A 'call to action' to sharpen local weather forecasting

The nation's weather services are at a crossroads. After three decades of improvements in daily global forecasts, it's time to meet the challenge of accurately forecasting the smaller-scale storms that are the most destructive and disruptive to human activities.

That is the essence of an unprecedented ''call to action'' issued this week by a consortium of 50 universities involved in atmospheric science, the University Corporation for Atmospheric Research (UCAR).

Entitled ''The National Storm Program: A Call to Action,'' the report is the result of two years of effort by more than 100 atmospheric scientists. It reflects a growing consensus within this community that the United States has the capability to significantly reduce the hundreds of lives and billions of dollars lost annually in severe storms. Less dramatic but equally as significant is the potential that more precise, two- to six-hour local forecasts hold for improving the nation's productivity.

''The time is right now to undertake a program to reduce the devastating loss of life and property and the economic and social impact of storm-scale weather. Technology is available to observe and track storm-scale events and to disseminate information and timely warnings; scientific methods are at hand to increase our understanding of storm-scale weather; and the incentive exists to apply this knowledge to operate our economy more effectively,'' the report states.

If the US government boosts spending on weather services by $60 million to $ 120 million annually over the next decade and redirects efforts toward this type of forecasting, UCAR scientists say property damage from severe weather could be reduced by 5 percent. This amounts to savings of about $1 billion annually. Improved hurricane forecasting has cut annual loss of life from 200 to 60 persons in 50 years. Based on this, the scientists estimate that it should be possible to halve the death toll from tornadoes, flash floods, and other severe storms by the end of the century.

Researchers admit that the current emphasis in Washington on reducing budget deficits makes a new weather forecasting initiative unlikely. In fact, the White House Tuesday announced plans to sell National Weather Service (NWS) satellites to private firms. The Reagan administration is also exploring other ways to turn some current NWS activities over to the private sector.

''Our purpose is to make decisionmakers in the government and the private sector aware of the tremendous opportunities,'' says George S. Benton of Johns Hopkins University, chairman of UCAR's Stormscale Operations and Research Meteorology (STORM) steering committee.

A pilot project conducted by the National Oceanic and Atmospheric Administration (NOAA) here illustrates some of the techniques that weather scientists say should be put into the hands of the nation's forecasters.

The three-year-old Program for Regional Observing and Forecasting Services (PROFS) uses computers to integrate, automate, and animate data from satellites, weather radar, and a dense local network of automated weather stations. The system helps forecasters spot conditions that generate thunderstorms, tornadoes, flash floods, and other destructive conditions.

''Currently, forecasters have access to only 1 percent of the weather data that are collected. PROFS allows them to use much more,'' says NOAA researcher Robert Bunting.

Local forecasters have access to large-scale weather forecasts prepared on large computers. They must compare these maps with what they see on local weather radar scopes. Each half-hour they receive printouts of the latest weather satellite images.

With PROFS, a properly programmed minicomputer can take these and a number of other sources of data, convert them into a single scale, and display them on one screen nearly instantaneously. This allows forecasters to superimpose satellite, radar, surface-wind diagrams, and other critical meteorological measurements. The computer can code these pictures in various colors to make important cloud features stand out vividly. It can take two satellite pictures of the same area and convert them into a stereo image or run a series of images to provide animation.

This system was tested last summer in the Denver area. It allowed forecasters to identify areas where thunderstorms were forming 30 minutes before they showed up on radar, and allowed them to narrow areas covered by severe storm warnings by 25 percent.

Another key to the STORM program is better dissemination of weather information. Sandy MacDonald, the director of PROFS, outlines one possible scenario:

It's the winter of 1995. You must drive from New York to Buffalo, and you want to know what the weather will be. So you sit down in front of your home computer and dial into a local information service like CompuServe or The Source. You call up its weather data base. You enter your itinerary and the time you intend to leave. After a brief pause, the computer might tell you that the weather will be fair for the first three hours of your drive. But 150 miles down the pike a light rain will begin to fall. This will turn into a freezing rain making driving hazardous. Your computer might even advise you to postpone your trip if extremely dangerous conditions appear to be in the offing.

Mr. MacDonald adds that a home computer also could act as an alarm to warn homeowners of impending severe weather, such as tornadoes.

''It's easy to concentrate on the extreme weather events,'' says UCAR's Edwin L. Wolff. He stresses that improved short-term, local forecasts also could have an economic payoff.

An illustration comes from a study by Harold C. Cochrane, Takayoshi Nakagawa, and Alan Harney of Colorado State University. For a single electric generating plant in Colorado, they estimate that improved four- to six-hour temperature forecasts could save as much as $72,000 in yearly operating costs.

The US shells out $1.7 billion annually to distribute weather information to the public. The question UCAR scientists are raising now is whether the nation's leaders are interested in spending 15 to 20 percent more to radically improve weather forecasting.