Seasonal weather prediction as an initial value problem

Using a climate version of a regional atmospheric model, we show that the s easonal evolution of weather is dependent on the initial soil moisture and landscape specification. Coupling this model to a land-surface model, the s oil moisture distribution and landscape are shown to cause a significant no nlinear interaction between vegetation growth and precipitation. These resu lts demonstrate that seasonal weather prediction is an initial value proble m. Moreover, on seasonal and longer timescales the surface characteristics such as soil moisture, leaf area index, and landcover type must be treated as dynamically evolving dependent variables, instead of prescribed paramete rs.