Investigation of the Effects of Dynamic Vegetation Processes on Global Climate Simulation Using the NCEP GFS and SSiB4/TRIFFID

To investigate the effects of dynamic vegetation processes on numerical climate simulation, two experiments are conducted globally by integrating  the National Centers for Environmental Prediction Global Forecast System (GFS) coupled with a biophysical model, simplified simple biosphere model (SSiB) version 2 (referred to as GFS/SSiB2) and with a biophysical and dynamic vegetation model, SSiB version 4/TRIFFID (referred to as GFS/SSiB4/TRIFFID) from 1948 to 2008. By assessing GFS/SSiB4T/TRIFFID and GFS/SSiB2 results against satellite-derived leaf area index (LAI) and albedo, as well as observed land surface temperature and precipitation, we identify the effects of dynamic vegetation processes on the simulations of precipitation, near-surface temperature and the surface energy budget at monthly and annual scales. The results show that compared to the GFS/SSiB2, the temporal correlation coefficients between globally averaged monthly simulated LAI and the GIMMS/GLASS LAI in the GFS/SSiB4/TRIFFID increase from 0.31/0.29 (SSiB2) to 0.47/0.46 (SSiB4). Meanwhile, the correlation coefficients between simulated and observed monthly mean near-surface air temperature increase from 0.58 (over Africa), 0.47 (over Southeast Asia) and 0.50 (over South America) to 0.66, 0.55 and 0.58 respectively. While the correlation coefficients between model-simulated and observed monthly mean precipitation increase from 0.31 (over Africa), 0.41 (over East Asia) and 0.21 (over Australia) to 0.38, 0.47 and 0.24 respectively. The most improvement occurs over arid and semi-arid areas. The spatial-temporal variability and changes in vegetation and ground surface albedo modeled by GFS with dynamic vegetation model are more consistent with observations, which contribute to the surface energy and water balances, and in turn improve the annual variations in simulated regional temperature and precipitation. The dynamic vegetation processes have the greatest influences on the temporal and spatial changes of latent heat fluxes. This study demonstrates that the dynamic vegetation processes in the Earth System model are capable to improve the climate mean status simulation significantly.