Michael Lowry

An update to the preliminary model track and intensity verification for Typhoon Champi, which is just making landfall in Taiwan. JTWC (Joint Typhoon Warning Center) is the official forecast, then I grouped my selection of models by type.

If you believe the models, we should finish off October without any additional tropical threats or development contenders. But what about November? In today's newsletter, I dive into just how common tropical threats are in the season's final month. ⬇️

Josephine formed yesterday but will be another Atlantic "shortie" as it quickly morphs into a powerful extratropical storm today into Friday. It'll also do little to pick the season off the mat, which continues to be one of the least active on record. ⬇️

The strengthening nor'easter, also known as Invest 93L, is trying to make a run at a third moniker – Josephine. We're seeing an uptick in storminess near the low pressure today but name and standing aside, the ongoing coastal impacts don't change. ⬇️

Glad to offer my thoughts in this @npr.org Morning Edition report that aired today about how AI forecast models are changing the nature of hurricane predictions. The AI models aren't always the answer key (see Simon last week) but they're advancing predictions at a pace we've not seen.

Josephine aside, whether our Mid-Atlantic nor'easter laps up enough Gulf Stream fuel later this week to transition into a subtropical storm is immaterial to the coastal impacts expected from the Outer Banks to Southern New Jersey over the next few days. ⬇️

Our better performing forecast models are starting to advertise possible tropical or subtropical development along a stalled front off the Carolinas next week. As NWS Morehead City notes, there's "some level of coastal concerns" regardless. More ⬇️

The Atlantic basin has seen reduced activity nearly everywhere this hurricane season, except along its edges (including the Gulf). Meanwhile the eastern and central Pacific couldn't look any different, especially around Hawaii. More in today's newletter ⬇️

With some chatter going around about how consistent the Google DeepMind forecasts were for Isaias, I decided to apply some work Andy Penny, a former NHC colleague of mine, and I did a few years back. Basically, we came up with a few measures of forecast consistency. (1/8)

Many versions of the current El Niño intensity have been shared… we’ve had to change the colour scale to appropriately show the strength of the anomaly.

When does the mainland U.S. usually see its last hurricane landfall? In today's newsletter I look at what the historical record tells us and why the next few weeks are especially meaningful for folks living in the hurricane zone. ⬇️

Quite the evolution in Isaias's winds from earlier in the day Friday when it was a much more symmetric hurricane with its strongest winds east. Sting jet pretty obvious here but exceptionally challenging to predict this evolution in real time.

Wasn't planning on this since the sample is tiny, but had two requests, so here are the track and intensity verifications for Hurricane CantPronounce. GDMI big winner for track. Mixed bag for intensity, but the physical hurricane models were positive contributors. (Name credit Marty Merzer.)

Preliminary track verification statistics for #Isaias show that the Google Deepmind model dominated the model guidance by a huge margin. The National Hurricane Center closely and wisely followed it. Details at bmcnoldy.blogspot.com/2026/10/isai...

Hurricane Isaias packed quite a punch as it struck Friday evening, worsened by a probable sting jet that developed on its back side, which sent winds topping nearly 120 mph rushing to the surface in the Pensacola area. My Saturday morning wrap-up ⬇️

Isaias definitely a gusty hurricane. Lots of stations with gusts north of 100 mph so far, with gusts topping almost 120 mph roughly 25 miles north of Pensacola. Not many storm surge stations in the worst hit area east of Fort Walton/Destin but Panama City Beach hit nearly 6 feet.

5 PM CDT Friday Oct 9 Update: Wind gusts approaching hurricane force along the Northern Gulf Coast. Follow the latest on #Isaias at https://hurricanes.gov https://x.com/NHC_Atlantic/status/2108680866255819051

* Updated Graphic* Hurricane #Isaias is expected to make landfall on the Central Gulf Coast later today, bringing a threat of heavy rainfall and flash flooding to the region. Numerous instances of flash flooding are likely, some of which will be significant.

Hurricane Isaias continues to strengthen this morning and is poised to be the first U.S. hurricane landfall in 2 years. The big question now is when does the strengthening stop? Latest models show it closing in on a Category 4 by this afternoon. My latest ⬇️

A reminder from @michaelrlowry.bsky.social... "You often don’t have to move inland very far to escape life-threatening storm surge. Roughly 90% of deaths in hurricanes historically come from water, not wind."

Hurricane Hunters are finding steadily falling pressures with Isaias this morning. Later today and on Thursday is when the National Hurricane Center forecasts rapid intensification. By Friday we could see major Hurricane Isaias bearing down on the northern Gulf Coast. My latest ⬇️

For those not on the other site, a good reminder from Matt Onderlinde (NHC scientist and weathernerds.org creator) about a quirk of the ECMWF tracker for systems originating west of 90⁰W longitude. I like the 100-member ECMWF probabilities as well here --> charts.ecmwf.int/products/med...

Lots to cover with a very healthy-looking and organizing Invest 92L. There's a growing threat to the northern Gulf of a tropical storm or hurricane by late Friday into Saturday. The latest Google DeepMind takes a notably aggressive stance. All the details ⬇️

While the Gulf Coast NWS comms issue appears resolved, with the growing tropical risk was seeing way more social media traffic about the radar outages that occurred over the weekend. Feel compelled to share a thread as seeing a lot of misinformation. (1/)

There are plenty of reasons to be concerned about the aging NEXRAD network, but this was NOT an issue with the radars. This was solely about comms. A break in the NWS network occurred somewhere that took out two NWS offices comms: WFOs Mobile and Tallahassee. (2/)

Tricky forecast ahead for the Gulf system, with big diasgreements in organization, timing, and placement of late week/weekend impacts. The AI models – especially Google's DeepMind – have turned very bullish but are they overcooking things? My Monday morning deep dive ⬇️

Models aren't sold on the frontal mess late next week over the southern Gulf spawning anything tropical (or subtropical). Chances aren't nil but they're low for now. Regardless, if anything makes a run, it should stay sloppy, lopsided, and disorganized. ⬇️

In other words: This meme is a surprisingly accurate scaled depiction of the magnitude of the 2026 El Niño event vs the 1997 event *if using the absolute index.* But if using the relative index, Danny DeVito is currently the same height as Arnold Schwarzenegger. And either pairing is striking... 👀

The development signal over the southern Gulf has ticked up slightly for late next week into the holiday weekend. For now, odds favor a sprawling, subtropical-like system if anything gets going. Given the Gulf warmth it's worth following the trends. ⬇️

Fay was declared done yesterday by the National Hurricane Center but it's trying to mount another comeback this morning. Today I discuss why we can't quite rule out Fay having been our first Atlantic hurricane. Plus an update on what we're watching late next week in the Gulf. ⬇️

If you're sick of hearing about Eastern Pacific hurricane records, I have bad news. The basin just set a record for the highest activity to date. Also we now have Hanna in the Atlantic but don't expect it to pick the basin out of the hurricane gutter. ⬇️

Hurricane Nolo is at Category 2 strength as it nears Hawai'i.

Polo's been a Category 5 hurricane for a total of 48 hours so far, longer than any other Category 5 hurricane on record for the Northeastern Pacific. 2026 also just tied 1997 and 1994 for the most Category 5 hours in a single season and should be the new recordholder by tonight.

The nor’easter continues to slowly, very slowly come into better focus. We explain the tidal flooding, wind, and heavy rainfall risks for New England and the Mid-Atlantic. We also rundown Hurricane Polo’s latest, as well as Nolo’s threat to Hawaii. Details below or at theeyewall.com.

Nolo is revving up south of Hawaii and threatens to be a record 3rd hurricane impact for the state this year. The Big Island has never recorded a major (Category 3 or stronger) hurricane as close as NHC is forecasting Nola to get this weekend. More ⬇️

In today's newsletter I recap all the Hurricane Polo superlatives from Tuesday, including gnarly record warmth hurricane hunters measured while passing through its eye. Plus, another Hawaii hurricane inbound? Could be one of the stronger approaches yet. ⬇️

An extraordinary high-resolution view of a Category 5 Hurricane Polo this afternoon, seen from NOAA-20.

If we use a 35-kt intensity reference point (minimum TS strength) to count hours to category 5 strength... - Linda 1997: 48 hours, 35->155 kt - Patricia 2015: 42 hours, 35->150 kt - Otis 2023: 45 hours, 35->145 kt - Polo 2026: 36 hours, 35->140 kt 140 kt = lowest 5-kt increment for cat 5 strength.

However, if *any* TS strength counts (35-60 kt), we get these stats... - Linda 1997: 30 hours, 55->155 kt - Patricia 2015: 24 hours, 60->150 kt - Otis 2023: 21 hours, 55->145 kt - Polo 2026: 24 hours, 55->140 kt By this definition, Otis was faster than (preliminary) Polo. Thresholds matter! :)