The IPCC assessment reports confirm that climate change will hit developing countries the hardest. Adaption is on the agenda of many countries around the world. However, before devising adaption strategies, it is crucial to assess and understand the impacts of climate change at regional and local scales. In this study, the impact of climate change on rain-fed maize (Zea mays) production in the Wami-Ruvu basin of Tanzania is assessed using process based crop model the Decision Support System for Agrotechnological Transfer (DSSAT). The model was calibrated using detailed field and household survey information of (crop yields, soil and management data inputs). Daily minimum and maximum temperatures, rainfall and solar radiation for current climate condition as well as future climate projections (2010-2039), (2040-2069) and (2070-2099) for two Representative Concentration Path ways (RCPs): RCP45 and RCP85 scenarios were used to drive the crop model. These data are derived from three high-resolution regional climate models (RCMs), used in the Coordinated Regional Climate Downscaling Experiment program (CORDEX). Impact of climate change on maize production is assessed by analyzing the changes in simulated maize yields for the period 2010-2039, 2040-2069 and 2070-2099 relative to baseline period 1971-2000. Projection results from different models showed that due to climate change, the length of growing season and future maize yields over Wami-Ruvu basin will decrease under both RCP4.5 and RCP8.5 at the current, mid and end of the centuries. However, the projected yields estimates and the length of growing season differ from model to model highlighting the uncertainties associated with the projections. Climate data from the ensemble average of five model members was constructed to address the issue of uncertainties from individual climate models and used to drive DSSAT. Results showed that due to climate change future maize yields over Wami-Ruvu basin will slightly increase relative to the baseline during current century under RCP 4.5 and RCP 8.5. Meanwhile, maize yields will decline in the mid and end centuries. The spatial distribution shows that more decline in maize yields are projected over lower altitude regions due to projected increase in temperatures and decreased rainfall in those areas. The eastern part of the basin will feature more decrease in maize yields, while central parts of the basin and the western side of the basin will experience increase in maize yields during current, mid and end centuries under RCP 4.5 and RCP 8.5. The main reason for decrease and increase maize yields is the projected increase in temperatures that will reduce the length of growing seasons and hence affecting maize productivity. It is therefore recommended that appropriate and adequate adaptation strategies need to be designed to help the communities adapt to the projected decrease in maize production.

The amount of rainfall and its distribution in time and space is dependent on the atmospheric instability conditions, and on its moisture content. The aim of this study was to determine the atmospheric instability conditions during January to March (JFM), February to April (FMA), March to May (MAM), and October to December (OND) rainy seasons over local climate zones in Tanzania. Zone area average seasonal Convective Available Potential Energy (CAPE), Convective Inhibition (CIN), Precipitable Water (PW) and Lifted index (Li) were calculated and analyzed. Results showed Li < 0 in JFM and FMA over whole Tanzania. During MAM and OND, Li < 0 over the Lake Zone, Western Highlands Zone and Central Zone only. CAPE ranged from 793 J/kg to 1183 J/kg during JFM, and 700 J/kg to 1080 J/kg during FMA. During MAM, CAPE ranged from 170 J/kg to 921 J/kg and from 173 J/kg to 833 J/kg during OND. Results also showed CAPE > 1000 J/kg over the Lake Zone, Western Highlands Zone, Island Zone, and Central Zone. These results show that the atmosphere was moderately unstable during the JFM and FMA and was weakly unstable during the MAM and OND. Therefore, the atmosphere is likely to be more convective during JFM and FMA seasons. Keywords: Lifted index, Convective inhibition, Precipitable water, Convective available potential energy, Atmospheric instability.