BI’s Article search uses Boolean search capabilities. If you are not familiar with these principles, here are some quick tips.

To search specifically for more than one word, put the search term in quotation marks. For example, “workers compensation”. This will limit your search to that combination of words.

To search for a combination of terms, use quotations and the & symbol. For example, “hurricane” & “loss”.

Login Register Subscribe



LONDON -- New initiatives are emerging to provide more advanced analysis and prediction of U.K. windstorm risks.

U.K.-based reinsurance broker Willis Faber Re and the U.K. Meteorological Office recently launched a new, joint venture computerized windstorm model.

It is the first time the U.K. Meteorological Office has joined forces with a company within the U.K. insurance and reinsurance industry to analyze windstorm risks, according to Oliver Peterken, executive director of research and development at Willis Faber Re.

Aon Group (UK) Ltd. also is looking to improve its current windstorm modeling techniques and plans to launch a new, more detailed model by the end of the year, said Matthew Eagle, associate director of research and development for Aon's U.K. non-marine reinsurance division.

Mr. Eagle said there is a general move within the United Kingdom to refine current windstorm modeling techniques.

"There is a recognition that the early models need to be updated, and we are moving to upgrade to a new generation," he said.

Mr. Peterken said Willis Faber Re, the worldwide reinsurance division of London-based Willis Corroon Group P.L.C., initiated the new project with the Meteorological Office because of Willis Faber Re's dissatisfaction with existing U.K. windstorm models.

Windstorm modeling is widely used throughout the U.K insurance and reinsurance industry, and Willis Faber Re has previously used a number of models, Mr. Peterken said.

But, he said, existing models used by U.K. brokers, including Willis Faber Re, and insurers, reinsurers and catastrophe modeling companies have a scientific weakness because of their reliance on observations of windspeeds at ground level.

Such observations are severely impacted by hills and buildings and do not provide accurate modeling data, according to Mr. Peterken.

He said Willis Faber Re decided to access the scientific experience and expertise of the Meteorological Office to develop a new approach based on modeling windspeeds in the upper atmosphere.

Windspeeds at such levels are calculated by a detailed, mathematical technique developed by the Office. The new model can transfer the upper atmosphere windspeeds of past windstorms to ground level to estimate their potential impact on current properties.

The new windstorm model is the result of 18 months of research and development by Willis Faber Re and the Meteorological Office. Neither would reveal its monetary investment in the project.

Aon's Mr. Eagle said Aon recognizes Willis Faber Re's criticism that current U.K. windstorm modeling techniques are not entirely accurate. But he said any windstorm modeling technique always will require some estimation.

Aon's current windstorm model is based on data from weather stations across the United Kingdom.

The new model will use digital terrain analysis to more accurately "fill in the gaps" between the weather stations, Mr. Eagle said.

He said the analysis will take into account factors such as ground elevation, distance from the coast and wind direction, and it will allow Aon to produce average windspeed grids of 1 square meter across the United Kingdom.

The new Aon model is being developed with advice from spatial statisticians from Oxford University.

Willis Faber Re's Mr. Peterken said another advantage of the new Meteorological Office model is it can be very geographically focused.

The Office has computer data on all extreme wind events in the United Kingdom dating back to the early 1970s.

Under the new model, that data is put into specially designed statistical analysis software supplied by U.K. computer company SPSS (UK) Ltd. of Surrey, England.

The software analyzes the past windstorm descriptions to create profiles of future windstorms in particular geographic areas. For example, the model can produce estimates of the maximum gust speeds likely to be experienced in individual postal code sectors throughout the United Kingdom.

By combining the Meteorological Office estimates with its own property damage and insurance records from past windstorms, Willis Faber Re aims to estimate the potential value of insurance claims arising from windstorms in specific areas of the United Kingdom.

Mr. Peterken said the estimates will enhance Willis Faber Re's consulting advice to its reinsurance clients.

He said the analysis produced by the new model also will be marketed to other companies within the U.K. insurance and reinsurance industry.

There also has been interest in the model from outside the insurance industry, such as utility companies, Mr. Peterken said.

Winter is the peak season for windstorms in the United Kingdom. The most recent catastrophic U.K. windstorm loss was an October 1987 storm that caused insured damage of about L1 billion ($1.68 billion). A series of severe windstorms in the United Kingdom last winter caused insured damage of about L650 million ($1.08 billion) (BI, Jan 19).

Aon's Mr. Eagle said there is some evidence there has been an increase in windstorm activity in the United Kingdom over the past 20 years.

"But I think it is too early to say whether it is a permanent increase or just a natural variation," he said.

Dan Hollis, an applied climatologist at the Meteorological Office, agreed it is difficult to say if there has been an increase or decrease in U.K. windstorm occurrence, as such trends are most accurately charted over a period of about 50 years.

"We don't have computerized data for that long, so it is difficult to compare," he said. "But what we can say is there has been a move towards greater variation and more extremes."

Willis Faber Re and the Meteorological Office eventually plan to expand their new model so it can produce estimates of the probability of future storm and gale occurrence in the United Kingdom.