Thoughts on the 2014 Atlantic Hurricane Season

   Posted by Levi at 10:57pm on March 14, 2014

Spring is upon us, and it is time to start looking toward this year’s Atlantic hurricane season. Readers should be instantly dubious of any hurricane season forecast, such as this one, due to their inconsistent skill. Last year, myself and every major forecasting agency that I know of predicted a more active hurricane season than normal, and we were all wrong. In fact, 2013 turned out to be one of the top 10 least active seasons since 1950, with only 2 hurricanes and an ACE of 34% of normal. In hindsight, there are a couple of possible reasons for this, such as an apparent faltering of the Atlantic thermohaline circulation and the local Atlantic Hadley Cell.

Regardless of how well hurricane activity is predicted in advance, you the reader should be aware that nobody can predict individual hurricanes months in advance, and no matter what the season as a whole is like, there is always the potential for a hurricane to form and impact you. Hurricane Andrew destroying south Florida during the otherwise completely quiet 1992 hurricane season is the poster child example of this. With this in mind, and despite forecasting failures like 2013, meteorologists can usually glean useful insight into how active the hurricane season will be before it has even begun. Here, I will offer my thoughts on what 2014 may bring.

The video above is there so that I don’t have to type a long post and so you don’t have to read a long-winded analysis. Some basic points are outlined here, but watch the video discussion for details.

Current global sea surface temperature (SST) anomalies reflect a decaying La Nina and negative PDO, which is now more like a positive PDO. An El Nino is expected to develop this summer due to an impressive Kelvin wave likely to bring warm water to the surface of the equatorial Pacific. Both of these climate regimes tend to shut down Atlantic tropical activity through enhanced wind shear over the main development region and displacement of upward motion away from the tropical Atlantic.

We are also in a unique place in the ENSO cycle wherein it has been 46 months since El Nino conditions were last seen, an impressively long period in the context of the last 65 years. We may expect, then, that at least some major aspects of the Atlantic hurricane season will be similar to other El Nino years that followed a long period of neutral or La Nina conditions. If we take 10 such seasons since 1950: 1957,1963,1968,1972,1982,1986,1991,1994,2002, and 2009, the average storm count is 8 (normal is 12) and the ACE averages ~52% of normal. This is likely a good first approximation for 2014. Below is a map showing the anomaly of tropical cyclone frequency (storms/year) in 1×1 degree grid boxes for the analog years mentioned. Note the great lack of activity in the deep tropics. However, an area of above-normal activity shows up off the SE U.S. coast, coinciding with the area of low wind shear during El Nino years. These years also tend to have warm SSTs in that area, and if we look at the SST map earlier in the post, we see very warm water off the SE U.S. contrasting the cold water off of Africa. This suggests a very quiet Cape Verde season and less activity than normal for the deep tropics in general, but perhaps a window for storms to form in the subtropics where the water is anomalously warm east of the U.S. and in the vicinity of Bermuda and the central Atlantic.

If we now look at the best composite analogs (using SST, MSLP, and Z500) for the North America region for the month of February, we see that several of the analogs similar to this winter are also in the ENSO analogs I gave earlier. If we take the intersection of these lists and look at the TC frequency anomaly, the same pattern is still evident, with a generally inactive Atlantic except for the SW Atlantic region.

The climate models generally agree with this type of pattern in the Atlantic this summer, with a very dry tropical Atlantic void of much tropical activity, but perhaps a slightly more favorable pattern off the SE US coast, promoting home-grown storms close to the coast, as well as storms out in the middle of the central Atlantic away from everyone except Bermuda.

With the likelihood of El Nino increasing, perhaps a strong one, and the climate models agreeing with the general pattern seen in the historical analogs I have outlined, a generally quiet Atlantic hurricane season is expected for 2014, similar to last year in many respects. Last year was so quiet, however, that based on probability alone, 2014 is likely to be perhaps slightly more active. The deep tropical breeding grounds are likely to be less active than normal, and storms that do form will likely tend to form farther north in the subtropics, either out in the middle of the Atlantic, or closer to the southeast U.S. coast, which would make the eastern seaboard the most likely region to see any potential landfalls. However, as always, a storm could make landfall anywhere during any season. Stay safe this year.

42 comments

« Previous Entry   Next Entry »

---