Reliable Evapotranspiration Predictions with a Probabilistic Machine Learning Framework

Evapotranspiration is often expressed in terms of reference crop evapotranspiration ($E{T}_o$), actual evapotranspiration ($E{T}_a$), or surface water evaporation ($E_{sw}$), and their reliable predictions are critical for groundwater, irrigation, and aquatic ecosystem management in semi-arid regions. We demonstrated that a newly developed probabilistic machine learning (ML) model, using a hybridized “boosting” framework, can simultaneously predict the daily $E{T}_o$, $E_{sw}$, & $E{T}_a$ from local hydroclimate data with high accuracy. The probabilistic approach exhibited great potential to overcome data uncertainties, in which $100\%$ of the $E{T}_o$, $89.9\%$ of the $E_{sw}$, and $93\%$ of the $E{T}_a$ test data at three watersheds were within the models’ $95\%$ prediction intervals. The modeling results revealed that the hybrid boosting framework can be used as a reliable computational tool to predict $E{T}_o$ while bypassing net solar radiation calculations, estimate $E_{sw}$ while overcoming uncertainties associated with pan evaporation & pan coefficients, and predict $E{T}_a$ while offsetting high capital & operational costs of EC towers. In addition, using the Shapley analysis built on a coalition game theory, we identified the order of importance and interactions between the hydroclimatic variables to enhance the models’ transparency and trustworthiness.