Carter et al. (2000) performed a comprehensive characterization of regional climate change projections for the 21st century. They assumed a range of atmospheric greenhouse gas (GHG) loadings according to the four draft marker scenarios developed for the IPCC Special Report on Emissions Scenarios (SRES), in combination with the IPCC range of climate sensitivities (1.5-4.5°CIPCC, 1996). Their method of analysis involved combining simple climate model estimates of the global mean annual temperature response to four combinations of GHG forcing/climate sensitivity (see Table 3-9) with regional patterns of seasonal temperature and precipitation change obtained from 10 general circulation model (GCM) simulations for the end of the 21st century relative to 1961-1990. The GCM patterns of change were scaled up or down so that the global mean temperature change from the GCM coincided with that obtained from the simple climate model. Ten patterns of change were obtained for each emissions/climate sensitivity combination, and each was averaged over subcontinental regions, including five representing the African continent: southern Europe/north Africa, the Sahara, west Africa, east Africa, and southern Africa (Carter et al., 2000; see Chapter 3). Ranges of projected rates of change in temperature and precipitation over these regions are depicted for each season, and projected changes over southern Africa by the 2050s are compared with modeled natural multi-tridecadal variability from the HadCM2 GCM 1,400-year control simulation (Tett et al., 1997) for the summer (DJF) and winter (JJA) months.
An analysis using the similar methodology also has been conducted specifically for Africa (Hulme et al., 2001). Future annual warming across Africa ranges from 0.2°C per decade (B1low scenario) to more than 0.5°C per decade (A2high scenario). This warming is greatest over the interior of semi-arid margins of the Sahara and central southern Africa. The intermodel range (an indicator of the extent of agreement between different GCMs) is smallest over north Africa and the equator and greatest over the interior of southern Africa.
Future changes in mean seasonal rainfall in Africa are less well defined. Under the lowest warming scenario, few areas experience changes in DJF or JJA that exceed two standard deviations of natural variability by 2050. The exceptions are parts of equatorial east Africa, where rainfall increases by 5-20% in DJF and decreases by 5-10% in JJA.
Under the two intermediate warming scenarios, significant decreases (10-20%) in rainfall during March to November, which includes the critical grain-filling period, are apparent in north Africa in almost all models by 2050, as are 5-15% decreases in growing-season (November to May) rainfall in southern Africa in most models.
Under the most rapid global warming scenario, increasing areas of Africa experience changes in summer or winter rainfall that exceed the one sigma level of natural variability. Large areas of equatorial Africa experience increases in DJF rainfall of 50-100% over parts of eastern Africa, with decreases in JJA over parts of the Horn of Africa. However, there are some JJA rainfall increases for the Sahel region.
Hulme et al. (2001) also analyzed future rainfall changes for three African regionsthe Sahel, east Africa, and southeast Africato illustrate the extent of intermodel differences for these regions and to put future modeled changes in the context of past observed changes (see Figure 10-3). Although model results vary, there is a general consensus for wetting in East Africa, drying in southeast Africa, and a poorly specified outcome for the Sahel.
Other reports in this collection |