%0 Bachelor Thesis %A Mifsud, Andrew (2010) %I University of Malta %D 2010 %G English %T Construction of future global climate scenarios : a comparative study %U https://www.um.edu.mt/library/oar/handle/123456789/100655 %X B.SC.(HONS)PHYSICS Using MAGICC/SCENGEN software global climate scenarios were developed for the coming 110 years, using different emission scenarios and regions. Projections of emissions and concentrations of key greenhouse gases, and the resulting global-mean temperature and sea level changes between 1990-2100 were obtained using the MAGICC climate model. The data for the global-mean temperature increase obtained using the HadCM3 climate model, were used to drive SCENGEN to obtain a regional pattern. Six world regions were considered, namely Africa, Australia, Europe, Greenland, the North Atlantic and the USA. Furthermore, the study employs a method for the choice of the best possible ensemble of AOGCMs over a specific region, for a given climate variable. On a regional scale, the climate variables studied were temperature, precipitation and mean sea level pressure, while the two emission scenarios employed were the no-climate-policy SRES AlB-AIM scenario and the WRE650 stabilisation scenario. Mean changes over the selected regions were obtained on a seasonal and annual basis for the years 2025, 2050, 2075 and 2100. The associated variability of the three climate variables was studied using SCENGEN, which was also used to investigate uncertainties in the results through signalto-noise-ratios. Overall the results of this study are in good agreement with the latest IPCC findings. The range of temperature increase obtained for 2100 with respect to 1990 is between 1.6 and 5.0°C, when employing a climate sensitivity of 3.0°C. The corresponding projected increase in global-mean sea level rise for 2100 ranges from 24 to 53 cm. Temperature increase was shown to have a strong dependence on the global-mean atmospheric concentration of C02. In turn projections for sea level rise were seen to follow similar patterns to those of temperature increase. Warming over many land areas is greater than global annual mean warming. The annual temperature increase by 2100 over the six regions considered, using the HadCM3 model with ...