Search

Future projection of air temperature and precipitation due to climate change plays an important role in shaping the water resources management plans in arid and semi-arid countries such as the United Arab Emirates (UAE). The effect of local topography, land use, and sealand contrasts are generally dwarfed in the coarse global circulation models [ocean atmosphere global climate models (OAGCMs)] results. Statistical downscaling (SD) is a statistical tool relating OAGCM outputs (predictors) to finer local observed weather data (predictands) to improve future predictions. SD is used in this study to downscale the Hadley Model (HadCM3) predictors using local observations at two stations representing the dominating bioclimatic zones in the UAE. The paper proposes a rigorous and systematic methodology to guide the selection of dominant predictors in the projection scenarios. The proposed methodology is applied for the UAE but can be easily adopted for any other area. The calibrated model, used to project future scenarios in the region, revealed a range of increase of the annual mean maximum temperature of 2.79-3.80°C and a range of reduction of annual precipitation of 16.80-37.00% by 2080 at the considered stations. Impacts of climate change on the UAE's water resources management are discussed in light of the downscaled estimates for temperature and precipitation. Because a major portion of the water resources is devoted to agriculture, temperature impacts are quite significant. However, the impacts of precipitation are hardly detectable at the national scale. Instead, the impacts of reduced precipitation will be of more localized significance in mountainous regions that are critically dependent on precipitation for water resources. [ABSTRACT FROM AUTHOR]