Can dynamic vegetation information improve the accuracy of Budyko's hydrological model?