Quantifying Observed and Modeled Evapotranspiration Uncertainty in California Woody Perennial Crops

Evapotranspiration (ET) plays a critical role in California's water management strategies, particularly under emerging groundwater regulations. Despite its importance, quantifying ET is challenging due to intricate atmospheric-land surface biophysical interactions and limitations in observational techniques, which are often constrained to small areas not well suited for heterogeneous landscapes. While modeling methods offer some solutions, they require extensive input variables and empirically calibrated parameters. Satellite and airborne remote-sensing technologies have advanced significantly over the past three decades, offering ET estimations over large areas at frequent intervals. However, these techniques, including those used to estimate ET based on near-ground sensors (e.g. eddy covariance method), often yield variable results, contributing to ET estimate uncertainty. This study aims to quantify this uncertainty by leveraging eddy covariance observations from the Grape Remote sensing Atmospheric Profile and Evapotranspiration eXperiment (GRAPEX) and the Tree crop Remote Evapotranspiration eXperiment (T-REX). We employ various micrometeorological approaches to assess ground-based ET. Likewise, airborne retrievals from these projects are used to model ET using different models such as the Two Source Energy Balance (TSEB) and the Model and Mapping EvapoTranspiration at high Resolution and Internalized Calibration (METRIC). Satellite-based ET products from the OpenET collaborative (https://openetdata.org) will also be used to evaluate modeling uncertainties across different methodologies. Our preliminary results aim to quantify the magnitude of ET uncertainty and identify contributing factors, including timing, landscape, weather, and other environmental conditions. These insights are crucial as we transition towards a new generation of ET tools for efficient water management.