Disturbed weather in Eastern Atlantic (Is invest 95L)

[cycloneye]

The 15 minute updated images from Central Africa and West Africa.

West Africa. You can see almost the whole wave at the right bottom.



Central Africa. You can see the back end of wave to the left of image.

[Vortex]

Showtime next week!!! Hope everyone is rested up

[KWT]

I implore folks not to talk fish when the system hasn't even formed or come off the continent. We don't know how this will evolve and we all know that they are usually all fish in the early going.

Maybe its premature BUT the pattern is very heavily hinting at a recurve, so much so I'd say odds are rather slim it becomes a threat to anywhere other then Bermuda, even more so then normal given how far north it sets up and the fact the W.Atlantic ridging is at best average, which isn't really enough when you've got the current shortwave pattern.

[ConvergenceZone]

Showtime next week!!! Hope everyone is rested up

Yes, it will be nice. I'm not use to everything in the tropics being this dead in the middle of august...
It will be nice to be able to finally track a long tracker, whether it recurves or not...

[cycloneye]

That wave located near the African coast will clean the path for our wave we are watching. The moist enviroment is represented by the green color.

[ColinDelia]

The 1 standard deviation error for the official NHC forecast is about 260 nm. If you look at the chart for the errors for each forecast period from
0-120 hours it is roughly linear. Maybe we can estimate a 500 nm 1 SD error for 10 days. Obviously that is pretty ballpark. (They don't do 10 day forecasts,
the track error on systems which haven't formed must certainly be even worse. And just a note - this isn't an NHC forecast either. )

Now a fair amount of that might be along the track error but that can change whether it recurves or not too. I think it's fun to speculate but I don't think it's anything but that as far as whether it will recurve or not - if we even get something strong enough to become a named storm (which, admittedly certainly seems reasonable but is not a forgone conclusion). Then there's the enormous error in forecast intensity. What if the intensity is much weaker than shown? That's quite possible. I think we all acknowledge how unreliable intensity forecasts can be. If it does become a tropical cyclone but it happens 3 days later than models show, then what?

Anyone have some comments or links on 10 day computer model forecast accuracy as far as strength and position of Atlantic highs? (or anything similar which might she some light on that).

I'm not saying it won't be a fish. I wouldn't even have a clue on how to handicap it. As such I wouldn't even place a bet on this horse one way or another.

[ColinDelia]

This is all new to me but a quick look at the literature shows that the 500 hpa anomaly correlation is what is used as a measure of model forecast skill.

According to this paper
http://people.sc.fsu.edu/~inavon/pubs/krish.pdf

“An anomaly correlation of 0.6 is generally regarded as an indication of a useful forecast. This threshold value
comes from experience gained watching the forecast charts. A forecast with a skill greater than 0.6 generally
implies that troughs and ridges at 500 hPa are beginning to be properly placed in that forecast.”

Such data can be found here for the GFS, UKM, CAN, EC and a few others (2009)
http://www.emc.ncep.noaa.gov/gmb/STATS/ ... .2009.html

For the GFS (Northern Hemisphere) the cross-over point of .6 is at 7.5 days. At 7 days the anomaly correlation is .65. At 8 days it is .55. At 9 days it is .45.

I’m not claiming to know much about this. I just read this in the past hour. But it seems to me the at 7.5 days the forecast position of the troughs and ridges at 500 hPa start to become unreliable and that at 9 days they are considerably unreliable.

[KWT]

The mean position is indeed something I'd not give too much attention to outside of 7 days as it does tend to move about...however I'd say you can get a very good idea of the general trend at least till 240hrs and sometimes even further out on the GFS. For example the pattern I'm seeing right now will probably lead to enough shortwaves to be present to allow the system to be scooped up before it becomes a real threat...the pattern for something down the line IMO maybe suggestive of more of a threat.

Whenever anything sets up at 15N or above that far east and strengthens, unless you have a 2007 type ridge, its going to likely struggle to get across...just common sense really. The wildcard is whether or not the models maybe being a little overagressive with development, I think though the 12z ECM is right on the money.

[ColinDelia]

As a quick point of reference since we look at forecasts at these timescales:

96 hrs is 4.0 days (anomaly correlation > .9)
144 hrs is 6.0 days (anomaly correlation =~ .65)
192 hrs is 8.0 days (anomaly correlation =~ .55)
288 hrs is 12.0 days (throw darts with H's and L's on them at a map)

[ColinDelia]

Whenever anything sets up at 15N or above that far east and strengthens, unless you have a 2007 type ridge, its going to likely struggle to get across...just common sense really. The wildcard is whether or not the models maybe being a little overagressive with development, I think though the 12z ECM is right on the money.

I certainly agree with you there.

[somethingfunny]

You know how people say a certain city or region is "due" for a hurricane?

And then everyone tells those people that there's no such thing as "due".

Well, I'd say that at this point in the season with the SSTs, ENSO phase and MLSP levels the way they are, that the Atlantic is "due" for a major hurricane very soon.

But there's no such thing as "due".

[HURAKAN]

I just want something to track. I don't care if it's a fish!

[cycloneye]

I just want something to track. I don't care if it's a fish!

Same here. I have charts that I want to start filling with tracks.

[StormClouds63]

Any kind of fish will do ... I'm partial to catfish or red snapper. Seriously, the first CV system has to got to be about ready to make an appearance.

[KWT]

Yeah I'd just like a hurricane from this, would be nice to get a fish hurricane that gets quite strong, not sure the upper conditions are going to be quite condusive enough for a major but a hurricane is certainly on the board IMO from this...

Make the most of this fish...I think a low rider isn't far off...

[cycloneye]

It already shows up at 84 hours on the 18z GFS. Run 9 in a row from GFS showing it.

[Vortex]

18Z GFS is indicating low pressure developing within 72 hours about 600 miles wsw of Cape Verde...I think this has to do with the wave about to exit Africa this evening...

H+72

http://www.nco.ncep.noaa.gov/pmb/nwprod ... n_072l.gif

[KWT]

Comes off very far north indeed, probably aboutn 18N looking at that and the Vort map...quite a strong system already looking at the 18z run by that point.

[Ivanhater]

Coming off that far north is almost guaranteed to be a fish, but well see

[expat2carib]

I just want something to track. I don't care if it's a fish!

There must be a hidden sadist in me...LOL I love to see those systems struggle and die.