Last week, the fourth named storm—Danielle—of the Atlantic hurricane season formed. It was the earliest such formation in any of the 165 seasons on record. Tropical Storm Colin preceded Danielle just days into the official season, which runs from June 1 through November 30, and was the earliest-on-record third named storm. And Tropical Storm Bonnie formed in May, prior to the season’s start, a pattern that seems to have increased in the past decade.
It makes one wonder: should the official Atlantic hurricane season be lengthened to accommodate the earlier storm formation? The season for Eastern Pacific tropical storms and hurricanes, which form off Mexico’s coast, already does. It runs from May 15 to Nov. 30, and almost like clockwork, the first storm of that season will typically appear midmonth or after.
In the past decade, half of the Atlantic’s seasons had “preseason” storms. In 2012, two storms—Alberto and Beryl—were named before the season officially started. And last year, Ana formed east of Georgia on May 7. Granted, it was initially a subtropical storm, a hybrid with both tropical features and features of midlatitude cyclones. But waters were warm and Ana became fully tropical in just days, and moved ashore in South Carolina on May 10.
In an e-mail exchange with James Franklin, branch chief of the Hurricane Specialist Unit of forecasters at the National Hurricane Center, he noted that the current 6-month Atlantic hurricane season was established in 1965 and was based on the formation dates of roughly 97% of the total annual tropical cyclone activity in the Atlantic basin, which includes the northern hemisphere Atlantic Ocean, Caribbean Sea, and Gulf of Mexico.
Franklin pointed out that tropical cyclones have formed in every month of the year, including Hurricane Alex this year in January. (Its formation was considered a very late entry in the 2015 hurricane season, despite its “A”-name designation in the calendar 2016 season.) He also pointed out that moving the season up to May 15 wouldn’t have prevented an out-of-season start in half of the recent early-season years (in addition to Alex and 2015’s Ana, 2007’s Andrea formed on May 9).
While extending the season might not catch all early storms, it would accommodate an increasing number. Weather Underground’s Jeff Masters and Bob Henson blogged last year with Ana’s early development that it isn’t all that rare to have early-season storms. Adjusting the numbers up by two for 2016, they conclude that 41 preseason tropical or subtropical systems have formed in 33 separate years since record-keeping began in 1851. Since the satellite era began in 1960, which improved detection of tropical systems basin wide, they find that there has been on average about one such system in the Atlantic every 2-3 years.
“Preseason named storms may be getting more common,” they wrote. Of note, they mentioned a 2008 paper published in
Geophysical Research Letters by Jim Kossin of the University of Wisconsin, titled “
Is the North Atlantic hurricane season getting longer?” that supports an earlier start to hurricane season. Kossin concluded that there is an “apparent tendency toward more common early- and late-season storms that correlates with warming sea surface temperature, but the uncertainty in these relationships is high.”
So the question remains: Is it climate change that necessitates a longer season, or natural variability?
Franklin states that there would need to be pretty convincing evidence “to go to the trouble of changing the official season. I don’t think we know whether this ‘uptick’ is real or apparent, or whether it will persist,” he writes. “I think we’d want to see a definitive trend in the long-term climatology before contemplating such a change.”
Makes sense. However, if we have too many seasons like these past several years, with multiple storms in May, starting hurricane season on June 1 may begin to appear arbitrary.
What do you think?
Karloski and Evans (2016), published in the January 2016 issue of the Journal of Climate, revisited Jim Kossin’s story over a longer period of record. They found no trend, statistically-significant or otherwise, in the length of the Atlantic hurricane season. They also identified the most common genesis pathways for early and late season storms while quantifying a small fraction of the large-scale variability responsible for such events.