Can Remotely Sensed Actual Evapotranspiration Facilitate Hydrological Prediction in Ungauged Regions Without Runoff Calibration?

Runoff prediction in ungauged catchments is a significant hydrological challenge. The common approach is to calibrate hydrological models against streamflow data from gauged catchments, and then regionalize or transfer parameter values from the gauged calibration to predict runoff in the ungauged catchments. This paper explores the potential for using parameter values from hydrological models calibrated solely against readily available remotely sensed evapotranspiration data to estimate runoff time series. The advantage of this approach is that it does not require observed streamflow data for model calibration and is therefore particularly useful for runoff prediction in poorly gauged or ungauged regions. The modeling experiments are carried out using data from 222 catchments across Australia. The results from the remotely sensed evapotranspiration runoff‐free calibration are encouraging, particularly in simulating monthly runoff and mean annual runoff in the wetter catchments. However, results for daily runoff and in the drier regions are relatively poor, and further developments are needed to realize the benefit of direct model calibration against remotely sensed data to predict runoff in ungauged catchments. Direct calibration of hydrological models solely against remotely sensed ET (RS‐ET) data is used to predict runoff in ungauged regionsThe RS‐ET runoff‐free calibration approach is tested on 222 catchments across AustraliaThe RS‐ET runoff‐free calibration can estimate mean annual runoff and monthly runoff series in wet catchments reasonably well