Five tropical or subtropical storms have impacted the United States in May since 2012. (NASA/WeatherTiger)

The authors, Ryan Truchelut and Erica Staehling, are co-founders of WeatherTiger, a weather consulting and seasonal forecasting company.


Tropical Storm Alberto, which slogged ashore just west of Panama City on Memorial Day with sustained winds of 45 mph, was unfortunately not the final or the most significant tropical impact on Florida’s Emerald and Forgotten Coasts of the 2018 hurricane season.

It was, however, the fifth tropical or subtropical storm in the past seven years to require tropical storm watches or warnings to be posted by the National Hurricane Center (NHC) for the continental U.S. coastline before the official start of hurricane season.

As meteorologists digest the lessons of a punishing 2018 hurricane season that punched well above weight given summer forecasts of a quiet year, there are ongoing conversations about how to improve public tropical cyclone threat messaging, particularly concerning extreme precipitation risks.

Based on both long-term climatological data and recent trends, moving the official start of Atlantic hurricane season to May 15 should be a part of this enhanced communication strategy.

The June 1 to Nov. 30 bounds of the official hurricane season predate important changes in operational tropical forecasting methods. According to James Franklin, former branch chief of NHC’s Hurricane Specialist Unit, that frame was selected in 1965 to capture 97 percent of the formation dates of Atlantic tropical cyclones. However, basin-wide storm monitoring via geostationary satellites did not begin until 1966, and the NHC did not classify subtropical cyclones, particularly common in May and June, until the 1971 season.

These methodological shifts augur an undercount of preseason activity relative to when hurricane season was defined. With almost 50 years of satellite-era Atlantic tropical and subtropical cyclone climatology now in the NHC’s database, WeatherTiger’s research team queried this data set to test how well a June 1 to Nov. 30 season fits real world observations.

We did this in several ways. First, we used the original parameter defining the Atlantic hurricane season — formation date — and determined what percentage of total development events occurred between June 1 and Nov. 30 for 1971 to 2018, along with an “optimal” six-month period capturing the most formations. This process was repeated for two measures of the human impact of tropical activity: the dates of continental U.S. tropical or subtropical storm landfalls and accumulated cyclone energy (ACE), a measure of storm duration and intensity, occurring near or over the continental U.S.

These tests show the current definition of hurricane season encompasses less than its intended 97 percent of storm development cases in the modern era. The June 1 to Nov. 30 period captures 96.2 percent of total tropical or subtropical formation events over 1971 to 2018, meaning 3.8 percent of storms develop in the purported offseason. A hurricane season beginning on May 15 would miss 2.6 percent of Atlantic genesis cases, cutting offseason systems by a third.

Even worse, June 1 to Nov. 30 is particularly late-shifted with respect to impacts, missing 3.1 percent of U.S. landfalls with wind speeds exceeding 39 mph (34 knots). In contrast, a six-month hurricane season beginning on May 23 would miss just 1.2 percent of landfalls, and a May 15 to Nov. 30 season would capture all but one out of 162 U.S. storms since 1971.

U.S. ACE tallies are dominated by major hurricane landfalls in late summer and early fall, but even in these terms, starting the season half a month earlier would decrease offseason U.S. ACE from 1.0 percent to 0.2 percent of the total.

The current hurricane season may become a yet poorer fit to observe activity with time. A trend toward earlier development of the Atlantic’s first tropical storms may exist over the past several decades, with seven preseason storms in the past seven years. While peer-reviewed studies reach mixed conclusions on this issue, any change would be particularly concerning as preseason and early season activity develops disproportionately close to the U.S. coastline.


Where tropical and subtropical storms have formed in May from 1900 to 2018. (WeatherTiger)

There are significant public safety implications to offseason tropical systems, when diminished risk awareness and preparation time may exacerbate realized impacts. Recent published studies have noted trends toward slower forward motion and heavier rainfall from tropical storms worldwide. It is certainly plausible given these findings that a flooding event like Tropical Storm Allison, which made landfall early in June 2001 and caused 41 U.S. deaths and nearly $10 billion in damages, could occur a few days earlier and fall outside of hurricane season.

Changing the Atlantic hurricane season start to May 15, matching the Eastern Pacific hurricane season, would be a prudent measure to reduce the risk that such an event occurs when the public is less primed to respond and react to emergency messages. An earlier start to the NHC’s Tropical Weather Outlooks, issued each six hours during the season, would be a further welcome protection against a “surprise” storm.

Overall, one of the major lessons of 2018 is that weather threats with no clear parallel in the historical record need to be anticipated and proactively mitigated. Changing the definition of Atlantic hurricane season is one simple way to do just that.