For those who are out of the loop, here's what happened in the past 24 hours: After hyping up Invest 91L as a substantial CV hurricane for days, the global models began dropping development one by one:
ECMWF at 0z 9/5,
ICON at 6z 9/5, and
GFS and CMC at 12z 9/5.
This immediately reminded me of
September 1, 2024. A tropical wave in the western MDR was expected to develop on all global model runs, with many becoming
Caribbean cruisers or Ida-like Gulf majors. Then the night of August 31 hit - and
everything fell apart.August 31 was also when
this infamous post, "Why is the 2024 Atlantic hurricane season underperforming expectations?", was made
(initially with a different title). Over there and in other threads, the sudden loss of model support immediately triggered comments like these:
On 8/31/24, Someone wrote:I’d say if we reach September 5th with no development expected within the following 2-3 days, then it’s clear that something is wrong with the basin, and another 2022 becomes more likely.
Hard to believe that a year with a 35 ACE July Cat 5 would end up under-performing, but that seems to be where we’re headed. Perhaps Beryl was simply a freak anomaly…which just makes its existence ever more unusual. Perhaps something like 2007: a first year La Niña after a single-year El Niño that produced Cat 5 Caribbean cruisers but not much else of significance and ended up as one of the weakest La Niña hurricane seasons.
On 8/31/24, Someone wrote:If these tropical waves are failing to develop, when they are finally coming off at the right latitude and surviving the entire track through the MDR, then it really speaks to some background unfavorability that we really don’t know when it will end. This year is already not favoring any subtropical spinups at all, and if we can’t get waves developing in the heart of the season then it begs the question of whether we will not just bust hyperactive forecasts, but potentially have a below average season when all is said and done. Lots of opportunity for late September and October mischief, but that remains theoretical. What isn’t theoretical is the lack of any consolidation.
There have been years where during the peak, convective blowups across the basin have led to surface lows which lead to weak systems, and tropical waves blow up into large puffballs where the forum is rife with speculation about surface level lows. Storms form that were barely even predicted by models a few days in advance. Compare it to now, where models are persistently pushing back the development of just about everything. Even Debby and Ernesto’s precedent waves were practically dead as they crawled across the Atlantic, and the monsoon trough breakdown has failed to currently produce anything so far. The frontal pattern across the US right now looks like October. What is going on?
On 8/31/24, Someone wrote:This IS season peak right now and some people on this board are expecting a trackable system and right now the basin should be ideal for development MJO or not.
This is sad to see areas of interest all strung out given the high expectations and the supposed peak right now.
On 9/1/24, Someone wrote:The way I see it, it looks like the ITCZ has lifted north semipermanently. This is likely the result of a strong WAM and the ASW that's persisted since 2015. This means no more MDR storms until it stops because the waves exit too far north to develop, and the northward-displaced ITCZ pumps a never-ending train of dry air parcels into the MDR, killing anything that even tries to organize as far west as the Caribbean. This isn't a season predictions thread, but IMO, Beryl was our only MH.

On 9/1/24, Someone wrote:you know a season is lackluster when even waves with not a horrible environment around them struggle to develop during the peak of the season. Unbelievable. I thought for sure we would have at least 3 or 4 "active" storms by now. . I think this will end up being a depression or weak TS storm heading into Mexico, but again we are talking about season peak, not June or July.
On 9/1/24, Someone wrote:With or without S2S forcing, that is a pretty unambiguous indicator that overall conditions are just simply not favorable for TCs this year & most certainly not favorable enough to support hyperactivity.
Conditions may change later in the season to a degree, but the writing is on the wall for this season as a whole (we aren’t getting a hyperactive year and an above average season isn’t a gimme), although some have would call this take premature for some reason
Honestly, even I myself was seriously in doubt about the rest of the season:
On 9/1/24, Teban54 wrote:A big red flag for me was that the 10/40 wave had everything going for it last night: stacked vorticity (free from the competing lobe problems that models were expecting to happen for days), a weak surface low, and relatively low shear. Yet, the wave simply couldn't maintain convection during DMAX. I think that was likely a main contributor to model downtrends during 0z and 6z, as it's already causing the system to decouple (even accounting for the well-modeled acceleration to the west today).
So what happened afterwards?The said wave
finally became Invest 91L almost a week later, leaving behind
the longest pre-invest thread ever with 54 pages, and eventually became Cat 2
Francine. But even then, people were still very uncertain about the season, to say the least.
My comment on 9/7/2024 in the 91L thread aptly captures the frustration everyone had felt by that point
(in terms of activity, not impact).
The last 3 months of the season, starting with
September 9, saw 13/8/4 and 106.8 ACE (itself an average to above-average season). Among the majors were Helene and Milton. Even the Cabo Verde season lasted as late as October, with Kirk becoming (one of?) the latest Cat 4 within the MDR on record.
(It was also widely believed that the "true" peak season didn't start until
September 24, when Helene formed. Francine and Gordon were "underwhelming" by peak-season standards.)
Overall, the 2024 season finished with 18/11/5 and 161.5 ACE, hitting the hyperactive ACE threshold (albeit just barely).
Are you saying 2025 will be the same?Of course not. 2025's pre-season indicators weren't nearly as favorable as 2024's, and the early season was also less active (39.3 ACE in 2025 so far, 54.7 ACE on this date in 2024).
I'm making this comparison almost solely because
the sudden drop in model support for 91L today was incredibly similar to pre-Francine. Both were monsoon trough breakdown scenarios, occurred at practically identical times of the season, and mid-level dry air was seemingly the culprit for the demise of both. (They even have the same invest number.)
On 9/1/2024, we had almost no precedents for seasons that had such an inactive early September, but still managed to turn around. But now, on 9/5/2025, we have one such precedent.