Climate Variability, Climate Forecasting, and Society
The 1997-1998 El Niño provides a dramatic example of the effects relatively short-term climatic variations have on society and the potential value of forecasting them. The key indicators of a strong El Niño, including a sharp rise in sea surface temperature in the tropical Pacific Ocean, were detected by March 1997. Sea surface temperatures in the Eastern Pacific reached record values of 5 degrees Celsius above normal by June, and researchers were comparing the strength of the event to the 1982-1983 El Niño and recalling the worldwide impacts of that event. What made the 1997-1998 El Niño different was that scientists were monitoring the event as it developed and making predictions of its evolution 3 to 6 months ahead. Although the forecasts disagreed somewhat on the intensity, timing, and geographic extent of the emerging event, there was sufficient agreement for several national meteorological services, including the National Oceanic and Atmospheric Administration (NOAA), to issue advance advisories. The media interest in these predictions was unprecedented, and a number of groups took steps to prepare for the impacts.
In the U.S. state of California, preparations for the predicted higher-than-average winter precipitation and unusually severe storms included government planning for emergency response, the reinforcing of hill slopes and coastal defenses, and insurance purchases and roof maintenance by homeowners. More than $100 million was spent on levee repairs and, in the last quarter of 1997, California flood insurance policies increased by 40 percent. Between January and May 1998, California re-
ceived 228 percent of normal precipitation (NOAA press release, June 8 1998, El Niño and Climate Change record temperature and precipitation), and by June 1998 the state was estimating $500 million in property damage (USA Today, June 12, 1998). In other regions of the United States, El Niño was blamed by some for an unusually high number of tornadoes, resulting in more than 120 deaths.
On the positive side, El Niño was credited for unusually warm winter weather in the Midwest and the Northeast that brought lower heating costs for consumers, downward pressure on oil prices, a longer construction season, decreased snow removal costs, and other benefits. On the East Coast, no hurricanes hit land in the 1997 hurricane season, which reduced disaster losses but increased fire risk in Florida. In the Southwest, where El Niño brought more winter rains, the increase in vegetation and wildflowers boosted tourism but increased allergies and concerns about diseases such as hantavirus.
In Latin America, as reported in The Economist (May 9, 1998), the costs attributed to El Niño were large. Drought caused water shortages, crop failures, and wildfires in Mexico, Central America, the Caribbean, Colombia, Venezuela, and northeast Brazil. Floods drenched Ecuador, Peru, Chile, Argentina, Paraguay, and Uruguay, and fall hurricanes struck Mexico's Pacific coast. El Salvador's coffee production dropped by 30 percent, and the Colombian government reported a 7 percent drop in agricultural output because of drought. In northeast Brazil damages were estimated at $4 billion. Nine million Brazilians suffered from food shortages, and more than 48,000 square kilometers of forest burned in the state of Roraima. In drought-stricken Central America and Colombia, urban areas relying on hydropower had long power cuts. In Mexico, 400 people died when Hurricane Pauline hit Mexico's Pacific coast in October 1997 (Gobierno de Oaxaca, 1997); the hurricane's intensity was widely attributed to El Niño. Forest fires caused by drought due to El Niño burned about 400,000 hectares in Mexico in spring 1998 (Comision Nacional Forestal, 1998).
Although the Peruvian government, heeding the forecasts, had prepared for rains by rebuilding dikes and reinforcing bridges at a cost of $300 million, the floods destroyed more than 300 miles of roads and 30 bridges and displaced 300,000 people. In Ecuador, infrastructure damage from floods exceeded $800 million. In southern South America, the Paraná and other rivers overflowed, displacing thousands of people, killing cattle and destroying crops. The losses in Argentina were estimated at $3 billion. Fish catches declined, particularly in the Chilean and Peruvian anchovy and mackerel fisheries.
In Asia, El Niño was associated with drought and vast forest fires in Indonesia and with heat waves in India. In Australia, it was associated
with unusually dry conditions. However, in southern Africa, where governments prepared by building food reserves, the devastating droughts that had occurred during the 1982-1983 El Niño were not repeated in 1997-1998, and the rainy season (until June 1998) was relatively normal.
From a health perspective, the extreme weather events were associated with many disease outbreaks. In Latin America, flooding was associated with significant upsurges in malaria and cholera in Ecuador, Peru, and southern Brazil. Heavy rains in the Horn of Africa precipitated outbreaks of cholera, malaria, and Rift Valley fever. In Asia, drought was associated with poor water quality and cholera. The massive forest fires in Indonesia, as well as in Brazil, Mexico, Central America, and Florida, inflicted widespread respiratory illness. Poor air quality also affected trade and tourism, and fires in tropical forests have adversely affected wildlife and ecosystem functioning, as well as releasing additional carbon into the atmosphere (Epstein, 1998; Stevens, 1998).
In addition, high sea surface temperatures have taken an enormous toll on sea life, especially marine mammals. During 1997-1998, significant marine mammal mortalities were linked to El Niño on the Pacific coasts of the United States, Peru, Venezuela, and the Galapagos Islands and in the southeast Pacific, New Zealand in particular (Epstein, 1998; Stevens, 1998). These effects may have been caused by the migration of food sources, enhanced blooms of toxic phytoplankton, and/ or changes in the immune systems of marine mammals.
In sum, the 1997-1998 El Niño had major negative impacts on many people and regions and also brought significant benefits to other people and regions. The availability of accurate forecasts of extreme weather led some people and organizations to act in ways that spared them even worse damage. However, many others in these areas did not hear or respond appropriately to the forecasts, and, in other areas, forecasts were wrong and some prepared for forecast disasters that did not arise.
The experience of 1997-1998 strongly suggests that there is great potential social value in the developing ability to forecast climateaverages of temperature, precipitation, and the likemonths to a year or more in advance. Improved forecast skill, that is, accuracy beyond annual and seasonal averages,1 may open up a vast array of possibilities for the use of climate information to reduce the risk of damage from unfavorable cli-
1 The term ''forecast skill'' has precise meanings in meteorology. Commonly, skill is measured by the correlation between the forecast and actual values of an index of some weather or climatic event or by the average of the root-mean-square error over the length of a forecast (National Research Council, 1996a). The concept of forecast skill is described further in Chapter 2.
matic events and to seize the benefits of favorable ones. A premise of this study is that improved climate prediction can reduce the negative effects and enhance the positive effects of seasonal-to-interannual climate variability.
However, it seems clear that only a portion of this potential has so far been realized. We do not fully know how people responded to the predictions of the most recent El Niño, how much benefit the forecasts brought to those who did respond, or how much additional benefit there might have been if responses had been more appropriate and widespread. We also know little about how seasonal climate forecasts should be organized and forecast information disseminated in order to have the best possible effects. This book examines what is known and what needs to be known to enable climate forecasting to achieve its potential value for society.
To address this issue, we raise and discuss a broad array of questions. How well adjusted are human systems to the various forms of seasonal and interannual climatic variation, from the commonplace fluctuations that people ordinarily expect and prepare for to infrequent, extreme events that cause major disruption? Which economic sectors, segments of populations, or regions seem most sensitive to seasonal-to-interannual climatic variability? What is the net impact of a major climatic event such as the recent El Niño, and how is it distributed among those who suffer or benefit? How can one separate the impact of such a climatic event from other simultaneous influences on economies, ecosystems, and societies? Are those who are sensitive to seasonal-to-interannual climate variability able to use improved climate forecasts to improve efficiency or reduce risk? If they are able, under what conditions do they use climate forecasts to improve their well-being? How did people and organizations respond to the most recent forecasts and interpret the uncertainties in them? Why did some countries, organizations, and individuals respond when others did not? What role did mass media coverage play in public perceptions and institutional responses to the event? How will the perceived success of the most recent predictions affect responses to future forecasts? What will be the effect of the forecasts' failure in some regions? How can future forecasts be made more useful than those of the past? This book considers how to develop a research program aimed at answering such questions. Such a research program would have two main goals:
Climate Variation And Society
The climate system is a fundamental natural resource of the earth. It is driven by the sun and contains the gases necessary for photosynthesis, and is thereby the foundation of all food chains necessary for human life. It keeps the temperatures on the earth's surface within the narrow range tolerated by life. It drives the biogeochemical cycles that distribute nutrients and water about the biosphere. It delivers the water for shipping, irrigation, municipal consumption, and hydroelectric generation. It generates wind to turn windmills and makes snow for skiers. It also provides warm, sunny days that please the senses. In short, climate is thoroughly involved in virtually every aspect of the environment and human activity.
Human beings and societies have always had to cope with variations in weathershifts of wind, temperature and precipitation that can be extreme and that are experienced on the time scales of minutes, hours, and days. Humanity has also always coped with variations in climateaverages of weather on longer time scales. Seasonal variations affect the need for clothering and the availability of food and water, and people have responded by varying their diets and clothing and developing systems of building construction and food and water storage. And, at least since biblical times, the potential to experience years of plenty followed by years of famineinterannual climate variabilityhas been a major issue for societies. Climatic variations have contributed to the rise and fall of societies throughout human history.
People can respond to climate in several ways. At the most general level, people adapt to the average or mean climate of the region in which they live, on the assumption that the average of past experience is the best guide to the future. Thus, people in desert regions develop irrigation, design housing, and adapt their lifestyles to cope with the hot, dry conditions they routinely expect. Farmers choose crops appropriate to the average local climate and its usual variability and develop agricultural calendars that give a recommended day for planting. People also respond to observed conditions of climate and weather after the fact. Farmers wait to plant until the rains actually begin or apply more irrigation on hot days. Households adjust home heating and air conditioning in response to observed temperature and humidity. And people respond to forecasts, both of weather and of climate, with a range of anticipatory actions that depend on the lead time and reliability of the forecast. A farmer may decide not to plant at all if a drought is forecast; a water manager may adjust plans for reservoir control.
In responding to climate, people may act both to minimize the risk of hazardous climate and to capitalize on climatic opportunities. Flood-
control dams, for example, minimize the risk of floods and also enable farming to take advantage of abundant sunshine and warmth in a dry growing season by adding stored water from previous, wetter seasons.
People have always sought the ability to predict the weather and climate in the belief that this ability would bring great benefits. Developments in weather forecasting over the past few decades have confirmed this belief. It is now possible, for example, to warn human populations about approaching hurricanes and tornadoes and thereby greatly reduce loss of life from these extreme weather events. Weather information can now be arrayed in forms that enable decision makers to fine-tune activities so as to get the best possible outcomes from the weather conditions they experience. The focus of this book is on how to achieve similar benefits from the recent impressive advances in understanding the mechanisms that regulate climatic variability on seasonal-to-interannual time scales in many tropical and some temperate regions and in skillfully forecasting climate on these time scales.
Use of Climate Knowledge to Improve Well-Being
Climatic resources are exploited best by human beings when human activities are attuned to the types of climatic variations (mean conditions, seasonal-to-interannual variability, and the frequency and magnitude of extreme events) that affect their outcomes. Although human societies are not perfectly attuned to the seasonal and interannual rhythms and anomalies of climate, societies have co-evolved with local climatic resources to the point that our species generally copes well with a range of expected climatic conditions. Humanity has developed a variety of coping systems that function within individuals, small groups, firms, industries, societies, and governments. At the individual level, people keep coats and umbrellas handy if they live in climates that get cold and rainy. Farmers grow a mix of crops that has proven profitable over the long run under expected climate conditions that include some outstanding years, some bad ones, and a lot in between. Engineers design a certain amount of excess capacity into reservoir operations to take account of natural variability in precipitation and thus are able to meet demands for water under most climate conditions. At the group level, many communities develop norms that require the sharing of resources to help those harmed by extreme climatic events.
Human beings also adjust their societies to the risks of a variable climate and codify these responses in human institutions. Nomadic pastoralism provides a basis of subsistence and a structure for society for some human groups living in climates with scarce and highly variable precipitation. When moisture is too scarce in one location, people move
to locations where it is adequate to produce required supplies of food and fiber. Early agricultural societies like those of the Nile delta were built around seasonal variations in water flow, which affected their technology, their social organization, and even their religious beliefs. In many modern societies, a hazard insurance industry and programs of disaster transfer payments from government have arisen to help offset social and economic loss from the extreme weather conditions that are part of a variable climate.
Knowledge about climate is used not only to respond to extreme eventsby reducing risk and exploiting climatic "windfalls"but also to make minor adjustments to improve efficiency when variations are less extreme. In the United States, for example, an entire industry of consulting climatologists has developed to provide tailored climatic information routinely to clients in sectors such as the hydroelectric power industry, which can use this information to make incremental adjustments to planning and operations.
When Climate Becomes Hazardous
Climate does not always stay within the limits that social institutions plan for, and human adjustment is not perfect. One-hundred-year floods occasionally occur in consecutive years in the same watershed. Killing frosts occurring days, even weeks, after the "95 percent probability of last frost date" may happen two years in three. In such situations, when conditions fall outside the range of the expected, climate can become a hazard. An additional recent concern is the possibility that global climate change may increase the frequency or magnitude of extreme climatic events such as heat waves and major storms, making the systems that societies have put in place to cope with such events no longer adequate.
Climatic hazards come in many forms, from rapid-onset, short-lived events such as hurricanes, hail storms, and blizzards to slow-onset, long-lived fluctuations such as droughts. When climatic knowledge is poor, preparedness is low, and coping systems inadequate, climatic hazards exact severe social, economic, and environmental costs. By the same token, departure from normal climatic conditions can create new opportunities to be exploited.
Climate Sensitivity and Vulnerability
The sensitivity of human well-being to climatic variation is the extent to which important outcomes change as a function of that variation. Sensitivity is mainly indirect, in that climatic effects on human health and socioeconomic systems are in large part mediated by climate-sensitive
biophysical systems. For example, human nutrition is sensitive to climate mainly because crop production is sensitive to climate, and crop production is sensitive because of climatic effects on such factors as local rainfall and the spread of crop pests and diseases. So both biophysical and socioeconomic systems may be sensitive to climate, and many of the socioeconomic effects are due in part to the biophysical ones.
The human consequences of climatic variation depend on the behavior of social systems as well as on biophysical events. To the extent that a society or social group understands or accurately anticipates climatic events and their biophysical effects, it may be able to buffer the negative effects of these events and take advantage of climatic opportunities, thus decreasing sensitivity on the downside while exploiting it on the upside. Modern production agriculture and those whose livelihoods depend on it remain sensitive to variability in temperature and precipitation despite decades of technical and social innovation aimed at reducing sensitivity by controlling access to water; limiting infestations of pests, weeds, and diseases; insuring against catastrophic loss; developing drought-tolerant and disease-resistant seed varieties; and the like.
Often sensitivity is greatest at ecological, economic, and social margins. Faunal and floral communities in areas straddling the margins (boundaries) of ecosystemsnatural and managedare less stable with respect to climate variability than communities safely in the interiors (Blaikie and Brookfield, 1987). Similarly, the poor, the elderly, the infirm, and other marginal segments of society often bear a disproportionate share of the total social costs of climatic variability (Blaikie and Brookfield, 1987). In such cases, a relatively minor climatic fluctuation may cause disproportionately large consequences. With appropriate policies in place, the most affected groups may therefore gain great benefits from the use of climate forecasts.
Our definition of sensitivity includes human efforts to adapt to climate in that it refers to outcomes after taking into account things people do to cope with expected climatic variations. This definition contrasts with that employed by some other writers, whose concept of sensitivity presumes that the human consequences of climatic events can be meaningfully analyzed independently of adaptive behavior. We do not find this approach useful because, as we elaborate in Chapter 3, human societies, and particularly the conduct of weather-sensitive activities, has coevolved with climate and has always included a range of adaptive strategies. Thus, sensitivitya measure of the functional relationship between climatic events and human outcomesis a property of human groups or activities that have particular adaptations in place. Changing the adaptations can change sensitivity.
We use the term "vulnerable" to refer to human groups or activities
that face the risk of extreme negative outcomes as a result of climatic events that overwhelm the adaptations they have in place. Vulnerability, like sensitivity, is a function of both climatic events and human adaptation. We use separate terms to reflect the special importance most societies give to the risk of catastrophic (i.e., extreme negative) outcomes. It is important to recognize that, as with sensitivity, human activities can increase or decrease vulnerability. For instance, urban development in hurricane-prone coastal areas increases the risk from hurricanes even when the frequency of hurricane events remains unchanged. Increasing population and affluence in the arid western United States have stimulated rising demand for essentially fixed water supplies; this has increased the risk from drought apart from fluctuations in precipitation. Systems of flood-control dams decrease vulnerability to flood damage from most major storms, but they may increase the damage caused by the most extreme ones. Actions that affect the distribution of income also affect the vulnerability of human populations to extreme negative climatic events by altering the resources people have to prepare and respond.
Sensitivity and vulnerability to climate variability constantly change over time. Some reduction or increase in sensitivity, and particularly in vulnerability to extreme events, may be the unintended result of fundamental structural social changes accompanying social development. For example, as the general level of affluence and technological sophistication rises in a developing country, changes in food preferences (for example, wheat over millet, meat over grain) may lessen (or strengthen) dependence on resources that are directly affected by seasonal-to-interannual climate variability. As people depend increasingly on world markets for food, their well-being becomes less sensitive to local climate variations, but perhaps more sensitive to distant climatic events that may threaten their supply lines.
The Potential Usefulness of Climate Forecasts
Climate forecasting can benefit people by allowing them to change the things they do to anticipate climatic events, thus reducing their sensitivity to negative events and perhaps increasing their sensitivity to positive events. The potential value of skillful climate forecasts may or may not be greatest in those regions where the predictive skill is the greatest. The greatest value may be found in the regions where climate variability has the largest economic impacts (positive or negative), or where vulnerability is greatest and adequate coping mechanisms can be provided. In regions where impact or vulnerability is very large, even a small increase in forecast skill may be of great value, even if the predictions are not as certain as in other regions. Therefore, a focus on improving forecast skill
for those regions where the physical links are strongest may provide the highest scientific payoff, but it may not provide the most significant economic or humanitarian payoffs. Such considerations may imply that there is much to be gained by shifting some predictive effort from regions such as Latin America and Southern Africa that are highly sensitive to the El Niño/Southern Oscillation (ENSO) phenomenon to regions such as Europe and West Africa, where outcomes may be highly sensitive to Atlantic climate variability or to monsoon predictions for Asia, even though predictive skill is currently very limited.
Improvements in the skill of forecasts, combined with the expectation that the new knowledge will not be used with perfect efficiency, means that it may be possible to deliver forecast information in ways that lead human groups to cope more effectively with seasonal-to-interannual climatic variability, reduce sensitivity to the downside of climatic variation, and take better advantage of climatic opportunities.
Therein lies the crux of our concerns here. The eventual value of improved forecasting skill will depend on how people and organizations deal with the new kind of information. Are they likely to pay attention to it? Will they understand what the climate models mean for them? Will they trust the messengers? How will mass media organizations and other messengers transmit forecast information, and how will their messages be interpreted? Are recipients likely to systematically misinterpret the information given to make it conform to their preexisting ideas? How will they respond to the false alarms and false reassurances that any imperfect forecasting system sometimes produces and to the inevitable simplifications offered by mass media and other messengers? And what can be done to transform potentially useful forecasts into information that is actually used to benefit society?
Structure Of This Book
This book examines the state of knowledge and the needs for further knowledge relevant to understanding the effects of seasonal-to-interannual climate forecasts and making them more useful. Chapter 2 examines the current state of scientific capability to make skillful climate forecasts on a seasonal-to-interannual time scale and begins to address the question of what it would take to make such forecasts more useful. The information on climate forecasting is meant primarily as background for those outside the forecasting community; the section on usable knowledge is addressed both to forecasters and other readers. Chapter 3 considers what is known about the strategies people and societies have developed to cope with two qualities of their environments: that climate is variable, and that (until recently) climate variations have been essentially
unpredictable. It summarizes the state of knowledge about the coping strategies used in specific climate-sensitive human activities and about human institutions, such as disaster insurance and emergency preparedness, that have developed to help cope with climatic variations.
Chapter 4 takes up the question, critical for making climate forecasts more useful, of how individuals and organizations are likely to respond, and how they might be led to respond more effectively, to the information in climate forecasts. It considers the ways in which climate forecast information might be useful and then considers available sources of information on how the coping systems people have developed for climate variability might respond to new information. These include actual responses to recent climate forecasts; research on how people assimilate information generally; and past experience with efforts to provide other kinds of scientific and technical information that people might use to improve their well-being, including information on practices to promote personal health and information from hazard warning systems. The chapter concludes by summarizing the state of knowledge and some promising hypotheses about how individuals and institutions are likely to respond to climate forecast information and how to make these responses more effective.
Chapter 5 examines the state of concept, methods, data, and knowledge that could be used to measure the human effects of climatic variability and the potential and actual benefits of skillful climate forecasts. It presents a conceptual framework and raises several issues that must be addressed to make such measurements, summarizes the state of scientific efforts to estimate the effects of climatic variations and the benefits of forecasts, and presents the panel's findings on these issues.
Finally, Chapter 6 summarizes the findings of the study and identifies a dozen scientific priorities—sets of research questions that, if pursued, will yield progress toward the ultimate goals of understanding and increasing the social value of seasonal-to-interannual climate forecasts. The questions fall into three broad categories: research on the potential benefits of climate forecast information, on improved dissemination of forecast information, and on estimating the consequences of climatic variations and of climate forecasts.