Scientists need more money from the insurance industry to help them be able to predict killer weather patterns more accurately, according to a leading tornado expert.

Charles Doswell III of the National Severe Storms Laboratory in Norman, Oklahoma made the comment while he was addressing a group of insurance brokers last week in Hamilton.

He was taking part in a workshop for members of the industry and leading US meteorologists on tornadoes and hail organised by the Risk Prediction Initiative -- part of the Bermuda Biological Station.

"We always want more money for research,'' said Mr. Doswell at the end of his talk on the value of statistics on the weather. "We have made a lot of progress and we should be proud of what we have done.

"It is possible that in the next 20 years or so we will get more accurate information about the distribution of tornadoes in the US.'' As tornadoes cause $420 million worth of damage to property in the US and American farmers pay $2.1 billion a year to insure against hail damage, they are of prime interest to the Island's insurers who gathered to hear a dozen scientists speak on their research into weather patterns.

Mr. Doswell spoke about how fallible the collection process for data on tornadoes is, varying from highly populated places where there are more eyewitnesses and more sightings recorded, to changes in personnel and their varying levels of dedication about recording the data and swells where tornado chasers -- like those shown in the movie Twister -- log and record data diligently.

And he went on to show how these factors could be smoothed out to allow an overall picture of where tornadoes are most likely to hit and how far they spread across the States.

He said of reading the data they had available: "I think we are better than we were a year ago, and I hope that a year from now it will be even better.'' But he said scientists needed more money to make the information gathering process more accurate.

Mr. Doswell was a guest of the Risk Prediction Initiative (RPI), which was set up to help members of the insurance world meet up with and speak to the scientists doing work on severe weather.

It allows the scientists to get funding for projects and the insurance world to have more tailored data for their risk assessment needs.

The high-level tornado and hail workshop looked at how science can help reduce the impact of severe weather on insurers.

Workshop participants included insurers, reinsurers and weather experts from the US, Canada and Europe.

The workshop was designed to introduce insurers to the latest in severe weather research. Care was taken to speak in lay terms and complicated methods of working out trends explained to the insurance men.

The issue of cloud seeding reducing hail risks, the climatology of tornadoes, and the effects of El Nino on weather in tornado areas was also looked at.

One of the aims of the workshop was to identify areas where further insurance-relevant research is most needed.

Studying the severe weather in North America is a fundamental part of the insurance industry. It is worth billions of dollars in premiums and has a heavy cost in lives -- this year alone tornadoes in the United States have killed 27 people and 30 in Canada.

The RPI holds regular workshops -- about three or four a year -- for its industry sponsors on a number of key insurance and reinsurance issues.

Corporate sponsors of RPI include XL Mid Ocean Re, General Re, Renaissance Re, ACE USA Inc., Zurich Group/Centre Solutions (Bermuda) Ltd, Employers Re, AIG/IPC Re, American Re, State Farm, USAA and Swiss Re.

The cost of hail: A scientist speaking to insurance experts at a workshop in Hamilton last week.

Twister: A tornado sends debris into the air in Fort Morgan Colorado in 1996.

The tornado destroyed an abandoned dairy and tore off part of the roof of a nearby house. The insurance industry needs data on tornadoes to be able to fairly evaluate the risk involved in insuring buildings, businesses and lives.

Scientists, funded through the Risk Prediction Initiative, say they are getting closer to predicting when and where weather patterns will strike.

BUSINESS BUC