Hi all! We'll be hosting experimental (and operational) model guidance on our AOML Hurricane Model Viewer again this season: https://t.co/PU340GUzlQ 2 models I particularly want to highlight are the ones we'll be helping to run: 1) HAFS-M V2.1 (HAFS multistorm, with multiple

Still a little ways out, but interesting to see AIFS trending towards other models with a stronger wave over the Caribbean next week, and trending away from its solution of burying the system into Central America.

Another step forward for DeFi. We’re teaming up with @SuiNetwork to bring the first SEC-registered stablecoin, YLDS, natively to Sui. As part of this, @DeepBookonSui's new margin system will integrate YLDS to transform idle USDC into yield-bearing assets.

Ensembles: some uncertainty in whether a low forms in the Caribbean and heads NE ahead of a trough or gets trapped under the ridge. GFS: porque no los dos?

Did a forecast discussion on the possibility of Caribbean development next week: https://t.co/JarYkozNZD

Sometimes this year, AI models have helped in giving a clear indication of the likely outcome for a system. Definitely *not* the case for this wave. Pretty much all possibilities on the table from FNV3 with a "squashed spider" ensemble track look: -Weak system into Nicaragua (a

DWWL - JUST SAY

The wave in the Central Atlantic is plugging along to the west. Convection looks fairly healthy this morning, though it's mostly ITCZ-related at this point. I don't expect much development for another several days, but a lot of models show a signal for development in the

Ever since the dry air behind the nor'easter, it's started to feel a little like fall down here! Just have a few more weeks of hurricane season to dodge and we're full speed ahead into the Chamber of Commerce time of year.

Another AI models discussion with @brightband and @danrothenberg, going around the world (including some long-range forecasts for 4 systems at once in the Indian Ocean). Check it out! https://t.co/3Ysty2Bzs7

The tropical wave in the East Atlantic continues to make its way west. Some scattered convection but nothing super organized yet. We're probably still a week or so from potential development (probably somewhere in the Central Caribbean if it happens), so nothing to worry about

Don't change your current workflow. Tembo meets you where you are.

GFS already develops huge (generally fake) hurricanes in the Caribbean every October regardless of the setup. Give it an actual wave with a favorable pattern and it's like a raccoon on meth.

GFS shade aside, I do think there's a general chance of development in the SW Caribbean sometime late next week as a tropical wave interacts with the monsoon trough. Development is probably contingent on the wave being robust enough and feeling a tug from a trough to avoid

Getting some good material for my "GFS fantasy 10-day hurricanes that verify as cloud-free days" Google slidedeck this week.

This tropical wave south of Cabo Verde is firing some disorganized convection. Models for the most part (other than the Canadian) have dropped development East of the Islands, but it could still be something to watch 7-10 days down the road in the Central Caribbean. The pattern

New project looking good... +65 models in one single chat. Text, code, image and video generation in a single chat.

Google DeepMind has been extremely impressive this year - these stats are consistent with what I've seen in some individual storm analyses. A few thoughts: 1) Beating NHC for track is quite impressive. AI could be a major tool to help track forecasts keep improving. 2) Given a

Forwarding these plots over from @FranklinJamesL’s BlueSky account. In 2025, the Google DeepMind model has outperformed everyone on track, including the NHC and the HCCA consensus. It is competitive with the NHC and HCCA on intensity as well, outperforming the hurricane models.

My latest discussion with @brightbandtech and @danrothenberg on the global tropics (as seen through the lens of AI model forecasts). There's a special focus on a possible new typhoon impacting the #Philippines and also a discussion on the signal (well, lack thereof) for Caribbean

As expected, GEFS is backing off on development of this wave in the MDR. GFS still blows it up after some unrealistic convective feedback, but I imagine it will eventually stop showing that. A sizable number of GEFS members now show a solution like the AIFS/ECMWF with a weak

1/ $PYPL is one of the few remaining high-quality, undervalued stocks. - Margins are expanding again. - Venmo growth reaccelerated to 20%. - Its advertising network is ramping up. Yet, it's trading at just 14 times earnings. Here's our $PYPL investment thesis: 🧵

AIFS has the least aggressive solution for the MDR wave moving off Africa. Has the wave under an anticylone near Barbados Sunday, but it's too low amplitude to do anything, and then land interaction limits further development as a strong ridge shoves it into Central America. I

The other wave of interest is emerging off Africa. It's a pretty low-amplitude/weak wave for now, and doesn't seem like it'll be in a hurry to develop. GEFS/GFS have, not surprisingly, backed off on development East of the islands. The best bet for development is probably going

#97L has become TS #Lorenzo overnight. Despite the upper low to the west, it has found itself in a pocket of low shear long enough to organize and get named. In fact, the upper divergence over it right now is fairly favorable. The long-term outlook is still quite murky, however.