Assessing the climate sensitivity of the global terrestrial carbon cycle model SILVAN

The growth rate of the atmospheric CO2 concentration exhibits interannual anomalous variations of 1–2 ppmV yr-1 which reflect the response of the global carbon fluxes to large scale climate fluctuations. The climate sensitivity of global carbon cycle models can be explored by the simulation of these variations. Here we test the climate sensitivity of the global terrestrial carbon cycle model SILVAN 2.3 using this approach. The model has a horizontal resolution of 0:5o, a 6–day time step and considers potential vegetation only. Important features are a model–generated water balance and physiological approaches to determine net primary productivity (NPP) and phenology. In the three sensitivity experiments SILVAN 2.3 was forced in addition to the monthly climatologies by: (A) observed temperature anomalies 1854–1993, (B) observed precipitation anomalies 1900–1993, and (C) observed anomalous temperature and precipitation as well as the atmospheric CO2 concentration increase 1765–1993. Simulated and observed anomalous CO2 fluxes into the atmosphere 1958–1993 are well correlated. The largest fraction of the modelled anomalous CO2 fluxes results from the temperature sensitivity of the physiological NPP model; the effect of the precipitation variations is relatively small. The simulated heterotrophic respiration is more sensitive to precipitation than to temperature. We discuss the extent to which the model response results additively from the anomalous CO2 fluxes generated by the temperature or precipitation anomalies only.