Improving Orchidee Simulations of Surface Energy Fluxes Assimilating the ESA-CCI LST Product

We assimilate the ESA-CCI Land Surface Temperature (LST) into the ORCHIDEE land surface model to optimize key parameters, improving the simulations of LST and surface energy fluxes. We first conduct a series of synthetic twin data assimilation experiments to find the best strategy for assimilating LST. For this purpose, we notably test: i) two methods of optimization (genetic and gradient-based) and ii) different diurnal cycle LST features and the 3-hourly LST observations assimilated into ORCHIDEE. We select the best strategy to assess in a second step the benefit of assimilating ESA-CCI LST data across 34 FluxNet sites in Europe. We demonstrate the effectiveness of this assimilation process, improving the accuracy of fluxes simulation (against in-situ measurements) up to 60% of RMSD reduction with a median of 20%. Future works will aim to refine the utilization of uncertainties provided by the ESA-CCI LST product in the ORCHIDEE assimilation system (ORCHIDAS).