Abstract: Reliable Aboveground Biomass (AGB) estimates for woody crops are essential for carbon accounting and for Measurement, Reporting and Verification (MRV) frameworks. However, it remains unclear how LiDAR modality and sampling geometry influence plot-scale and tree-scale AGB predictions in intensively managed Mediterranean orchards. In this study, we benchmarked four LiDAR modalities, namely open national airborne laser scanning from the Spanish National Aerial Orthophotography Plan (PNOA/ALS), a dedicated Riegl airborne laser scanner (ALS), unmanned laser scanning (ULS) and mobile laser scanning (MLS), across three woody-crop sites in Córdoba (southern Spain): IFAPA, Doña María, and Villaseca. Plot-level LiDAR metrics (mean height, 95th height percentile, maximum height, and canopy cover proxies) were extracted from normalised point clouds and related to field AGB using Random Forest and XGBoost regression models, together with an ensemble predictor, under an 80/20 train–test split. In parallel, TreeQSM-based Quantitative Structure Models (QSMs) were evaluated as an independent tree-level three-dimensional reconstruction approach. XGBoost achieved the lowest errors at IFAPA (RMSE = 0.400 Mg ha⁻¹; R² = 0.994) and Villaseca (RMSE = 0.872 Mg ha⁻¹; R² = 0.995), whereas PNOA/ALS was competitive at Doña María (RMSE = 0.725 Mg ha⁻¹; R² = 0.994). TreeQSM closely matched field inventory at the low-biomass IFAPA site but tended to overestimate biomass at Doña María and Villaseca, and only 28% of scanned trees yielded usable reconstructions. The results support the use of cross-platform LiDAR for orchard AGB and carbon mapping and identify the conditions under which open national LiDAR can enable scalable MRV of Mediterranean woody crops.

Abstract: Remote sensing has become a core technology for environmental and climate monitoring, supported by expanding sensor constellations, advanced processing capabilities, and coordination frameworks established by the European Space Agency (ESA), Global Earth Observation System of Systems (GEOSS), and Committee on Earth Observation Satellites (CEOS). Ensuring consistency across missions requires robust geometric and radiometric calibration and validation. However, traditional reliance on ground control points (GCPs) is limited by sparse global coverage, temporal instability, and dependence on surveyed accuracy. While alternative geospatial datasets, including satellite and aerial imagery, Light Detection and Ranging (LiDAR) point clouds, and vector databases, can serve as references, challenges remain in data access, automation, and cross-sensor applicability. This study proposes a generative adversarial network (GAN)–based approach to generate geometrically consistent image chips from vector maps. Two models were trained at 50 cm and 10 m resolution within the ESA-supported Generative Ground Control Point (GenCP) study, using Sentinel-2 and very-high-resolution RGB imagery. The generated GenCP image chips are evaluated using image-based similarity (radiometric consistency), geometric, and model-performance metrics. Results demonstrate their suitability for automated Cal/Val workflows and their potential as scalable, fit-for-purpose reference datasets.

Abstract: Accurate monitoring of soil salinization in arid oasis regions is crucial for agricultural sustainability and ecological security. However, existing deep learning-based approaches often suffer from insufficient utilization of multi-scale information and inadequate modelling of feature interactions, limiting their accuracy in retrieving complex salinity patterns. To address these limitations, this study proposes a scale-attention optimized hybrid deep learning model that integrates multi-scale 1D convolutional neural networks (1D-CNN), bidirectional gated recurrent units (Bi-GRU), and Transformer mechanisms. The model employs a multi-scale feature extraction module to capture remote sensing signals across different scales, a scale attention mechanism to adaptively weight the most informative features, and a Bi-GRU-Transformer module to explore complex sequential and global feature relationships. The proposed framework is applied to the oasis irrigation zone in Weili County, Xinjiang, using hyperspectral data from the ZY-1E satellite, topographic indices, and spectral-derived variables. Experimental results demonstrate that our method achieves a coefficient of determination (R²) of 0.952 and a root mean square error (RMSE) of 0.867 g·kg⁻¹ on the test set, outperforming conventional 1D-CNN, GRU-Transformer, and other benchmark models with improvements of 2.8% in R² and 18.9% in RMSE.

Abstract: Accurate simulation of hyperspectral cloud radiance remains challenging under optically thick cloud conditions, where conventional layered radiative transfer (RT) models tend to underestimate cloud-induced backscattering and return radiance in the visible to shortwave infrared (VIS–SWIR) range. In this study, we propose an extinction-dependent interlayer reflective augmentation within a Curtis–Godson (CG)–based layered RT framework. Instead of introducing explicit cloud-top or cloud-bottom boundaries, the method adds a reflective coupling term at all discretized sublayer interfaces, scaled by local extinction properties, to compensate for the underrepresented backward radiative contribution in standard solvers. The proposed approach is designed for optically thick, plane-parallel cloud conditions and aims at improving forward radiance simulation rather than detailed microphysical retrieval. The formulation is constructed so that the reflective augmentation vanishes as the local extinction decreases, although the present experiments focus on optically thick cloud cases. Validation using Gaofen-5A (GF-5A) hyperspectral observations further confirms improved spectral fidelity of simulated cloud radiance in real scenes. Compared with conventional layered RT, the proposed method provides a favorable balance between computational efficiency and accuracy, making it suitable as a fast forward module for hyperspectral cloud radiance simulation of optically thick cloud scenes.

Abstract: Timely detection of missing persons is critical for successful Search and Rescue (SAR) operations, especially under challenging environmental conditions. Modern SAR efforts utilize both manned helicopters and unmanned aerial systems (UAS), often equipped with electro-optical (EO) and infrared (IR) sensors, while helicopters may also employ visual observers. Despite their widespread use, limited empirical data exists on how these platforms, sensor types, and search techniques perform across varying terrain and vegetation densities.This study presents results from the SAVIOUR 2024 field experiment, conducted during a large-scale SAR exercise in Rogaland, Norway. Twelve professional SAR aircrews (six helicopters, six UAS teams) conducted 48 search sorties across sectors with low, medium, and high vegetation density, targeting 251 human subjects. Key metrics were Probability of Detection (POD) and Time-to-Detection.Both platforms achieved high detection rates (mean POD >83%), with 54% of sorties reaching 100% POD. Vegetation density was the strongest predictor of POD, with reduced performance in high-density forest (helicopters: 71.4%, UAS: 73.3%). Platform type did not significantly affect POD when controlling for vegetation. Helicopters detected targets faster, likely due to initial sweep strategies. UAS teams favored systematic detailed searches, resulting in longer detection intervals. Sensor-based searches outperformed visual-only methods, though visual-only data were limited.We propose that coordinated, vertically separated operations—helicopters at high altitude and UAS at low altitude—can enhance efficiency through concurrent coverage. These findings offer guidance for integrated SAR practices and highlight future research needs, including AI-assisted detection and performance evaluation under diverse thermal and geographical conditions.

Abstract: Coastal urbanization is increasingly constrained by legacy land-use patterns and escalating climate risks, yet long-term morphological trajectories remain poorly quantified due to the absence of multispectral data in pre-satellite archives. This study presents a scalable deep learning pipeline that bridges this century-scale domain gap, enabling automated reconstruction of urban expansion from panchromatic historical aerial imagery (1920–1971) and digital aerial photographs (1997) to contemporary very-high-resolution satellite data (2024) in Les Sables-d’Olonne, France. Spectral restoration was performed using an attention-enhanced Pix2Pix generative adversarial network with hybrid inference, achieving high fidelity (PSNR 35.21 dB, SSIM 0.9762). Semantic segmentation was conducted with U-Net++, yielding strong performance on modern data (mIoU 0.9789). However, direct transfer to historical periods suffered from severe domain shift due to radiometric variations.To overcome this limitation without extensive manual annotation, few-shot adaptation was applied on year-specific calibration sets, producing reliable building footprints (mIoU 0.53–0.65) despite degradation. Multi-scalar analysis of the reconstructed footprints revealed constrained anisotropic expansion: early saturation of the coastal historic core, followed by rapid inland peri-urbanization post-1971 driven by geographic barriers. This spatiotemporal shift has entrenched spatial lock-in, placing recent development in retro-littoral zones vulnerable to submersion and characterized by severe vegetation loss. This framework unlocks previously inaccessible historical archives for quantitative urban monitoring, providing critical insights into legacy effects of unconstrained growth and informing resilient coastal planning under climate change.

Abstract: Shallow, turbid coastal environments (Case 2 waters) challenge optical remote sensing due to the complex, non-covariant interaction between dissolved and particulate constituents. This study quantifies the relationship between the effective (K_eff) and the diffuse attenuation coefficient of downwelling irradiance (K_d) across 14 stations in the southern Baltic Sea, representing a transition from estuarine to open coastal waters. Using K-Means clustering and Random Forest regression, we characterised Optical Water Types (OWTs) and decoupled the specific drivers of attenuation. Results indicate that K_eff consistently exceeds K_d by a factor of 2-3, with the spectral ratio (K_eff/K_d) significantly surpassing the theoretical geometric limit of 2, particularly in the 500-650 nm window. Although total suspended matter (TSM) is the primary driver for both coefficients, K_eff exhibits heightened sensitivity to coloured dissolved organic matter absorption at 440 nm (a_CDOM (440)) due to the geometric rejection in the collimated beam; in contrast, K_d remains coupled to the broad-band scattering effects of phytoplankton. We conclude that assuming a fixed geometric relationship (K_eff ≈ 2K_d) leads to systematic errors in scattering-dominated waters, and propose a robust empirical relationship (K_eff ≈ 1.71K_d + 1.44; Pseudo R² = 0.4) to improve subsurface retrievals in shallow and optically complex coastal zones.

Abstract: UK forestry faces increasing drought risk under climate change, raising concerns about the resilience of Sitka spruce, the UK’s dominant commercial conifer. This study assessed whether hyperspectral vegetation indices can detect intraspecific drought responses to support resilience screening. An eight-week controlled drought experiment was con-ducted on six clonal groups, using needle-level hyperspectral reflectance to derive indices of chlorophyll status, photoprotective pigments, and water content, alongside chlorophyll fluorescence (Fv/Fm). Drought responses were detected across multiple indices, with pigment-based and red-edge indices showing the earliest and strongest sensitivity, while water-related indices captured later-stage hydraulic decline. Significant clonal variation was observed in the timing and magnitude of pigment regulation, water loss, and photosynthetic impairment, indicating contrasting drought response strategies. These results demonstrate that hyperspectral approaches enable rapid, non-destructive detec-tion of physiologically meaningful drought responses and can support the identification of drought-resilient genotypes for climate-adaptive forest management.

Abstract: Marine-terminating glaciers are major contributors to sea-level rise, yet their frontal ablation—the combined loss from ice discharge and terminus retreat—remains poorly constrained. This study presents a monthly time series of ice discharge for 40 marine-terminating glaciers in Alaska from 2015 to 2021, derived from Sentinel-1 velocity data, and reconstructed ice thickness information. Frontal ablation was calculated as the sum of ice discharge and terminus mass loss, from manually delineated terminus positions. The mean annual ice discharge was 11.81 ± 5.35 Gt a⁻¹, dominated by Hubbard, Columbia and Yahtse glaciers, which together accounted for ~70% of Alaska’s total ice discharge. Terminus retreat contributed an additional 1.30 ± 0.07 Gt a⁻¹, resulting in a total frontal ablation of 13.11 ± 5.35 Gt a⁻¹. Most glaciers exhibited late-summer velocity minima likely indicating seasonal changes in subglacial drainage efficiency, while interannual variability corresponded with El Niño-Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO) phases. These findings confirm that Alaska’s marine-terminating glaciers currently lose relatively little mass through frontal retreat compared to their regional mass balance, suggesting that most glaciers have passed their phase of rapid retreat. The presented analysis also provides fundamental information for refining sea-level rise projections.

Abstract: Accurate estimation of crop yield from remote sensing remains challenging due to the crop-specific nature of yield drivers and the difficulty of interpreting spectral indicators across agronomic systems. While many studies prioritise predictive accuracy through complex models, fewer explicitly examine the stability and physiological relevance of in-dividual spectral and phenological indicators under controlled analytical conditions. This study investigates yield–spectral relationships in wheat and cotton using a harmonised Sentinel-2 indicator framework applied across multiple growing seasons in a Mediterra-nean agricultural environment. A consistent set of spectral and thermal indicators was derived from two phenologically targeted Sentinel-2 acquisitions per season and analysed using correlation analysis, univariate regression, constrained multivariate modelling, and recurrence analysis within an identical workflow for both crops. Distinct crop-specific patterns were observed. Wheat yield was most strongly associated with water-sensitive and canopy-related indicators, with NDWI-based metrics reaching Pearson correlations up to r = 0.85 and multivariate models explaining a substantial proportion of yield varia-bility (up to R² ≈ 0.82) under controlled analytical conditions. In contrast, cotton yield var-iability was dominated by thermal accumulation, with growing degree day indicators showing correlations up to |r| = 0.59 and multivariate performance reaching R² = 0.76. Recurrence analysis confirmed the stability of these indicator families across analytical stages. Overall, the results indicate that parsimonious, physiologically interpretable indi-cator combinations can account for a substantial proportion of yield variability without reliance on black-box modelling, supporting crop-aware indicator selection for precision agriculture applications.

Abstract: With an ever-growing recognition of Land Surface Temperature (LST) as a key Essential Climate Variable (ECV), it becomes utmost important to have such a variable at both the fine spatial and temporal scales of urban spaces and dynamics. Sentinel-3 provides coarse LST (1 km, daily) based on thermal imagery acquired by its Sea and Land Surface Temperature Radiometer (SLSTR) as well as fine Spectral Directional Reflectances (SDR, 300 m, every two days) synergically inferred from both SLSTR and the optical bands acquired through the Ocean and Land Colour Instrument (OLCI), which gives opportunity for using the latter as predictor in the downscaling of the former. Herein, two scale-invariance-based architectures were developed: a single-timestamp model, trained with the coarse data of the timestamp whose fine target it tries to infer; and a multi-timestamp one, trained with several timestamps and that can infer for any other. While for the case of the multi-timestamp architecture, Machine Learning (ML) models besides Linear Regression (LR) were trained, solely LR was considered for the single-timestamp architecture due to the smaller amount of data available, making it less suitable for hyperparameter tuning. The models were developed over four Danish Functional Urban Areas (FUAs) between 2020 and 2023 using SRD-derived indices, seasonal and geospatial predictors. From 112 Sentinel-3 scenes, 105 were used for training and 7 for validation against Landsat data. While Gradient Boosting (GB) achieved the best coarse-scale performance (test set Root Mean Square Error, RMSE, of 1.56 K), fine-scale predictions showed degraded performance, indicating scale-invariance breakdown. Tree-based models performed poorly due to extrapolation limitations, whereas Neural Net (NN) and LR proved more robust. After residual correction, single-timestamp LR achieved the best fine-scale performance (test set RMSE of 1.40 K), making it the most reliable and operationally recommended architecture.

Abstract: Weed infestation significantly threatens crop productivity and quality, highlighting the need for accurate and scalable monitoring approaches. Recent advances in unmanned aerial vehicle (UAV) remote sensing and deep learning provide promising tools for field-scale weed detection. This study evaluates and compares two state-of-the-art instance segmentation models, Mask R-CNN and YOLOv8, for species-level weed detection in wheat fields under Mongolian agro-ecological conditions. The experiment was conducted in a 4 ha wheat field in Tuv Province, Mongolia, using high-resolution RGB imagery acquired from UAV flights in July 2025. Three dominant weed species were annotated and analyzed. Model performance was evaluated using mAP@0.5:0.95, Precision, Recall, F1-score, and mask IoU. At IoU thresholds of 0.25 and 0.5, both models demonstrated moderate detection performance (IoU = 0.25: Precision 0.49–0.76, Recall 0.20–0.77, F1-score 0.32–0.75; IoU = 0.5: Precision 0.42–0.67, Recall 0.18–0.75, F1-score 0.28–0.69), with variation among weed species. Mask R-CNN achieved higher Recall and more precise boundary delineation, improving weed coverage estimation, whereas YOLOv8 provided faster inference (≈11 ms per image, ~90 FPS) and higher precision, making it more suitable for large-area and near-real-time monitoring. These findings demonstrate the potential of UAV-based instance segmentation for weed detection in Mongolia and provide practical guidance for model selection in precision agriculture applications.

Abstract: Airborne LiDAR data acquired before and after the 2024 Noto Peninsula earthquake in Japan were used to estimate three-dimensional (3D) ground-surface displacements based on the Iterative Closest Point (ICP) algorithm. Digital elevation (terrain) models (DEMs) were generated from pre-earthquake point cloud data acquired by Ishikawa Prefecture and compared with post-earthquake DEMs developed by the Forestry Agency of Japan. Three-dimensional coseismic displacements were derived from the spatial correlation between pre- and post-event DEMs for 50 m × 50 m tiles. The results depend on tile size and are influenced by ground movements within and surrounding each tile. Therefore, moving-average windows of 250 m and 550 m were applied to the 50 m tiles to obtain continuous 3D displacement fields across the ground surface. A comparison between GNSS-measured displacements and the corresponding moving-average estimates for tiles containing triangulation points and continuously operating reference stations (CORSs) showed that the accuracy of the estimated displacements in all three components was within 0.2 m in terms of root mean square error (RMSE).

Abstract: Remotely Piloted Aircraft (RPAs) equipped with multispectral sensors have emerged as promising tools for estimating foliar nitrogen content (FNC). In this context, this study applied a methodological approach aimed at simulating UAV multispectral data using hyperspectral leaf data obtained in a controlled environment, with the objective of evaluating its predictive potential and its transferability to field data collected by UAVs for FNN estimation. To this end, spectral bands and indices equivalent to those of UAV-mounted sensors were simulated based on hyperspectral data acquired by a benchtop sensor, and subsequently used in modeling via Partial Least Squares Regres-sion (PLSR) and Random Forest (RF). The results showed similar performance across the levels, with R² values of 0.75 and 0.76 for PLSR and RF on the UAV data, and 0.75 and 0.74 for PLSR and RF on the simulated data, respectively. The RF model also performed well in cross-domain validation, with R² = 0.70 when calibrated with simulated data and ap-plied to UAV data. Furthermore, the simulated data maintained high predictive power even with a reduced sample size. It is concluded that spectral simulation constitutes a viable strategy for expanding the applicability of nutritional monitoring using multi-spectral sensors.

Abstract: Background: The intersection of machine learning (ML) and deep learning (DL) with thermal remote sensing (TRS) has undergone a transformative expansion since 2018, driven by the proliferation of high-resolution satellite missions and open-source deep learning frameworks. Despite this rapid growth, to the best of our knowledge, no comprehensive PRISMA-compliant systematic review has synthesised ML/DL applications specifically within the thermal RS domain across the post-2018 period.Objectives: This review maps the complete landscape of ML/DL applications in thermal RS from January 2018 to March 2026 with five primary objectives: (i) quantify publication trends; (ii) classify the taxonomy of ML/DL architectures; (iii) map application domain coverage; (iv) appraise methodological quality and open science practices; and (v) identify research gaps and future directions.Methods: A systematic electronic search was conducted across Scopus and Google Scholar. Following PRISMA 2020 guidelines (Page et al., 2021), records underwent a structured multi-stage screening process implemented in Python. This consisted of five main screening stages after initial deduplication, followed by a final full-text eligibility assessment for open-access records retrieved via the Unpaywall API. Data extraction employed a structured template covering bibliographic metadata, sensor platforms, ML/DL architecture, application domain, performance metrics, and open science practices.Results: A total of 193 peer-reviewed studies met the inclusion criteria, of which 93 were available as open-access full texts and 100 were accessible through title, abstract, and structured metadata only due to institutional access restrictions. CNNs (43.7%), LSTM/BiLSTM (33.0%), and SVR/SVM (29.1%) were the dominant architectures across the 93 open-access full-text studies from which comprehensive architecture data were extracted. Application domains concentrated on SST forecasting, LST retrieval, LST downscaling, and gap-filling, leaving wildfire detection, evapotranspiration estimation, and permafrost monitoring relatively underrepresented compared to core domains. Code availability was reported in fewer than 5% of included studies.Conclusions: This review reveals a maturing but architecturally conservative field with transformative opportunities in physics-informed neural networks, transformer-based models, and underserved application domains. The persistent open science deficit represents a structural reproducibility challenge that warrants urgent community attention.

Abstract: Accurate estimation of aboveground biomass (AGB) in mountainous forest ecosystem remains a significant challenge due to complex terrain, the high cost and limited applicability of traditional field-based methods. To address this issue, a remote sensing–based AGB estimation framework integrating intelligent optimization and machine learning was developed for Mount Tai in eastern China. Sentinel-2 multispectral data were selected to derive 105 candidate variables, including spectral bands, vegetation indices, texture features, and topographic factors, from which 17 key variables were selected using Pearson correlation analysis for model construction. A Support Vector Machine (SVM) optimized by the Pigeon-inspired optimization (PIO) algorithm was developed to adaptively determine optimal hyperparameters, and its performance was compared with that of Random Forest (RF) and standard SVM models. The results demonstrate that the PIO-SVM model achieved the best overall performance. For the training dataset, the model obtained an R² of 0.85. For the test dataset, the R² reached 0.73, outperforming RF (0.70) and standard SVM (0.72). The spatial distribution of AGB derived from the optimal model shows higher AGB values in the central and northern regions characterized by dense forest cover, in close agreement with field observations. These findings indicate that the PIO algorithm effectively enhances SVM hyperparameter optimization in complex parameter spaces, significantly improving the accuracy and stability of AGB estimation in mountainous forest. This study provides a reliable and efficient framework for regional-scale monitoring of forest biomass and carbon sink dynamics.

Abstract: In recent years, salient object detection in optical remote sensing images (ORSI-SOD) has garnered increasing research attention. However, in practical applications, issues such as blurred target edges under low contrast and complex background interference continue to restrict the accuracy and robustness of detection. To address these problems, this paper proposes the Phase Congruency-Guided Cross-Scale Contextual Fusion Network (PCFNet). Specifically, we design a novel Phase Congruency Enhanced (PCE) Module to solve the problem of low contrast between targets and backgrounds. It acquires multi-scale phase features via Fourier decomposition, fuses them with Transformer shallow features and uses a tailored loss weighting mechanism to weight phase congruency learning for better PCE module adaptation. To address complex background interference, we design a novel Dynamic Residual Fusion (DRF) Module. It leverages dynamic spatial attention and residual connections to refine multi-scale features and enables the model to accurately capture effective target features under complex background interference. Experiments on ORSSD, EORSSD, and ORSI4199 benchmarks show that PCFNet outperforms 24 state-of the-art methods in core metrics, and ablation studies further confirm the effectiveness of each module.

Abstract: Automated building damage assessment from satellite imagery has become increasingly critical for rapid disaster response and humanitarian relief operations. However, current state-of-the-art deep learning models exhibit significant generalization challenges when deployed to geographically and environmentally diverse regions. This study investigates the nature and extent of geographic bias in building damage detection systems, revealing that model performance degradation stems primarily from geographic and structural characteristics rather than insufficient training data representation. Through systematic evaluation of top-performing xView2 competition solutions across 17 disaster locations spanning multiple climate zones, we found that even state-of-the-art models struggle with generalization, particularly for Minor and Major damage classes, and exhibit strong geographic biases toward certain regions. Strikingly, Nepal despite having the largest training dataset (15,234 images) achieves the worst performance, demonstrating that geographic and structural characteristics dominate generalization behavior more than training data quantity. To address these fundamental limitations, we explore Fusion Augmentation, a novel methodology that enhances edge detection and structural feature representation by integrating auxiliary information channels with standard RGB imagery. Experimental results demonstrate substantial improvements of 7.1% overall F1 score, with dramatic gains for intermediate damage categories such as Minor and Major damage. Domain adaptation experiments on three unseen locations show that combining Fusion Augmentation with supervised fine-tuning yields 40.8% and 60.0% improvements over Minor and major classes, while unsupervised CORAL achieves 24.2% and 39.5% improvements over Minor and major damage classes compared to benchmarks. These findings challenge prevailing assumptions about data-driven generalization in remote sensing AI systems and demonstrate that structural feature enhancement combined with domain adaptation is essential for robust detection across geographically diverse deployment scenarios, providing practical strategies for globally deployable damage assessment systems.

Abstract: Mining generates various alterations to the environment, affecting flora, fauna, morphology, and soil. To contribute to solving this problem, this study measured land cover (LC) changes induced by open-pit mining in Zacatecas, Mexico. Remote sensing techniques were applied using multitemporal Landsat 5 and 8 satellite imagery, along with supervised classification, to detect land cover variations. Tests performed on eleven classification algorithms showed that the Spectral Angle Mapper (SAM) obtained the best results, with an accuracy of 85.16% and a Kappa coefficient of 0.79. Measurements of changes in land cover revealed an increase in the surface area of ​​water bodies of +556.83 Ha, in mining cover of +1729.35 Ha, in infrastructure of +2.61 Ha, and of bare soil of +1488.15 Ha, and a loss of soil of -2372.49 Ha, of scrubland of -1444.59 Ha, and of vegetation of -9.45 Ha. The use of supervised classification for multi-temporal satellite imagery allowed for the measurement of alterations to land cover. These alterations highlight the need for sustainable management strategies, environmental restoration, and the importance of continued monitoring for informed decision-making. It is recommended to explore new categories and techniques, such as deep learning, to improve the accuracy of land cover classification.

Abstract: Accurate estimation of latent heat flux (LE) and sensible heat flux (H) is essential for determining actual crop evapotranspiration (ET_a) and optimizing irrigation water management. However, uncertainties in characterizing the zero-plane displacement height (do) often limit H and LE model accuracy. This study introduces a novel approach to characterize do using a dynamic fractional vegetation cover and a new proposed canopy porosity (Φ_dp) term derived from Unmanned Aerial System (UAS) imagery. Field experiments were conducted in 2024 near Greeley, Colorado, USA, at a research farm using fully and deficit-irrigated maize fields. Eddy covariance (EC) systems, handheld multispectral radiometry, and PlanetDove mini-satellite imagery were used in the land surface energy balance (EB). A dynamic heat flux footprint area was implemented based on crop height, atmospheric stability, and wind conditions, to align and integrate those measurements with measured EC heat fluxes. Results indicated that both developed do models noticeably outperformed existing methods. The new do models reduced the normalized root mean square errors (NRMSE) for H estimation by up to 21.1% in the fully irrigated (FI) field and by 16.9% in the deficit-irrigated (DI) field. Furthermore, a higher agreement index of up to 0.74 reflected an improved do model vs. observation correlation. These findings highlight the potential of incorporating a dynamic canopy porosity and vegetation fractional cover to refine EB-based ET_a modeling and advance agricultural irrigation water management based on remote sensing inputs.