Introduction

Why should society care about weather?

Several points stand out. First, it may come as a surprise that the largest loss of life in recent years has been associated with extreme temperatures. Second, among the first four phenomena listed in the box, floods result in the most deaths, followed by winter storms and tornadoes, and finally hurricanes. Lightning has perennially been associated with a large loss of life.

In terms of the economic losses associated with extreme weather, a conservative estimate of national losses is on the order of $300 million per week. The number is conservative because it neglects the effects of inflation, covers only the direct impacts of extreme events, and leaves out the costs associated with extreme temperatures, which are certainly significant. The actual total economic impacts associated with extreme weather events is likely to be several times that estimated here.

Trend data can mislead. Workshop participants pointed out that underlying the data on extreme event impacts are sub-trends in climate patterns and changes in society. For instance, hurricane damages have risen almost exponentially in recent decades during a period of decreased hurricane activity. The reason for the increase in damages is the enormous coastal growth, placing more people and their property in vulnerable locations. As a consequence of societal change, historical impacts data is likely to underestimate today's vulnerability to weather.

In recognition of our relatively poor understanding of trends in impacts and the underlying vulnerability to those impacts, participants at the Workshop put together a qualitative summary of trends in impacts, event frequency and intensity, and causes for the trends. This figure is shown in Table 2.

The figure shows that impacts associated with all extreme events, on a roughly 20-year time scale, are perceived as increasing in terms of both deaths and dollars (with the single exception of tornado-related deaths). Additionally, the figure shows that participants perceived that the increase in impacts was largely due to societal factors with respect to each phenomenon, expect for perhaps floods. Participants identified a need to better understand the interrelationship of climatological and societal factors which underlie the trends in impacts.

Economic efficiency and competitiveness
Participants felt strongly that the cumulative impacts of more typical day-to-day weather were significant, and probably much larger than the attention-getting extreme events. Day-to-day weather associated with temperature, precipitation, winds, etc. has the potential to disrupt decision making, thereby adding to the "cost of doing business." The impacts of weather on the day-to-day costs of business are thus a matter of economic efficiency and competitiveness.

Present at the meeting were experts from the following industries: oil and gas, electric power, surface transportation, agriculture, aviation, and insurance. They related numerous examples of the effects of weather that decision makers face in their day-to-day operations and the potential value of improved weather information and information use. Examples of impacts and their potential reduction include:

Oil and gas exploration:

• improved forecasts of tropical weather conditions (wind, waves, disturbances) can reduce delays in drilling operations at a cost of up to $250,000 per rig per day (several thousand rigs in the Gulf).

• improved hurricane track predictions could reduce days of production shutdown, each day of which costs the industry and the U.S. treasury a combined $15,000,000.

• improved temperature and precipitation forecasts can lead to greater efficiency in chemical spraying (e.g., pesticides), which costs $10-$15 per acre per application for hundreds of thousands of acres.

• on a national scale the annual cost of lost production to the vegetable processing industry, primarily due to weather, is $42,500,000.

• a single hurricane could lead to more than $50,000,000,000 in damages.

• weather-related catastrophes have led to more than $48,000,000,000 in property insurance claims over the period 1989-1993.

• it costs $2,000 per hour to stop a train. A single tornado warning covering 15 miles of track for 15 minutes can lead to seven stopped trains.

• most weather-related derailments cost $1,000,000 to $5,000,000.

• using improved thunderstorm forecasts could save one utility $200,000 annually in reduced outage time.

• using "good QPF forecasts" could save one utility $2,000,000 over five years.

• using improved temperature forecasts could save "hundreds of millions annually nationwide for the utility sector".

• every avoided cancellation saves $40,000, every avoided diverted flight saves $150,000.

• for the 16 members of the Air Transport Association, delays and cancellations cost $269,000,000 annually.

How can we better understand the use/value of weather information to society?

Available Tools
Social scientists utilize a range of methodologies to research the use and value of information to decision makers. Examples include:

• prescriptive studies assume that decision maker behave in a manner consistent with prescribed principles, such as maximization of expected utility. Such studies typically involve formal modeling of decision situations.

• descriptive studies focus on the context in which actual decision makers operate in order to describe their behavior. Such studies can use models, but also can rely on narrative techniques.

• decision analysis focuses on choosing among a set of alternative actions based on expected probabilities and values associated with various outcomes.

• institutional analysis focuses on the role that various institutional factors such as goals, rules, norms, authority, control play in creating circumstances for effective or less-than-effect decision making.

• microeconomics is the area of economics that focuses on the behavior of individuals and firms in a market setting.

• macroeconomics is the area of economics that focuses on the entire economy in an aggregate sense.

• survey techniques are used to assess the value of certain information or services to decision makers through, for example, assessing their willingness to pay.

Value to Society is a Function of Information Quality AND Information Use

It has long been recognized that information acquires value through influencing the behavior of decision makers. A perfect forecasts is of no value if it is unavailable to or unusable by a decision maker. Consequently, attention must be focused on the use and value of weather information in parallel to ongoing efforts to improve the quality of the information. Participants at the Workshop noted that, typically, greater attention is paid to improving the quality of information than it is to improving the use of information. As a result, decision makers do not use existing information products to their full potential. This theme was reinforced by the representatives of the "user community" at the Workshop who felt strongly that existing weather information is not as well used as it could be.

Meeting Recommendations

1. Establish an ongoing user-based group to advise the USWRP (and others) on needed research and feedback information on research value.

2. Focus a complementary research effort, parallel to research focused on improving weather information, in the area of the use and value to society of weather information. Two areas are particularly important:
   1. Conduct a comprehensive assessment of the use and value of weather information in a particular community (population <100,000) in order to systematically identify opportunities for and constraints on improved use of information by various decision makers. Work with similar initiatives under way by the Federal Emergency Management Agency, Insurance Institute for Property Loss Reduction, and Center for Disease Control. Such an assessment could ultimately lead to the development of "tool kits" for decision makers to help guide use of weather information. Such "tool kits" would be tested as prototypes prior to operational use.

   2. Support focused evaluations of the use and value of weather information in various decision contexts, in particular public and private sectors in order to establish an understanding of "what works" and "what is needed." Such evaluations could be sponsored through a modest grants program.