Obtaining evapotranspiration (ET) estimates at high spatiotemporal resolution is a fundamental prerequisite for clarifying the patterns and controlling factors of agricultural water consumption in drylands. However, most existing ET products are provided at daily or coarser spatial–temporal scales, which limits the ability to capture short-term variations in crop water use. This study developed a novel hourly 10-m ET estimation method that combines remote sensing with machine learning techniques. The approach was evaluated using agricultural sites in two arid regions: the Heihe River Basin in China and the Bekaa Valley in Lebanon. By integrating hourly eddy covariance measurements, Sentinel-2 reflectance data, and ERA5-Land reanalysis meteorological variables, we constructed an XGBoost-based modeling framework for hourly ET estimation, and incorporated the SHapley Additive exPlanations (SHAP) method for model interpretability analysis. Results demonstrated that the model achieved strong performance across all sites (R² = 0.86–0.91, RMSE = 0.04–0.05 mm·h⁻¹). Additional metrics, including the Nash–Sutcliffe efficiency coefficient (NSE) and percent bias (PBIAS), further confirmed the model’s robustness. Interpreting the model with SHAP highlighted net radiation (Rn), 2-m temperature (t2m), and near-infrared reflectance of vegetation (NIRv) as the dominant factors controlling hourly ET variations. Significant interaction effects, such as Rn × NIRv and Rn × t2m, were also identified, revealing the modulation mechanism of energy, vegetation status and temperature coupling on hourly ET. The study offers a practical workflow and an interpretable framework for generating high-resolution ET maps, thereby supporting regional water accounting and land–atmosphere interaction research.

Wheat is a vital staple crop worldwide. However, its production has been increasingly challenged in Lebanon in recent years by multi-factorial crises intersecting socio-economic, political, security, and climate factors. This study provides the first national, multi-year monitoring of wheat area in Lebanon (2017–2025), linking satellite observations with crisis impacts. We conducted a multi-temporal supervised classification across the 2017–2018 to 2024–2025 seasons, using Google Earth Engine, Sentinel-2 imagery, and Random Forest classifier, achieving an 87% overall accuracy. We then estimated wheat area based on a random stratified sample. Interannual changes were related to major crises and input-price dynamics. Wheat area increased during 2019–2021 but dropped sharply in 2021–2022 as subsidies weakened and input costs surged. Indeed, during the transition toward economy dollarization, production costs in USD peaked in 2021–2022 and then eased consistently with the 2022–2023 area rebound. In 2023–2025, the crop area decreased again dramatically (−34% in 2023–2024 and −38% in 2024–2025) due to the conflict with Israel and associated widespread displacement that likely constrained field access and reduced sowing, particularly in southern Lebanon. Ultimately, our findings highlight the critical need for ground-truth data for accurate area estimation, alongside yield and socio-economic metrics to better understand interannual wheat area variability.

Wheat stands as one of the most important staple crops worldwide. However, the vital role of this crop has been increasingly challenged in Lebanon, in recent years by multi-factorial crises from socio-economic, political, security and climate factors, threatening agricultural stability and food supply. Consequently, monitoring wheat production is crucial for managing import and export activities, developing effective policies, achieving resilient agricultural development, and ensuring food security. This study provides the first national, multi-year monitoring of wheat area and production in Lebanon (2017-2025), linking satellite observations with crisis impacts. We conducted a multi-temporal supervised classification from 2017-2018 to 2024-2025 seasons, usinging Sentinel-2 optical images and Random Forest classifier. We estimated wheat area based on a random stratified sample achieving an overall accuracy of 87%. Interannual changes were then related to major crises and input-price dynamics. Wheat area increased during 2019-2021 but dropped sharply in 2021-2022 as subsidies weakened and input costs surged. Indeed, during the transition toward economy dollarization, the computed indicator of production cost expressed in USD peak in 2021-2022 and then ease consistently with the 2022-2023 area rebound reaching the highest level observed during the study period. In 2023-2025, the crop area decreased again dramatically (-34% in 2023-2024 and -38% in 2024-2025) in relation to the conflict with Israel and associated widespread displacement of population that likely constrained field access and reduced sowing, particularly in southern Lebanon. The derived wheat cover maps were then used as input of the agronomic model AgriCarbon-EO to assess biomass and grain yields variability.