Meteorological determinants of growing season onset in grasslands

The exchange of the trace gases between the land and atmosphere is highly influenced by vegetation. Therefore, the representation of phenological properties in global carbon models plays a key role in understanding and predicting the global carbon cycle. Phenological parameters such as Leaf Area Index (LAI) and fraction of photosynthetically active radiation absorbed (fPAR) are often calculated or estimated based on remote sensing measurements, which can be biased by clouds, aerosols, or snow. Alternatively, we can prognose vegetation phenology through the use of models that predict vegetation status based on meteorological conditions. Here our goal is to provide better understanding of carbon dynamics as a function of phenological parameters and their dependence on meteorological forcing. We evaluate phenological characteristics and their influence on carbon dynamics at several grassland sites. Modeled carbon flux, as a function of both diagnosed (from satellite) and prognosed phenological state are confronted with data from flux towers. Remotely-sensed phenology has a tendency to expand the growing season, and does not reflect the rapid response of vegetation in rain-green biomes as well as the prognostic phenology model does. These differences in phenology are reflected in modeled fluxes of energy, moisture, and carbon.