The 2012 Atlantic tropical storm and hurricane season looked like it was going to be relatively benign as far as the insurance, reinsurance and catastrophe bond sectors were concerned until Sandy came along and turned the season on its head. Up until that point at the end of October it was easy to think that the season had been relatively quiet, with only a few storms threatening the U.S. mainland yet again. In reality though, the 2012 season was much more active than the majority of forecasters had predicted.
Readers will be aware that we always follow the season and have a page dedicated to it with tracking maps, forecast details and satellite imagery (our 2012 hurricane season page can be found here). This year, the forecasts that we watch were pretty much in agreement that there would be around 14 named tropical storms over the course of the season, around 6 would become hurricanes and 2 or maybe 3 would become major hurricanes. Now, with the Atlantic hurricane season officially ending today, it runs from 1st June to 30th November every year, it’s a good time to see how close to reality those forecasts were.
The table below shows the four forecasts that we tend to follow, as they are from recognised forecasters who are generally widely trusted, and also shows the actual 2012 season numbers for comparison.
|Forecaster||Named storms||Hurricanes||Major hurricanes|
|NOAA||12 to 17||5 – 8||2 – 3|
|Colorado State University||14||6||2|
|Tropical Storm Risk||14||6||3|
So forecasters didn’t get as close as you might have thought. It’s normal for there to be some deviation between forecasts and actual numbers of storms but this year has seen the forecast a little further out than normal. The forecast numbers above are all the latest forecasts issued by each team and their predictions for numbers of storms had risen steadily from their pre-season April forecasts up to the August forecasts you see above. In fact, if you consider the early forecasts then Accuweather was closest as the numbers above were their forecast from the end of May and they didn’t change it through the season. For comparison, the average is 12 named storms, 6 hurricanes and 3 major hurricanes.
Forecasting storm formation, storm tracks and storm impacts in advance is an extremely difficult and also inexact science. Changes in the planet’s atmosphere and jetstream can have a major influence on the number of storms that form, where they track towards and whether they impact land. It’s worth remembering that forecasters struggled in 2011 as well, when the jet stream had moved and storms followed an unusual track. This year the jet stream was in a more normal position but still forecasting the number of storms has proved difficult.
2011 also saw 19 tropical storms form, and that was the third highest total for a season since records began in 1851, and well above the average of 11 storms. 2011 only saw 7 storms make it to hurricane status, compared to the 10 in 2012 and 3 storms reached major hurricane status in 2011 compared to just 1 in 2012.
The Colorado State University tropical forecasting team, who’s forecast we always include as it is always widely referenced, have published a review of the season (which you can download here). They say that while the 2012 season had more storms than had been forecast, the season was notable for having a high number of weak, high latitude cyclones and activity that occurred in 2012 was anomalously concentrated in the northeast subtropical Atlantic.
Two interesting numbers which tell the tale of the 2012 hurricane season are the number of named storm days and the ACE or Accumulated Cyclone Energy for the season. The Colorado team had forecast 52 named storm days from 14 named storms in their August update. The season actually had 99.5 named storm days from 19 storms which suggests that, on average, storms churned in the Atlantic for longer than expected which was typified by the number of storms which recurved in the northern Atlantic this year. ACE, which denotes the energy within the storms over the season, was forecast as 99 but actually was only 129. So while storm days was nearly double ACE was only 25% higher, showing that 2012 had more, longer lasting but weaker energy tropical cyclones over the course of the season.
The image below from the NOAA shows the tracks of every tropical storm and hurricane over the 2012 season.
The image above clearly shows that a large amount of the 2012 seasons tropical storm activity was focused in the mid to northern Atlantic and a large number of storms curved north long before reaching the Caribbean or U.S.
Four storms hit the U.S. coastline during the 2012 season, which is much higher than seen for the last few years. Tropical storm Beryl was the first to strike the U.S. mainland, impacting the Jacksonville area of Florida, but insured losses were minimal. Tropical storm Debby was the next to impact the U.S., and also impacted the Caribbean and Central America, again insured losses were minimal. Tropical storms Ernesto and Helene both impacted the Caribbean and Central America or Mexico with heavy rains and some damage, but again this didn’t significantly impact the insurance markets.
The first storm to worry the reinsurance sector was hurricane Isaac which impacted the Caribbean, side-swiped southern Florida and then came ashore in Louisiana. Isaac worried the catastrophe bond market for a while as it threatened the Pelican Re cat bond, but in the end of the cat bond was deemed safe. There was some nervous trading in the secondary market for a short time after this storm while damage estimates were firmed up. In the end insured losses from hurricane Isaac were thought to be somewhere over the $2 billion mark, certainly not enough to trouble most industry loss positions.
Another storm, Leslie, impacted the Caribbean lightly but then travelled all the way north to then impact Canada, which is very unusual. Damage from this storm was minimal.
Finally, the last storm to cause any damage was of course hurricane Sandy. Sandy caused damage all the way along the storms track, from the Caribbean, via the Bahamas and a side-swipe for Florida, tracking along the U.S. east coast where tropical storm winds caused some minor impacts and then making landfall on the mid-Atlantic coast as a post-tropical super-storm where Sandy is thought to have caused around $20 billion of industry losses. Sandy is now the largest insurance, reinsurance and catastrophe bond impacting event in the U.S. since Katrina and the uncertainty around the final loss totals is still significant. Sandy was a very unusual storm and the report from Colorado State Uni contains an extensive analysis of the meteorology that caused the super-storm. Find more coverage of Sandy and the storms impact on the reinsurance and cat bond sector.
So 2012 was an unusual season, with more impacts than in recent years, but lower energy across the whole season and then finishing off with one of the most complex natural catastrophe events on record in the form of Sandy which will become a major economic and insurance loss. This quite unusual season shows that no matter how benign the forecasts you cannot be complacent when it comes to preparation and risk transfer for hurricanes. Imagine the impact if the climate reverted back to the pattern seen in the late 90’s and early 00’s when storms regularly tracked into Florida. Once again, 2012 saw Florida escape with no direct impact from a hurricane, only a side-swipe with tropical storm winds. That won’t happen for ever and one day we will see a repeat of a hurricane Andrew which could really test both the reinsurance and catastrophe bond markets to their limits.
The animation below shows the entire 2012 hurricane season in four and a half minutes. It’s worth watching as you can clearly see the patterns that emerge with storm tracks heading north.
Also for interest, here are some unusual facts about the 2012 Atlantic hurricane season, taken from the Palm Beach Post here.
- 2 tropical storms formed in May, Alberto and Beryl, before the season had officially even begun.
- Ten hurricanes formed; only five other years since 1944 have had more. But the season recorded only one major hurricane, lowest since 1997, and it was at that strength for only one fourth of one day, lowest since 1994.
- For the fifth straight year, there were no Category 5 hurricanes; that hasn’t happened since 1993-1997.
- None even reached Category 4, for the first time since 2006.
- For the seventh year, no major hurricanes made a U.S. landfall. Since 1878, the U.S. has never had such a seven-year stretch.
- The top sustained wind speed by any storm this year was Category 3 Michael’s 100 knots (115 mph). This is the lowest since 1994′s Florence with 95 knots (109 mph).
- At 60 knots (69 mph), Beryl became the strongest off-season tropical cyclone on record to make U.S. landfall, on May 28th near Jacksonville.
- Hurricane Nadine tied Hurricane Ginger (1971) for the most days as a named storm (21.25) since aircraft reconnaissance began in 1944.
- Sandy, by then a post-tropical storm, generated the lowest barometric pressure ever recorded in the Northeast at landfall (943 mb), breaking the record set by 1938′s Great New England Hurricane/Long Island Express.
You can see details of every storm of the 2012 season on our 2012 Atlantic Hurricane page.