Atlantic Tropical Action 2013

[dtk]

Thanks for starting to clear this up... and this seems to correspond well to the blog that I cited earlier today. It sounds like vortex bogusing in the GFS is rare and only typically occurs during the genesis of a TC, if at all. However, just investigating some of the text files from nomads, vortex relocation (while very minor) seems to happen quite frequently with nearly every model cycle if there are TCs. 

 

My apologies if I wasn't clear earlier.  Vortex relocation is the default action prior to assimilation for any/all TCs of at least TD strength as classified by the appropriate center (NHC, JTWC, etc.).  The way the process works, and I think there is a separate blog post somewhere based on information I passed to Gary Lackmann and Mike Brennan, is as follows:

 

We run a "cyclone tracker" algorithm on the 0-9hr forecast to attempt to "find" a TC in the model forecast.  This is based on several parameters (minimum SLP, vorticity at multiple levels, etc.).  The details aren't that important, but based on the output from the tracker algorithm,

a. The storm is found.  In this case, we remove the vortex from the "environmental flow" and relocate it to the observed/estimated position.  This is done prior to assimilation.

OR

b. The storm is not found by the tracker.  In this case, we generate observations for a bogus vortex and assimilate the winds.  In other words, we don't actually bogus a vortex, but instead assimilate synthetic winds along with all real observations.

 

Step b is pretty heavy handed, as it results in vortices that are typically "too coherent' relative to the observations.  This is especially true given that this typically happens during very weak, heavily tilted, and/or genesis events. 

 

For any active storm (TD and above) for which the responsible center is creating advisories, we apply either step A or B above (and never both).  There are rare occasions where the forecast TC does not require relocation (i.e. the forecast position is within a threshold distance of the observed position).

[phil882]

My apologies if I wasn't clear earlier.  Vortex relocation is the default action prior to assimilation for any/all TCs of at least TD strength as classified by the appropriate center (NHC, JTWC, etc.).  The way the process works, and I think there is a separate blog post somewhere based on information I passed to Gary Lackmann and Mike Brennan, is as follows:

 

We run a "cyclone tracker" algorithm on the 0-9hr forecast to attempt to "find" a TC in the model forecast.  This is based on several parameters (minimum SLP, vorticity at multiple levels, etc.).  The details aren't that important, but based on the output from the tracker algorithm,

a. The storm is found.  In this case, we remove the vortex from the "environmental flow" and relocate it to the observed/estimated position.  This is done prior to assimilation.

OR

b. The storm is not found by the tracker.  In this case, we generate observations for a bogus vortex and assimilate the winds.  In other words, we don't actually bogus a vortex, but instead assimilate synthetic winds along with all real observations.

 

Step b is pretty heavy handed, as it results in vortices that are typically "too coherent' relative to the observations.  This is especially true given that this typically happens during very weak, heavily tilted, and/or genesis events. 

 

For any active storm (TD and above) for which the responsible center is creating advisories, we apply either step A or B above (and never both).  There are rare occasions where the forecast TC does not require relocation (i.e. the forecast position is within a threshold distance of the observed position).

 

Awesome... thanks for the additional details! With a) when you remove the vortex and relocate it in the estimated position, you don't do any modifications to the intensity of that vortex correct? For example if the TC is estimated at 50 knots, but the GFS only shows a 30 knot circulation, you just reposition the 30 knot vortex to the correct position and leave the intensity as is. 

 

Just going through a few days of nomads data, it appears that the GFS did do a "bogusing" of the vortex for Chantal, but not for Dorian. 

 

Again as always, thanks for the extra info, as this is a nice treasure trove to refer back to when we actually have Atlantic TCs to observe

[Bryan Wood]

Awesome... thanks for the additional details! With a) when you remove the vortex and relocate it in the estimated position, you don't do any modifications to the intensity of that vortex correct? For example if the TC is estimated at 50 knots, but the GFS only shows a 30 knot circulation, you just reposition the 30 knot vortex to the correct position and leave the intensity as is. 

 

Just going through a few days of nomads data, it appears that the GFS did do a "bogusing" of the vortex for Chantal, but not for Dorian. 

 

Again as always, thanks for the extra info, as this is a nice treasure trove to refer back to when we actually have Atlantic TCs to observe

 

Yep. Chantal was bogused.

 

http://cimmse.wordpress.com/2013/07/12/chantal-gfs-bogusing-and-how-you-can-monitor-tc-initialization-with-the-gfs/

[Ed Lizard]

WMBas76.png

 

Just based on TPW, the dominant feature is Northeast of that, South of the DR.  However, it looks like the models are mosre enthusuastic about that circulation.

 

All talk of models aside, I see nothing suggesting more than a high end TS or minimal hurricane, and that is from most robust FIM model.  Most guidance is not even that enthusiastic, and usually reliable Euro still has a very weak and surpressed system.  Most 6Z GEFS are weak and far South.  Hard to argue with the 5 day 20% lemon at this point. With even genesis questionable, track and intensiity is, to use the phrase, a crap chute*.

 

 

*sic(k)

[dtk]

Awesome... thanks for the additional details! With a) when you remove the vortex and relocate it in the estimated position, you don't do any modifications to the intensity of that vortex correct? For example if the TC is estimated at 50 knots, but the GFS only shows a 30 knot circulation, you just reposition the 30 knot vortex to the correct position and leave the intensity as is. 

 

Just going through a few days of nomads data, it appears that the GFS did do a "bogusing" of the vortex for Chantal, but not for Dorian. 

 

Again as always, thanks for the extra info, as this is a nice treasure trove to refer back to when we actually have Atlantic TCs to observe

Correct, the relocation is designed as not to mess with intensity or size.  This helps mitigate spin-up/spin-down issues.

[Srain]

RECON has been tasked to investigate the potential tropical disturbance in the Western Caribbean Sea:

 

WEATHER RECONNAISSANCE FLIGHTS
CARCAH, NATIONAL HURRICANE CENTER, MIAMI, FL.
1015 AM EDT MON 12 AUGUST 2013
SUBJECT: TROPICAL CYCLONE PLAN OF THE DAY (TCPOD)
         VALID 13/1100Z TO 14/1100Z AUGUST 2013
         TCPOD NUMBER.....13-073

I.  ATLANTIC REQUIREMENTS
    1. SUSPECT AREA -- WESTERN CARIBBEAN
       FLIGHT ONE -- TEAL 70
       A. 13/2100Z
       B. AFXXX 01BBA INVEST
       C. 13/1645Z
       D. 12.0N 82.5W
       E. 13/2030Z TO 13/2300Z
       F. SFC TO 10,000 FT
    2. OUTLOOK FOR SUCCEEDING DAY.....NEGATIVE.

[Ed Lizard]

Life is unfair, entire state of Texas is yellows, oranges, reds and browns on the drought monitor, Florida is clear. 

 

 

ETA

 

 

GFS depiction at hour 120, ~30 knot SW at 250 mb over top, drier air just West of system as seen on PW, 700 mb and 500 mb RH, a low end TS doesn't seem radically bad, assuming system forms and tracks as GFS forecasts.  Looks like this GFS does focus in on PW surge seen on CIMSS product South of Hispaniola over small circulation North of Columbia

 

Again, may all be academic.  Shall wait to the ~19Z release of the 12Z Euro, if it still doesn't see a meaningful system, with GFS somewhat unstable in its solution between runs...

[Minnesota Meso]

For entertainment purposes only, the GEM once again when it sniffs something it develops it.

 

[GaWx]

 OTOH, the lowest Euro pressures for Irene were above 30N. For Irene above 30N, it had the following as the lowest for the 8/22/11 12Z through 8/26/11 0Z runs (in mb), respectively: 926, 936, 923, 927, 925, 921, 924, and 918. The lowest for 30N or higher verified to be 945 mb.

 

1. To clarify: The Euro studies I did in 2010 and 2011 were based entirely on Alan Huffmann's Euro lowest SLP output just in case other sources have shown different SLP's, which I assume may occur due to different resolutions.

 

2. To phlwx: no problem. I didn't take it negatively!

[Ed Lizard]

Euro consistency on very weak, Southerly and land impeded, can't go half full.  3/8ths on a Gulf TC 4-7 day range.  Models with a system, excluding Canadian, fairly weak, sheared and dry air issues, half again on it being more than a TS.

 

Euro is consistent, and the GFS alternates solutions. 

[Mike.Ventrice]

Euro consistency on very weak, Southerly and land impeded, can't go half full.  3/8ths on a Gulf TC 4-7 day range.  Models with a system, excluding Canadian, fairly weak, sheared and dry air issues, half again on it being more than a TS.

 

Euro is consistent, and the GFS alternates solutions. 

 

Something interesting to look at is how the European op handled Utor last Thursday and Friday for today. Not even a closed circulation... I'm thinking something is wrong with the European op with regards to TCG but I could be wrong....

[wxmx]

Don't forget that the Ukmet backs up the Euro, so it's not the only outlier. Very interesting considering that models start to differ markedly as soon as +48 hours.

[Mike.Ventrice]

is there anywhere we can see the Ukmet data free of charge? The pouch boys at the post-graduate navy school use to include the Ukmet in their pouch tracking diagnostics but these are now down

[Minnesota Meso]

is there anywhere we can see the Ukmet data free of charge? The pouch boys at the post-graduate navy school use to include the Ukmet in their pouch tracking diagnostics but these are now down

 

Try here, click on Ukmet and chose Northern Hem.  

 

http://meteocentre.com/models/models.php?mod=gemglb&map=na&run=00〈=en

 

Or here

 

http://wxcaster.com/conus_0012_foreign_models.htm

[Derecho!]

is there anywhere we can see the Ukmet data free of charge? The pouch boys at the post-graduate navy school use to include the Ukmet in their pouch tracking diagnostics but these are now down

 

 

 

In addtion to the already posted Meteocentre site, this is the UKMET Text tropical guidance (when they perceive the model has resolved a TC, gives lat-longs and an intensity):

 

http://www.metoffice.gov.uk/data/tropicalcyclone/wtnt80.txt

[wxmeddler]

is there anywhere we can see the Ukmet data free of charge? The pouch boys at the post-graduate navy school use to include the Ukmet in their pouch tracking diagnostics but these are now down

Like GRIB data or graphics of different parameters? What parameters are you looking for?

[Ed Lizard]

HPC disco favors Euro/UK, and new 5 day forecast followed 17Z coordination call with NHC.

 

 

[wxsmwhrms]

The Euro definitely has a defined vorticity max.  Because it never deepens it much, though, it just keeps it going W across the Yucatan and into MX.

[phil882]

The key differences between the GFS and the ECMWF are where they are keying in on the initial source of vorticity. The ECMWF is clearly focusing on the weak broad llc that is already located in the southwestern Caribbean as the main impulse. The GFS meanwhile has a much stronger low-level reflection back with the advancing tropical wave. Look at satellite imagery today, its clear the advancing tropical wave is a more substantial wind axis than the ECMWF is giving it credit for, so I'd actually lean towards the GFS on this set of guidance given the presence of more pronounced secondary vorticity maxima associated with this wave. 

 

[GaWx]

1. The current and forecasted MJO are within the circle per the GFS and Euro during the next five days. Whereas August MJO genesis stats back to 1995 suggest somewhat less than climo avg. TC genesis probabilities per day (10% per day for within circle vs. 14% for all MJO including within circle), there still were 25 TC's that formed during August while the MJO was within the circle since 1995. These include US H hitters Irene of 2011 and Katrina of 2005 as well as the Yucatan smasher Dean of 2007 fwiw.

 

2. The latest weekly Nino 3.4 SST anomaly dropped back slightly from -0.2 to -0.3. This only increases the chance that he ASO averaged 3.4 anomaly will be within the 0 to -1.0 range, which along with a cool August would suggest the likelihood of a major H hit on the US sometime later this season per stats back to 1950.

[Sunny and Warm]

1. The current and forecasted MJO are within the circle per the GFS and Euro during the next five days. Whereas August MJO genesis stats back to 1995 suggest somewhat less than climo avg. TC genesis probabilities per day (10% per day for within circle vs. 14% for all MJO including within circle), there still were 25 TC's that formed during August while the MJO was within the circle since 1995. These include US H hitters Irene of 2011 and Katrina of 2005 as well as the Yucatan smasher Dean of 2007 fwiw.

 

2. The latest weekly Nino 3.4 SST anomaly dropped back slightly from -0.2 to -0.3. This only increases the chance that he ASO averaged 3.4 anomaly will be within the 0 to -1.0 range, which along with a cool August would suggest the likelihood of a major H hit on the US sometime later this season per stats back to 1950.

While I don't necessarily disagree with your overall thoughts (you've provided a lot of good data the last few weeks), I'm not sure you can say a Cat 3 or greater hurricane is "likely" this year.  To me, likely is greater than 60% chance, and that's some serious odds on something occurring that hasn't in 7 years now.  Maybe I'm misreading your post.

[icebreaker5221]

Something interesting to look at is how the European op handled Utor last Thursday and Friday for today. Not even a closed circulation... I'm thinking something is wrong with the European op with regards to TCG but I could be wrong....

 

I don't think there's something wrong with it, it's just that genesis and intensity forecasts are inherently difficult.  Overall, the ECMWF has had a few misses, but the GFS/CMC/NAVGEM have been having too many false alarms. 

 

You're right that the ECMWF forecast for Utor was quite bad (much too weak), but I don't think it is a systematic bias.  A week ago the ECMWF had a run or two that kept Henriette a moderate to strong hurricane as it crossed south of Hawaii, while in reality it dissipated completely. 

[Derecho!]

You're right that the ECMWF forecast for Utor was quite bad (much too weak), but I don't think it is a systematic bias.  A week ago the ECMWF had a run or two that kept Henriette a moderate to strong hurricane as it crossed south of Hawaii, while in reality it dissipated completely. 

 

There are some ridiculous SST gradients near Hawaii (which is why you need a very precise W to the south then sudden 90 degree turn N path to strike anywhere in the Hawaiian Islands with a significant storm) so very slight errors in a track forecast can generate a significant intensity forecast error.

[wxmx]

is there anywhere we can see the Ukmet data free of charge? The pouch boys at the post-graduate navy school use to include the Ukmet in their pouch tracking diagnostics but these are now down

Not much options...and where there are, they run for either 3 days max or up to 7 days for 500dm heights. Fortunately we're in window now.

 

[wxmx]

There's a big area of low level vorticity in the SW Caribbean...there's also the low level vorticity associated to a TW currently moving in the C Caribbean. The latter will move in just N of the former, interacting. Which will be the dominant one? Euro and Ukie says the SW Caribbean piece of energy... GFS and CMC say the TW energy. Down the road, very different forecasts, with the european models not developing much and hugging the coast of C America and MX, the american models aiming at the Yucatan channel and the N GoM.

 

 

The key differences between the GFS and the ECMWF are where they are keying in on the initial source of vorticity. The ECMWF is clearly focusing on the weak broad llc that is already located in the southwestern Caribbean as the main impulse. The GFS meanwhile has a much stronger low-level reflection back with the advancing tropical wave. Look at satellite imagery today, its clear the advancing tropical wave is a more substantial wind axis than the ECMWF is giving it credit for, so I'd actually lean towards the GFS on this set of guidance given the presence of more pronounced secondary vorticity maxima associated with this wave. 

 

 

Yep, I agree about the key differences.

[Ed Lizard]

Yep, I agree about the key differences.

 

 

Just based on TPW, the dominant feature is Northeast of that, South of the DR.  However, it looks like the models are more enthusuastic about that circulation.

 

All talk of models aside, I see nothing suggesting more than a high end TS or minimal hurricane, and that is from most robust FIM model.  Most guidance is not even that enthusiastic, and usually reliable Euro still has a very weak and surpressed system.  Most 6Z GEFS are weak and far South.  Hard to argue with the 5 day 20% lemon at this point. With even genesis questionable, track and intensiity is, to use the phrase, a crap chute*.

 

 

*sic(k)

 

 

I think I agree.

[phil882]

Yep, I agree about the key differences.

 

I missed your post earlier Jorge. Good to see we are both on the same page .  What both models agree on is the rapid replacement of strong upper-level westerlies associated with an elongated TUTT with favorable upper-level easterlies as a upper-ridge and builds to take its place over the Western Caribbean. For the most part, latitude and land interaction seem to be the main reasons preventing TCG in the ECMWF and UKMET as there is healthy disturbance that appears to be getting its act together before a Central American landfall. The reason why the GFS has a TC is simply because the disturbance gains enough latitude to both remain over water and also start to feel the mid-level weakness between the Mexican and Bermuda ridge. 

[icebreaker5221]

There are some ridiculous SST gradients near Hawaii (which is why you need a very precise W to the south then sudden 90 degree turn N path to strike anywhere in the Hawaiian Islands with a significant storm) so very slight errors in a track forecast can generate a significant intensity forecast error.

 

That's probably part of the story but not the whole thing.  Henriette went slightly north of the Euro forecast for 24-36 h over cooler waters, but the track forecast was very good in the medium to long range.  The Euro showed noticeable weakening as it traversed cooler waters followed by rapid deepening once Henriette re-entered 27.5+ C waters (which it eventually did), whereas in reality there was continuous steady weakening, even after the tracks lined up again. 

 

Still makes the point that the bias is not consistently on the weak side.  If you'd like to discuss it further we should probably take this discussion to the E. Pac thread.

[thewxmann]

From my observations: the Euro tends to have a deep bias with large-size systems, weak bias with small-size systems. Of course there are exceptions to the rule (Utor, Henriette notable examples). 

[jordanwx]

Correct, the relocation is designed as not to mess with intensity or size.  This helps mitigate spin-up/spin-down issues.

Well...if you change the location of a storm you will change it's odds of forming. Guessing most of the adjustments are pretty minor though.