Scarce water resources present a major hindrance to ensuring food security. Crop water productivity (WP), embraced as one of the Sustainable Development Goals (SDGs), is playing an integral role in the performance-based evaluation of agricultural systems and securing sustainable food production. This study aims at developing a cloud-based model within the Google Earth Engine (GEE) based on Landsat -7 and -8 satellite imagery to facilitate WP mapping at regional scales (30-m resolution) and analyzing the state of the water use efficiency and productivity of the agricultural sector as a means of benchmarking its WP and defining local gaps and targets at spatiotemporal scales. The model was tested in three major agricultural districts in the Lake Urmia Basin (LUB) with respect to five crop types, including irrigated wheat, rainfed wheat, apples, grapes, alfalfa, and sugar beets as the major grown crops. The actual evapotranspiration (ET) was estimated using geeSEBAL based on the Surface Energy Balance Algorithm for Land (SEBAL) methodology, while for crop yield estimations Monteith’s Light Use Efficiency model (LUE) was employed. The results indicate that the WP in the LUB is below its optimum targets, revealing that there is a significant degree of work necessary to ameliorate the WP in the LUB. The WP varies between 0.49–0.55 (kg/m³) for irrigated wheat, 0.27–0.34 for rainfed wheat, 1.7–2.2 for apples, 1.2–1.7 for grapes, 5.5–6.2 for sugar beets, and 0.67–1.08 for alfalfa, which could be potentially increased up to 80%, 150%, 76%, 83%, 55%, and 48%, respectively. The spatial variation of the WP and crop yield makes it feasible to detect the areas with the best and poorest on-farm practices, thereby facilitating the better targeting of resources to bridge the WP gap through water management practices. This study provides important insights into the status and potential of WP with possible worldwide applications at both farm and government levels for policymakers, practitioners, and growers to adopt effective policy guidelines and improve on-farm practices.

High-resolution spatial–temporal root zone soil moisture (RZSM) information collected at different scales is useful for a variety of agricultural, hydrologic, and climate applications. RZSM can be estimated using remote sensing, empirical equations, or process-based simulation models. Machine learning (ML) approaches for evaluating RZSM across numerous spatial–temporal scales are less generalizable than process-based models. However, data-driven ML approaches offer a unique opportunity to develop complex models of soil moisture without making assumptions about the processes governing soil water dynamics in a given study region. In this study, comparisons were made between two models, pySEBAL and EFSOIL, which were based on evaporation fraction (EF) and soil properties, and a data-driven model based on the Random Forest (RF) ensemble algorithm. These approaches were evaluated to demonstrate their capabilities for RZSM estimation. The EF obtained from Landsat images was used after validation with eddy covariance measurements as the major input to all three models, along with other meteorological and soil physical properties. The RF model was trained using in situ soil moisture data from Time Domain Reflectometry (TDR) sensors installed in a vineyard from 2018 to 2020. The predictor variables comprised of meteorological, soil properties, EF, and a vegetation index. The results reveal that there was a strong correlation between the in situ measured soil moisture and the RF predicted soil moisture at all sensor locations. Due to the complexity of the physical processes involved in soil water flow, the empirical models pySEBAL and EFSOIL were unable to reliably predict RZSM values at all monitored locations. The high RZSM predicted by pySEBAL demonstrated the presence of possible bias in the model’s algorithm used to estimate soil moisture. We also demonstrated that ML based on the RF algorithm may be used to predict spatially distributed RZSM when a few soil moisture ground measurements are combined with remote sensing to produce soil moisture maps.