Costa Rica Country Case Study:
Impacts and Responses to the 1997-98 El Niño Event
|
Team Leader: Team Members: |
 |
Executive Summary
Introduction
With the announcement of the 1997-98 El Niño at the beginning of 1997, Costa Ricans began a massive response at the national level. National media, government ministries, agriculture and fisheries began preparations based on prior experiences with El Niño phenomena since 1982. As such, Costa Rica represents an intriguing case study of early warning and response to climatic anomalies. Significantly, not all prevention strategies worked as planned, and some had unforeseen consequences. These experiences provide a revealing window on the possibilities of response to early warnings, and the limitations of such strategies.
One initial conclusion of the review of the 1997-98 El Niño response is that Costa Rica in fact has adopted a "culture of preparedness," as recommended by international disaster prevention efforts, such as the International Decade for Natural Disaster Reduction. This preparedness is especially clear in contrast to the lack of preparation on the part of other Central American countries.
A more sobering observation is the delicate policy decision of climate prediction. As probabilistic statements, forecasts have a likelihood of error; the impact of mispredictions must be carefully weighed against the benefits of early preparation. Uncertainties surrounding El Niño's forecasts led to very different approaches on the part of the various organizations involved in the the early analysis of the 1997-98 event, with different institutions favoring more and less speculative forecasts.
Global and historical context of El Niño for Costa Rica
El Niño-Southern Oscillation (ENSO, or ENOS in Spanish) is a climate phenomenon arising out of the interaction of global scale air masses and the tropical Pacific Ocean. While first noted and named El Niño by Peruvians and Ecuadoreans because of the coincidence of coastal ocean warming with the Nativity (El Niño – The Child) the phenomenon extends in some form along a great length of the American shorelines of the Pacific.
The connection between El Niño and extreme climate variations in Costa Rica was not explicitly recognized until the mid-1980s, when a fortuitous conversation between Dr. Michael Glantz, a leading El Niño researcher, and Patricia Ramirez, a leading Costa Rican climate researcher,1 identified the coincidence of periodic Costa Rican droughts with years of El Niño. At that time, Ramirez was studying droughts in western Costa Rica for the purpose of improved agricultural adaptation to climate variability, since periodic droughts were the cause of losses and economic instability in the agricultural sector. At approximately the same time, the Laboratory for Oceanography was established at the National University in Heredia (Costa Rica), a center that would eventually provide needed researchers and information for national climate analyses.
As one of the areas first and most directly affected by the tropical Pacific Ocean warming, Costa Rica is a site of research activities into the historical evolution of the El Niño phenomenon (e.g., tree ring research). The role of El Niño in global climatic variability and its impact on global climate change have become key questions for understanding current and future global climate conditions.
El Niño has now become a topic of public discourse, which is reflected in the publications of Costa Rica's national institutions. The Bulletin of the National Meteorological Institute IMN (spanish acronym), published as a vehicle for communicating technical meteorological information to the public, shows a concern with the ENSO phenomenon. At least 3 bulletins in 1991 and 1992 explicitly describe El Niño or associated phenomena.2 A more recent bulletin discusses the decadal oscillation in the Pacific (the PDO).3
Costa Rican Climate Patterns
Costa Rica is characterized by two typically tropical climate patterns, on its Caribbean and Pacific coasts. The coasts are separated by a volcanic cordillera, which divides the types of weather patterns. The central plateau and intermountain valleys at an elevation of 1200 meters is home to the major cities and a large part of the Costa Rican population. The populated central plateau is significantly cooler than the coastal regions, but generally follows the weather pattern of the Pacific coast.
The most widely recognized climate pattern is the rainy climate of the eastern, Caribbean watershed, which is responsible for Costa Rica's well known humid tropical rain forest vegetation. In contrast, the Northern Pacific watershed has a distinct monsoonal pattern, with intense rains followed by months of drought. This area is characterized by dry tropical forest.
Costa Rican climate patterns, and their variations, are a product of its inter-oceanic and inter-hemispheric location. As a narrow land bridge dividing the Caribbean/Atlantic and the Pacific, Central America in general, and Costa Rica in particular, are directly affected by ocean/atmosphere interactions. The east-west Walker Circulation which covers the breadth of the Pacific tends to cause subsidence of dry air on the western coast of the isthmus; the volcanic mountain range which divides the isthmus tends to create a rain shadow, so that even when pressure gradients favor the introduction of moist Caribbean air, most precipitation remains in the Caribbean watershed.
The other major influence on the Costa Rican climate is the Inter-Tropical Convergence Zone (ITCZ), which marks the dividing line between the north-south and south-north cycling Hadley cells, within which warm air rises at or near the equator, and subsides some 30 degrees north and south of the equator. The ITCZ migrates south in the northern winter months, and north in the northern summer months, crossing Costa Rica; the convective currents of the ITCZ are an important source of moisture for both the Pacific and Caribbean watersheds.
Obviously, Costa Rica's climate is extremely sensitive to subtle changes in the relationships between the Pacific and Caribbean air masses, and the northern and southern air masses. Disruptions caused by El Niño have a direct impact on day to day, and season to season climate. However, since the impacts depend on the precise location of the ITCZ and the relative pressures of the ocean air masses, the climate forecast is easily mistaken.
The Identification and Announcement of 1997-98 El Niño Event
By the time the 1997-98 El Niño event was first detected early in 1997, the phenomenon had been linked in the public's mind to earlier events of the 1980s and 1990s. The linkage was so strong that the Director of the Costa Rican Meteorological Institute felt a need for extreme caution in the possibly premature identification and offical announcement of an "El Niño Event," as it would have sown panic among some sectors of the population. Even with this caution exercised, cases arose where actions were taken which went beyond reasonable and prudent precaution to unnecessary caution to the detriment of the Costa Rican population.
Ironically, the initial announcement of the El Niño event in official circles was made by regional organizations, rather than the national institutions. In March of 1997, the head of the Regional Commission for Hydraulic Resources (CRRH) communicated the possibility of an El Niño event beginning later that year. [NB: The CRRH is part of the regional economic integration structure collectively known as SICA, the Secretariat for Central American Integration. The governing commission for CRRH is made up of the heads of the meteorological institutes for each country.] This note was sent first to another regional institution, CORECA (Regional Council for Agricultural Cooperation), and from there it was communicated to the ministers of relevant institutions in the different countries of the isthmus. In view of the apathetic response to the announcement from most of the Central American countries, the regional institutions may have been justified in their somewhat speculative initial announcement, which relied on the one and only climate model which predicted the onset of El Niño (more than a dozen climate models are commonly used, and none of the others at the time predicted the event).
The actual wording of the CRRH/CORECA announcement was quite reserved, indicating the possibility of an El Niño event; it did not predict with certainty, and it did not suggest the possibility of a strong event. Nevertheless, it was a more forceful indication than the National Meterological Institute considered appropriate; the NMI only issued its first advisory when there was a much more positive forecast in May and June.
Preparation for the 1997-98 Event
Of all the countries of Central America, Costa Rica's prompt and active response to the announcement was unique. The initial speculative announcement prompted the formation of El Niño commissions within national institutions, and the creation of an interinstitutional coordinating committee.
In the ministry of agriculture, the El Niño commission began reviewing contingency plans for drought. In some cases, preparation simply consisted in activating existing plans which had been developed even before Costa Rica's periodic droughts had been associated with El Niño. The expected impact of El Niño was an early onset of the dry season in November of 1997, and its extension into August or September of 1998; the normal monsoonal dry season lasts from November until May.
The preparation effort was mostly directed toward consciousness-raising among the agricultural population during the latter part of 1997, in preparation for drought in mid-1998. Public informational meetings were convened, and conducted by ministry officials. In some cases the meetings were attended by high officials in the NMI, the Oceanographic Laboratory of the National University of the CRRH.
The consciousness-raising was hindered by the then-current weather conditions; a heavy rainy season beginning in June of 1997 made producers skeptical. In some cases, the drought warning was issued in the middle of torrential downpours and floods, which seemed to limit the credibility of the prediction. To make matters worse, the first predicted impact of the El Niño was an early cessation of rains at the end of 1997, which did not occur. This warning had been issued in public meetings, and caused great embarrassment for the meteorologists, as it made their entire message suspect.
The most active and intrusive preparation for the El Niño event was the recommendation to reduce the cattle herd in the Pacific monsoonal climate zones of Guanacaste and the Central Pacific Coast, as it was expected that the extended drought would push producers past the limits of their customary water and food supplies. As a result, there was a massive sell-off of the western herd, pushing down prices in the market, and redistributing animals to areas where the impact of the drought was expected to be less severe.
In the water sector, the major preparation was the accelleration of existing irrigation and well-boring projects to help ensure water supplies in potentially affected zones.
The national insurance agency, INS, cut off credit to farmers on the west coast, apparently in response to a similar action by the national agency in Panama. Protests by farmers led to a careful meteorological review of specific regions, and the re-establishment of credit in zones where there was a minimal possibility of drought.
Impacts of the 1997-98 El Niño
At the outset, it must be stated that it is extremely difficult to differentiate El Niño impacts from normal climate variability in Central America. Given the torrential nature of the rains and the limited capacity of watersheds, floods are common throughout the region on an annual basis. Further, drought is a normal feature of Central American climate. Traditional methods have been developed to respond to the annual drought, but a crisis may ensue with the extension of the ‘normal' dry season by a few weeks.
The official report of the UN FAO Economic Commission for Latin America identified approximately US$100 million in damage, combining impacts in agriculture, fisheries and forestry. Drought, irregular rains and water shortage lead to losses principally in the drier zones of the country. The regional ocean warming typical of the El Niño event leads to a decrease in the fish catch in national waters – an exact parallel to the well-known impacts on fish populations in Peru.
Losses in forestry are of special note. An extended dry season leads to an accumulation of combustible material in dry tropical forest areas, which makes the control of fire extremely difficult. Fortunately, in Costa Rica the use of fire for agricultural clearing has been largely eliminated, and while the occurrence of forest fires did increase, the frequency was nowhere near the crisis levels achieved in other Central American countries.
Surprisingly, one of the major impacts of the El Niño was in animal production, not in the most drought-prone areas but in the less prone areas where animals had been moved in preparation for the extended drought. Costa Rica's Northern Plain is an anomolous climate area; following the traditional division of climate zones along the continental divide, the Northern Plain falls in the Caribbean zone. However, in parts it is much closer to the Pacific coast than to the Caribbean, and the continental divide is so low as to be non-existent. While it generally follows the Caribbean pattern of having virtually no dry season, its driest period is drier than that of other areas in the Caribbean climate area.
For the 1997-98 El Niño event, the climate of the Northwest portion of the Northern plain was much more like the Pacific than the Caribbean. Ironically, the relatively benign traditional climate had not required that ranchers of that area develop contingency plans to prepare for drought. In fact, Ministry of Agriculture officials reported that the ranchers responded inadequately to the impending drought because they could not believe it would come to pass as predicted. They continued to expect that the rains would begin ‘next week', and took inadequate action to protect their cattle. To make matters worse, the climate prediction for El Niño did not include drought for the northern zone, so ranchers felt confident on the basis of historical patterns and official predictions that the drought would not be overly extended.
Another ironic impact of the El Niño "drought" was the increased water damage to crops and flooding. Long-time meteorological researcher and former head of the National Meteorological Institute Patricia Ramirez found that total precipitation during El Niño years does not vary significantly from long-term averages. The apparent drought is caused by a change in rainfall pattern, where rain tends to cluster in very short time periods. Aquifers do not recharge, plants and infrastructure are damaged by heavy showers, and there follows a relatively long period before the next rain.
A bittersweet impact of the 1997-98 event was the restructuring of the mandate of the National Emergency Commission (CNE, in Spanish). Along with the Ministry of Agriculture, the CNE began preparations for El Niño through the allocation of disaster response budget to areas of expected impacts. This initiative was curtailed by a Supreme Court ruling, which confirmed that its disaster-response mandate required that it take no action until the disaster had actually occurred, and its planned activities were postponed until El Niño impacts actually manifested themselves. In the wake of El Niño, national legislation was changed to permit the CNE to prepare for predicted disasters.
Conclusions and Lessons Learned
The principal conclusion of this El Niño event is that there is a willingness and capability to respond to early warnings of climatic events in Costa Rica, even when these are not entirely certain. Government officials and producers were able to take raw information and transform it into meaningful responses based on their previous experience, and the expected nature of the El Niño impacts.
On the negative side, El Niño forecasts were too imprecise to permit specific preparations for impacts. Major errors were made in the forecast that had negative economic impacts for farmers who had taken actions based on the forecasts. Unfortunately, impacts are very localized, and farmers and government officials must make decisions on specific farms, rather than larger regions. While there is no doubt that the heightened alertness of the producers in response to the knowledge about the El Niño event is positive, it is not prudent to make long-term plans on that forecast, since it may be generally correct or incorrect, and in addition, may be specifically correct or incorrect. The proper response to the forecast is to hedge bets and be prepared for a range of contingencies.
Several efforts are currently proposed and underway to improve predictive capabilities of major climatic events such as El Niño. The major recommendation to such efforts is "Don't be satisfied with generalities." While meteorologists may find probabilistic statements satisfying and appropriate in their forecasts, decision makers require more clear-cut (detailed) information, which helps them make correct decisions and avoid foreseeably bad decisions. This lesson may less to do with forecasting than with the development of improved, detailed descriptions of meteorological processes within a country, and with the development of methods of communicating relevant information to non-professional users of climate-related information.
The potable water sector has determined that important forecast data are air temperatures. During the 1997-98 El Niño the national water service AyA (Aguas y Alcatarillados) found a surprising increase of about 5% in water consumption, which they attribute to increased temperatures.
At the same time, there is an important niche for the more generalized seasonal predictions. On an isthmus, hydrological cycling is greatly accelerated. Improved seasonal forecasting can be critical for water-control decisions for flood control, hydroelectric generation, and irrigation. Similarly, these generalized forecasts may be useful to farmers for generalized management decisions (whether to plant corn or sorghum, for example). The eventual integration of seasonal forecasts into production planning requires a certain period for learning and development of communication techniques, as farmers and other users learn how to interpret seasonal forecasts in the context of their own activities.
Lessons Learned
- In some sense, the issuing of a warning raises an expectation of responsibility on the part of the warning agency. As an economic disaster, recovery is politically complex since, without a loss of life, a "disaster" takes on a different aspect in the public's mind. Some parts of the economic disaster are directly caused by preventative measures; for example, the forced sale of cattle in Costa Rica in the face of an El Niño-related drought drops the price of cattle and impoverishes the farmers. There is an expectation on the part of those warned that the warning agencies will remedy the problems, which arise. The justice of this expectation is questionable, since it might be argued that the farmers were saved from greater losses. In effect, the warning allowed them to "cut their losses" in anticipation of a potentially greater misfortune. Unfortunately, this is impossible to demonstrate, and one might always argue that the warning was in some sense misleading, thereby causing an economically disadvantageous decision on the part of the farmer. The lesson learned is that the warning of a possible disaster causes a series of actions, which may be negative over the long run, and there is an expectation of responsibility for the warnings because of these negative impacts.
- The level of detail in an El Niño forecast is so low that in some cases it tends to mislead rather than inform. While the forecast of a drought in Guanacaste was correct, cattle were moved to an area where no drought was forecast, but an area in which one later occurred, creating an intensified crisis. Since the forecast drought turns out to be functional rather than absolute (i.e., an effective drought as a result of rain at the wrong time rather than too little rain), the forecast of a drought may cause counterproductive water conservation measures in water control and hydroelectric facilities. If water is maintained at maximum levels in expectation of drought, short intense storms can cause flooding because of improper preparation. This type of flooding has not actually occurred, but it is a possibility.
- The determination of the economically appropriate response to a potential El Niño event is difficult because of local variability in climate, and uncertainty as to the extent of an expected drought. As a result, the proposed measures may be too extreme and could cause more harm than good. A year after the end of the 1997-98 event, some in Costa Rica questioned whether the recommendation to sell off cattle had in fact been justified, in view of the economic losses that it caused. The alternative of keeping the animals in situ and having the cattle lose weight, and having some die, may have been the more appropriate response.
- The response to the forecast of El Niño must be geared to each national situation. In Costa Rica, a simple announcement resulted in an immediate response on an official level. The National Meteorological Agency exercised extreme caution in its announcements and advisories to avoid an extreme public overreaction. In other Central American countries, the same type of announcement resulted in very little preparation, possibly requiring a more emphatic announcement of the forecast and possible consequences of El Niño than that required for Costa Rica.
- Much of the discussion regarding the forecast of El Niño uses concepts and language which are not entirely meaningful to the general public. More effort is necessary to interpret meteorological information about El Niño and to present it in language that is specific to the different sectors, such as agriculture, hydroelectricity, or potable water services.
- As an occasional event (i.e., irregular but recurring), there is a tendency for a lack of continuity in preparedness for El Niño. Since Costa Rican governments change every 4 years, and personnel in ministries change as well, the technical awareness of preparation and response methods has a tendency to dissipate.
-
Ramirez, P., 1992: Descripción de situaciones climatologicas que pueden producir desastres en Costa Rica. Revista Geográfica de America Centrál, Nos. 25-26, 153-169.
[ Return to Executive Summary ] -
IMN, 1991: Boletín #2, "Sobre la evolución de un fenomeno de ENOS entre 1991 y 1992." 12 Sept 91; IMN, 1991: Boletín #10, "Evolucion y perspectivas de la posible presencia de un evento ENSO durante 1991." 20 Feb 91; IMN, 1992: Boletín #8, "Sobre la evolucion de un fenomeno de ENSO entre 1991 y 1992." 18 March 92.
[ Return to Executive Summary ] -
IMN, 2000: Boletín Meteorologico Mensual. Oscilación Decadal del Pacifico. January.
[ Return to Executive Summary ]
