Prediction of future climate in Ngerengere river catchment, Tanzania

Abstract

Prediction of future climate forms one of the tenets of assessing impacts of climate change in various aspects of livelihoods. The ability to validate climate variables generated by relatively coarse global models, at the desired scales, not only adds value to the usefulness of such models but also engenders research on impacts of climate change in various domains. In this study, one of the protocols for downscaling climate variables from coarse Global Circulation Models (GCM), the Long Ashton Research Station Weather Generator (LARS-WG) was applied in the Ngerengere catchment in Tanzania. LARS-WG is a stochastic weather generator which can be used for generation of weather data at single site under both current and future climate condition.The LARS-WG model was calibrated and validated using the observed daily rainfall data for the period 1971–2000 (30 years) in Ngerengere catchment. LARS-WG calibration results indicate a good fit between observed and generated average daily rainfall and temperature indices. The calibrated model was then used to downscale climate attributes from GCMs with appreciable results where it was observed that LARS-WG performed well in generating future rainfall and temperature data series. The results indicated an increase in minimum and maximum temperature between 0.2 and 2.6 °C in the 2050s and between 2.7 and 4.4 °C in the 2080s. The lowest increase in future temperature was noted to occur in October and November and the highest increase to occur in May, June and July. Future rainfall was predicted to decrease by 12–37% in April, May, June and July, while rainfall in the remaining months, was predicted to increase by 3–58%.