Upcoming week - October 25-31

[Andrew92]

Evaluating last week

Well, I predicted no tropical cyclones to develop last week, including Invest 99L. Unfortunately, that one did develop, and then some – becoming Richard, and then a hurricane before making landfall in Belize. So that gets an F right there. Elsewhere, nothing else developed. Invest 90L was designated off the coast of Africa, but nothing consequential has come from it so far, or from anything else. Therefore, my prediction for the rest of the Atlantic certainly gets an A. The Richard part of the forecast has to be given the most weight though, I’m afraid. Anytime a hurricane forms and makes landfall anywhere, it’s big news for where it hits at the time, even if it’s a C1 and ends up causing minimal damage, which I pray is the story that comes out of this storm. Overall, my grade for this week is a potentially very generous D.

How is this week looking? Let’s take a look…

Current situation and models

Hurricane Richard made landfall earlier this evening in Belize and will weaken over land. The consensus from the NHC and from models is that in about 36 hours, this storm will reach the Bay of Campeche as a dying tropical depression. There, it will encounter significant wind shear that will tear it into shreds. Therefore, the Richard part of this week’s prediction is appearing, on the surface, to be fairly cut-and-dry.

Invest 90L is still spinning over the open Atlantic. Its hours for development are clearly numbered as it heads northwest over cooler waters – though shear will remain moderate. That said, only the NOGAPS seems to do anything serious with this disturbance.

The remainder of this week appears to be quiet until the weekend. By that point and into early next week, some reliable models such as the GFS and Euro are hinting on a development of some sort in the Caribbean, not far behind Richard but a few days distant. The area of development appears, for now, to be around the central or eastern part of the Caribbean.

Recent history

Since 1995, these storms have developed during the last week of October:

Tanya in 1995
Katrina in 1999
Unnamed subtropical storm in 2000
Lorenzo in 2001
Michelle in 2001
Beta in 2005
Noel in 2007

That’s a total of seven storms during this upcoming week. When looking at where Invest 90L is currently situated, only one of these formed in a comparable location: Lorenzo. That storm only mustered tropical storm intensity, gathering maximum sustained winds of no higher than 40 mph. The last storm before Lorenzo to pull off development in a similar area during the last week of October was Tropical Storm Jose in 1981 – and it too was weak, having maximum sustained winds of 50 mph at one point at its highest.

But when looking at the Caribbean, it’s a different story. Four of these seven storms developed in the Caribbean: Katrina (1999, not the famous one we all know and hate), Michelle, Beta, and Noel. Of this bunch, only Katrina failed to become a hurricane, simply due to forming too close to Central America and then running into it – though Noel didn’t become a hurricane until it was already out of that area and skirting the East Coast, in the first couple days of November. The calendar also crossed into November before Michelle had a chance to become a hurricane. The last storm to develop in this area during the last week of October before this mini-batch was quite a distance back, and a weak one too: Tropical Storm Helena in 1963, in the northeastern part.

However, in the past 15 years, one storm was still active in the Caribbean, and a big one at that: Mitch. In fact, it was during this upcoming week when Mitch was making headlines as a rare C5 hurricane in October, and for its devastating mudslides in Central America.

So what does this all tell us?

Invest 90L has really got its hands full. Shear may be only moderate, but waters are cooling, and history is not exactly in favor of it developing. The disturbance also just plain doesn’t look like it’s in any hurry to develop out there. If it does, per history and what it looks like, there is very little chance of anything intense coming out of it in the tropics.

The Caribbean tends to gradually calm down during late October, but a late development here cannot be ruled out, and is still the likely place to look for it. It usually happens in the western part, but Noel proved it can still happen further east. Also recently, some of the storms that have developed have gone on to become major hurricanes – though the sample size is admittedly very small.

Back to looking ahead

First, let’s get to the top story, Hurricane Richard. I predict Richard will weaken to a tropical storm on Monday morning as it crosses into Mexico, and a tropical depression late that night as it nears the Bay of Campeche. Richard will reach the bay on Tuesday morning and then run into the strong shear that will ultimately spell its demise. Richard will become a remnant low that night. Overall, due to it being a small storm, effects will be somewhat localized, but still severe in Belize and the southern part of the Yucatan Peninsula. The primary threat will be heavy rainfall. Confidence is 75%.

Now, Invest 90L’s prediction, and a simple one at that: I predict no tropical cyclone development for Invest 90L. Confidence is 80%.

Now, the tricky part. It’s been my undoing in recent weeks to nail the last part of each upcoming week, and with a new development possible next weekend, I have another tough call to make. History is not really against development in the Caribbean, and models are supporting something coming out of there. This means it is likely safest to predict something to happen around that time. But a couple of questions loom in terms of this week’s prediction. First, will it pull it off? And if so, when will it pull the trigger so I can try to make a decent track and intensity forecast? Once again, not easy, but I have history, the models, and my gut, with my gut hopefully being the most reliable since models really seem to enjoy stinking it up for me. Having nothing of note in that area at the present time also toughens this prediction. But with the information I have, here is what my gut says: I predict a new tropical depression will develop on Sunday, and become Tropical Storm Shary before the week is up, gathering maximum sustained winds of 40-50 mph as it heads westward, away from the Lesser Antilles. Confidence is 60%.

Elsewhere, I predict no tropical cyclone development. Confidence is 95%.

Historical footnote

With Hurricane Richard’s landfall in Belize, I thought I’d look back at the most recent storms to hit Belize. The last hurricane to make a direct landfall in that country was Hurricane Iris in 2001 – also, the last storm to do so in October! The last storm that would count as a hurricane strike for Belize was Dean in 2007, as it made landfall just north of its border with Mexico.

And the last hurricane to make landfall within a week of Hurricane Richard in Belize? That would be Hurricane Hattie, on October 31, 1961.

-Andrew92

[Migle]

Off by four days on Shary and we may even end up getting Tomas before Sunday, but I always love reading these things. Keep up the good work

[hurricaneCW]

We got Shary and Tomas, we got Shary a few days before Sunday. It was a good write up though, I don't think anyone could have predicted that we would see Shary and Tomas quickly develop into TS, especially Tomas.

[Andrew92]

Yeah, no way I even saw 92L coming, let alone develop. Still not good, but not as blatant of an error as some I've had recently, given few saw it coming and it was a fish. But I definitely made the right call not to develop 90L and also right on 91L, just the timing was off which led to the intensity being off.

All things considered, this week was an improvement for what's been a very rough October for me. I'll do the "formal" post when the 11 PM advisory comes out.

-Andrew92

[snowcane180]

Its fun reading your predictions.

[Andrew92]

Well, with the 11 PM advisory on Tomas posted, it's time to evaluate.

Really, I can sum October in a nutshell as being an overall very rough month. This past week was not all that great either, but was not as bad as a couple other weeks. I did accurately predict no development for Invest 90L, and I also accurately predicted a storm in the eastern Caribbean by the week’s end. Oh, and Richard also fizzled in the Bay of Campeche as expected, when it was expected.

Since I’ve also already described how I did on Richard, I’ll give myself a lowly-weighted A there. I also have already covered my accurate prediction for 90L never developing, so that too gets an A, and that one gets a bit more muscle as that one, albeit a fish, was a bit harder to pin down (though still not the bulk).

That sounds pretty darn good on the surface. However, going deeper is where I faltered more than a little. I predicted the eastern Caribbean storm to be Shary, and it was named Tomas. Now, the idea of development and gradual intensification is really what matters more than just getting the name wrong. However, it also shows that I was wrong somewhere else. Indeed, Invest 92L developed further north of the Virgin Islands and became Hurricane Shary on its way out to sea before this particular entity had any chance of developing. And yeah, Shary managed to defeat some enormous odds and become a hurricane anyway from that disturbance that seemed to sneak up on some people. Either that, or it wasn’t actually 90L that models were developing in that area, but something else instead. In any event, if I’m going to entirely miss a storm altogether, at least it was one that was little threat to land, merely providing just a little rain to Bermuda on its way out to sea. For not seeing this part of the past week coming, I have to give myself an F. However, with this being such a little-consequence event (if that term even exists), weighting will be a bit lower than elsewhere (except for maybe Richard dissipating).

Now comes the meaty part of this week, which was saved for the end. I did mention an accurate prediction of a storm on Sunday in the eastern Caribbean heading westward away from the Lesser Antilles. That is ongoing with now downgraded Tropical Storm Tomas, after it was a hurricane prior. For such a long-range prediction, the track was executed exceptionally well. The real problem is the timing of development, which usually also leads to intensity being well off. That is no exception here, as I predicted this storm would develop on Sunday in the Caribbean. Instead, it developed about 36-48 hours before I said it would, and it was still east of the Windward Islands. It also developed quickly from there, into a hurricane the next day after passing by Barbados towards St. Lucia. A weakening has since taken place, but the damage in this area was already done. Predicting the development of storms has been a difficulty all season for me, and end-of-week developments especially has really been an albatross. This one is no exception. Since I at least called for it to happen, and the positioning on Sunday was pretty good, I cannot ding myself too much in this area. Having the timing off as egregiously as it was, though, is not something I can exactly hang my hat on too much either. I will potentially be generous and split the difference of an A (for seeing it coming and its positioning at week’s end) and a D (timing and intensity issues) with a C grade for this area.

Wow, what a complicated evaluation. I totally didn’t see Invest 92L becoming Shary between Bermuda and the Virgin Islands. I did accurately predict no 90L development, and something at week’s end in the Caribbean. But the Caribbean development’s timing was also off. Overall, the Tomas portion of the forecast was the most significant by far, and this week will be weighted on there (the name of the storm itself doesn’t matter as much in my opinion here, only shows I messed up somewhere else). This is appropriate, as the Shary failure and the successes of Richard’s dissipation and no 90L development can probably offset each other. Therefore, I give myself a C for this past week.

The first week of November will be up shortly.

-Andrew92