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

14.3. Synthesis

There is ample evidence of climate variability at a wide range of time scales all over Latin America, from intraseasonal to long term. For instance, at decadal scales, multiple climate records throughout the region consistently exhibit a shift in the mean during the mid-1970s, which could be a consequence of sudden changes in the climatology of the Pacific Ocean. These changes have important socioeconomic and environmental consequences that could be enhanced by global warming and its associated climate change.

Precipitation changes in Latin America do not follow a consistent trend. In northwestern Mexico there is a clear tendency for more winter precipitation, which has resulted in positive trends in river-water level. In north and northwestern Nicaragua, there is a negative trend in rainfall. In the Amazonian region, the most important finding is the presence of periods with relatively wetter or drier conditions that are more relevant than any unidirectional trend. Rainfall in north-northeast Brazil exhibits a weak positive trend. Precipitation in subtropical Argentina, Paraguay, and Brazil increased abruptly in the 1956-1990 period after a dry period from 1921 to 1995. In the Pampas, there was a positive trend in precipitation over the 1890-1984 period. At higher elevations in northwestern Argentina, records suggest an increase in precipitation over the past 200 years.

The Southern Oscillation is responsible for a large part of the climate variability at interannual scales in Latin America. The region is vulnerable to El Niño, with impacts varying across the continent. For example, El Niño is associated with dry conditions in northeast Brazil, northern Amazonia, the Peruvian-Bolivian Altiplano, and the Pacific coast of Central America, whereas southern Brazil and northwestern Peru exhibit anomalously wet conditions. Extensive studies of the Caribbean watersheds of Mexico and other countries show compelling evidence of more winter and less summer precipitation; in addition, the most severe droughts in Mexico in recent decades have occurred during El Niño years. In Colombia, La Niña is associated with heavy precipitation and flooding, whereas El Niño is associated with negative anomalies in precipitation and river streamflow. Drought also occurs in southern Brazil during the positive phase of the Southern Oscillation. If El Niño or La Niña were to increase, Latin America would be exposed to these conditions more often (see Table 14-6).

Table 14-6: Variability and impacts of El Niño and La Niña on several Latin American countries and subregions.
or Country
El Niñoa Severe droughtsb in recent decades
Mexico Magaña and Conde, 2000

Severe droughts Northeast Brazil Silva Dias and Marengo, 1999 1901-1997
Decrease in precipitation Central America (Pacific)
Magaña and Conde, 2000 1958-1999
Increase in precipitation Central America (Atlantic)
Magaña and Conde, 2000 1958-1999
Decrease in precipitation,
soil moisture, river streamflow
Colombia Poveda and Mesa, 1997
Carvajal et al., 1999
Poveda et al., 2001
Increase in precipitation and floods
Northwest Peru Marengo et al., 1998

Decrease in precipitation
during rainy season
Northern Amazonia
and northeast Brazil
Aceituno, 1988; Richey et al., 1989; Marengo, 1992; Uvo, 1998
Negative large anomalies of
rainfall during rainy season
Northeast Brazil Silva Dias and Marengo, 1999
Hastenrath and Greischar, 1993
Increase in precipitation
during November-January
time frame

(Pampas region)
Nobre and Shukla, 1996 1900-1996
Intense snowfalls in high
Andes mountains
Central western Argentina
and central Chile
Barros et al., 1996; Tanco and Berry, 1996; Vila and Berri, 1996; Vila and Grondona, 1996; Magrin et al., 1998

Increase in runoff

Chile and central
western Argentina

Canziani et al., 1997 1906-1994

Heavier precipitation and floods


Poveda and Mesa, 1997
Carvajal et al., 1999
Decrease in precipitation during October-December time frame

(Pampas region)
Barros et al., 1996; Tanco and Berry, 1996; Vila and Berri, 1996; Vila and Grondona, 1996; Magrin et al., 1998 1900-1996

Increase in precipitation, higher runoff

Northern Amazonia
Northeast Brazil

Marengo et al., 1998
Meggers, 1994
Severe droughts

Southern Brazil

Grimm et al., 1996, 2000 1956-1992
Negative anomalies of rainfall Chile and central western Argentina
Compagnucci et al., 2000 Paleoclimate
a Extremes of the Southern Oscillation (SO) are responsible in part for a large portion of climate variability at interannual scales in Latin America. El Niño (or ENSO) events represent the negative (low) phase of the SO.
b Prolonged periods of reduced summer soil moisture.
c Six El Niño events occurred during this period.
d La Niña is the positive (high) phase of the SO.

Warming in high mountain regions could lead to disappearance of significant snow and ice surfaces, which could affect mountain sport and tourist activities. Because these areas contribute to river streamflow, it also would affect water availability for irrigation, hydropower generation, and navigation, which represent important sources of income for some economies. It has been well-established that glaciers in Latin America have receded in recent decades.

It is well-established that Latin America accounts for one of the Earth's largest concentrations of biodiversity, and the impacts of climate change can be expected to increase the risk of biodiversity loss. Some adverse impacts on species that can be related to regional climate change have been observed, such as population declines in frogs and small mammals in Central America. Maintenance of remaining Amazonian forest is threatened by the combination of human disturbance and decreased precipitation from evapotranspiration loss, global warming, and El Niño. Fire in standing forest has increased in frequency and scale in Amazonia as a result of greater accumulation of deadwood in the forest from logging activity and from trees killed by past fires, more human settlement providing opportunities for fire initiation, and dry conditions during ENSO events. Neotropical seasonally dry forest should be considered severely threatened in Mesoamerica.

Mortality of trees has been observed to increase under dry conditions that prevail near newly formed edges in Amazonian forests. Edges, which affect an increasingly large portion of the forest with the advance of deforestation, would be especially susceptible to the effects of reduced rainfall. In Mexico, deciduous tropical forest would be affected in approximately 50% of the area presently covered by these forests. Heavy rain during the 1997-1998 ENSO event generated drastic changes in the dry ecosystems of northern Peru's coastal zone. Increases in biomass burning in Amazonia affected by human activity and by climate increase wind-borne smoke and dust that supply nutrients to the Amazon forests. Global warming would expand the area that is suitable for tropical forests as equilibrium vegetation types. However, the forces driving deforestation make it unlikely that tropical forests will be permitted to occupy these increased areas. Land-use change interacts with climate through positive feedback processes that accelerate loss of humid tropical forests.

Sea-level rise will eliminate the present habitats of mangroves and create new tidally inundated areas to which some mangrove species may shift. Coastal inundation stemming from sea-level rise and riverine and flatland flooding would affect water availability and agricultural land. These changes can exacerbate socioeconomic and health problems in critical areas.

On a scale of decades to centuries, it is well-established that changes in precipitation and catchment runoff may have significant effects on mangrove forest communities. This also would affect the region's fisheries because most commercial shellfish and finfish use mangroves as nurseries and for refuge.

Studies in Argentina, Brazil, Chile, Mexico, and Uruguay, based on GCMs and crop models, project decreased yield for several crops (e.g., maize, wheat, barley, grapes), even when the direct effects of CO2 fertilization and implementation of moderate adaptation measures at the farm level are considered. Predicted increases in temperature will reduce crops yields in the region by shortening the crop cycle. Although relationships between the amount of precipitation and crop yields are well-established, the lack of consistency in the results of different GCMs means that confidence in the estimated impacts of future precipitation on crop production is necessarily limited.

Over the past 40 years, the contribution of agriculture to the GDP of Latin American countries has been on the order of 10%. Agriculture remains a key sector in the regional economy because it employs 30-40% of the economically active population. It also is very important for the food security of the poorest sectors of the population. Subsistence farming could be severely threatened in some parts of Latin America (e.g., northeastern Brazil).

Evidence is established but incomplete that climate change would reduce silvicultural yields because water often limits growth during the dry season, which is expected to become longer and more intense in many parts of Latin America.

The scale of health impacts from climate change in Latin America would depend primarily on the size, density, location, and wealth of populations. Evidence has been established but incomplete that exposure to heat or cold waves has impacts on mortality rates in risk groups in the region.

Increases in temperature would affect human health in polluted cities such as Mexico City and Santiago. Ample evidence provides high confidence that the geographical distribution of VBDs in Peru and Cuba (e.g., cholera, meningitis) would change if temperature and precipitation were to increase, although there is speculation about what the changes in patterns of diseases would be in different places. It is well-established that weather disasters (extreme events) tend to increase death and morbidity rates (injuries, infectious diseases, social problems, and damage to sanitary infrastructure)—as occurred in Central America with Hurricane Mitch at the end of 1998, heavy rains in Mexico and Venezuela at the end of 1999, and in Argentina in 2000. It is well-established that the ENSO phenomenon causes changes in disease-vector populations and in the incidence of water-borne diseases in Brazil, Peru, Bolivia, Argentina, and Venezuela. It has been speculated that weather factors and climate change may affect the incidence of diseases such as allergies in Mexico. Increased temperature, ultraviolet radiation, sea-level rise, and changes in pest ecology may threaten food production in Argentina.

In summary, climate change already has a diverse array of impacts in Latin America. Projected future changes in climate, together with future changes in the vulnerability of human and natural systems, could lead to impacts that are much larger than those experienced to date.

Other reports in this collection