Ms. Clark compared insuring cat risks today to insuring fire risks in the early 1900s. She said that fire used to be an unknown and costly risk, but today, the risk is stable for insurers. Now catastrophe risks are growing, Ms. Clark said, and much like fire risks earlier, society needs the insurance industry's help to deal with them.

Three recent factors led insurers to doubt their abilities to adequately write catastrophes, Ms. Clark said: the losses from the 2004-2005 hurricane seasons, updated models and the global warming debate.

With respect to the bad hurricane seasons, Ms. Clark said that a storm like Katrina should not have been a surprise to the industry, as intense storms causing even more damage have been predicted for a while.

She noted that insured losses from Hurricane Andrew in 1992 reached $15 billion but would have been four times higher if the storm had hit nearby Miami instead of Homestead, Fla. That, she said, would have put the damages at $60 billion--$20 billion higher than the insured losses from Katrina.

Ms. Clark said that because of reactions from insurers, reinsurers and rating agencies to Katrina, the industry became too reliant on catastrophe models, which were updated after the 2005 hurricane season.

Insurers "stopped thinking about risks independently," she said. Rating agencies also demanded modeling numbers, asking about insurer exposure to 1-in-100-year storms and 1-in-50-year storms.

But Ms. Clark--who founded the first catastrophe modeling company, Applied Insurance Research (AIR), which later became AIR Worldwide Corp.--said that models are only best estimates and are not designed to replace underwriters or to be the final word in which risks are acceptable and which ones are not.

She noted that models can change, and if an insurer's business is based solely off modeling results, then that insurer's business model will have to keep changing as well.

"We've become a modeling society," Ms. Clark said, noting that model results should not replace common sense. She pointed out that some insurance company executives have admitted to her that model results have looked wrong, but the company cancelled risks based on the numbers anyway. The key to using models effectively, Ms. Clark said, is to understand the models and use them as a tool, rather than a truth.

She also pointed out that models can contain uncertainties, limitations and inaccuracies, and she added that unknown factors can greatly alter the models' results.

As an example, Ms. Clark said the current subprime/credit crisis happened because of an overreliance on models.

Many smart people, she said, developed models showing how money can be made off of mediocre loans, and it worked well until the market experienced a factor that the models did not account for--the housing market slowdown. Once this occurred, Ms. Clark said, the models and the business decisions based on them fell "like a house of cards."

Ultimately, Ms. Clark said, there is much about atmospheric conditions that scientists and modelers do not understand, and that has an influence on modeling results. Insurers, she stated, have to figure out how to make decisions under uncertainties.

Regarding global warming, Ms. Clark said there will likely be repercussions as the Earth's temperature rises, but the 2004-2005 hurricanes, including Katrina, were not the result of global warming, and need to be put in perspective.

She said that 2004 and 2005 were bad years, but not unprecedented, and the real reason for the higher insured losses had more to do with sharply increasing property values than the actual storms.