Towards sub-daily satellite-based interception loss estimates

Abstract

Terrestrial evaporation (E) plays a crucial role in the Earth system, acting as a link between the water and carbon cycles and playing a major role at the complex interplay between land and atmosphere. Therefore, accurate monitoring of E and its different components is crucial. However, since E cannot be observed directly from satellite sensors, current E retrieval algorithms are largely indirect and satellite-based E estimates remain highly uncertain, especially in what respects to the partitioning of evaporation into its different components. In particular interception loss (Ei), the volume of precipitation captured by plant surfaces and evaporated back into the atmosphere without reaching the ground, remains one of the most uncertain components in the global water balance. Moreover, current existing E datasets only deliver daily satellite-based Ei estimates, being unable to resolve precipitation event scales.  The research presented here is focused on estimating Ei on a sub-daily scale. To do so, the Global Land Evaporation Amsterdam Model (GLEAM; Miralles et al., 2011) is used. GLEAM is an E model that simulates the different components (transpiration, soil evaporation, interception loss) using satellite data, including microwave observations of surface soil moisture and vegetation optical depth (VOD). We adapted GLEAM to function at sub-daily resolution, by (1) relying on sub-daily satellite-based forcing data and (2) extending the recent interception model presented by Zhong et al. (2022) by following a Rutter approach (Rutter et al., 1975) to make it applicable at sub-daily scales. This interception model calculates a running balance in time of rainfall, throughfall, evaporation and changes in canopy storage, whereby Ei is the evaporation from the wet canopy. The model is driven by satellite-observed vegetation dynamics, potential evaporation and precipitation. The sub-daily Ei estimates are compared to existing daily estimates and diurnal cycles are analyzed, and this at different spatial scales.