The model was calibrated using hourly data from a 2.5-year period of observation in the river basin from June 1996 through November 1998. The simulated stream discharge for the two water years of 1997 and 1998 is consistent with the observed hydrograph of the basin although the Karst geology of the basin is the likely cause of the minor differences between the hydrograph and the simulation. The two-year simulation result was used as the baseline in comparison to model results using climate and land-cover change scenarios developed from historical climate data and from the possible vegetation changes in the basin.
Results of the study suggest that even a small percentage change in land-cover in the basin can have a drastic impact on annual stream discharge. Reduction of forested land in the basin can increase the frequency and/or magnitude of extreme hydrological events, i.e., floods and low flow during droughts, in the basin. Compared to the land-cover change effect, climate change has a relatively small influence on stream discharge and its annual distribution. In addition, any climate effect will result from the slow process of climate change as compared to the often rapid process of land-use change. Results from further investigation of the stream discharge change using an interactive relationship between changes of climate and vegetation physiology in the basin is presented.