Lessons for the Societal Application of Climate Information
Kenneth Broad
International Research Institute for Climate Prediction
Columbia University
Palisades, N.Y. USA
kbroad@iri.ideo.columbia.edu
Definitional Issues
It became clear from discussions that multiple definitions and categorizations of La Niña (and of El Niño as well) co-exist. There is no agreed upon index for La Niña. Within the scientific community, this uncertainty may be a healthy reality which encourages discussion, analysis, debate, etc. In fact, to this group, it may seem perfectly reasonable that there seems to be a "lingering El Niño and an incipient La Niña" occurring right now. This uncertainty, however, becomes problematic as it gets played out in public forums, where the media and decision makers are less versed in the scientific nuances that surround the variaty of classifications.
Scientific Versus Societal Perceptions
Scientists and society differ in their perceptions of what the categories of events (i.e., weak , moderate, strong, extraordinary) imply. Scientific experts judge events by a relatively bounded set of parameters that can be measured objectively. Society in general, however, is not interested in whether the events is "strong" in terms of, for instance, thermocline depth in the western or eastern tropical Pacific. They will judge events by the socio-economic and environmental impacts in their particular areas. Thus, confusion arises as "we" (scientists) categorize and predict environmental parameters, and decision makers interpret these as predictions of impacts in their regions.
This confusion with meteorological forecast versus impacts forecast is perhaps more acute in the case of La Niña, simply because we have less information on the impacts of La Niña events and in recent decades there have been fewer La Niña events to assess. While there may be some degree of collective memory of El Niño impacts in parts of the world that have robust teleconnections, the awareness of La Niña’s effects (both by scientists and decision makers) seems much less. Even the more popular impacts attributed to La Niña, such as the costly 1989 drought in the US Midwest, is still up for debate. At the conference we saw how La Niña impact was explained to the public by the media: simply as the opposite of El Niño (El Niño’s "evil sister") -- thus, bringing the opposite impacts to the same regions.
Probability Issues
As one of the Summit speakers put it: "once we imply attribution, we imply predictability." The scientific community is aware of the statistical insignificance of statements about impacts based on a sample of two or three, or tenuous proxy-data reconstructions. However, the public has no sense of the skill of predictions based on such evidence. Clearly, the media do not sell newspapers based on uncertainty, and the most popular magazines and newspapers are even more prone to sensationalism. A lack of understanding of uncertainty can lead to poor policy making.
Whose Responsibility
Scientists should not be expected to be specialists in education, public relations and marketing. Nonetheless, there are certain things that can be done to minimize the potential for the misuse of climate information by the public (in addition to improving reliability of information and forecasts).
From the point of view of the public, it is often difficult to distinguish forecasts from predictions from observations. Someone surfing the Interent, may be as likely to find an experimental prediction from an unvalidated model as a forecast which takes into account several models and other more local climatic conditions besides just ENSO. This begs the question as to whether there should be some sort of quality control for publically distributed information.
Results that are intended for sharing with colleagues, for instance, could be password protected, while forecast products which have been approved by the community as a whole could be distributed more widely in the proper form. This approach could minimize the burden of media relations, difficulties relaying probability limitations, qualified statements, etc. by reducing the amount of information publicly released. This would allow the information to be tailored so as to minimize misinterpretation and distortion of the information.
It is only in the last few years that there has been targeted social science research geared toward looking at the societal impacts, potential uses, and sociopolitical constraints on the uses of climate information in various regions in the world. Much has been done in the context of El Niño, while collecting only anecdotal information about La Niña’s effects. Efforts should be geared in this direction. Researchers could work with those producing forecasts to determine: 1) where the most consistent teleconnections are; 2) current forecast capabilities in these areas; 3) research on what type of information decision makers at all levels need; 4) what are possible negative, or unintended, consequences of introducing such information into the decision-making processes of various groups.
As our skills and our observational and communications networks increase, there will be better synthesis of climate information to compliment model-generated predictions. Hopefully, these tools, informed by social science, will enable us to better predict impacts that occur throughout the ENSO cycle of warm and cold events, and to disseminate more effectively this information to affected societies.