Towards the assimilation of microwave vegetation optical depth into global land surface models

Abstract

Only a few recent studies have considered assimilating microwave vegetation optical depth (VOD) to improve estimates of vegetation states in land surface models (LSMs) with dynamic vegetation. These studies showed that VOD assimilation can improve estimates of GPP, ET, and CO2 concentrations. We intend to assimilate VOD from merged products, as well as from multiple individual passive and active microwave sensors, using the NoahMP and SURFEX-ISBA LSMs. This poses several challenges, in particular developing observation operators for VOD for the different sensors, or characterizing model and observation error structures. We will present various linear and nonlinear approaches to map variables simulated by the models to VOD observations. These will be evaluated with respect to their performance with merged VOD products from the VOD Climate Archive (VODCA) for different bands (Ku/X/C/L), and on single-sensor VOD retrievals from LPRMv6. Additionally, we will assess the added value of including plant water status related variables compared to using only optical vegetation indices as predictor variables for VOD. Acknowledgements: This work is part of the projects CONSOLIDATION, funded by the Austrian Science Fund (FWF), and the CCN1 CCI+ Soil Moisture Scientific Evolution, funded by ESA.