|Title||Analysis of Leaf Area Index in the ECMWF land surface scheme and impact on latent heat and carbon fluxes: Applications to West Africa|
|Publication Type||Technical memorandum|
|Secondary Title||ECMWF Technical Memoranda|
|Authors||Jarlan, L, Balsamo, G, Lafont, S, Beljaars, A, Calvet, J-C, Mougin, E|
Within the Observatory of Natural Carbon of the GEOLAND project (GEOLAND, 2004), the land surface model of the European Centre for Medium Range Weather Forecast (hereafter CTESSEL) has been modified to include an interactive vegetation module. Following this improvement, a land data assimilation system has been developed and the performance of the assimilation of Leaf Area Index (LAI) satellite products is presented in this paper. The CYCLOPES and MODIS LAI satellite derived data sets are compared together with the open-loop simulation of the CTESSEL model. The land data assimilation scheme is based on a simplified 2DVAR. It is theoretically evaluated thanks to a twin experiment. The 2DVAR is then tested over an experimental site of the AMMA project located in Mali. Simulations after and before data assimilation are confronted to ground measurements of LAI and above-ground biomass. The MODIS products are finally used to analyse the CTESSEL LAI over a large window covering the West Africa from 2001 to 2005. The data assimilation of LAI leads to a significant improvement of the simulated phenology of the vegetation. In addition, it is shown that the impact of the LAI data assimilation on Latent Heat Fluxes is limited whereas LAI is one of the main driving force for the CO2 Net Ecosystem Exchanges (NEE). It is concluded that a LAI climatology is sufficient to simulate latent heat fluxes with acceptable errors whereas an interactive LAI coupled to a data assimilation system of satellite derived products is necessary for NEE fluxes.