A simple model for estimating the sensitivity of runoff to long-term changes in precipitation without a change in vegetation

Forecasts of changes in precipitation (P) and potential evaporation (PE) ca n be applied to hydrologic models calibrated on existing conditions to obta in predictions of changes in runoff. This study describes an alternative ap proach, which uses a simple soil-moisture accounting model with a small num ber of independent and physically based parameters to explore the sensitivi ty of runoff to climate change for three simplified climates. The climate t ypes chosen initially are those for which a piecewise analytical solution c an be obtained so that computer programmes involving numerical solutions ca n be verified before being applied to field data. Sensitivity factors are c alculated for the various cases and their relationships with climatic condi tions and soil conditions are explored. Breakpoint Values were determined f or each type of climate studied. These correspond to situations in which th e soil becomes momentarily saturated once during the seasonal cycle but doe s not remain saturated for any finite duration. For humidity ratios greater than the breakpoint, the sensitivity of runoff to precipitation increases abruptly. For the climates studied, the sensitivity factor approaches the v alue of the soil parameter c as the humidity index approaches zero. The oth er climates studied exhibit the same sensitivity at this limit. A particula r feature of the model is that analytical solutions can be determined in ma ny cases to check and confirm the results of the numerical simulations. (C) 1999 Elsevier Science Ltd. All rights reserved.