ATL: KATIA - Post Tropical - Discussion

[USTropics]

192 nogaps is going to be interesting. I can't believe how incredibly differential both nogaps and gfs are in PP. That is quite and wide range of possibilities. Not going to take either seriously of course cause it is way too far out, but from what I understand of peoples opinion (and nhc) gfs is usually most accurate and is more renowned. But I could be all wrong here.

Well using 2010 and 2009 statistic here is a breakdown of model performance. The single best-performing model over the past two years has been the ECMWF. In fact, it out performed the NHC forecast for the 3-4 day range in 2010. The next best performing models were the GFS and GFDL models, followed by the UKMET. The NHC also has access to a now private model (the FSU MM5), which is believed to be the most accurate of the global models. The NHC also uses what is referred to as consensus models (the TCOA and the TVCA for example), which blends the four most reliable model outputs together to come up with a track.

[Hurricaneman]

It looks like if you're looking at 108hrs to 144hrs the UKMET is the best, but shorter range from 12 to 96 hrs the Euro is useful

[USTropics]

Here is a quick breakdown (and a poorly drawn image) of what we're talking about. I'll use the latest GFS run at 114 hours:

http://gbimg.org/p.php?q=9EFWO

As you see, up to the north over the Great Lakes/eastern Canada is our trough moving east. It is indicated by the 'U' type of shape and the low pressures accompanied with it (as well as the yellow/orange/green colors). To our right is our Atlantic ridge indicated by higher pressures (as well as the purple colors). Higher pressures act to block a tropical system from moving north, but as our trough moves towards the east, it will erode the ridge and allow for, in this case Katia, to be pulled north by the low pressures associated with the trough.

OOO okay, danke, that makes sence. So the ridge acts as a wall practically forcing the tropical system to move westwardly more so. The trough is it's weakness and causes it to weaken when near. Alright. I see that now, what in this forecast indicated this more westward movement models are leaning towards if that ridge is well, from your image, looks destined to weaken since it interacts with that trough.

Each model creates it's own upper level synoptics. A lot of variables differ from model to model, even run to run. That is why model accuracy is still sub-par at best (especially the further out in time you go). The strength of a trough, amplitude, timing, speed, and other variables are looked at. On top of that, the strength of the ridge is weighed in, and the strength, movement, and speed of a tropical system also must be added. Each model does different things with these variables, and therefor different outcomes.

[meriland23]

192 nogaps is going to be interesting. I can't believe how incredibly differential both nogaps and gfs are in PP. That is quite and wide range of possibilities. Not going to take either seriously of course cause it is way too far out, but from what I understand of peoples opinion (and nhc) gfs is usually most accurate and is more renowned. But I could be all wrong here.

Well using 2010 and 2009 statistic here is a breakdown of model performance. The single best-performing model over the past two years has been the ECMWF. In fact, it out performed the NHC forecast for the 3-4 day range in 2010. The next best performing models were the GFS and GFDL models, followed by the UKMET. The NHC also has access to a now private model (the FSU MM5), which is believed to be the most accurate of the global models. The NHC also uses what is referred to as consensus models (the TCOA and the TVCA for example), which blends the four most reliable model outputs together to come up with a track.

Interesting, I did hear great things about euro and their accuracy. I guess I will take their word for it before anyone else LOL. I don't remember seeing model runs of Euro today at all, missed it I think,I know one is coming up (weehoo). But anyway, you don't have to show me a loop or image, but what was Euro's 12z or 18z model run projection like? Just out of complete curiousity of course.

[USTropics]

192 nogaps is going to be interesting. I can't believe how incredibly differential both nogaps and gfs are in PP. That is quite and wide range of possibilities. Not going to take either seriously of course cause it is way too far out, but from what I understand of peoples opinion (and nhc) gfs is usually most accurate and is more renowned. But I could be all wrong here.

Well using 2010 and 2009 statistic here is a breakdown of model performance. The single best-performing model over the past two years has been the ECMWF. In fact, it out performed the NHC forecast for the 3-4 day range in 2010. The next best performing models were the GFS and GFDL models, followed by the UKMET. The NHC also has access to a now private model (the FSU MM5), which is believed to be the most accurate of the global models. The NHC also uses what is referred to as consensus models (the TCOA and the TVCA for example), which blends the four most reliable model outputs together to come up with a track.

Interesting, I did hear great things about euro and their accuracy. I guess I will take their word for it before anyone else LOL. I don't remember seeing model runs of Euro today at all, missed it I think,I know one is coming up (weehoo). But anyway, you don't have to show me a loop or image, but what was Euro's 12z or 18z model run projection like? Just out of complete curiousity of course.

Here is a link to the ECMWF model output. It only runs twice a day (00z and 12z), and they start running around 3:00am/3:00pm.

http://raleighwx.americanwx.com/models/ecmwf.html

[Wthrman13]

Incidentally, the reasons (probably) that the ECMWF model is superior (most of the time) to the GFS are due to the following:

ECMWF spends pretty much their entire focus on this one model, utilizes the state-of-the-art in data assimilation techniques, and has pushed their model to higher resolutions than the GFS. The U.S., on the other hand, has a more divide-and-conquer approach to numerical weather prediction, with many different specialized models focusing on different needs, of which the GFS is only one. As such, because of limited computational resources, they can't afford to improve the GFS as much as the ECMWF (though the technical and scientific know-how is certainly there!). I tend to think that both approaches can learn from each other, and I personally would like to see the U.S. move toward a more unified approach to NWP, without giving up more specialized configurations for regional and special-needs modeling (such as for hurricanes).

[meriland23]

latest loop of euro 0z

http://raleighwx.americanwx.com/models/ ... _loop.html

[USTropics]

Euro has shifted to the right, very similar to the GFS run with a direct hit on Bermuda. Slower though by about 24 hours (in regards to a Bermuda landfall compared to the GFS).

168 hrs


192 hrs

[meriland23]

latest loop of euro 0z

http://raleighwx.americanwx.com/models/ ... _loop.html

Wow, yeah, pattern is much more different than 12z, as you can see. It takes it SHARP to the east way fast.

[somethingfunny]

The NNE jog that spares Bermuda is caused by a 996mb low suddenly appearing where there was a 1027mb high just 24 hours earlier, off the coast of Nova Scotia. Take that 0z ECMWF run with a grain of salt...

[meriland23]

This is why you guys need to keep a eye out on this storm and not pish posh it so surely. This is a serious strm with an array of possibilities and enormous changes fraying from different miniscule changes. I just came across this run of the CNC at 00z today, and it is the first run I have seen that COLLIDES with the north east coast. Not to scare and not putting my eggs in one basket or telling you to at all, just saying, it is foolish to be so sure of something so uncertain.


http://raleighwx.americanwx.com/models/ ... MLoop.html

[WxGuy1]

The NNE jog that spares Bermuda is caused by a 996mb low suddenly appearing where there was a 1027mb high just 24 hours earlier, off the coast of Nova Scotia. Take that 0z ECMWF run with a grain of salt...

That surface low develops because of a strong short-wave trough that the ECMWF has progressing off the northeastern U.S. coast. In other words, I'm not sure the "cause" of Katia's forecast northward turn is because of that newly-developed low as much as it is because of the upper-level trough along the eastern U.S. coast (which itself is helping to develop that secondary low. The next result doesn't really change, though. Personally, it doesn't look very dubious to me, but I'm not going to put too much confidence is the presence of what's a pretty short wavelength trough in a 6-day forecast. Instead, I'll speak briefly on the situation at a little larger scale.

The 00z GFS (like previous runs of the GFS) also shows a trough (actually, a series of troughs that progress eastward) along the east coast of the U.S. -- Katia is seen on the far right side of the image below. Now, the GFS is a bit faster with Katia, but the point remains consistent:



I think the only chance Katia really has to make a run at the U.S. is if it stays as far south as possible (e.g. just N of Puerto Rico and/or Hispaniola) -- the closer it gets to the strong westerlies aloft, the better chance it has a recurving (or perhaps affecting the Canadian maritimes). In the 00z GFS forecast (above), even if it plows westward, it would experience some appreciable shear that would likely weaken the cyclone. There are occasionally some model runs that show Katia staying quite far south, where Katia is more removed from the effects of the troughs moving through the eastern U.S. and western Atlantic. Again, though, there may be issues with stronger wind shear at these low latitudes near the coastline.

Now, we're still looking 160-200+ hours into the future, and the models (CMC, GFS, and ECMWF) are handling the various troughs in the eastern U.S. and northwestern Atlantic differently. These details, however, are expected since we're looking at a forecast so far into the future. The signal remains that there will likely be series of troughs (most of the models show 2-3 shortwave troughs moving through the area in the area through the next 7-9 days) across the eastern U.S. coast.

[bexar]

remains at 65 KT for 5AM

[meriland23]

remains at 65 KT for 5AM

yup, only thing different is change in direction. Heading W now instead of NNW

[meriland23]

you can also see in comparison to 11 pm, the PP has curved slightly to the WSW around 60W..

[dwsqos2]

I am really sick of seeing stratocumulus/dry stable air all but surround virturally every tropical cyclone east of 60W this year.

I would just like to see a vaguely decent fish; I am not even sure that this will become a major hurricane in the short-term. Models tend to weaken upper troughs way too quickly (Or, at least that was a common problem in the past.)

The posts in this forums are NOT official forecast and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or storm2k.org. For official information, please refer to the NHC and NWS products.

[alhddar]

[alhddar]

KATIA.65kts.987mb.14.8N.43.6W.

[bexar]

I am really sick of seeing stratocumulus/dry stable air all but surround virturally every tropical cyclone east of 60W this year.

I would just like to see a vaguely decent fish; I am not even sure that this will become a major hurricane in the short-term. Models tend to weaken upper troughs way too quickly (Or, at least that was a common problem in the past.)

The posts in this forums are NOT official forecast and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or storm2k.org. For official information, please refer to the NHC and NWS products.

I am almost convinced that this season is not very friendly for CV development just like 2005. I also doubt this would reach MH status given the huge uncertainties between the models. Aside from Irene, so far this season has been a blah.

Personal Forecast Disclaimer:
The posts in this forum are NOT official forecast and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or storm2k.org. For official information, please refer to the NHC and NWS products.

[meriland23]

I am really sick of seeing stratocumulus/dry stable air all but surround virturally every tropical cyclone east of 60W this year.

I would just like to see a vaguely decent fish; I am not even sure that this will become a major hurricane in the short-term. Models tend to weaken upper troughs way too quickly (Or, at least that was a common problem in the past.)

The posts in this forums are NOT official forecast and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or storm2k.org. For official information, please refer to the NHC and NWS products.

I am almost convinced that this season is not very friendly for CV development just like 2005. I also doubt this would reach MH status given the huge uncertainties between the models. Aside from Irene, so far this season has been a blah.

Personal Forecast Disclaimer:
The posts in this forum are NOT official forecast and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or storm2k.org. For official information, please refer to the NHC and NWS products.

You are right about the uncertainties in the models. At first they were all in pretty good agreement that this storm was going to head on up to the flemish cap. The interesting thing is that the more time progresses, the more models skew anywhere between the FC and the florida panhandle. This is kind of suspicious considering you would naturally trust a newer model versus the one before it since more time has lapsed and they can get a better outlook. I think this season is very active and is going to proceed activity. Look at the gulf, the typhoon and now this. We might even have a Fuji on our hands which is a rare occurance when two cyclones connect and the less significant cyclone orbits around the bigger one. If that is not incredible activity for 2011, I do not know what is.

(note: I said possible fuji and not probable. Even so, I can not remember the last time two cyclones danced in these parts.)