This study presents the model results using the NCEP Global Forecast System (GFS). The NCEP EMC is among the AGCM groups participating in the Global Land-Atmosphere Coupling Experiment (GLACE), an AGCM inter comparison study aimed to examine the degree to which the atmosphere responds to anomalies in land surface states. A total of three ensembles of GFS simulations are conducted for the 1994 boreal summer using observed SST throughout the integration. The first set is a typical AMIP runs. For the second set, each member is forced to maintain the same time series of surface prognostic variables. The third set is same as the second set except the surface variables are allowed to evolve freely, as only root zone soil moisture is forced to be identical among the member simulations.

The impact of prescribed land states on near surface fields is evident, e.g., the time series of latent heat flux and near-surface temperature show great coherence among the members. However, the evolution of precipitation among the members shows a broad disparity and only a small fraction of the precipitation variance can be explained by soil moisture anomalies. In fact, the NCEP GFS is among these AGCMs that show very weak response of precipitation to prescribed land states. The lack of sensitivity of precipitation to soil moisture constraints is likely due to the inherent characteristics of the land surface scheme and how the moisture convection is formulated in the GFS.