Willis Re Tornado Damage Survey Report: A Question of Distance Across the Path?

Tornado-Damaged House

While researching the damage caused by the wave of tornadoes that struck the Midwest U.S. last month for our report, Willis Re May 2013 Tornado Damage Survey Report, the Willis Re team made several interesting observations.

One thing that struck us in Moore and Texas was the rapid decrease in the degree of damage moving away from the centerline of the tornado. In Moore, we observed 80% damage to residential buildings at about .1 mile across from the centerline of the storm; the damage ratio dropped to 10% at .25 miles. At about .35 miles across from the centerline of the storm, the damage ratio was reduced to 1%.

This was also reflected in the Tornado Damage Isolines (TDIs) that Willis Re developed for each surveyed event. TDIs are contours of comparable building damage levels within a tornado path.

Increasing Understanding

We believe that TDIs can help companies to depict the impact of tornadoes from a loss perspective. Official tornado intensity categories are often difficult for companies to relate to damage or loss to the exposure on the ground.

We also noted that the damage gradient moving away from the centerline of the storm in Granbury, TX and Cleburne, TX is relatively steeper than in Moore, OK. One reason for this could be the damage caused by wind-borne debris in Moore. Breach of building envelopes by flying debris can exacerbate the damage from wind forces.

Another observation was that the rotating column of wind along the centerline of the tornado produced significant damage compared to damage from somewhat straight-line winds on the periphery of the storm.

Something to Look Into

We observed the peak intensity of the storm to have some influence on the decay rate of damage moving away from the centerline of the tornado. Because of this, we recommend further research into understanding the relationship between tornado intensity and the damage gradient away from the centerline of the tornado paths.

Based on what I saw, modeled risk could be underestimated or overestimated. It really depends on the assumptions the catastrophe modelers use in assessing the intensity vs. damage gradient across the storm path. I would also exercise caution in using a single constant damage profile across the storm path for all intensity levels, as it too can lead to over estimation of the risk from less intense storms.

About Prasad Gunturi

Prasad Gunturi is Senior Vice President of Willis Re Analytics, where he leads the North American catastrophe model…
Categories: Reinsurance, Weather risk | Tags: , , ,

Leave a Reply

Your email address will not be published. Required fields are marked *