Upcoming week - August 23-29

[Andrew92]

Evaluating last week:

Last week was a mixed bag. Very bad in one area, about average in another, and very good elsewhere. Let’s look at each area separately and try to come up with the best grade for this week.

First, ex TD-5, partly just to get it over with but also because it basically kicked off the week. To sum it up bluntly, I blew this one. I predicted re-development, a tropical storm, and a landfall closer to Texas and Louisiana! Man, it didn’t even become a depression again, and its would-be landfall (if it were to count) was in eastern Louisiana. The rainfall threat was sure correct, but that’s all I can say that’s any remote silver lining for this forecast. I definitely flunked this part of the forecast, grade F.

Now let’s head to the African coast. I did predict development, and this happened. I also predicted it would take about a day from the time of the depression forming to seeing a tropical storm come out of it, and that also happened exactly to my predictions. My issues though are the timing and which one developed. I thought the first one would develop, doing so on about Thursday. The second one I thought would not be ready just yet, but it is the one of the two that ended up doing so. This had a bit of an effect on the intensify forecast as well, but like I said, the first 24 hours of intensity I also did a generally good job on. The track was also well-executed, as it has moved west-northwest – even making a little jog northwest for a bit. Overall, not a bad forecast, but not all that good either. I’ll give myself a C+ here.

In the Caribbean, the NOGAPS (again) forecast a possible development in the western part. I was adamant that this would not happen. Other models, however, began to jump onboard and try to make me a little nervous. Fortunately, this didn’t happen, and although there was a tropical wave that passed through, it never developed here. I also predicted no other development elsewhere throughout the tropics. When you combine these two together, I’d say that’s worth an A here.

So overall, I blew ex TD-5, but I also did a great job in not predicting the western Caribbean to have any development. The only other area I was also looking at was off the west coast of Africa. The forecast there was about average given the time I was making it – and trust me if you think this is easy, why don’t you try this once and see if you’re right every time! If it hadn’t been for my badly blown ex TD-5 forecast, this grade might easily have been a B or higher. But that one really brought my grade down, and although elsewhere was overall OK, I have to give myself a C- for the week.

And it’s supposed to only get tougher from here, right? Well, let’s take a look.

Current situation and models

Tropical Storm Danielle is churning off the west coast of Africa. There is a massive model consensus that it will gradually intensify into a hurricane and re-curve and be no more than a “fish” storm. A threat to Bermuda is not being ruled out, however. Another wave right behind Danielle is of interest by many models for possible development. This wave is looking fairly healthy as it emerges off the African coast. Yet another item of interest has also been talked about for possible development in the Gulf of Mexico by a fair amount of members here.

Recent history:

Rather than just look at every single storm that has formed or was active, I’ll simply hone in on each area for potential development, starting with the west coast of Africa. Storms in this area during this week since 1995 have included:

Humberto in 1995
Iris in 1995
Karen in 1995
Luis in 1995
Edouard in 1996
Fran in 1996
Gustav in 1996
Danielle in 1998
Cindy in 1999
Emily in 1999
Dolly in 2002
Fabian in 2003
Frances in 2004

There are only a few big names on this list. Among them are Luis, Fran, Fabian, and Frances (therefore, watch out for the F-storm if comes this week!) Even among these four, only Frances was a major hurricane by the end of this week, although Luis was also the only one not yet a hurricane. Over the rest of this sample, only two others also were major hurricanes as well: Edouard and Cindy. Of the remaining seven storms, three of them would become hurricanes during the week: Humberto, Iris, and (ironically) Danielle. The rest of the storms just could not get off the ground, though all but Dolly moved on northwesterly paths early (and Dolly was just forming at week’s end in an El Nino year). Adding it all up, 12 storms have occurred in this region since 1995 during this week, and 8 were hurricanes, with 3 being majors.

Additionally, a lot of the storms that were immediately consecutive saw one storm become respectably strong, and the other stay very weak, at least for a while. Humberto and Iris is one example, followed by Karen. Humberto was the first to intensify, and while Iris quickly followed suit, it also quickly weakened back to a tropical storm before re-intensifying later. Karen was in the middle of it all and could just not get off the ground. Even Luis struggled early in its life. Fran had similar issues as Iris for a while as it followed Edouard before intensifying more later, and Gustav could be compared to Karen, unable to get off the ground. Finally, in 1999, Emily appeared far away from anything at first glance, but actually was too close behind Dennis before then getting too close to Cindy – close enough to be absorbed by it, in fact!

As for the Gulf of Mexico, are you ready for something you may not be expecting? Not a single storm has developed between August 23-29 in this area since 1995. That’s right, not a one. However, the last one to successfully do so was a biggie: Elena in 1985. There have been some storms forming close to the August 23 or 29 since then, but not during the actual week. Of course, the two biggest names of all took place in the Gulf of Mexico this week, Andrew and Katrina. However, the idea here is new development in the Gulf of Mexico, not an already-developed storm reaching the area, which Andrew and Katrina were (Katrina having admittedly forming this week, but in the Bahamas).

So what does this all tell us?

It would not be a big surprise if a major hurricane does not take place this week. Conditions certainly are more favorable for development off the coast of Africa, and some of the first big names in the region start emerging as well. Most of the ones that do form, however, tend to become “fish” storms, many staying even east of Bermuda. Still, two-thirds of them go on to become hurricanes, but most are not (yet) major hurricanes by the end of the week. Most of the majors during the week, or storms that would go on to become majors, move on generally westward to west-northwestward paths, posing at least somewhat of a threat early on to the Lesser Antilles. The only one that did not do this was Cindy in 1999, making her a bit of an outlier in that area. If storms come back-to-back-to-back, at least one storm typically struggles.

The Gulf of Mexico is just not a favorable area to see new developments during the week. A couple of juicy names have occurred in this area in the last 20 years, but they were already well-established storms. Therefore, it is hard to be sold on anything forming in this area based on history alone. That said, I don’t see where models indicate development here, either – only speculation from members on the board.

Back to looking ahead:

The eastern Atlantic is the area to be watched this week. With Danielle out there, a wave that lots of models develop right behind it, and history on the side of both of them, there is little reason other than some weak shear in the area to not predict intensification for at least Danielle. The other one may struggle a bit. Of course, if it were to stay weak enough, there is the theory that it could keep pushing west, while Danielle could re-curve. If the western Atlantic is more favorable, that would make this one potentially the one to watch. But is that what I really predict? I must admit, that part of the forecast will be tricky.

Let’s start with Danielle: I predict Danielle will continue west-northwest over the next three days. It will take some time to intensify, given that it is not all that well-organized right now to begin with. Giving it time, it will intensify and become a hurricane on about Tuesday. By then, Danielle will already at a latitude equal to the northernmost Virgin Islands. Danielle will then become a C2 hurricane with maximum sustained winds of about 100-105 mph on Wednesday and Thursday. During this time period, the storm will also begin gradually curving to the northwest, passing safely northeast of the Leeward Islands. The re-curve will then begin within 5 degrees of Longitude 55 west, and the storm will stay east of Bermuda. On Friday, Danielle will begin to weaken, and by week’s end, the storm will have weakened to a minimal C1 hurricane with maximum sustained winds of 75-80 mph, well south of Nova Scotia and Newfoundland and heading northeast away from land. The best analog for this storm, in my opinion, will be Humberto from 1995. Overall confidence in this forecast is 75%.


Now the wave behind Danielle: I predict this wave will take a bit of time to develop. However, it will do so probably on Thursday. The struggles will continue, but Tropical Storm Earl will form on Friday. Then, as Danielle begins to lift out later on Friday, Earl will finally begin to intensify a little more, but at a slower pace than Danielle as it moves along a similar, west-northwest path. By Sunday, maximum sustained winds will have reached 50-60 mph and the storm will be a little northeast of the Leeward Islands, slowly now heading northwest. There are many possible analogs for this storm; therefore, this forecast is very challenging and this prediction just as much bold. Confidence is 45%.

As for the Gulf of Mexico, the system is decent-looking, but has not shown this all that consistently. Models do not seem to support any development, and climatology definitely does not. If it were to develop, I honestly think the best analog for this storm would be Grace from 2003, which you can guess what that means, to me, for this system. I do not predict this system to develop this week. Confidence is 85%.

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

-Andrew92

[Migle]

You didn't do to bad with Earl here. Of Course Danielle became a Category 4 so that part was a little off, but go the no GOM system. Can't wait to see this weeks.

[Andrew92]

Well, I kinda have a personal rule to wait until the last possible advisory unless it's obvious nothing is going to happen that's earth-shattering. Overall though, this was actually a pretty good week looking back - and if 97L doesn't develop tonight (and I think it won't), then the scare I got there (for no development besides Earl already-established Danielle) will be that and only that: a scare.

Cindy in 1999 ended up, despite appearing to be an outlier, being Danielle's best analog. Intensity was off, but that is still only one element to the forecast. The track and timing were almost perfectly executed (recurve area off by one degree but that's probably forgivable). Earl wasn't too bad either considering the low-confidence forecast, though becoming a hurricane today will hurt the grade a little. But like Danielle, track was pretty good, albeit a tad south of what I thought.

The next one will be a special. I will be going all the way through Labor Day itself to account for the long weekend. Also, as a word of warning for now, this new upcoming week could have a very alarming prediction. Disclaimers will most definitely be included for August 30-September 6.

Official evaluation in about six hours! You'll see the grade I give myself then.

-Andrew92

[Andrew92]

OK, let's evaluate....

Let me set this straight: this was not a perfect forecast. In both cases of Danielle and Earl, the intensities were off. I predicted Danielle would only become a C2 hurricane, and Earl to be a 50-60 mph tropical storm by the time this week was up. I underestimated both by a decent amount. There were other lesser discrepancies in both forecasts that I’ll bring up below.

Now, on with the good stuff, starting with Danielle! I’ve already mentioned how it became a C4 hurricane and that part was off, but let’s dissect the rest, with quotes from my prediction from last week. I said it would “continue west-northwest over the next three days” for Monday through Wednesday. That’s exactly what happened in terms of track those three days – to even the point that it was, by Tuesday “already at a latitude equal to the northernmost Virgin Islands.” I did predict a hurricane on Tuesday, but it came Monday. Still, the early fluctuations forgive that element a little bit. I then predicted Danielle to “become a C2 hurricane with maximum sustained winds of about 100-105 mph on Wednesday and Thursday” and “During this time period, the storm will also begin gradually curving toward the northwest, passing safely northeast of the Leeward Islands.” Could I have timed that part of the forecast any better? Wednesday was when it became a C2 hurricane, and on that day and Thursday it was moving generally northwest. Friday is when it got more intense than I had predicted. Also, I did predict the re-curve within 5 degrees of Longitude 55, and it occurred at 61, so I was one degree off, but since it was far away from land, I won’t dock myself too much there. Besides, I still predicted it would stay east of Bermuda. But then I really nailed it: “On Friday, Danielle will begin to weaken, and by week’s end, the storm will have weakened to a minimal C1 hurricane with maximum sustained winds of 75-80 mph, well south of Nova Scotia and Newfoundland and heading northeast away from land.” Friday is when peak intensity took place, and weakening began that evening. Now, Sunday night, the storm is heading rapidly northeast, positioned well south of Nova Scotia and Newfoundland, and its maximum sustained winds are 80 mph. I mean, I hate to brag too much, but to predict this so many days in advance…you can’t do much better than that that far out! Really, the intensity of C4 does hurt my overall grade for Danielle. However, when couple with an otherwise near-perfect forecast for this storm (the timing of becoming a hurricane and the recurvature point are minute mistakes when taken into comparison, all things considered), I have to give myself a B+ here (and would be an A if I had called for a C4 hurricane, and probably even an A- if I had picked a C3).

And how did I do on Earl? I predicted a slow development, taking place on Thursday, with a tropical storm forming on Friday. Well, both took place on Wednesday. That said, the slow development did continue from there, and I predicted as such, as it didn’t intensify again until Saturday. But I predicted that would happen on Sunday instead, so that too was 24 hours off, and as a result it was able to become a hurricane, and a C2 at that tonight. The track and timing of it, however, were excellently done. I predicted it would be a little northeast of the Leeward Islands, and it is in fact just approaching those islands, but the latitude itself is off by about two degrees or so – again, about a week in advance. A lot of forecasters would call that a pretty good forecast, though I did think it would be starting to head northwest by this time. It is, however, moving west-northwest and gradually bending right, so even that part wasn’t far off either. Overall, there were some mistakes with Earl’s forecast, but the track was near perfect. Considering my 45% confidence in this forecast and the slight disclaimer that came with it, this was a good forecast. I give myself a B here.

But Danielle and Earl were not the only ones being talked about. Several members on this board talked about a Gulf of Mexico disturbance that might develop. I gave it consideration, but also refuted that it would not develop, and that’s exactly what happened. I also predicted no other development elsewhere the rest of the week. Well, 97L sure gave that element a good scare, didn’t it?! But, as of right now, the end of the week, it has still not developed into a tropical depression yet. Therefore, on these two areas, I have no choice but to give myself an A.

So overall, two good forecasts on Danielle and Earl, and also accurately predicting no other developments the rest of the week. Sure, I was off on some areas of Danielle and Earl’s forecasts (*cough*INTENSITY*cough*), but there were too many elements that were spot-on nailed, even over five days in advance. When you throw all that with an accurate (and looking back, very bold!) prediction of no other developments and the week being a pretty tough one, my overall grade for this week is a B+.

LONG writeup for this upcoming week shortly.

-Andrew92

[Macrocane]

Congratulations one of your best forecast.

[Andrew92]

Thank you Macrocane!