partner login

BMS News

Hurricane Wilma’s 8th Anniversary

As we approach the end of the 2013 Atlantic hurricane season and take in the media attention around the anniversary of Superstorm Sandy, it is also important to mark the 8th anniversary of Hurricane Wilma’s landfall, which occurred October 24, 2005. This was the last major hurricane to make landfall on the U.S. coastline. It has now been 2,938 days without a major landfalling hurricane – remarkable given the changes scientists said might result from warmer sea-surface temperatures in the Atlantic Ocean. The U.S. landfalling hurricane event data set is one of the best meteorological records that exist in the U.S. In looking at the historical landfall record, the longest period without a major landfalling hurricane stands at 3,316 days (August 11, 1860 – September 8, 1869). If a major hurricane doesn’t make landfall in the U.S. next year, we will surpass the longest period without one.

Unless we are in some very unusual climate state that has not been discovered, there is a growing disconnect between overall Atlantic Basin activity and landfalling named storms. While the average overall Basin numbers are higher than normal since 2006, with every passing year since then the U.S. has seen only 19 named storms make landfall, and only six hurricanes – with no major hurricanes making landfall. This translates to a landfall rate of 0.75.

Using the landfall data from 1900, in a given year the expected landfall rate of a hurricane impacting the U.S. coastline is 1.5, with a 77% probability of at least one hurricane impacting the U.S. coastline. For major hurricanes the rate is 0.5 with a 40% probability – so the U.S. landfall rate is significantly below average.

Given this landfalling hurricane drought, the United States coastline has been lucky. Although insurance companies have been suffering losses of other types over it, the average annual hurricane loss during this drought has been just $4.9 billion, according to Property Claims Services. This is below the long-term average annual loss of $6.4 billion as calculated using the insured historical loss data from Dr. Pielke Jr., a database that attempts to normalize hurricane damages in the United States. Accounting for Superstorm Sandy in 2012, which was not a hurricane at landfall, this average annual loss since 2006 would increase to $7.7 billion.

With a below normal landfall rate of only 0.75 hurricanes since 2006, in the future the trend for more landfalls should correct back closer to the long-term rate if we assume that hurricane landfalls follow a poisson distribution and we are not in some unknown climate regime. After all, the probability of not having a major hurricane make landfall over a 9-year period is a very low 1%, meaning insurance companies should expect an increase in losses from hurricanes in the future. Something to ponder as we await next year’s forecast.

Wet East Coast Increases Loss Concern this Hurricane Season

As we approach the peak of the 2013 Atlantic hurricane season (which peaks around September 10), the forecasts (for an active season) made back in April have partially panned out – if you consider the total number of named storms. Looking at the last 50 years (1976-2012), the average formation dates of the fourth named storm and first hurricane are August 19 and August 3. So in terms named tropical storms, the season is ahead of par with climatology, but slipping behind on the occurrence of the first hurricane for 2013. In terms of Accumulated Cyclone Energy (ACE), 2013 is essentially average for this date. We’re at 6.6 compared to the 1981-2010 average of 9 for this date.

With August being the month where typically the number of tropical systems ramps up and Colorado State University’s August forecast calling for an active landfalling season, the likelihood of the season’s wet soil conditions leading to increased losses from a landfalling named storm must be considered.

The Ohio Valley and East Coast were much wetter than average. June precipitation totals for 18 states – from Georgia to Maine – ranked among their 10 wettest in the historical record. The fact that this weather continued into July undoubtedly creates concern over a named stormed impacting these rain-soaked areas.

 

 

NOAA Advanced Hydrologic Prediction Service view of the past 60-day Percent of Normal Precipitation. Over the past 60 days, much of the East Coast has seen 150% – 350% of normal precipitation.

Research shows that past hurricanes have demonstrated the combined effects of subsequent excessive rainfall and a named-storm wind speed that can have a major impact on insured losses. These losses might not otherwise be represented if the subsequent seasonal rainfall was normal or below normal. This loss increase is primarily associated with increased basement leakage and tree fall. (When soil is saturated, the connection between the root plate of a tree and the soil is lessened, which can increase tree fall.) Research also shows that the average wind speed expected to snap a hard- or softwood tree trunk is a 90 mph gust. Therefore, while more intense winds wouldn’t necessarily increase the loss, lower wind speeds might – given the weakened condition of saturated soil.

Recent examples of events that might have seen increased losses due to higher soil moisture are hurricane Isabel 2003 and Irene 2011. With New Jersey and Delaware having had their wettest June on record and 18 other eastern states having had Junes ranking in their top 10 wettest, this August has seen some of the highest soil moisture levels ever recorded. And this could increase the risk of river and basement flooding as well as tree fall if a named storm were to impact the area.