<p>To bypass the thermal data requirement for actual evapotranspiration (ET<sub>a</sub>) estimation in satellite remote sensing, two general approaches have been taken into practice based on previous efforts: (1) Multi-sensor data fusion for thermal sharpening and (2) the use of the process-based models such as the Penman-Monteith and Shuttleworth-Wallace equations augmented with satellite-based crop parameters. To address this issue, this study introduced an optical satellite data-based ET<sub>a</sub> estimation model, OPTRAM-ET, based on the optical trapezoid model (OPTRAM) estimates of soil moisture. The new model has been applied to Sentinel-2 and Landsat-8 images over 16 eddy covariance flux towers in the United States and Germany. The flux towers were chosen in a way to cover different ranges of landcover types, e.g., agriculture, orchard, permanent wetlands, and foothill forests. In order to assess the model in comparison to a thermal-based conventional method, the land surface temperature (LST)-vegetation index (VI) model was utilized. The results of the proposed OPTRAM-ET model showed promising performance in all the studied regions. While agricultural sites showed higher correlation due to their wider range of ET<sub>a</sub> values, error indicators were lower in foothill forests because soil moisture changes were smaller compared to irrigated and wet lands. In addition, the OPTRAM-ET model showed comparable performance to the conventional LST-VI model. The OPTRAM-ET model however does not need thermal data, and it benefits from higher spatial and temporal resolution data provided by ever-increasing drone- and satellite-based optical sensors to predict crop water status and demand. It is worth noting that the thermal sharpening step was excluded in this model which subsequently makes the model substantially less computationally demanding than a thermal-based model. Unlike the LST-VI model, which needs to be calibrated for each satellite image, a temporally-invariant region-specific calibration is possible in the OPTRAM-ET model. Importantly, the model requires further enhancement due to limitations caused by the simplified basic assumptions.</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.