13.2.2 Weather and climate stresses

Over the past three decades, Latin America has been subjected to climate-related impacts of increased El Niño occurrences (Trenberth and Stepaniak, 2001). Two extremely intense episodes of the El Niño phenomenon (1982/83 and 1997/98) and other severe climate extremes (EPA, 2001; Vincent et al., 2005; Haylock et al., 2006) have happened during this period, contributing greatly to the heightened vulnerability of human systems to natural disasters (floods, droughts, landslides, etc.).

Since the TAR, several highly unusual extreme weather events have been reported, such as the Venezuelan intense precipitations of 1999 and 2005; the flooding in the Argentinean Pampas in 2000 and 2002; the Amazon drought of 2005; the unprecedented and destructive hail storms in Bolivia in 2002 and Buenos Aires in 2006; the unprecedented Hurricane Catarina in the South Atlantic in 2004; and the record hurricane season of 2005 in the Caribbean Basin. The occurrence of climate-related disasters increased by 2.4 times between the periods 1970-1999 and 2000-2005 continuing the trend observed during the 1990s. Only 19% of the events between 2000 and 2005 have been economically quantified, representing losses of nearly US$20 billion (Nagy et al., 2006a). Table 13.1 shows some of the most important recent events.

In addition to weather and climate, the main drivers of increased vulnerability are demographic pressure, unregulated urban growth, poverty and rural migration, low investment in infrastructure and services, and problems with inter-sectoral co-ordination. The poorest communities are among the most vulnerable to extreme events (UNEP, 2003a), and some of these vulnerabilities are caused by their location in the path of hurricanes (about 8.4 million people in Central America; FAO, 2004a), on unstable lands, in precarious settlements, on low-lying areas, and in places prone to flooding from rivers (BID, 2000; UNEP, 2003a).

Natural ecosystems

Tropical forests of Latin America, particularly those of Amazonia, are increasingly susceptible to fire occurrences due to increased El Niño-related droughts and to land-use change (deforestation, selective logging and forest fragmentation) (see Box 13.1; Fearnside, 2001; Nepstad et al., 2002; Cochrane, 2003). During the 2001 ENSO period, approximately one-third of the Amazon forests became susceptible to fire (Nepstad et al., 2004). This climatic phenomenon has the potential to generate large-scale forest fires due to the extended period without rain in the Amazon, exposing even undisturbed dense forest to the risk of understorey fire (Jipp et al., 1998; Nepstad et al., 2002, 2004). Mangrove forests located in low-lying coastal areas are particularly vulnerable to sea-level rise, increased mean temperatures, and hurricane frequency and intensity (Cahoon and Hensel, 2002; Schaeffer-Novelli et al., 2002), especially those of Mexico, Central America and Caribbean continental regions (Kovacs, 2000; Meagan et al., 2003). Moreover, floods accelerate changes in mangrove areas and at their landward interface (Conde, 2001; Medina et al., 2001; Villamizar, 2004). In relation to biodiversity, populations of toads and frogs are affected in cloud forests after years of low precipitation (Pounds et al., 1999; Ron et al., 2003; Burrowes et al., 2004). In Central and South America, links between higher temperatures and frog extinctions caused by a skin disease (Batrachochytrium dendrobatidis) were found (Dey, 2006).

Agriculture

The impact of ENSO-related climate variability on the agricultural sector was well documented in the TAR (IPCC, 2001). More recent findings include: high/low wheat yields during El Niño/La Niña in Sonora, Mexico (Salinas-Zavala and Lluch-Cota, 2003); shortening of cotton and mango growing cycles on the northern coast of Peru during El Niño because of increases in temperature (Torres et al., 2001); increases in the incidence of plant diseases such as ‘cancrosis’ in citrus in Argentina (Canteros et al., 2004), Fusarium in wheat in Brazil and Argentina (Moschini et al., 1999; Del Ponte et al., 2005), and several fungal diseases in maize, potato, wheat and beans in Peru (Torres et al., 2001) during El Niño events, due to high rainfall and humidity. In relation to other sources of climatic variability, anomalies in South Atlantic sea-surface temperatures (SST) were significantly related to crop-yield variations in the Pampas region of Argentina (Travasso et al., 2003a, b). Moreover, heatwaves in central Argentina have led to reductions in milk production in Holando argentino (Argentine Holstein) dairy cattle, and the animals were not able to completely recover after these events (Valtorta et al., 2004).

Water resources

In global terms, Latin America is recognised as a region with large freshwater resources. However, the irregular temporal and spatial distribution of these resources affects their availability and quality in different regions. Stress on water availability and quality has been documented where lower precipitation and/or higher temperatures occur. For example, droughts related to La Niña create severe restrictions for water supply and irrigation demands in central western Argentina and central Chile between 25°S and 40°S (NC-Chile, 1999; Maza et al., 2001). In addition, droughts related to El Niño impacts on the flows of the Colombia Andean region basins (particularly in the Cauca river basin), causing a 30% reduction in the mean flow, with a maximum of 80% loss in some tributaries (Carvajal et al., 1998), whereas extreme floods are enhanced during La Niña (Waylen and Poveda, 2002). In addition, the Magdalena river basin also shows high vulnerability (55% losses in mean flow; IDEAM, 2004). Consequently, soil moisture and vegetation activity are strongly reduced/augmented by El Niño/La Niña in Colombia (Poveda et al., 2001a). The vulnerability to flooding events is high in almost 70% of the area represented by Latin American countries (UNEP, 2003c). Hydropower is the main electrical energy source for most countries in Latin America, and is vulnerable to large-scale and persistent rainfall anomalies due to El Niño and La Niña, e.g., in Colombia (Poveda et al., 2003), Venezuela (IDEAM, 2004), Peru (UNMSM, 2004), Chile (NC-Chile, 1999), Brazil, Uruguay and Argentina (Kane, 2002). A combination of increased energy demand and drought caused a virtual breakdown in hydroelectricity generation in most of Brazil in 2001, which contributed to a GDP reduction of 1.5% (Kane, 2002).

Coasts

Low-lying coasts in several Latin American countries (e.g., parts of Argentina, Belize, Colombia, Costa Rica, Ecuador, Guyana, Mexico, Panama, El Salvador, Uruguay and Venezuela) and large cities (e.g., Buenos Aires, Rio de Janeiro and Recife) are among the most vulnerable to climate variability and extreme hydrometeorological events such as rain and windstorms, and sub-tropical and tropical cyclones (i.e., hurricanes) and their associated storm surges (Tables 13.1 and 13.2). Sea-level rise (within the range 10-20 cm/century) is not yet a major problem, but evidence of an acceleration of sea-level rise (SLR) rates (up to 2-3 mm/yr) over the past decade suggests an increase in the vulnerability of low-lying coasts, which are already subjected to increasing storm surges (Grasses et al., 2000; Kokot, 2004; Kokot et al., 2004; Miller, 2004; Barros, 2005; Nagy et al., 2005; UCC, 2005). Moreover, some coastal areas are affected by the combined effects of heavy precipitation, landward winds and SLR (for example, ‘sudestadas’ in the La Plata river estuary), as already observed in the city of Buenos Aires (EPA, 2001; Bischoff, 2005).

Human health

After the onset of El Niño (dry/hot) there is a risk of epidemic malaria in coastal regions of Colombia and Venezuela (Poveda et al., 2001b; Kovats et al., 2003). Droughts favour the development of epidemics in Colombia and Guyana, while flooding engenders epidemics in the dry northern coastal region of Peru (Gagnon et al., 2002). Annual variations in dengue/dengue haemorrhagic fever in Honduras and Nicaragua appear to be related to climate-driven fluctuations in the vector densities (temperature, humidity, solar radiation and rainfall) (Patz et al., 2005). In some coastal areas of the Gulf of Mexico, an increase in SST, minimum temperature and precipitation was associated with an increase in dengue transmission cycles (Hurtado-Díaz et al., 2006). Outbreaks of hantavirus pulmonary syndrome have been reported for Argentina, Bolivia, Chile, Paraguay, Panama and Brazil after prolonged droughts (Williams et al., 1997; Espinoza et al., 1998; Pini et al., 1998; CDC, 2000), probably due to the intense rainfall and flooding following the droughts, which increases food availability for peri-domestic (living both indoors and outdoors) rodents (see Chapter 8, Section 8.2.8). Prolonged droughts in semi-arid north-eastern Brazil have provoked rural-urban migration of subsistence farmers, and a re-emergence of visceral leishmaniasis (Confalonieri, 2003). A significant increase in visceral leishmaniasis in Bahia State (Brazil) after the El Niño years of 1989 and 1995 has also been reported (Franke et al., 2002). In Venezuela, an increase in cutaneous leishmaniasis was associated with a weak La Niña (Cabaniel et al., 2005). Flooding produces outbreaks of leptospirosis in Brazil, particularly in densely populated areas without adequate drainage (Ko et al., 1999; Kupek et al., 2000; Chapter 8, Section 8.2.8). In Peru, El Niño has been associated with some dermatological diseases, related to an increase in summer temperature (Bravo and Bravo, 2001); hyperthermia with no infectious cause has also been related to heatwaves (Miranda et al., 2003), and SST has been associated with the incidence of Carrion’s disease (Bartonella bacilliformis) (Huarcaya et al., 2004). In Buenos Aires roughly 10% of summer deaths may be associated with thermal stress caused by the ‘heat island’ effect (de Garín and Bejarán, 2003). In São Paulo, Brazil, Gouveia et al. (2003) reported an increase of 2.6% in all-cause morbidity in the elderly per °C increase in temperature above 20°C, and a 5.5% increase per ºC drop in temperature below 20°C (see Chapter 8).