Climate Change 2001:
Working Group II: Impacts, Adaptation and Vulnerability
Other reports in this collection Future: Climate Scenarios

Climate modeling has proven to be extremely useful in building projections for climate change and scenarios of future climate under different forcings. General circulation models have demonstrated their ability to simulate realistically the large-scale features of observed climate; hence, they are widely used to assess the impact that increased loading of the atmosphere with greenhouse and other gases might have on the climate system. Although there are differences among models with regard to the way they represent the climate system processes, all of them yield comparable results on a global basis. However, they have difficulty in reproducing regional climate patterns, and large discrepancies exist among models. In several regions of the world, distributions of surface variables such as temperature and rainfall often are influenced by the local effects of topography and other thermal contrasts, and the coarse spatial resolution of GCMs cannot resolve these effects. Consequently, large-scale GCM scenarios should not be used directly for impact studies, especially at the regional and local levels (von Storch, 1994); downscaling techniques are required (see TAR WGI Chapters 10 and 13).

At the large scale, rates of mean annual temperature changes in the Latin American region for the next century are projected to be 0.2-2°C (Carter and Hulme, 2000) under the low-emissions scenario (B1) produced as part of the IPCC Special Report on Emission Scenarios (SRES). The warming rate could range between 2 and 6°C for the higher emissions case (A2). Most GCMs produce similar projections for temperature changes on a global basis; projected changes in precipitation remain highly uncertain.

For impact studies, it is crucial to have a projection of concurrent changes of temperature and precipitation at the regional scale. Various scenarios of climate change for Latin America have been put forward on the basis of GCM projections under the IS92a scenario. Most of these regional scenarios are based on GCM experiments that are downscaled through statistical techniques. Derived climate change scenarios for Mexico suggest that climate in Mexico will be drier and warmer (Perez, 1997). Several hydrological regions in Mexico are highly vulnerable to decreased precipitation and higher temperatures (Mendoza et al., 1997). A regional climate change scenario for central Argentina in response to CO2 doubling under the IS92 scenario for the year 2050, also obtained through a statistical downscaling approach, shows a smaller increase in minimum temperature as compared to the maximum and larger increases for summer than for winter months, which generates enhanced temperature amplitudes (Solman and Nuñez, 1999). In addition, a decrease in precipitation is projected over the region, which is larger for summer (12%) than for winter months (5%). This result highlights an important consequence in the rainfall regime over the region: A large decrease in rainfall projected for the rainy season will seriously affect soil moisture, hence agricultural production in the region.

Several climate change scenarios for other parts of Latin America rely on linear interpolation of GCM output to estimate increases in surface temperature and precipitation (Mata, 1996; Carril et al., 1997; Hofstadter and Bidegain, 1997, Paz Rada et al., 1997; Centella et al., 1998; MARENA, 2000). In the case of Costa Rica (MINAE-IMN, 2000), under the IS92a scenario for the year 2100, the results show a small increase in precipitation for the southeastern Caribbean region and an important decrease—close to 25%—in the northwestern Pacific region. This latter region already experiences water problems as a result of El Niño and an increasing demand from infrastructure for tourism and irrigation. Under the same climate scenario, mean temperature in Costa Rica is expected to rise by more than 3°C by 2100, and tendencies in actual climate series (1957-1997) show already an increase of 0.4°C every 10 years for the more continental Central Valley areas. This last estimation may reflect signals other than the one related to climate change.

Results from climate scenarios for Nicaragua imply an additional pressure on productivity sectors and human activities. Under the IS92a emissions scenario, mean temperature for the Pacific watershed would be expected to rise, ranging from 0.9 for the year 2010 to 3.7°C for the year 2100, and precipitation would decrease by 8.4% for the year 2010 and 36.6% for the year 2100. For the Caribbean watershed, mean temperature would increase, ranging from 0.8°C for the year 2010 to 3.3°C for the year 2100, and precipitation would decrease in a range between 8.2% for the year 2010 and 35.7% for the year 2100 (MARENA, 2000).

Potential effects of climate change in Brazil suggest changes of 4-4.5°C in surface temperature as a result of increased CO2 concentrations (de Siqueira et al., 1994, 1999). Central and south-central Brazil may experience increases of 10-15% in autumn rainfall; reductions could appear during December, with high risk of drought during summer, affecting crops (see Table 14-1).

Table 14-1: Estimated changes projected under IS92 scenario for some countries within Latin America region.

Costa Rica

- Pacific sector
- Southeast Caribbean sector  
small increase
- Pacific sector
- Caribbean sector
- Central and south central sector
+10 to +15% for autumn
reductions for summer
Central Argentina
summer: +1.57°C(+1.08-2.21°C)
winter: +1.33°C (+1.12-1.57°C)
summer: -12%
winter: -5%

Analysis of climate variations during the instrumental period and evidence suggested by paleoclimatic and other proxy climate information suggests that climate variations and change have been found in several regions in Latin America. Most climate records cover the past century; at this time scale, there have been indications of multidecadal and interannual variability, some linked to extremes of the Southern Oscillation. The lack of continuous and long-term records from the past does not allow one to identify climate patterns with a high degree of confidence to determine whether these climates were similar to or much different from that of present times—particularly with respect to the frequency and intensity of extreme events such as drought, floods, freezes, heat waves, and especially hurricanes and tropical storms. However, multidecadal variations have been identified in rainfall and streamflow records in the region, although no clear unidirectional trend indicators of climate change have been identified.

Other reports in this collection