Search Results (3,482)

Mountainous watersheds often suffer from incomplete and spatially biased landslide inventories, which limit the reliability of conventional susceptibility modelling. This study examines whether the erosion coefficient Z of the Gavrilović Erosion Potential Method can provide a process-oriented indicator of slope-instability predisposition in the Portaikos watershed, Central Greece. The Z coefficient was derived from geospatial layers representing vegetation protection, lithological erodibility, erosion-process expression and slope gradient, using Copernicus land-cover products, tree-cover density data, Sentinel-2 imagery, FABDEM and national soil–geological information. A landslide inventory of 46 mapped occurrences from the Hellenic Survey of Geology and Mineral Exploration was then used as an independent reference layer. Erosion severity was classified into five classes and compared with the landslide distribution through Frequency Ratio analysis. Most of the basin was assigned to moderate, very slight and slight erosion classes, covering 34.7%, 30.1% and 28.2% of the area, respectively. By contrast, severe and excessive erosion occupied only 6.7% and 0.4% of the watershed, but contained a much larger proportion of the mapped landslides: 58.7% and 10.9%, respectively. This disproportion was also reflected in the Frequency Ratio analysis. When the severe and excessive classes were considered together, they occupied approximately 7.1% of the watershed but contained 69.6% of the mapped landslides, corresponding to an FR value of 9.79. The separate excessive class showed the highest FR, but it was interpreted cautiously because of its very limited spatial extent and small landslide count. These results indicate that high Z values coincide with terrain sectors where lithological weakness, steep slopes, reduced surface protection and erosion-related sediment-source conditions jointly favour slope instability. The Gavrilović Z coefficient should therefore not be interpreted as a substitute for rainfall-threshold analysis or inventory-based predictive models. Rather, it may serve as a useful first-order screening layer for field verification, spatial prioritization and ecosystem-based mitigation planning in data-scarce Mediterranean mountain watersheds. Full article

Crop monitoring is an important aspect of agricultural management, as it provides insights into cultivated area, crop health, growth patterns, and yields potential. Mapping cultivated areas and identifying crop types was historically conducted through field surveys and manual mapping, which are time-consuming and labor-intensive. Remote sensing classification has transformed large-scale land cover mapping, including crop identification. This work aims to: (1) compare the performance of Dynamic Time Warping (DTW) and two machine learning families (artificial neural networks and decision trees) for crop classification using Sentinel-2 data; (2) assess whether reflectance data, spectral indices, or both yield better classification results; and (3) evaluate the effect of hyperparameters on model performance. Among the DTW variants evaluated, dynamic time warping without a time constraint performed the best, with an overall accuracy of 0.921 using the combination of both reflectance and spectral indices. Most machine learning methods outperformed DTW. Although the convolutional neural network reached the highest single accuracy (0.948), the transformer was selected as the best model overall (accuracy of 0.944), as it combined a comparable accuracy with the lowest sensitivity to hyperparameter variations, making it a reliable option when testing machine learning architectures applied to crop mapping. This work also provides insights for model architecture development based on an exhaustive hyperparameter search for the machine learning models. Full article

Accurate landslide susceptibility mapping (LSM) is important for hazard prevention and land use planning in mountainous regions. Existing machine learning and deep learning methods mainly use raster-based conditioning factors. They often ignore landslide-related attribute information and spatial context. To address this issue, this study proposes an image–tabular joint deep learning framework for regional-scale LSM. The framework is based on a FiLM-conditioned U-Net. The model combines raster patches with an estimated soft attribute-prior vector and uses FiLM to guide condition-aware spatial feature learning. The proposed framework was tested in the Tacheng region, Xinjiang, China. The dataset includes a landslide inventory and conditioning factors related to terrain, hydrology, vegetation, geology, land cover, and human activities. Model performance was evaluated using stratified five-fold cross-validation, an independent test set, buffer-radius sensitivity tests, and spatial hold-out validation. FiLM-U-Net achieved the best performance among the tested models. It obtained an accuracy of 89.73%, an F1-score of 89.41%, and an AUC of 0.953 on the independent test set. In the spatial hold-out validation area, the model achieved an AUC of 0.921. Feature importance analysis showed that distance to roads, rainfall, NDVI, and terrain factors provided important predictive information. These results suggest that the proposed image–tabular joint framework can improve condition-aware feature learning and support regional landslide susceptibility assessment. Full article

The Brazilian Amazon is a contested frontier, shaped by destructive and conservationist forces. Most forest clearing is driven by agro-extractivism, an agrarian pathway based on raw commodity production. The socio-biodiversity economy (SBE) has emerged in response, widely regarded as a transformative agrarian pathway capable of reconciling environmental conservation and rural livelihoods. However, recent research suggests that, as socio-biodiversity products scale up, agro-extractivist dynamics can be reproduced within the SBE. We examine this tension in the cocoa frontier of the Transamazon, where cocoa is institutionally promoted as an SBE alternative. We conduct an exploratory, mixed-methods study combining a geospatial analysis of land use and land cover (LULC) change (2020–2025, random forest classification) with a qualitative analysis drawing on participatory mapping and 87 semi-structured interviews with farmers, cooperatives, buyers, and institutional actors. Using a gradient framework, we read LULC transitions and farmers’ land use practices along an agro-extractivism–SBE continuum. The cocoa frontier emerges as a hybrid geography. The landscape is predominantly stable but internally reorganizing: anthropogenic forest declines while full-sun monoculture expands over pasture, with intensification concentrated in peri-urban areas and restoration in remote ones. Land use practices form five recurring configurations, two firmly anchored at the socio-biodiversity or agro-extractivist poles and three whose alignment with the SBE depends on access to markets, knowledge, and institutions. We argue that the frontier contestation unfolds not only between distinct economies but within the cocoa economy itself, and we identify the policy areas relevant to sustaining SBE-oriented practices in the Transamazon. More broadly, this study suggests that the classification of Amazonian forest-based economies as inherent alternatives to agro-extractivism should be treated as an empirical question rather than an assumption. Full article

The rapid decline of Arctic sea ice is increasing the feasibility of trans-Arctic shipping along the Northern Sea Route (NSR)—a considerably shorter and potentially lower-emission alternative to the Suez Canal route—for which the Bering Strait is the first gateway chokepoint for vessels departing the Republic of Korea. Continuous optical monitoring of this region is fundamentally limited by cloud cover and polar night during the ice season; in 2025, no usable Sentinel-2 MultiSpectral Instrument (MSI) scene was available for January (mean cloud cover 56.9% in November). This study therefore used all-weather Sentinel-1 dual-polarization (VV, VH) C-band SAR, acquired monthly through Google Earth Engine for January–December 2025, to classify open water, thin ice (<30 cm), and first-year ice (FYI, ≥30 cm) with a Support Vector Machine (SVM); land was masked prior to classification. Three kernels (linear, radial basis function (RBF), and polynomial) were compared after $5\times 5$ Lee speckle filtering, using VV, VH, and the VV/VH ratio (in dB) as input features. The RBF kernel achieved the highest accuracy (overall accuracy 97.0%, $\kappa =0.954$), and the classification was cross-referenced against the U.S. National Snow and Ice Data Center (NSIDC) Multisensor Analyzed Sea Ice Extent (MASIE) mask. The monthly ice maps were then reclassified into World Meteorological Organization (WMO)/POLARIS ice types and converted into a POLARIS Risk Index Outcome (RIO) cost surface using simplified proxy weights consistent with the POLARIS risk ordering rather than the full Risk Index Value table; least-cost paths and a seasonal navigability assessment were derived for three representative ship ice classes—a non-ice-class vessel, an ice-class 1A merchant vessel (≈PC7), and the icebreaker Araon (≈PC5)—under risk-index, water-depth, and coastal-buffer constraints. For a representative refreezing-onset scene, the routing outcome was strongly ice-class dependent: the non-ice-class vessel and the ice-class 1A vessel were blocked (after 42 and 139 km, respectively), whereas only the icebreaker Araon completed the transit along a 124 km least-cost path within a 2 km safety corridor. The framework demonstrates how an interpretable SAR-based sea-ice classification can be coupled with a recognized international risk standard to support risk-informed navigation through the Bering Strait gateway. By indicating when shorter, lower-emission Arctic transits are feasible while limiting the risk of ice-related accidents in this sensitive polar environment, the framework also contributes to the safety and environmental sustainability of trans-Arctic shipping. Full article

Multi-sensor fusion of optical and synthetic aperture radar (SAR) imagery is widely used for land cover classification, yet most studies treat heterogeneous sensors as a uniform feature pool, leaving class-dependent differences in sensor contribution insufficiently quantified. To address this gap, this study integrates optical and SAR imagery within a Random Forest (RF) classifier in Google Earth Engine (GEE) and applies a combined SHapley Additive exPlanations (SHAP)—Remove and Retrain (ROAR)—bootstrap framework to disentangle, for each class, which sensor provides which discriminative information and how faithful those attributions are. The dataset included Sentinel-1/2, Korea Multi-Purpose Satellite (KOMPSAT)-3A/5, and Landsat-8, representing a range of spatial resolutions and spectral characteristics. An RF-based machine learning (ML) model was employed to perform multi-sensor data fusion and classification. To address the inherent opacity of ML models, we employed the SHAP algorithm, an explainable artificial intelligence (XAI) method, to interpret classification decisions. SHAP analysis indicated that visible and near-infrared (NIR) bands, along with vegetation indices, were the dominant contributors to land cover classification in this study area, while SAR data provided complementary structural information for spectrally ambiguous targets, such as roads—a class-dependent role reflected in the stable rank ordering of the Sentinel-1 VV contribution across bootstrap replications, rather than in a formally significant magnitude difference. ROAR results were consistent with the top-ranked SHAP features being those on which the classifier relies, supporting a physically plausible interpretation of ML-based remote sensing classification. These attributions were robust to estimator choice (SHAP-permutation ρ = 0.89) and spatial partitioning (mean ρ = 0.97 across folds). The contribution of this study is methodological rather than algorithmic: it provides an integrated analytical framework that applies SHAP, ROAR, and bootstrap confidence intervals jointly to a specific multi-sensor land cover problem, demonstrating that interpretability and high classification accuracy can be reported together. The results demonstrate that optical and SAR fusion contribute differently across land cover classes rather than uniformly, providing practical, class-specific guidance for sensor selection in operational land cover mapping and improving the interpretability of machine learning-based mapping workflows. Full article

Accurate and transferable land cover mapping from satellite imagery is a prerequisite for national-scale agrienvironmental monitoring and climate change impact assessment. The joint use of Sentinel-1 synthetic aperture radar (SAR) and Sentinel-2 multispectral imagery offers complementary structural and spectral information, yet systematic evaluation of fusion strategies and their geographic transferability remains limited. We trained and tested five U-Net fusion architectures S1-only, S2-only, early (input-level), feature-level (middle), and decision-level (late) alongside a SegFormer-b2 transformer baseline over two German cities (Munich and Berlin) using the Multi-Sensor Land Cover Classification (MSLCC) dataset (single-date 2017 Sentinel-1B/Sentinel-2A acquisitions) at 10 m resolution. Three cross-city transfer protocols (Munich → Berlin, Berlin → Munich, and combined training) quantify model transferability across contrasting urban–rural gradients. Early fusion achieved the highest in-city macro-averaged F1 score among U-Net variants (0.8278), a small but statistically significant improvement over the optical-only baseline (0.8236; patch-level paired bootstrap, $p=0.006$); feature-level (middle, 0.8164) and decision-level (late, 0.8196) fusion were, by contrast, significantly worse than the optical-only baseline ($p<0.001$ and $p=0.030$, respectively), and the SAR-only model (0.6864) trailed substantially. The built-up class was the primary beneficiary of SAR inclusion under early fusion. SegFormer-b2 (0.8214) was numerically close to, but statistically significantly below, the best convolutional configuration ($p=0.005$), and exhibited strong cross-city transfer (0.8632–0.8608 macro-F1), consistent with the geographic invariance conferred by its ImageNet-pretrained encoder. Combined training across both cities improved over the single-direction transfer average by 0.012 macro-F1 points for U-Net and 0.006 points for SegFormer, offering a practical route to national-scale deployment without requiring explicit domain adaptation. Spectral index augmentation (NDVI, NDWI, ExG) and SE channel attention did not significantly improve over plain early fusion when derived from percentile-normalized inputs, with the best variant statistically indistinguishable from the baseline at macro-F1 = 0.8268 ($p=0.365$); the result is attributable to a specific preprocessing dependency: NDVI, NDWI, and ExG are only physically meaningful when computed from calibrated reflectance, whereas here they were derived after scene-level 2nd–98th percentile stretching, which strips the absolute radiometric referencing the indices rely on; practitioners combining spectral indices with percentile-normalized (rather than physically calibrated, e.g., Level-2A surface-reflectance) inputs should expect a similar null result. Full article

Regional landslide susceptibility mapping commonly relies on static conditioning factors, including terrain, geology, hydrology, land cover and human activity. These factors describe long-term instability settings but cannot directly represent recent or ongoing ground deformation. Interferometric Synthetic Aperture Radar (InSAR) can provide spatially distributed deformation information, yet mountainous InSAR evidence is affected by uneven observation availability, vegetation decorrelation, terrain-induced geometric distortion and confounding between observation support and static environmental conditions. These issues make it difficult to determine whether radar-derived variables represent deformation signals or mainly indicate where observations are reliable. This study develops a scale-sensitive and confounding-audited Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) evidence representation framework for regional landslide susceptibility mapping. In Pingwu County, China, a 1607-record landslide inventory was converted into 1579 unique 30 m landslide cells, with 1393 for training and 186 for inventory-concentration validation. Fourteen static factors formed the baseline model. Deformation evidence from 98 Sentinel-1A descending acquisitions was represented using point-based line-of-sight (LOS) variables, neighbourhood component descriptors, a compressed deformation-intensity and observation-availability index, and a separated deformation intensity (DI) plus observation availability (RI) representation. Results show that the static factors already provided a strong first-order baseline. Adding SBAS-InSAR evidence did not produce uniform paired improvements across models, metrics or neighbourhood scales. In 10 km spatial-block cross-validation, random forest models using static factors, 300 m component descriptors and 300 m DI + RI features achieved similar mean area under the receiver operating characteristic curve values of 0.948, 0.949 and 0.950, with mean Matthews correlation coefficient values of 0.791, 0.795 and 0.795. Broader neighbourhood representations may expand the attainable performance boundary, but only as diagnostic evidence. SBAS-InSAR-derived information should therefore not be treated as a simple additional conditioning factor; its value depends on jointly interpreting deformation intensity, observation availability and their confounding with static context. Full article

Climate variability represents a major threat to agricultural systems, particularly in arid and semi-arid regions such as the Beni-Amir irrigated perimeter, located in the Tadla plain in central Morocco. In this perimeter, olive trees are exposed to multiple environmental and management-related factors that are associated with variations in phenology and vegetation vigor, such as successive drought episodes. This study represents a spatio-temporal assessment of drought impact on olive using satellite- derived vegetation indices from Sentinel-2 imagery and precipitation satellite data from CHIRPS over the period 2015–2024. The Standardized Precipitation Index (SPI-12) was used to identify wet and dry phases over this period. The results indicate an alternation of dry and wet periods between 2015 and 2021, followed by a predominance of dry conditions from September 2021. Over the same period, the time series of the Normalized Difference Vegetation Index (NDVI) and the other vegetation indices reveals marked interannual variability and a progressive degradation of olive tree phenological cycles. A land cover map derived from a supervised support vector machine (SVM) under three classification scenarios achieved high overall accuracies exceeding 94%. Post-classification change detection highlights a substantial reduction in mapped olive-growing areas between 2016 and 2024, with an estimated 72% loss of the initial area. The findings reported in this study indicate that the succession of drought episodes may have contributed to olive grove degradation, including disruptions in phenological cycles and a decline in maximum NDVI values. Even the most resilient olive groves appeared affected following the severe drought period after 2021. The study underscores the usefulness of satellite-derived vegetation indices and drought indicators for the effective monitoring of drought-related stress and supporting improved management practices under climate change. Full article

Hyperspectral remote sensing provides rich spectral information and has been widely used in fine-grained land-cover classification and forest monitoring. However, accurate tree species classification remains challenging due to subtle interspecific spectral differences, similar spatial structures among related species, redundant spectral bands, and the limited ability of existing methods to model long-range spatial–spectral dependencies efficiently. In addition, many existing hyperspectral image classification methods rely on patch-based inputs and sliding-window inference, which often lead to redundant computation and insufficient utilization of global image context. To address these issues, this paper proposes MambaHSINet, a dual-branch bidirectional state space network for full-image pixel-wise hyperspectral classification. Specifically, the proposed network employs a spectral branch and a spatial branch to explicitly extract complementary spectral responses and spatial structural features. Subsequently, a bidirectional Mamba global modeling module based on selective state space modeling is adopted to capture long-range contextual dependencies in both forward and backward directions with linear computational complexity. Unlike conventional patch-based methods, MambaHSINet takes the entire hyperspectral image as input and produces full-resolution pixel-wise classification maps, thereby avoiding repeated cropping and redundant sliding-window inference. We also construct two well-annotated subsets of a UAV-borne hyperspectral dataset dedicated to tree species classification. Experimental results on self-collected and public hyperspectral datasets demonstrate that the proposed method achieves excellent classification accuracy, inference efficiency, and generalization performance. It exhibits great potential for practical tree species classification and other general hyperspectral application scenarios. Full article

In response to Saudi Arabia’s amended White Land Fees Law, which imposes charges of up to 10% of land value on undeveloped urban plots, this study presents TerraVision, an intelligent framework for large-scale White Land detection and urban land monitoring using high-resolution satellite imagery and deep learning. The proposed framework aims to support sustainable urban development by enabling municipalities and planners to identify underutilized urban land, improve land-use efficiency, and support evidence-based planning decisions. Satellite imagery was acquired through the Esri ArcGIS platform at a spatial resolution ranging from 0.31 to 0.34 m per pixel. The Riyadh study area was divided into 1317 geographic tiles, of which 80 tiles covering approximately 180 km² were manually annotated to construct the training and evaluation dataset. Ten segmentation models representing four architectural families were evaluated, including encoder–decoder networks, transformer-based architectures, YOLO segmentation models, and the zero-shot Segment Anything Model 3 (SAM3). Six fine-tuned semantic segmentation models achieved Intersection over Union (IoU) scores between 0.94 and 0.96 on the held-out test set, with SegFormer achieving the highest performance at an IoU of 0.9563. A post-inference geoprocessing pipeline was developed to reconstruct city-scale prediction maps, estimate neighborhood-level White Land availability, and export results into GIS- and web-compatible formats. The framework was further integrated into a bilingual (Arabic and English) decision-support dashboard that enables visualization and spatial analysis of vacant land distribution. The results demonstrate that semantic segmentation models provide an accurate solution for monitoring undeveloped urban land that scales to city-wide inference across Riyadh, and can support preliminary screening for strategic urban planning and sustainable city development initiatives in Riyadh. Full article

The Qinghai Lake Basin (QLB) serves as a crucial ecological barrier on the Qinghai–Tibet Plateau, making high-precision mapping of land use/land cover (LULC) essential for eco-hydrological research within the basin. In this study, multitemporal Sentinel-1 radar and Sentinel-2 optical imagery from 2024, DEM-derived terrain information, and features derived from these sources were used to produce a 10-m resolution LULC map for the QLB using a support vector machine classifier. The Level-1 and Level-2 LULC datasets of QLB (QLBLC-10) achieved sample-based apparent overall accuracies (OAs) of 91.95% and 91.24%, respectively, and kappa coefficients of 0.90 for both. In contrast, the area-weighted apparent overall accuracy (OA_w) decreased to 81.50 ± 2.09% (95% confidence interval), indicating that class-area imbalance and small-area classes affect map-level performance. The ablation study confirms the contribution of multisource temporal information and terrain constraints to alpine LULC classification. The OA increased from 77.07% with single-temporal Sentinel-2 to 91.24% when multitemporal Sentinel-1/2 data and DEM-derived features were added, while the kappa coefficient increased from 0.75 to 0.90. The comparison with existing products shows that QLBLC-10 outperforms existing global and regional LULC datasets in representing alpine land cover patterns in the QLB. The LULC system proposed in this study is tailored to the QLB, and the presented LULC classification strategy enhances discrimination among major alpine vegetation types, including temperate and alpine steppes, alpine meadows, and alpine shrublands. It provides an up-to-date (2024) LULC dataset for ecosystem monitoring and land management across the QLB. Full article

Reliable and harmonised soil information remains critically limited across Africa, constraining soil monitoring, climate-resilient agriculture, and evidence-based land management. Existing soil resources are often fragmented, spatially uneven, outdated, or derived from legacy observations, limiting their usefulness for contemporary continental-scale assessment. The Soils4Africa project implemented a coordinated field campaign across 33 African countries between 2022 and 2025 to establish a harmonised soil monitoring framework for agricultural lands. Using a hierarchical probabilistic sampling design, 24,951 soil samples were collected from 14,311 locations, supported by standardised field protocols, digital data capture, QR-based sample traceability, and centralised quality control. This paper presents the conceptual, operational, and data-management framework underpinning the survey and reports baseline field observations on farming systems, land management, vegetation structure, and soil physical constraints. The framework achieved more than 70% of planned sampling coverage despite major logistical, environmental, and security-related constraints. Baseline observations show that African agricultural landscapes remain dominated by smallholder systems, low external input use, limited soil and water conservation, and widespread dependence on rainfed production. Field indicators also reveal sparse woody vegetation cover and common physical constraints, including compaction, coarse fragments, shallow effective rooting depth, and subsoil barriers. Unlike earlier continental resources based largely on legacy profiles or site-based surveillance, Soils4Africa provides a contemporary, harmonised, spatially structured field-survey framework designed to support future laboratory-based soil assessment, digital soil mapping, land suitability analysis, and long-term soil monitoring. The study therefore provides a scalable model for coordinated soil monitoring across diverse African agroecosystems and establishes an operational baseline for subsequent analytical studies. Full article

Unplanned urban expansion accompanied by a decline in green infrastructure poses significant challenges for sustainable land-use planning in semi-arid, water-constrained secondary cities. Quetta, Pakistan, exemplifies these challenges due to rapid population growth, ecological degradation, water scarcity, and the absence of an updated master plan. This study develops a GIS-based spatial decision-support framework to evaluate residential development and green infrastructure priorities and to identify areas of conflict, synergy, and balanced planning opportunities. Sentinel-2A imagery acquired in May 2023 was used to generate a land-use/land-cover map, while residential and green-infrastructure suitability factors were standardized using fuzzy membership functions and integrated through an AHP Weighted Linear Combination approach. The resulting Residential Suitability Index (RSI) and Green-Infrastructure Suitability Index (GSI) were normalized and combined through a rule-based Residential–Green Infrastructure Trade-off Index (RGTI). Unlike conventional suitability assessments that evaluate development and ecological priorities independently, the proposed framework explicitly identifies zones of residential dominance, ecological dominance, and shared planning potential. Five planning-priority categories were delineated, comprising Very High Green Infrastructure Priority, Moderate Green Infrastructure Priority, Shared Zone, Moderate Residential Expansion Priority, and Very High Residential Expansion Priority. A spatial consistency assessment demonstrated that the identified planning zones correspond closely with existing land-use patterns and available land resources, supporting the plausibility of the proposed framework. The results provide a practical basis for delineating ecological conservation areas, residential development zones, and integrated planning zones capable of balancing urban growth and environmental sustainability. The framework offers a transparent and transferable approach for supporting land-allocation decisions in arid, data-scarce, and rapidly urbanizing cities facing competing development and ecological pressures. Full article

Accurate monitoring of oil palm plantations is important for balancing economic development with environmental conservation in Southeast Asia. However, existing plantation maps often suffer from low spatial resolution and a lack of recent temporal coverage, impeding effective surveillance of rapid land-use changes. In this study, we propose a deep learning framework to generate 10 m resolution oil palm plantation maps for Indonesia and Malaysia from 2020 to 2024, utilizing Sentinel-2 imagery without requiring new manual annotations. To address the resolution mismatch between coarse 100 m historical labels and 10 m imagery, we employ a U-Net architecture optimized with Determinant-based Mutual Information (DMI). This approach effectively mitigates the influence of label noise. We validated our method against 2058 manually verified points, achieving overall accuracies of 70.64%, 63.53%, and 60.06% for the years 2020, 2022, and 2024, respectively. The gradual decline in accuracy with time is consistent with a growing temporal mismatch between the 2016 historical reference labels and the later prediction years. At the regional scale, the mapped oil palm area suggests a peak in 2022 followed by a lower mapped extent in 2024. Land cover transition analysis further indicates exchanges with cropland and flooded vegetation, which should be interpreted together with the reported accuracy and uncertainty. Given the moderate per-year accuracies and the temporal mismatch between the 2016 supervision and the later prediction years, these results should be interpreted as regional-scale indicators rather than pixel-level change maps. The generated maps can support regional monitoring, sustainability assessment, and prioritization of areas for further validation. Full article