Upcoming week - September 7-12

[Andrew92]

Evaluating last week:

It may have been an extended week by a day, and therefore more challenging there by default. However, this was easily the most difficult week we have faced yet this season, with Earl and Fiona forming as expecting, but also Gaston and Hermine came along. These two I did not see coming, and are going to hamper my grade a fair amount. I did not predict much other than Earl and Fiona for this past week, and boy was that wrong! (Even if Gaston hasn’t amounted to much)

Starting with Earl: This forecast was fairly decent, I have to say. Earl did indeed graze past the Leeward Islands as a C2 hurricane on Monday. It also became a C3 hurricane as it left the area later that day, which I also correctly anticipated. A northwest turn and intensification to a C4 hurricane were predicted for Tuesday, and that’s exactly what happened. The storm made it a little further west on Wednesday than I thought (halfway between the Bahamas and Bermuda), but that error means little considering it wasn’t really affecting land other than for surf at the time. I predicted maximum sustained winds to reach 140-145 mph, and they did so on Wednesday night and Thursday morning; I actually predicted this for 24 hours earlier, on Tuesday night and Wednesday morning. I did think about the possibility of an eyewall replacement cycle, but I thought it would start Wednesday as opposed to Tuesday. Still, this error is only a slight one. I did predict weakening to a C3 on Thursday as it passed east of the Outer Banks, which indeed happened, but it came more rapidly than anticipated as it also weakened to a C2 that same night. Friday’s track of just south of New England also materialized, but the more-rapid-than-predicted weakening continued; I predicted a C2 hurricane passing back, but it was a tropical storm. Still, I predicted tropical storm force conditions in this area, and that did happen. Then on Saturday, Earl hit Nova Scotia (again, correctly predicted), but as a tropical storm, albeit with 70-mph maximum sustained winds; not horrifically outside the 75-85 mph window I gave it. It then became extratropical that night, as predicted! So looking back overall, for such a difficult forecast (with the possibility of moving onshore or not affecting the East Coast at all other than surf), this was a darn good forecast! Not perfect, but despite some mistakes, it’s hard to have been able to do much better here. I give myself an A- for this part of the forecast (also sealing a passing grade for the week).

Now to Fiona: This forecast was (thankfully) badly botched. I ultimately predicted a major hurricane to make landfall out of this! Still, the earliest part of the forecast sees a little silver lining. However, it became a tropical storm a day before I thought it would; that said, slow development was the call before the Leeward Islands, and that did happen. I predicted a close approach to the Leeward Islands on Wednesday with maximum sustained winds of 50-60 mph, which is also exactly what happened. But from there, well if you read last week and know what actually happened, I horribly blew this part. Fiona never really got off the ground after that and was only a minor threat at worst to Bermuda; quite a contrast from the major hurricane threat I foresaw for Labor Day weekend! Early part of the forecast gets an A, the latter part gets an F, but considering the drastic latter part of the forecast, these do not average to a C. Even given the difficulty of this forecast, it is still best to give myself a potentially still very generous D for this part of the forecast.

The Danielle portion of becoming extratropical and continued weakening was correct. I get an A there but this counts to like 1% or so of this week’s total. I predicted no other developments, and while Gaston probably wouldn’t lower my grade too much (maybe to a B), Hermine definitely does, even if it came late in the week. I give myself a C- here.

Overall, with a very good Earl forecast, a good early Fiona forecast that became terrible after that, and not accurately seeing Gaston or Hermine, this wasn’t the best week for me. Still, the Earl forecast was the crucial one as it turned out, so my grade for this week will be a bit higher than if it didn’t take place. I give myself a C+ for last week (but would be a D or so without Earl).

So can I do better this week? Or do I want to do better? Here we go…

Current situation and models

Tropical Storm Hermine is heading inland tonight and will continue to weaken. However, a flood threat has to be considered for Texas as it does so. The remnants of Gaston are still swirling near the Leeward Islands. More and more models are starting to lose this system, and even the NHC is losing faith in it tonight (chance of development has dropped to medium for the first time in over three days). One potential alternative being discussed is the potential for re-development in the Caribbean, so I’ll consider that in this week’s prediction. There’s really only one other area that looks to be seriously be talked about though: a wave set to come off of Africa on Tuesday that models near unanimously develop later this week.

Recent history:

The last year to not have any tropical storms or hurricanes this upcoming week was 1992, which was an El Nino year and a terrible analog for 2010. Gaston may not re-develop and Hermine will dissipate inland, but the odds are too great for nothing to take place beyond these two systems. In other words, it’s best to expect something this week.

One of the hottest beds for development in recent years is, indeed, in the MDR. Looking back since 1995, we see new developments in that region which include:

Marilyn in 1995
Hortense in 1996
Floyd in 1999
Gert in 1999
Felix in 2001
Gordon in 2006
Helene in 2006
Ingrid in 2007
Fred in 2009

OK, so “only” eight newbies have formed in this area since 1995, about one every other year. Also of note, I haven’t forgotten about Georges, but he developed on September 15, just barely outside this week. If we were to include that as being close, it does indeed enhance the sample size. Other storms already active in the same area included:

Isabel in 2003
Ivan in 2004
Florence in 2006
Josephine in 2008

When combining all of these storms together, only two failed to become hurricanes in their lifetimes: Ingrid and Josephine. Further, only one other storm failed to become a major hurricane, that was present in this region at some point during this week: Florence. Two storms, Ivan and Isabel, went on to become C5 hurricanes, and Floyd came darn close. There are also a few big names on the list, such as Marilyn, Hortense, George, Isabel, and Ivan. The Caribbean is certainly a favored area for storms to hit, but only Georges and Ivan would hit the US.

But since some are talking about Gaston possibly re-developing in the Caribbean, what do we find there for new developments? What if I told you that not a single storm has developed in this region during this upcoming week since 1995? However, one could be correct in a theory perhaps that’s because the waves that could have, were already developed by that time. Marilyn, Hortense, and Ivan all certainly fit this bill. One might also point to Jeanne in 2004, which formed on September 13 in this region. So if we consider Jeanne as the only one since 1995, what about before then? There was Debby in 1994 (failed to get off the ground from there). Before that is a pretty big name: Gilbert in 1988. Danielle was also close in 1986, but more similar to Debby than Gilbert.

So what does this all tell us?

Our best chance for development is off the coast of Africa. If the tropical wave comes as advertised, there is almost no reason for it not to develop after the passages of Danielle, Earl, Fiona, Gaston, and other weak disturbance that couldn’t get off the ground. I have also mentioned that there is usually a lull period for development in this area after intense hurricanes. The lull can last about a week or a little longer, before one more big one makes its debut. This wave has a chance during the typical “peak” of hurricane season.

Gaston’s chances are very slim. With few new developments in the Caribbean, it is most likely that Gilbert was an outlier, and that we can’t use Marilyn, Hortense, Georges, and Ivan as good analogs for Gaston. Debby and Danielle showed it can be done in this area, but they usually dissipate quickly or don’t develop at all. If as many models take down Gaston as is present on top of that, his chances of development are near zero.

Back to looking ahead

First, let’s knock out Hermine since we kind of know what will happen. I predict Hermine to weaken to a tropical depression over Texas late Monday afternoon. There will be heavy rains and some flooding along the Rio Grande valley before the remnants move faster to the north, into Oklahoma by Wednesday and Kansas on Thursday, before dissipating altogether. Confidence is 70%.

Looking at Gaston, it’s been popularly talked about for re-development. Recent history and models are against that, though, and the NHC may be starting to follow at least the models. Another undiscussed aspect in this post has been interaction with land, and the remnants could be on a crash course towards the high mountains of Hispaniola in a couple of days. But if the remnants don’t even do that, the chances of further development aren’t good, and would probably be little more than a moderate tropical storm. I predict that Gaston will not re-develop this week. Confidence is 75%.

And now to the new tropical wave: Models like this wave to develop, and recent history suggests that almost every year, there can be something in this part of the world. If it’s not there at the present, it could easily come later by…developing! I predict a tropical depression to develop on Thursday night and become Tropical Storm Igor on Friday, while a little west of the Cape Verde Islands. Development will be gradual, but a C1 hurricane with maximum sustained winds will be on our tracking charts on Sunday, with maximum sustained winds of 75-85 mph. Confidence is 75%.

Elsewhere, I predict no tropical cyclone development. After last week, though, confidence is 80%.

-Andrew92

[Andrew92]

Let's evaluate...

It was a shortened week due to Labor Day. Still, the last 24 hours of the forecast proved to be might unpredictable. I said at week’s end we would have Hurricane Igor and that was it. I also thought Igor would be just forming into a hurricane today, with maximum winds reaching 75-85 mph. Yeah, um, that worked out all right. I definitely, even if I had the hindsight to predict a somewhat stronger storm at least, would not have thought to predict a storm nearing C5 intensity. Also, Tropical Storm Julia has developed. So in short, late Saturday night through now have been pure disaster.

But can one bad day make a bad weekly prediction? Many elements to my forecast before the last 24 hours were very well executed. Of course, Hermine dissipating while producing a flood threat was pretty much a given to happen. However, some models were insistent that Gaston would still find a way to re-develop. I countered that argument by mentioning that this time of year is not an ideal time for storms to develop in the Caribbean, and the ones that do typically dissipate quickly. Such was the case, as Gaston never re-developed. The formation of Igor was interesting. It developed 24-36 hours before I predicted it would, but then oddly weakened to a tropical depression…at about the time I predicted the incipient depression to develop. It then re-intensified to a tropical storm on Friday. Even though it quickly reached maximum winds of 70 mph on Saturday, its intensification slowed down from there and I thought my forecast was looking like a solid A-, because the track was pretty much nailed near perfectly. Then, came the aforementioned last 24 hours.

So I give myself an A- for the first five days, and a D- for the last 24 hours. When it averages out, this week was still at least a fairly decent prediction. I give myself a B for this week.


This new week is coming up in just a couple minutes....and there will be disclaimers.

-Andrew92

[JtSmarts]

Let's evaluate...




This new week is coming up in just a couple minutes....and there will be disclaimers.

-Andrew92

Sounds interesting, can't wait! I have been thoroughly enjoying your predictions by the way. I also found it interesting that Igor redeveloped back into a TS right around the time you said it would develop in the first place. Excellent job on everything up until the aforementioned "disaster period".