The Core Service biogeophysical Parameter (CSP) uses the CYCLOPES LAI derived from VEGETATION data to elaborate customized product according to the Observatory Natural Carbon (ONC) requirements. ONC needs 0.5° x 0.5° tiled averages (mean and standard deviation values) maps of LAI for 8 vegetation classes ("conifer evergreen forest", "deciduous forest", "broadleaf evergreen forest","grassland C3", "grassland C4", "crops C3", "crops C4"). The below animation displays variations of mean LAI for the 8 vegetation types. LAI is also provided to the Observatory Food Security and Crop Monitoring (OFM) for crop yield studies.

Global LAI at 0.5° resolution derived from VEGETATION measurements for years 2002 and 2003

The ONC partners, CEA/LSCE and ECMWF, used also the version 3 of CYCLOPES LAI to analyse the impact of satellie products assimilation on simulation of carbon fluxes by different models : ORCHIDEE model (Krinner et al., 2005) at CEA/LSCE, and CTESSEL model (Lafont et al., 2007) at ECMWF.
At CEA/LSCE, the method relies on the correction, every 10 days, of model forecast by the satellite data (and its associated error if available). Maps below show the impact of CYCLOPES and MODIS LAI on the Gross Primary Productivity (GPP) forecast over Europe. Differences are located over same areas, and their magnitude depends on the assimilated satellite product. Thus, over the boreal forest where the CYCLOPES LAI is lower than the MODIS LAI, the yearly GPP is strongly decreased (Demarty et al., 2007).

Impact of LAI assimilation (MODIS LAI at left, CYCLOPES LAI at right) in the ORCHIDEE model on the GPP (gC.m-2.y-1) assessment for the year 2001.

At ECMWF, prior to the data assimilation, the satellite derived LAI data sets (CYCLOPES LAI and MODIS LAI) are compared to the CTESSEL open-loop simulations over the West Africa area of the AMMA project. The satellite products are significantly lower than the CTESSEL simulations over the whole study region. The CTESSEL simulations exhibit a shifted phenology with regards to satellite derived LAI. The growth is quite low and the senescence is late over Sahel and savannah. Before assimilating them, the negative bias of the satellite derived LAI is reduced thanks to a histogram adjustment method. The data assimilation method is a simplified 2DVAR which presents the advantage to avoid the development of the adjoint and the tangent linear of the CTESSEL model. This study is presented in Jarlan et al., 2007. First,over the local sahelian site of the AMMA project in Mali, with regards to the open-loop, the CTESSEL simulations constrained by the analysis of LAI are more in agreement with the ground measurements in terms of LAI and above ground biomass. Second, the simplified 2DVAR approach is applied to West Africa. The slow growth and the late senescence of the CTESSEL open-loop LAI simulations are strongly improved thanks to the data assimilation system. The impact of the data assimilation is much higher on CO2 fluxes than on the latent heat flux LE. This analysis concludes that whereas an interactive vegetation module seems a requisite to simulate CO2 fluxes variability, being able to simulate a LAI responding to climate is of secondary importance for LE.

Demarty, J., F. Chevallier, P. Ciais, A.D. Friend, N. Viovy and L. Jarlan, Assimilation of satellite-derived biophysical products within a terrestrial biosphere model : A study case over Europe, to be submitted to Remote Sensing of Environment, 2007.

Jarlan, L., G. Balsamo, S. Lafont, A. Beljaars, J.C. Calvet, and E. Mougin, Analysis of Leaf Area Index in the ECMWF land surface scheme and impact on latent heat and carbon fluxex : Application to West Africa, submitted to Journal of Geophysical Research, 2007.

Krinner, G., N. Viovy, N. de Noblet, J. Ogée, P. Friedlingstein, P. Ciais, S. Sitch, J. Polcher, and I.C. Prentice, A dynamic global vegetation model for studies of the coupled atmosphere-biosphere system. Global Biogeochemical Cycles 19 : Art. No. GB1015 FEB262005. 10.1029/2003GB002199, 2005.

Lafont, S., M. Voogt, A. Beljaars, B.J.J.M. Van den Hurk, L. Jarlan, and G. Balsamo, Description and evaluaation of the Carbon version of TESSEL : CTESSEL model, ECMWF technical memorandum, 37p., 2007.