A comprehensive approach to building a continuous hydrologic model with Soil Moisture Accounting using Earth Observation data

Abstract

In recent years, both availability and interest in Earth Observations (EO) have increased due to their ability to provide information with extensive spatial coverage, which is valuable for data-scarce regions. This study provides a roadmap to exploit EO/satellite data (EO-based model) for continuous hydrological simulation using the Hydrologic Engineering Center–Hydrologic Modelling System model with a soil moisture accounting (SMA) component. As a case study, we consider the Boeotikos Kephisos River basin, in Greece. The SMA component is calibrated using the HiHydroSoils dataset, to map its parameters to the regionally varying soil hydraulic properties and a comparison is made to an alternative parameterization, following the standard literature approach (literature-based model). The effectiveness of satellite data in enhancing model performance is further assessed, by comparing three different satellite precipitation datasets, as model drivers, and by using satellite-based soil moisture for model initialization. The discussion extends to the potential for integration of EO/satellite data at the operational level, by simulating a significant precipitation event. Yet, the most promising result pertains to the opportunity to exploit satellite-derived estimates of soil hydraulic properties to base the calibration of the data-intensive SMA scheme, with the EO-based model significantly outperforming the literature-based model parameterization.