TS Nate

[HurricaneJosh]

Definitely improving a bit overnight. Not amazing-- the convection is still mostly S and W of the center-- but better than the complete wretchedness of yesterday afternoon:

[phil882]

The 4 am CDT package is out, and I have to say, this one is like watching paint dry!

No movement, no change in intensity, no change to the forecast or the reasoning behind it. Nada. (Well, there's one small detail of note: the satellite presentation is slowly improving.)

The latest track suggests a Cat-1 landfall between La Pesca and Tampico sometime Tuesday afternoon.

Actually, they lowered the intensity at Days 4 & 5 and added this caveat:

THIS...AND THE RELATIVE LACK OF CONVECTIVE ORGANIZATION...WOULD ARGUE AGAINST ANY SIGNIFICANT INTENSIFICATION IN THE SHORTTERM. LATER IN THE FORECAST PERIOD...SHIPS GUIDANCE STILLINDICATES THE PRESENCE OF A RELATIVELY DRY LOW- TO MID-LEVEL AIRMASS AND A SUDDEN INCREASE IN SOUTHWESTERLY SHEAR ASSOCIATED WITH ATROUGH OVER THE SOUTHEASTERN UNITED STATES. THIS SUGGESTS THAT THESTRONGER DYNAMICAL GUIDANCE...E.G. HWRF...MAY BE IN DOUBT.

Needless to say, liking the Phil-cast more than the [official] Forecast.

I going to go on a long rant about the shear mentioned in the intensity forecast that really bugs me.

The Latest SHIPS guidance shows the shear increasing to 22 knots at 84 hours. This was mentioned in the 4am CDT discussion.

SHIPS GUIDANCE STILLINDICATES THE PRESENCE OF A RELATIVELY DRY LOW- TO MID-LEVEL AIRMASS AND A SUDDEN INCREASE IN SOUTHWESTERLY SHEAR ASSOCIATED WITH ATROUGH OVER THE SOUTHEASTERN UNITED STATES. THIS SUGGESTS THAT THESTRONGER DYNAMICAL GUIDANCE...E.G. HWRF...MAY BE IN DOUBT. THEOFFICIAL NHC OFFICIAL IS SIMILAR TO THE PREVIOUS ONE AND PUTS MOREWEIGHT ON THE STATISTICAL INTENSITY MODEL OUTPUT.

So that's all well and good. However, I'm wondering if the dramatic increase in "Southwesterly shear" is actual related to the fact that the 00z GFS is significantly further south than the latest NHC forecast track. DeMeria et al. (2005) mentions that the SHIPS track is determined by the previous NHC forecast track adjusted for the new initial position. However, it is also mentioned that most of the atmospheric variables that SHIPS uses to calculate forecast intensity (including vertical wind shear) are directly derived from the 5 day GFS model fields.

This can lead to problems, namely the fact that the GFS forecast track of a tropical cyclone won't always lie along the same track as the NHC forecast track. This gets occasionally mentioned in a forecast discussion when the forecast feels that its prudent to say. This doesn't only affect vertical wind shear, but it also impact Relative Humidity, Sea Surface Temperatures... you name it!

This has not gone unnoticed. An AMS Conference Paper written by Rhome and Knabb (ams.confex.com/ams/pdfpapers/108729.pdf), explicitly show that there are certain instances where SHIPS preforms poorly. The first three points are critical.

Inaccurate forecasts by the SHIPS model can

generally be attributed to the following four sources: 1)

incorrect or incomplete input environmental data (e.g.,

shear, RH); 2) unrepresentative sampling of the

environmental data (i.e., chosen horizontal areas and/or

vertical levels/layers); 3) inaccurate track forecasts upon

which the SHIPS model relies; and 4) regression

equations that oversimplify or do not properly show the

true dependence of intensity on the environmental

characteristics in particular scenarios.

Now that we have a basis that SHIPS could potentially be inaccurate in cases where the GFS center is significant off the NHC forecast track, lets go back to the present and compare the shear on the 00z GFS highlighting the two different forecast center locations. Here I will highlight the interpolated 84 hour NHC center location in Blue, while the GFS center location is in red.

So wow, there are MAJOR differences in the shear depending on where you fix the center. In addition, it seems that the GFS is depicting the cyclonic circulation of Nate existing at or even above the 200mb level. This is common characteristic with vertically deep tropical cyclones. Note there is an anticyclone over the system, but it is not as well defined due to this circulation. Either way, there seem to be two rather glaring problems with the SHIPS shear analysis. First, the center location is significantly different from the GFS, with the SHIPS "seeing" very

quick easterly 850mb with lighter 200mb easterly flow superimposed on top. This is what is "creating" the southwesterly shear the model is seeing. Second, the cyclonic circulation with Nate extends up to 200mb and beyond, which is rendering the traditional shear products (from 850-200mb difference) rather useless. You often see this in CIMSS analysis as well when the upper level flow near a tropical cyclone starts to become cyclonic near the center.

So the end point of my rant here is that SHIPS in this particular scenario should be discounted due to the glaring differences in the center positions of the GFS and the NHC, along with the cyclonic circulation messing up the shear calculations. It should have been recognized that such a large position difference could cause problems in the derived SHIPS value. This is a great example where its worth looking directly at a forecast plot to see whats going on the synoptic scale rather than blindly following SHIPS guidance.

[phil882]

I'm also going to pretend that the 06z GFS never happened.

[HurricaneJosh]

I'm also going to pretend that the 06z GFS never happened.

How come?

[phil882]

How come?

Because its ridiculous... not that its got a 0% chance of verifying, but I mean seriously? It just completely re-wrote the tracks of Maria and Nate by taking them towards New Orleans and Miami respectively.

[HurricaneJosh]

Because its ridiculous... not that its got a 0% chance of verifying, but I mean seriously? It just completely re-wrote the tracks of Maria and Nate by taking them towards New Orleans and Miami respectively.

[Ed Lizard]

The 6Z GFS could be right, no rain for Texas. Suggestion of a little Fujiwara action, with its depiction of Nate briefly dropping back into the Gulf around Hour 192.

But I'll defer to the trained professionals.

[wxsmwhrms]

The Euro is slower/more easterly track at first, but looks like it will still hit Mexico south of Brownsville. However, it is awfully close to picking Nate up, just barely misses the trough which is a bit slower/further west. This one is a nail biter.....

A nail biter to say the least. I got up this morning and looked at the EC and ECMWF ensembles. They definitely looked to me like they were trending weaker with the trough over the southern CONUS (as predicted by Phil) which made me think it was time to start deferring to the Mexico track. Then I look at the GGEM, which is completely consistent on the MOB track, then at the 06Z GFS which has completely reversed course and shows a landfall near MOB!! You could see that the 00Z GFS Nate just barely got missed by the trough, and now the GFS has went to fully picking it up. Unbelievable...

[SnowGoose69]

GOM storms are very tricky, we tend to see huge errors on systems in their early stages of organizing, not always but its happened more often than the Atlantic.

[wxsmwhrms]

Because its ridiculous... not that its got a 0% chance of verifying, but I mean seriously? It just completely re-wrote the tracks of Maria and Nate by taking them towards New Orleans and Miami respectively.

Something else interesting...the experimental T254 GFS ensembles had made a dramatic trend yesterday going from nearly half of the members showing a movement to the NE on the 00Z run to none by the 18Z run. However, on the new 00Z run, nearly half are once again back to the NE, although several of those are left behind over the central Gulf when the trough lifts out and don't make it to the NC Gulf coast.

[wxsmwhrms]

Not surprisingly given the change in the GFS, the 06Z GFDL intensifies it to a cat 2/3 and takes in to Morgan City, LA in 5 days. The HWRF also intensifies it but stalls is at 26N, 92W.

[U_Thant]

Another interesting note...this 08/06Z run of the GFS looks an awful lot like the 07/00Z run of the ECM. I didn't buy that run, and I'm surely not buying this one.

[SnowGoose69]

I have to say it....even the CRAS model tracks Nate W or WSW into Mexico. I'm not sure I have ever seen it agree with most other guidance on anything.

[U_Thant]

Everybody 's to make forecasts for TC's that are trapped in a COL, right? Mitch, anyone?

[wxsmwhrms]

Well, one thing that will be interesting to see is with the GFS' defection, there is really not any model that shows what NHC is showing as far as a track, except maybe the UK. The EC is well right of what NHC has, getting the system up to almost 25N, 95W before turning it WSW. The GFS is obviously way to the right of that in the shorter term. Will be interesting to see how the play this in the 10 am advisory.

Anyway, looks like it's getting better organized. I would put the center around 20.1N, 92.5W, maybe starting to show a hair of a drift N now.

[dan88]

Recon was updated to add another flight overnight tonight, as an addition to this afternoon mission and the NOAA mission this evening.

[wxmx]

What an interesting turn of events. Now it's Euro/Ukie vs GFS/CMC/GFDL with the HWRF and 0zGFS EnKF in between (gets to around 27N before dipping SSW) ... *sigh*

[wxsmwhrms]

The 00Z GFS EnKF finally came in. Weird to say the least. Takes it south to almost the Mexico coast in the short term, then straight north to almost 30N, then at the end of the period turns it around and goes straight back S.

After looking closely at the data, one thing I am a little unsure of is whether the system will be far enough west to get driven into Mexico like the ECMWF shows. The 00Z EC trended about 1.5 to 2 degrees east of the 12Z run with the track of the system, and the only real ridging that is there after the trough moves out is the ridging to the west over Mexico. It will have to be far enough west to get under the influence of that to get into Mexico, or otherwise there is not really any ridging north of the system to turn it left. This is just based on the models - obviously, Phil has talked about the potential for the ridging to be stronger than what the models show, and that would change this. The EC has trended weaker with the troughing over the southern CONUS, but as Cheeznado said it's farther west, so if the system is just a bit farther east than what the EC shows, it could get involved with the trough.

Looking at the latest visible images, I think the system is about half a degree SE of what the models and NHC were showing, and in this sensitive of a setup, who knows if that could be an impact.

[wxmx]

The 00Z GFS EnKF finally came in. Weird to say the least. Takes it south to almost the Mexico coast in the short term, then straight north to almost 30N, then at the end of the period turns it around and goes straight back S.

After looking closely at the data, one thing I am a little unsure of is whether the system will be far enough west to get driven into Mexico like the ECMWF shows. The 00Z EC trended about 1.5 to 2 degrees east of the 12Z run with the track of the system, and the only real ridging that is there after the trough moves out is the ridging to the west over Mexico. It will have to be far enough west to get under the influence of that to get into Mexico, or otherwise there is not really any ridging north of the system to turn it left. This is just based on the models - obviously, Phil has talked about the potential for the ridging to be stronger than what the models show, and that would change this. The EC has trended weaker with the troughing over the southern CONUS, but as Cheeznado said it's farther west, so if the system is just a bit farther east than what the EC shows, it could get involved with the trough.

Looking at the latest visible images, I think the system is about half a degree SE of what the models and NHC were showing, and in this sensitive of a setup, who knows if that could be an impact.

I think most models had this slight SE drift before turning N, NW... at least I remember them with the GFS which almost had a LF in the Yucatan yesterday, the GFS EnKF, the HWRF, the GFDL and the Euro

[wxmx]

Yep, yesterday's 18z GFS, for example...maybe a bit extreme... the 0z and 6z had an E drift

[wxmx]

Nate looks pretty decent this morning...it's certainly > 40kts

[wxsmwhrms]

Yep, yesterday's 18z GFS, for example...maybe a bit extreme

Yep. I have to be honest, I was looking closer at the 06Z GFS, and thinking it had to be too fast in pulling the system out of the BOC, but to be honest it's not that much different in latitude than the ECMWF, and is even a hair slower in its northward motion. The difference is in longitude. At 00Z Sunday, the EC is at 24/94, the GFS is at 23.6/90.4.

[wxmx]

Yep. I have to be honest, I was looking closer at the 06Z GFS, and thinking it had to be too fast in pulling the system out of the BOC, but to be honest it's not that much different in latitude than the ECMWF, and is even a hair slower in its northward motion. The difference is in longitude. At 00Z Sunday, the EC is at 24/94, the GFS is at 23.6/90.4.

Yep, there's a big difference in longitude...the 06z GFS has a slightly deeper trough than the Euro. Also, looking at the 3hour increment Euro, it shows a SE drift to near 20.1N 92.1W which abruptly ends between 15Z and 18Z today (in a few minutes that range will start) and then moves NE/N/NW

[Srain]

12Z...

[wxmx]

12Z...

NHC will probably slow down the track and move it further north, in line with the TVC* and TC* consensus models

[Ivanhater]

Good discussion from the NHC on the latest model split

AFTER 36-48 HOURS...THE

MODEL SPREAD BECOMES INCREASINGLY LARGE...AS SUBTLE DIFFERENCES IN

THE STRENGTH AND ORIENTATION OF A MID/UPPER-LEVEL LOW AND TROUGH

EXTENDING FROM THE OHIO VALLEY SOUTHWESTWARD HAVE SIGNIFICANT

IMPLICATIONS ON THE FUTURE TRACK OF THE CYCLONE. THE LATEST GFS AND

GFDL MODELS...WHICH SHOW A SLIGHTLY DEEPER TROUGH...MOVE NATE

NORTHWARD TOWARD THE NORTHERN GULF COAST. THE ECMWF AND NOGAPS

MODELS DEPICT A WEAKER TROUGH THAT DOES NOT CAPTURE NATE AND ALLOWS

IT TURN TURN WESTWARD TOWARD MEXICO TO THE SOUTH OF A BUILDING

RIDGE...ALTHOUGH IT SHOULD BE NOTED THAT THERE IS CONSIDERABLE

SPREAD AMONG THE ECMWF ENSEMBLE MEMBERS.

[wxsmwhrms]

Good discussion from the NHC on the latest model split

AFTER 36-48 HOURS...THE

MODEL SPREAD BECOMES INCREASINGLY LARGE...AS SUBTLE DIFFERENCES IN

THE STRENGTH AND ORIENTATION OF A MID/UPPER-LEVEL LOW AND TROUGH

EXTENDING FROM THE OHIO VALLEY SOUTHWESTWARD HAVE SIGNIFICANT

IMPLICATIONS ON THE FUTURE TRACK OF THE CYCLONE. THE LATEST GFS AND

GFDL MODELS...WHICH SHOW A SLIGHTLY DEEPER TROUGH...MOVE NATE

NORTHWARD TOWARD THE NORTHERN GULF COAST. THE ECMWF AND NOGAPS

MODELS DEPICT A WEAKER TROUGH THAT DOES NOT CAPTURE NATE AND ALLOWS

IT TURN TURN WESTWARD TOWARD MEXICO TO THE SOUTH OF A BUILDING

RIDGE...ALTHOUGH IT SHOULD BE NOTED THAT THERE IS CONSIDERABLE

SPREAD AMONG THE ECMWF ENSEMBLE MEMBERS.

Yep, excellent discussion. Kind of disturbing, though, that the NOGAPS is the model that is supporting the ECMWF.

[wxmx]

It comes down to the strength of the trough. Looking at most of the guidance available, I'm more certain than yesterday that Nate will miss the trough...probably 70-30. One of the teleconnections that Phil uses, and so do I, is the Pacific, and it is in full Niña mode right now, with the SOI averaging a healthy +10 since August 26th... this would tend to enhance heights in the SE, hence preventing the trough to dig deeper. Also, the nowcast shows a farther S Nate, more closely to what yesterday's GFS showed, than what it showed at 06z. Even the Ukie will miss the forecast farther to the North a good deal in the very short term. Maybe slight differences, but that maybe all what it takes in this "on the edge" scenario.

[am19psu]

It comes down to the strength of the trough. Looking at most of the guidance available, I'm more certain than yesterday that Nate will miss the trough...probably 70-30. One of the teleconnections that Phil uses, and so do I, is the Pacific, and it is in full Niña mode right now, with the SOI averaging a healthy +10 since August 26th... this would tend to enhance heights in the SE, hence preventing the trough to dig deeper. Also, the nowcast shows a farther S Nate, more closely to what yesterday's GFS showed, than what it showed at 06z. Even the Ukie will miss the forecast farther to the North a good deal in the very short term. Maybe slight differences, but that maybe all what it takes in this "on the edge" scenario.

I was kinda surprised that NHC bit off on the 6z GFS based guidance so hard. I'd have thought they would wait to see what the foreign centers came out with at 12z.

[wxsmwhrms]

It comes down to the strength of the trough. Looking at most of the guidance available, I'm more certain than yesterday that Nate will miss the trough...probably 70-30. One of the teleconnections that Phil uses, and so do I, is the Pacific, and it is in full Niña mode right now, with the SOI averaging a healthy +10 since August 26th... this would tend to enhance heights in the SE, hence preventing the trough to dig deeper. Also, the nowcast shows a farther S Nate, more closely to what yesterday's GFS showed, than what it showed at 06z. Even the Ukie will miss the forecast farther to the North a good deal in the very short term. Maybe slight differences, but that maybe all what it takes in this "on the edge" scenario.

Excellent discussion, I agree with you except for one thing, I also think the strength of the ridge to the east of the system is playing a key role here. The stronger that ridge gets, the more poleward motion the system can gain in the shorter term. Along with trending the trough weaker, it seems like the positive SOI you and Phil have been discussing is also trending that ridge stronger, potentially resulting in more poleward motion early on.

I am 50/50 right now, but that is a trend down from 60/40 on the northward motion yesterday. The main thing that keeps me concerned is the amount of latitude the ECMWF gains in the shorter term. If the models were just showing it staying down in the BOC and eventually guided west, I could buy that pretty easily. The problem is all of them except the UK and NOGAPS show it making nearly 25N, and I think if it gets that far north it is going to be very vulnerable to being picked up, even by a weaker trough. I think it needs to stay S of 23/24N to be sure of staying on the more westward solution.