetic Gaussian noise levels, and ten paired realizations, the proposed method achieved macro-average component-wise position and velocity RMSEs of 20.64 m and 0.0463 m s⁻¹, respectively. The corresponding values for a supervised observation-conditioned GRU-Q/R point estimator were 20.42 m and 0.0462 m s⁻¹. The proposed method used six trainable parameters per checkpoint, compared with 8886 for GRU-Q/R, and had a median Day-7 filtering time of 13.37 ms versus 40.35 ms. Re-baselining reduced switch-local errors, while the tested initialization, training seed, and window settings produced limited variation for the proposed method. No numerical divergence was observed during the examined 24-h S1A–25 m runs. The learned scales are effective uncertainty parameters rather than uniquely identifiable physical disturbances. These findings are limited to the controlled synthetic observation protocol. Full article

Center pivot irrigation systems (CPISs) are modern irrigation technologies that are widely installed and are strong indicators of intensified industrial agriculture and large-scale investment in agriculture. Currently, these systems are being increasingly deployed across Africa’s hyper-arid regions, where scarce surface water and minimal rainfall force heavy reliance on groundwater extraction. However, despite the substantial groundwater demands of such large-scale irrigation, detailed annual data on the spatial extent of CPIS-equipped cropland remains unavailable, hindering the understanding of the spatiotemporal development of CPISs in these extreme regions. In this study, we mapped CPISs from both Landsat and Sentinel-2 archived satellite imagery in Africa’s hyper-arid regions annually from 1972 to 2025. Using a Cascade Mask R-CNN instance segmentation model with a Swin Transformer backbone, we achieved an average precision of 83.62%. Our mapping results reveal that over the past five decades, the total area of irrigated cropland equipped with CPISs in Africa’s hyper-arid regions expanded from 1110 ha in 1972 to 803,038 ha in 2025. The rate of expansion has not been uniform, showing a gradually increasing trend prior to 2008, followed by increasing rates that have continued to accelerate since then. As agriculture is already the largest consumer of water in Africa, developing large-scale industrial agriculture in areas with limited freshwater resources could bring the risk of unsustainable development in the future. The results of our study highlight the potential risks associated with the fast-increasing investments in large-scale irrigation projects within the hyper-arid regions of Africa, particularly in the absence of demonstrable long-term viability. Full article

The 2021 eruption of Cumbre Vieja on La Palma, Canary Islands, produced extensive lava flows and major landscape transformation, providing an opportunity to evaluate remote sensing approaches for mapping the extent of an emplaced lava flow. This study assessed direct spectral, classic Rao’s quadratic entropy (RaoQ), and multidimensional RaoQ approaches using satellite observations acquired before and after the eruption. Optical, radar, thermal infrared, and night-time radiance datasets were evaluated within a common change-detection framework implemented in Google Earth Engine. Difference maps were converted into binary change maps using a histogram-based thresholding procedure calibrated on the reference delineation and evaluated against the Copernicus Emergency Management Service (CEMS) lava-flow reference and no-change validation areas derived from ESA WorldCover using multiple accuracy metrics. Because the change reference is the final CEMS lava-flow delineation and the no-change samples lie outside a 100 m buffer around it, the accuracy figures reported here quantify the mapping of lava-flow extent and not of other eruption-related effects such as ash deposition or vegetation damage beyond the flow margins. Among the direct spectral approaches, the NHI_SWIR index achieved the highest overall classification performance. Among the individual Sentinel-2 bands, B12 achieved the highest overall accuracy, whereas B8A achieved the highest true skill statistic; both exceeded the multidimensional RaoQ configurations in mean prevalence-independent discrimination. Within the classic RaoQ approach, MIRBI produced the strongest single-variable heterogeneity-based results. The best multidimensional configurations combined Sentinel-2 B8A and B12 with Sentinel-1 VV, demonstrating that radar backscatter provided complementary information to optical observations. Although multidimensional RaoQ did not surpass the best direct spectral variables, it produced competitive and spatially coherent representations of lava-flow disturbance. The evaluated thermal infrared and night-time radiance products did not provide competitive discrimination under the selected spatial and temporal conditions for different reasons: a thresholding limitation in the case of the Landsat thermal product, and an unfavourable ratio of pixel size to flow width in the case of the night-time radiance products, while the MODIS product returned no valid validation points and could not be evaluated. These product-specific explanations rest on a small number of comparisons and are provisional. These results show that carefully selected Sentinel-2 SWIR variables remain the strongest benchmark for detailed mapping of fresh lava-flow disturbance, while multidimensional RaoQ provides a framework for optical–radar integration that requires no training data or prior classification. Because the evaluation covers a single eruption in a single landscape, transfer of the framework to other events and settings remains to be demonstrated. Full article

Remotely sensed surface temperature data are widely used by researchers and municipal practitioners to characterize urban thermal patterns and assess the impacts of increasing urban heat. While satellite-derived land surface temperature (LST) data effectively identify broad-scale patterns, their 30 m spatial resolution limits their ability to resolve fine-scale thermal variability. Downscaling approaches have been developed to generate higher-resolution LST products; however, in the absence of independent, high-resolution surface temperature observations, it remains difficult to determine whether the additional spatial variation introduced by downscaling reflects meaningful landscape-related thermal structure or model-generated variability. Our objective was to evaluate whether the additional spatial variation introduced by downscaling LST from 30 to 10 m was statistically consistent with independently derived land cover composition and landscape compositional heterogeneity. Using an open-source random forest model, we downscaled Landsat 8/9 LST to 10 m resolution using Sentinel-2-derived predictors over a summer season in Kelowna, BC, Canada. We quantified the additional thermal variation introduced through downscaling and used Generalized Additive Models (GAMs) to relate this variation to high-resolution land cover characteristics summarized within 100 m grid cells. Land cover composition and landscape compositional heterogeneity together explained 65.5% of the additional variation introduced by downscaling. Added thermal variation was most strongly associated with canopy and impervious cover, two landscape characteristics consistently identified as major controls on urban land surface temperature. These findings suggest that the additional spatial variation introduced through downscaling is consistent with independently derived landscape structure known to influence urban thermal patterns. Rather than providing direct validation of downscaled temperatures, this analysis offers an indirect means of evaluating whether downscaled LST produces interpretable, landscape-consistent thermal variability that may be useful for urban analysis and planning. Full article

Satellite active-fire products may miss fires that occupy only a small fraction of a pixel and produce limited absolute thermal responses. In this study, weak thermal anomalies are operationally defined as confirmed fire-affected VIIRS pixels with relatively low absolute $B{T}_4$ and $\mathsf{\Delta }BT$ responses but positive deviations from their recent temporal and local spatial backgrounds. STAR-FD (Stratified Spatiotemporal Adaptive Residual Fire Detection) was developed to detect such anomalies from VIIRS 375 m observations. The method reconstructs a 14-day recent thermal background and jointly evaluates single-pixel temporal residuals and spatial-contrast temporal residuals. Stable non-fire samples are stratified by day/night condition, season, macroclimate zone, land cover, and view zenith angle, and residual thresholds are estimated within each stratum. Validation was conducted in Heilongjiang, northwestern India, and California using Sentinel-2 MSI, Landsat 8/9 OLI, and available external fire information. STAR-FD detected 927 overpass-level thermal-anomaly events, of which 855 were confirmed as Fire, corresponding to a confirmation rate of 92.23%. VNP14IMG detected 343 events, of which 333 were confirmed as Fire (97.08%). STAR-FD identified 522 more confirmed fire-related thermal-anomaly events than VNP14IMG, corresponding to a 156.76% increase, while retaining 88.05% of VNP14IMG events. In northwestern India and Heilongjiang, STAR-FD-added daytime confirmed fire pixels had median $B{T}_4$ values 9.49 K and 11.88 K lower, respectively, and median $\mathsf{\Delta }BT$ values 8.10 K and 10.78 K lower than detections shared by both methods. These results show that STAR-FD provides complementary detection of confirmed fire-related thermal anomalies with weaker absolute thermal signals. Full article

Air quality modeling of sulfur dioxide (SO₂) concentrations remains challenging due to the high variability of both natural and anthropogenic emission sources, as well as the complexities associated with its multiphase chemistry. The data assimilation (DA) of satellite observations is a promising technique for constraining model uncertainties by combining the strengths of high-resolution and dense satellite retrievals with physical consistent model outputs. However, existing SO₂ DA applications have primarily focused on volcanic events while this study addresses them together with other emissions. The implementation of an SO₂ DA framework within the MINNI regional chemical transport model using an Ensemble Adjusted Kalman Filter (EAKF) via the DART framework is presented. The performances of the DA assimilation framework were tested using Sentinel-5P/TROPOMI SO₂-COBRA total column retrievals over continental Europe for August 2023. The filter constrained the ensemble variance to capture plumes from power plants and volcanic activity. The ensemble considered 20 members and perturbations of emissions and boundary conditions. On a monthly basis, the mean correction for the total column averaged over the domain was 2 × 10⁻⁵ mol m⁻², with localized maximum adjustments reaching 3.3 × 10⁻⁴ mol m⁻². At the surface level, domain-averaged corrections of concentrations reached up to 2.6 µg m⁻³. Despite current limitations related to ensemble size, static vertical localization, and the typical temporal fading of initial condition corrections, validation against in situ data confirmed the system’s ability to transfer column information to near-surface levels. These results demonstrate the feasibility and added value of integrating mixed-source SO₂ satellite retrievals into regional air quality simulations, contributing to more accurate, observation-driven atmospheric monitoring. Full article

In densely vegetated and topographically complex mountainous areas, the applicability of SAR data for potential landslide hazard identification depends not only on whether slopes are visible to the radar, but also on whether stable interferometric coherence can be preserved under vegetation and terrain constraints. To clarify the applicability differences between L-band Lutan-1 and C-band Sentinel-1 in such environments, this study focused on Hanyuan County, Sichuan Province, China. Ascending and descending SAR images acquired by the two satellite systems from 2024 to 2025 were processed using stacking-based Interferometric Synthetic Aperture Radar (Stacking-InSAR) and Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) to extract regional deformation anomalies and time-series deformation characteristics of representative landslides. DEM, LiDAR, optical imagery, fractional vegetation cover (FVC) derived from Sentinel-2, and field investigation data were further integrated to establish a comparative framework linking geometric visibility, interferometric coherence, and landslide identification results. The results show that both Lutan-1 and Sentinel-1 provided favorable geometric observation conditions after combining ascending and descending tracks, with joint visibility proportions of 98.48% and 97.76%, respectively, indicating limited differences in geometric coverage within the study area. However, at a unified grid scale, the mean coherence and valid grid-cell proportion of Lutan-1 reached 0.564 and 72.49%, respectively, substantially higher than those of Sentinel-1, which were 0.320 and 24.26%. As FVC increased, coherence decreased for both datasets, but Lutan-1 maintained higher coherence in densely vegetated areas, suggesting stronger adaptability to vegetation-induced decorrelation. Based on integrated interpretation of multi-source remote sensing data, 77 potential landslide hazards were identified in the study area, including 74 detected by Lutan-1, 17 detected by Sentinel-1, and 14 jointly detected by both datasets. Comparisons of representative landslides further show that Lutan-1 provided a higher density of valid deformation points in densely vegetated and small-scale landslides, with deformation patterns corresponding well to slope geomorphic boundaries and local deformation zones. Sentinel-1, with its higher temporal sampling density, can provide complementary information for time-series verification and multi-source cross-validation of key landslides. These results indicate that Lutan-1 is more suitable for spatial identification of potential landslide hazards in densely vegetated, topographically complex mountainous areas, while the joint use of Lutan-1 and Sentinel-1 can better balance landslide identification detail and time-series monitoring continuity. Full article