In this paper gross primary production (GPP) predicted by FORUG is compared with GPP calculated from eddy covariance measurements for a beech forest in France (Hesse). Two photosynthesis formulations at leaf level are compared: the biochemically supported approach described by Farquhar et al. [Farquhar, G.D., Von Caemmerer, S., Berry, J.A., 1980. A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta 12B, 549–587] (BCF) and an empirical light response curve (LRC). Five consecutive years (1996–2000) of measured GPP are compared with FORUG model predictions. Results did not discriminate between both model formulations, but good agreement between modelled and measured GPP support the reliability of FORUG for both photosynthesis approaches. Although some discrepancies appeared, the parameterization combining literature and fitted parameters can be considered as a useful strategy. Residualswere analysed to find explanations for discrepancies between model predictions and data. The increase in residuals over the years, indicate that interannual variability of GPP is not only determined by direct climatic effects. Due to interfering long-term effects, a combination of several climatic factors (drought, temperature), acclimation, environmental and management impacts account for the interannual variation in GPP. However, the longterm effect of drought appeared to be the most important driver of the interannual variability in GPP. Taking into account these long-term climate effects will be an essential step in development of better performing ecosystem models.