Robert Muir-Wood, chief research officer of Moody’s, has attended the Rendez-Vous de Septembre many times over the course of a more than 25-year career in catastrophe modeling.

But he’s hard-pressed to remember one that took place when hurricane-force winds weren’t whipping up along the U.S. coast or in the Caribbean, he noted in a blog post on Friday, Sept. 6 (“Melodramatic Hurricanes” | Moody’s). suggesting that predictions of an active hurricane season from NOAA, Colorado State University, Tropical Storm Risk and other forecasters in recent months might not play out.

“We should be prepared for the possibility of a seasonal forecast miss,” he wrote.

“It now looks possible there will be no significant hurricanes as a backdrop to the RVS. And while there could be a dramatic catch-up in activity through the second half of September and into October, the available window is reducing,” he wrote.

(Editor’s Note: At the time of this writing, meteorologists were reporting that Tropical Storm Francine had formed off the coast of Mexico and forecasting a possible Texas or Louisiana landfall as a Category 1 hurricane on the last day of RVS, Sept. 11.)

Muir-Wood, after reviewing some of the hurricanes that churned in North America in recent prior years while reinsurers, insurers and brokers met in Monaco, including Hurricane Irma making landfall in 2017 as part of the year’s devastating Harvey, Irma, Wilma trio, reported that five potential sites of Atlantic tropical cyclone circulations indicated by the National Hurricane Center on Sept. 5 had quickly fizzled out a day later. Two dropped off the forecast, and the other three had low chances of tropical cyclone formation.

In a separate blog item posted the same day (“2024 North Atlantic Hurricane Season: A Mid-Season Slumber but What Next?”| Moody’s), Sarah Hartley, director, Moody’s RMS Event Response, and James Cosgrove, assistant director, described the factors that are contributing to the quieter-than-expected hurricane season, while noting that the most recently updated forecasts from Colorado State University, Tropical Storm Risk, the U.K. Met Office, and the European Centre for Medium-Range Weather Forecasting (ECMWF) are still calling for an active one.

And during their pre-Monte Carlo briefings, reinsurance brokers and rating agency analysts were still addressing questions about property-catastrophe reinsurance conditions that would prevail if the high forecasted level of activity plays out rather than the reverse scenario (questions of whether prices or terms would ease up more than they have to date if the season is inactive.)

Hartley and Cosgrove noted that the North Atlantic basin did not produce a named storm between August 13 and September 2 for the first time in 56 years—dating back to 1968.

Why not? After an active start to the hurricane season that saw five named storms form between mid-June and mid-August—including Hurricane Beryl, the earliest Category 5 major hurricane in the basin on record—why is there a “mid-season slumber” during a period of time that history records as the most active?

The blog summarizes some technical details, starting with the conditions that still seem favorable for tropical cyclone formation:

• Sea surface temperatures in the central equatorial North Atlantic (the Main Development Region, or MDR) that are one and two degrees Celsius higher than average.
• Continuing ENSO-neutral (El Niño-Southern Oscillation) conditions present in the equatorial Pacific, with La Niña conditions—which typically lead to above-average North Atlantic hurricane activity.
• A very active West African monsoon, which is also typically correlated with above-average activity.

Working in the other direction, however, there has been “a complex interaction of several sub-seasonal atmospheric factors across the North Atlantic and West Africa.” The blog sets forth a series of such factors, including anomalous low pressure developing over West Africa during June, July, and August. The low pressure resulted in low-level southerly winds across the eastern subtropical North Atlantic, which transferred dry air into the far eastern Main Development Region, inhibiting cyclone formation, Moody’s said.

The blog post and a report from Colorado State University also describe anomalous moist air in the Sahel and Sahara regions of West Africa, and a northward-shifted West African monsoon trough which resulted in African Easterly waves (AEWs) emerging too far north to reach hurricane-conducive conditions. AEWs are atmospheric disturbances that are otherwise thought to be a key building block for tropical storm activity in the Atlantic basin.

Also noted is the fact that the atmosphere across the Atlantic basin has been much more stable than expected. Basin-wide above-average sea surface temperatures and high ocean heat content would typically result in greater atmospheric instability, Moody’s said.

Still, in spite of the anomalies, CSU and NOAA and others are sticking with forecasts of an active season.

Hartley and Cosgrove note that forecasts of activity may be influenced by factors such as a return a phase of climate cycle known as Madden-Julian Oscillation, which typically promotes tropical activity in the North Atlantic.

NOAA’s August update of 17-24 named storms, of which 8-13 could become hurricanes (winds of 74 mph or greater), including 4-7 major hurricanes is almost identical to its May outlook. So far, four named storms have happened, including two hurricanes (Beryl, a major hurricane).

Only the upper bound of NOAA’s named storms forecast has been lowered—to 24 from 25 previously. The Moody’s bloggers note that NOAA’s August update also sees “above-normal” activity as more likely—with a probability of 90 percent now vs. 85 percent in May. (NOAA: Climate Prediction Center – Atlantic Hurricane Outlook)

CSUs shows its latest forecast for 23 named storms, 12 hurricanes and six major hurricanes (Category 3 or greater) on a chart that shows those numbers largely unchanged throughout the year, and higher than averages of 14.4 named storms, 7.2 hurricanes and 3.2 major ones.

“We are anticipating below-normal activity over the next two weeks due to both unfavorable Madden-Julian oscillation conditions and a relative lack of tropical cyclone activity forecast by the ensembles. However, the Atlantic hurricane season does not end in mid-September. We still have an extremely warm Atlantic and a tropical Pacific that will likely be trending more towards La Niña as the season progresses,” the CSU report published on Sept. 3 says.

NOAA’s August 8 forecast also points to the likelihood of a La Niña event. “The official NOAA ENSO Outlook from early August indicates about even odds for the ENSO-neutral to transition to La Niña during ASO,” the forecast document says, referring to the peak months of August, September and October. “That is lower than the odds for the development of La Niña that coincided with the May seasonal hurricane outlook. So while La Niña is anticipated to develop, there is uncertainty whether it will become established concurrent with ASO to align with the peak of the Atlantic Hurricane Season to enhance the local signals (warm SSTs, weak tradewinds, active West African Monsoon).”

“Many active seasons have occurred during ENSO-neutral conditions,” concluded the forecasters who issued the August edition of NOAA’s 2024 Atlantic Hurricane Season Outlook.

The CSU report concludes by noting that if the season “were to have no named storm activity through 9 September, and then had an average season for the remainder of the season, it would end up classified at the high end of NOAA’s average Atlantic hurricane season.” CSU explains that NOAA classifies hurricane seasons based on an Accumulated Cyclone Energy index, a measure of the collective strength and duration of tropical storms. (“The ACE index is essentially a wind energy index, defined as the sum of the squares of the estimated 6-hourly maximum sustained wind speed (knots) for all named systems while they are at least tropical storm strength,” according to NOAA.)

Definitions available from NOAA’s Climate Prediction Center, reveal that an above-normal season is one with an ACE index above 126. By CSU’s calculations, an average last-half of the season would put the total ACE for the whole season at 122 (55 observed + 67 average from Sept. 10 onwards).

A hyperactive season is one with an ACE above 159.6 ACE, which would imply an ACE of 104 for the rest of the season—something that has happened in seven years in the satellite era, according to CSU. “So at this point, while it would be highly unlikely for CSU’s seasonal hurricane forecasts to exactly verify, there is still potential that a hyperactive season could be reached,” the CSU September report says.

Hartley and Cosgrove at Moody’s noted that Sept. 10 marks the climatological midpoint of the hurricane season, adding that “there’s a long way to go before the hurricane season officially concludes” on Nov. 30. “It’s important to monitor how the next few months unfold, as regardless of overall activity and storm count, it can only take one landfalling event to make a season costly or memorable,” they wrote.

Muir-Wood reminded readers in his blog post that seasonal forecasts “rely on a small number of parameters that are easily measurable and well-constrained, ” including sea surface temperatures and the state of the ENSO, “while assuming that other less understood parameters are in an average state….”

Offering his own observations about weather phenomenon that developed in unexpected ways (accompanying sidebar), Muir-Wood said, “above all this reveals why catastrophe modeling and reinsurance pricing apply a longer-term average for determining activity rates. Models help to explore the probability of no landfalling hurricanes during a season, through to probabilities of a very intense, damaging season.”

Shorter-term seasonal forecasts, on the other hand will continue to fall above or below actual activity with parameters beyond sea surface temperatures and the state of ENSO not normally considered. Referring to the added complexity that “needs to be unraveled” for better short-term predictions, he suggested that for reinsurers, “it not worth holding your breath awaiting a reliable formula.”