Accurate evaluation of water resource systems is essential for informed planning and decision-making. Evapotranspiration (ET), a key component of water resource management, is often estimated using remote sensing techniques; however, such estimates can be subject to significant uncertainties under certain conditions. In this study, we present a novel approach to improving the accuracy of ET estimates in composite terrains. The methodology involves optimizing the Surface Energy Balance Algorithm for Land (SEBAL-OPT) by integrating ground-based eddy covariance (EC) flux tower data into the satellite-based ET retrieval process. The approach was evaluated at four sites in California, each representing different land uses. Parameter optimization was achieved through Bayesian inference using the Differential Evolution Adaptive Metropolis (DREAM) algorithm, which minimized discrepancies between ET estimates derived from Landsat 8 and 9 imagery and the observed ET from EC measurements. Results from the global sensitivity analysis identified solar radiation and hot/cold pixel selection as the most sensitive parameters in the SEBAL algorithm, highlighting their critical role in reducing uncertainty in ET estimates. SEBAL-OPT demonstrated significantly improved accuracy, with root mean square error (RMSE) values ranging from 0.72 mm to 1.33 mm, compared to the original SEBAL parameterization (SEBAL-ORG), which produced RMSE values between 1.03 mm and 2.14 mm. This approach highlights that, when properly calibrated, the model can be effectively applied across diverse agricultural landscapes, regardless of the specific land use at individual sites. These findings have significant implications for water resource planning, agricultural water management, and water rights adjudication and could be applied to other remote sensing of ET models.@en

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.

@en