Seasonal variation of photosynthetic model parameters and leaf area index from global Fluxnet eddy covariance data

Global vegetation models require the photosynthetic parameters, maximum carboxylation capacity (V(cm)), and quantum yield (alpha) to parameterize their plant functional types (PFTs). The purpose of this work is to determine how much the scaling of the parameters from leaf to ecosystem level through a seasonally varying leaf area index (LAI) explains the parameter variation within and between PFTs. Using Fluxnet data, we simulate a seasonally variable LAI(F) for a large range of sites, comparable to the LAI(M) derived from MODIS. There are discrepancies when LAI(F) reach zero levels and LAI(M) still provides a small positive value. We find that temperature is the most common constraint for LAI(F) in 55% of the simulations, while global radiation and vapor pressure deficit are the key constraints for 18% and 27% of the simulations, respectively, while large differences in this forcing still exist when looking at specific PFTs. Despite these differences, the annual photosynthesis simulations are comparable when using LAI(F) or LAIM (r(2) = 0.89). We investigated further the seasonal variation of ecosystem-scale parameters derived with LAI(F). V(cm) has the largest seasonal variation. This holds for all vegetation types and climates. The parameter alpha is less variable. By including ecosystem-scale parameter seasonality we can explain a considerable part of the ecosystem-scale parameter variation between PFTs. The remaining unexplained leaf-scale PFT variation still needs further work, including elucidating the precise role of leaf and soil level nitrogen.