Lessons From the 2017 Atlantic Hurricane Season: Open Models, Real-Time Tracking Are Valuable Tools

Open models and tools to track hurricanes in real time can help boost modeling accuracy and aid the claims process after storms, according to a new Karen Clark & Co. analysis of the 2017 Atlantic hurricane season and its aftermath.

“With the ability to customize damage functions to their unique policy conditions and claims handling practices, insurers can further enhance the accuracy of their modeled loss estimates,” KCC argued.

Part of this involves KCC touting the success of its own technology. The firm pointed out, for example, that it provided a best estimate of private insured losses for each storm, while competitors give a wide, broader range of estimates. KCC predicted that Hurricanes Harvey, Irma, Maria and Nate would cause a combined $70.5 billion in private insured losses in the U.S. and Caribbean, something that is generally “proving to be quite accurate,” the firm noted, along with modeled loss estimates for individual insurers.

KCC said that the 2017 hurricane season warrants study, in part because it was so unusual as the most active hurricane season since 2005 based on the Accumulated Cyclonic Energy index, which measures seasonal intensity and duration.

When the season highlights are looked at in a combined way, the massive storm events offer examples to study in preparation for modeling future weather event possibilities. Among the KCC report highlights:

• Ten hurricanes formed in the Atlantic basin and six became major hurricanes.
• Two major hurricanes made U.S. landfall, after 13 years of no major landfalling storms.
• Every Gulf Coast state dealt with at least one hurricane during the season.
• Two Category 5 hurricanes caused catastrophic damage to the Caribbean Islands and Puerto Rico in particular.
• Jose, Irma and Maria, along with other storms, showed extreme wind speeds and/or duration.

Source: Karen Clark & Co. white paper 2017 Hurricane Season: Review and Analysis