Abstract
In this study, large‐scale atmospheric variables are downscaled to meteorological variables at local scale for the daily time step to assess hydrological impacts by climate changes. Large‐scale atmospheric modelling was by the HadCM3 General Circulation Model (GCM) while downscaling and water balance modelling was through the Statistical DownScaling Model and the HBV semi‐distributed rainfall‐runoff model, respectively. The area of study was the Gilgel Abay catchment that drains in Lake Tana. A selection of large‐scale atmospheric variables by the HadCM3 GCM are downscaled by a multiple linear regression model, were minimum and maximum temperature and precipitation for future time horizons are calculated. Climate scenarios as developed for the A2 (medium–high emission) and B2 (medium–low emission) scenarios for a 100‐year period based on the mean of 20 ensembles have been selected for this study. In addition, a synthetic incremental scenario was tested for a wide range of changes in climatic variables. Stream flow simulations by the HBV model were carried out for the 2020s (2011–2040), 2050s (2041–2070) and 2080s (2071–2099) to define hydrologic impacts. The result of downscaled precipitation reveals that precipitation does not manifest a systematic increase or decrease in all future time horizons for both A2 and B2 scenarios unlike that of minimum and maximum temperature and related evaporation. For the future horizons significant changes and variations in the seasonal and monthly flows are to be expected and for the 2080s the runoff volume in the rainy season will reduce by approximately 11·6 and 10·1% for the A2 and B2 scenarios. Results from synthetic incremental scenarios also indicate sensitivities to climate change. As much as 33% of the seasonal and annual runoff is expected to reduce when temperature increases by 2 °C and when rainfall decreases by approximately 20%. Copyright © 2009 John Wiley & Sons, Ltd.