In brief: Invest 98L is looking healthy this morning. We should probably anticipate a depression or tropical storm by later today or tomorrow. From there, it may stay buried in the Caribbean — or as Google’s AI modeling suggests, it will stall and intensify and come northward toward Hispaniola. Whatever the case, 98L bears close watching in the Caribbean this week.
Invest 98L in the Caribbean
Invest 98L remains a rather robust tropical wave this morning, probably the most impressive Caribbean wave this year.
Invest 98L moving into the Caribbean this morning. (Weathernerds.org)
I don’t know that this is yet a depression or storm, but let’s just say we’ve seen lesser systems be upgraded this season. I wouldn’t be shocked to see something happen here before the end of today.
For now, NHC odds are up to 80 percent. It’ll be interesting to see what happens here. The general rule of thumb is that a system that intensifies quicker will be more apt to turn poleward (to the north) faster. A weaker system would be more likely to remain buried in the southern Caribbean. If this is indeed intensifying and forming, we could quickly see a tap on the brakes, stall out for a bit, intensify, and turn due north or north-northeast back toward Hispaniola. You can see the split in model guidance this morning, with Google’s AI ensemble stronger and turning more northward faster than many European ensemble members that are generally weaker.
Google AI ensemble members are tracking farther north toward Hispaniola compared to the weaker Euro ensemble which is mostly remaining in the Caribbean. (Weathernerds.org)
The AI ensemble may have the edge here given the satellite presentation of 98L this morning, because it’s been the best performing model this season, and from what we’re seeing in SHIPS guidance today, showing over 4 times greater than average risk of rapid intensification in the 72 hour forecast timeframe.
Odds of rapid intensification of 65 kts (75 mph) in 72 hours is over 4 times climatology with Invest 98L, not an outlandish number but certainly with higher than average odds. (Polarwx.com)
While the idea of a weaker, farther south and west tracking system is valid, I think the odds definitely favor the potential for a much stronger system right now. The big question is whether the fair bit of westerly shear it is experiencing right now will fundamentally hold back the system from developing rapidly. Something like lots of bark and a little bite. In that instance, the system could also turn north as well, but it would be sloppier and more ragged.
Whatever the case, it is prudent right now for Hispaniola to begin taking this system quite seriously. While none of the explicit modeling forecasts a major hurricane right now, we know from history and observation of the current warm water in the Caribbean that intensity forecasting could be underdone, perhaps significantly. This will continue to bear close watching over the next few days.
A sample of storms that have tracked from a similar area over the years in October and November. (NOAA)
This is not an uncommon pathway to impact. We’ve seen memorable storms come out of this part of the Caribbean this time of year, including Sandy in 2012, Matthew in 2016, Hazel in 1954, and Tomas in 2010. Obviously, when noting Sandy and Matthew and Hazel, some folks on the East Coast may get jitters, but at this point in time, we expect this system to follow some of the eastern-most tracks on the map above, keeping it away from the East Coast.
We will continue to monitor Invest 98L and its potential impacts and have another update for sure tomorrow and perhaps this evening if necessary.
In brief: Caribbean development odds from Invest 98L are increasing this week, but there are far too many questions than answers on its possible future in the Caribbean. We also take a look today at what could be a forthcoming significant pattern change for much of the Lower 48 just before Halloween, with an atmospheric river in the Northwest and the coolest air of the season so far in parts of the East and Central U.S.
Caribbean cruising with Invest 98L
The area we’ve been watching for the Caribbean over the last week or so is now up to a 60 percent chance of development and is classified as Invest 98L.
60 percent odds of development exist in the Caribbean later this week. (NOAA/NHC)
As this races into the Caribbean, it probably won’t develop. However, by the time it gets about halfway across the Caribbean it will begin to slam on the brakes. This is a fairly common behavior of a tropical wave this late in the season. Once it slows down, then all bets are kind of off. We could see slow development into a tropical storm or hurricane. We could see slow, erratic movement. We could see nothing happen at all.
By the time we get to Thursday evening, Invest 98L will be somewhere in the central or southwestern Caribbean. (Google Weather Lab)
Based on the combination of Euro ensemble and Google AI modeling, it would appear that by Thursday evening, whatever Invest 98L will be is somewhere between the southwest Caribbean and Hispaniola. Some ensemble members keep hooking this north and northeast across Hispaniola or Cuba and out to sea. Others keep it buried in the Caribbean for a few days, eventually threatening Honduras or Nicaragua. Others still keep just drifting it northwest toward the far western Caribbean before eventually turning back north and northeast. It is unlikely to get into the Gulf given the time of year.
Given how warm the Caribbean is, and given the uncertainty of the track, it would be best for all folks from Nicaragua and Honduras up to Cuba, the Caymans, and Jamaica over to Hispaniola to monitor the progress of this disturbance through the week. Yes, large storms can form in the Caribbean in late October. Stay tuned
Major U.S. pattern change on the horizon?
We’re starting to see signs of a major shift in the weather during Halloween week, so beginning in about 8 days or so. You can see the change in the upper air pattern when we look at the Euro ensemble about 20,000 feet up from day 5 through day 11:
A significant push of moisture and cool air in the West (blue) will expand into the central and Eastern U.S. by the 29th or 30th.
The 8-to-14-day outlook is highlighting a major precipitation event in the West that seems plausible with that upper air pattern.
The 8-to-14-day hazards outlook. (NOAA CPC)
In fact, you can see that precipitation anomalies as forecast by the European model are much above average in the Pacific Northwest for the 7 day period ending on the evening of October 29th.
Precipitation anomalies for the 7-day stretch ending on the evening of October 29th. (Tropical Tidbits)
An atmospheric river will get things going here. Eventually, that will send a disturbance across the country, allowing for colder air to dump in from Canada. Much of the South has yet to have that “Wow, here’s autumn!” front yet, while much of the North has had some periodic colder temperatures. But it does appear that things will change next week with this system and cold front. Kansas City hit 43 this morning, their coolest morning of the season, but the current blended model forecast for Halloween is calling for 30s.
Kansas City should have their coolest morning of autumn right around Halloween. (Open Meteo)
We’ll see just how far south and east this can get, but it’s possible we see 40s in Dallas and Atlanta as well from this front (though Atlanta will probably see 40s tonight as well for the first time). Either way, it appears that at least a temporary pattern change is on the way for the Lower 48. Whether that sustains into November or we fall back into a milder stretch again remains to be seen.
In brief: Today’s update looks at the time remaining in the Atlantic season, which is dwindling. We also dig into a tropical system approaching the Caribbean Sea that will need to be watched closely in the coming days.
The finish line is in sight
Officially, the Atlantic hurricane season runs from June 1 to November 30 every year. Storms can form outside of these arbitrary dates, of course, and they often do. But the vast majority of tropical activity, and especially hurricanes, occur during these six months. Looking at the calendar this morning, we have a little more than 43 days left until November 1. So the finish line is in sight, and overall activity is definitely trending downward as the Atlantic basin cools. But we’re not there yet, so let’s jump in.
Southern blob
Let’s start with the southernmost blob on the National Hurricane Center’s forecast map, because that is of far more interest to pretty much all concerned. The forecasters at the hurricane center give this tropical wave a 30 percent chance of becoming a depression or tropical storm over the course of the next week as it moves westward, into the Caribbean Sea.
What we can say for sure is that the tropical system will bring a round of stormy activity, including heavy rainfall, to the Windward Islands this weekend. However we don’t expect tropical storm-force sustained winds or a significant storm surge. The question then becomes, what next? We of course don’t have any absolute answers for you. However, there is general agreement among most of our models that a low pressure system will be in the Central Caribbean Sea by around next Wednesday.
A mix of AI and global model ensembles for tropical low locations at 2 pm ET on Wednesday. (WeatherLab)
After this, well, we can’t really say much intelligent. There is potentially a corridor that could open to allow the tropical system to lift north during the middle of next week, toward Hispaniola or Puerto Rico. Alternatively, the system may just slowly meander westward in the Caribbean Sea (perhaps reaching Central America, and perhaps not) for a few days before finally moving north or northwest. I think the biggest point of concern I have is that, although we have reached mid-October, the Caribbean Sea remains plenty warm to support the development of a hurricane if other conditions are favorable. So we are going to keep a close eye on this one in the coming days.
Some of the warmest water in the Atlantic basin, well above 85 degrees F, remains in the Caribbean Sea. (NOAA)
Northern blob
This is non-tropical low pressure system that has a very slight chance of becoming an extra-tropical system during the next couple of days before reaching much colder water in the Northern Atlantic. Really, I think some forecaster at the hurricane center must have fancied the color yellow, and wanted to add a little more to the map, because this system is likely to be inconsequential.
In brief: Today we take a fresh look at the chances of Caribbean development next week. A lot of newsy stuff at the bottom today, including some discourse on how a lack of weather balloons may have impacted the storm forecast in Alaska this past weekend. Plus some good news and bad news as it relates to fallout from July’s devastating, tragic flooding in Texas.
Caribbean risks remain non-zero but still somewhat low
With the tropics now quiet, we can focus entirely on rampant speculation, which is to say the Caribbean. Let’s go way out to day 8, which is next Thursday evening.
Combination of the Euro ensemble and Google AI ensemble position of low pressure forecasts on Thursday evening the 24th. (Weathernerds.org)
Forgive the somewhat sloppy formatting, but I combined both the Euro ensemble and Google AI ensemble modeling on one map here. You can see that a minority of ensemble members support development in the Caribbean, and exactly where and how strong they get with it is also in question. What you’d typically see in this sort of pattern, I think, is that we’d have disorganized thunderstorms basically “fester” in this area for several days, and eventually they could congeal into a system that is slow moving in the southwest Caribbean. From there, it could drift inland into Central America and fizzle out, or it could come north or northeast. A lot of options are on the table. It could develop significantly or it could just be plagued by other issues to hamper its organization. Whatever the case, it’s something to watch but it remains mostly speculative at this point, without any real concrete risks to explain. We’ll follow along.
Newsy Bits
Our first real severe weather event of autumn may descend this weekend. Rather than reinvent the wheel, I will direct you to Alan Gerard’s writeup on this in his Balanced Weather Substack this morning. Fall often brings severe weather, especially to the Mid-South and Deep South, so it’s that time of year to begin paying attention.
Saturday’s severe weather risk extends primarily from NE Texas and northern Louisiana through the Missouri Bootheel to just south of St. Louis. (Pivotal Weather)
Alaska: Recovery and weather balloons
Recovery continues after this weekend’s horrific storm in western Alaska. There are still two missing and one person that was killed from the village of Kwigillingok. Schools are still being used as shelters as the recovery period begins. Thinking about recovery in rural, remote western Alaska, you have to reset your mind a bit. This isn’t going to be like a recovery virtually anywhere in the Lower 48. Life is challenging on a good day in some parts of Alaska, and this is the sort of recovery that is a logistical challenge. I strongly encourage you to read Rick Thoman’s piece linked there.
One thing we have discussed at times since spring is that the degradation in consistent weather balloon launches was not a question of “if” but “when” it would come back to bite us. While we were fortunate (to this point) to not see a hurricane in the Gulf, which particularly concerned me as the scenario we’d get burned on, it is entirely plausible, if not likely that the forecast for Halong’s impacts in Alaska was negatively impacted by a substantial loss of upper air data. As Rick points out, the storm deviated from model forecast tracks and sped up. Would upper air balloon data from St. Paul Island for example have helped better forecast the directness of the impacts? We need experiments to prove it, but circumstantially this is almost exactly what we would have expected to see with a lack of consistent upper air weather data. This is why a strong, fully funded and fully staffed National Weather Service is important. We currently do not have that.
Louisiana: Flood mitigation plans should be regional
Map of Louisiana watersheds. (Putnam-Duhon, Lorna. (2016). Nitrogen and Phosphorus Trends of Long-Term Ambient Water Quality Monitoring Sites in Louisiana. 10.13140/RG.2.1.1631.5920.)
Mississippi coast: The cost of home ownership due to disasters
Mississippi Today has a deep dive on the insurance challenges and costs of home ownership near the Mississippi coast. While there have been tremendous strides in the wake of Hurricane Katrina 20 years ago, the vulnerability remains extreme. Large carriers like Allstate and Progressive have stopped writing wind policies there. And there’s now a vicious cycle where people cannot afford repairs and upgrades to help make their properties more resilient to storms, while their premiums keep rising, and the state’s lax approach to any new regulation hangs over it all.
Texas: Further investigating this summer’s flood tragedy
Some encouraging news in Texas. The summer special session of the Texas Legislature was completely overshadowed by the redistricting situation, but there were some notable, good elements related to this past summer’s flooding tragedy that came from it. Is it enough? No. But there will be more discussion and testimony on the event and trying to piece together more about what went wrong. The state is convening what they are calling the Senate and House General Investigating Committees on the July 2025 Flooding Events. This will be a joint panel on digging into what happened that will probably be able to move more slowly and with more detail than was allotted this past summer. Presumably, additional recommendations and guidance will be implemented as a result of what they uncover. Overall, I call it encouraging because it would be easy to say “We did something” after this summer’s session and move on. This suggests that there is more to come, and that’s a good thing.
Texas: Questions about FEMA eligibility
Some discouraging news in Texas from Emily Foxhall of the Texas Tribune. Only about 20 percent of all FEMA disaster applicant requests from Kerr County have been deemed eligible since the July 4th floods. Sarah Lebowitz who researches disaster for the Carnegie Endowment found that between 2015 and 2024, about 40 percent of FEMA applications are deemed eligible. 54 percent of applications from Kerr County haven’t even been reviewed beyond the initial stage. The Texas Tribune also found that the rate in Kerr County is well under the rates of other affected counties in Texas that are dealing with similar processes in working FEMA’s system. One reason they speculate on may be that people are struggling to work an online system in an area with spotty internet that’s still disaster ravaged. FEMA isn’t sending people door to door to help victims work through the system and process as they have in the past. This is likely due to budget cuts enacted under the current administration. The current shutdown is likely not helping matters.
Whatever the exact explanation, the story is worth your time to understand how nonprofits are helping fill (some) of the gaps. They can’t do it all however, and the process of navigating a post-disaster landscape can be maddening and full of additional hardship, particularly for people who lost everything and/or friends and family during the flood.
