By Sean Henahan, Access Excellence

BOULDER, CO (May 29, 1996)- Supercomputers and hurricane-hunting high altitiude aircraft are among the tools used by scientists at the National Center for Atmospheric Research (NCAR) who are investigating how storms develop and how they might better be predicted.

Very small differences in air temperature and moisture can make the difference between a sunny day and major thunderstorm. Indeed, a temperature difference of only 1/16 of a degree C, which falls within the error of most weather observations, can spell the difference between potentially severe storms and no rain at all. Similarly small discrepancies in moisture can have the same effect.

NCAR scientists are using computer models of the atmosphere to conduct forward-sensitivity experiments. These bring a set of slightly varying scenarios forward in time to see how a given weather situation might evolve. Wind, temperature, and moisture values were derived from surface, radar, and radiosonde (weather balloon) observations. On average, these data had temperature errors of around 1/16 degree C and errors in liquid water content of around 1 gram per kilogram (g/kg). The latter is about 7 to 10% of the total atmospheric moisture present on a typical warm, humid day.

The researchers found that a temperature decrease of 1/16 degree C was enough to shut off storm development entirely, while a 1/16 degree increase led to a fourfold increase in rainfall. Similarly, rainfall dropped by 80% when liquid water content was lowered by 1 g/kg, while it more than doubled for a 1 g/kg increase. These findings reinforce the difficulty in forecasting where and when thunderstorms might develop on days when conditions are borderline. However, knowing the possible impact of observational error should help forecasters better quantify the uncertainty in a forecast.

NOAA Graphic: Tornado Alley

NCAR scientists have also developed a forecasting tool called "Auto-nowcaster." The automated thunderstorm forecasting tool is now undergoing its first major real-world test this summer. The Auto-nowcaster, which spots incipient storms and predicts their development up to a half-hour in advance, will be tested beginning in June at the Federal Aviation Administration's Integrated Terminal Weather Support Site near Memphis, Tennessee.

A cumulus cloud can blossom into a thunderstorm in as little as 10 to 20 minutes. Although operational computer forecast models have proven useful in predicting large-scale weather developments 12 to 48 hours in advance, they do not have the resolution to make accurate forecasts on the thunderstorm scale. With help from new techniques for analyzing Doppler radar data, the auto-nowcaster looks for gust fronts and other lines of converging air on which storms might be induced to form. These boundaries cannot be simulated directly by larger-scale computer models.

Other parts of the auto-nowcaster examine whether atmospheric conditions are sufficient to support storms once formed and how storm motion might evolve over time.

"It has been a major team effort to get the auto-nowcaster ready for field testing," says project manager Jim Wilson, "and we are very anxious to get feedback from this summer's test. We expect the auto-nowcaster will be issuing thunderstorm advisories on its own in the near future."

NCAR researchers will also be using airplanes and mobile radar to probe the chemistry of Colorado's notoriously intense thunderstorms. The project is aimed at documenting the chemical, dynamical, and electrical interchange between thunderstorms and their environments.

Entitled "Deep Convection and the Composition of the Upper Troposphere and Lower Stratosphere," the field program will take place in northeast Colorado during late June and July. It is one of three parts of STERAO, the Stratosphere-Troposphere Experiments: Radiation, Aerosols, and Ozone. STERAO is a multiyear study of the chemistry and dynamics of the upper troposphere (the atmosphere's lowest 15 kilometers, where our weather is shaped) and the lower stratosphere (the sensitive zone between 15 and 45 kilometers where the earth's protective ozone layer resides).

Water vapor and nitrogen are of particular interest in STERAO. Thunderstorms bring vast amounts of water vapor from the lower to the upper troposphere, but the exact trajectories are uncertain. Lightning is a significant source of active nitrogen, which can lead to the production of ozone, but the nitrogen's sources and sinks are not yet fully understood.

A high-altitude WB-57F aircraft, recently acquired by the National Science Foundation and operated by NCAR, is expected to make its research debut at STERAO. Also in the plans are a P-3 "hurricane hunter" aircraft from NOAA and a Citation operated by the University of North Dakota.

Among the instruments probing the Colorado storms will be: two mobile Doppler radars that can gather data from within several kilometers of severe storms; a lightning interferometer that will make unique three-dimensional observations of lightning channels; and a variety of devices for air sampling and analysis aboard the aircraft to assess the chemical make-up of air in and near the storms at both high and low altitudes.

Details on these research projects were presented in May 1996 at the 18th Conference on Severe Local Storms, the American Meteorological Society, in San Francisco.