Rapid Intensification "Study" (More Like Observations)

[brunota2003]

I would love some opinions from promets, but I believe I may of found something that shows when a hurricane is possibly about to go through a period of rapid intensification. Now, I do not have very many examples, but they all seem to be similar, and if it helps identify cases where a hurricane is about to explode, then I will be glad my time was well spent!

The first few examples I have are all from QuikSCAT, which as we know has sadly passed away...but I also found an example in microwave (MIMIC to be specific) imagery (with Hurricane Igor) and depending on what Hurricane Karl does over the next few hours, perhaps IR imagery as well (and both look very similar to the QuikSCAT images).

Here is how I will break this down, I will post an image that has not been edited, then will post a link to an edited version of the image showing what I am referring to...I am calling this the "Hook" effect (I suppose ). Forgive me if I misinterpreted the times on the QuikSCAT images...they can be a pain to figure out! I do know to subtract the hours down at the bottom of the images, so will do my best! I also know that a lot of these images are rain contaminated, I hope that does not detract from what I am trying to show!


Okay, this is an image from 2123 UTC (5:23 pm EDT) on the 18th of August 2009. This is Hurricane Bill over the Atlantic, during (or right before) a RI that was picked up by a NOAA Recon aircraft.



Same image as above, highlighting the "Hook" (used Paint, sorry if it is sucky! also, to keep the images down, I am only linking to the edited images)
http://img195.imageshack.us/img195/7643 ... 18hook.png

NHC Discussion at 1500 UTC (11 am EDT) Tues, Aug 18 (90 knots):
http://www.nhc.noaa.gov/archive/2009/al ... .013.shtml?
NHC Discussion at 2100 UTC (5 pm EDT) Tues, Aug 18 (95 knots):
http://www.nhc.noaa.gov/archive/2009/al ... .014.shtml?
Update from the NHC regarding the RI of Bill at 0030 UTC (8:30 pm EDT):
http://www.nhc.noaa.gov/archive/2009/al ... 0030.shtml?
NHC Discussion at 0300 UTC (11 pm EDT Aug 18) Wed, Aug 19 (110 knots):
http://www.nhc.noaa.gov/archive/2009/al ... .015.shtml?

Main page: http://www.nhc.noaa.gov/archive/2009/BILL.shtml?

Hurricane Bill strengthened from 110 mph at 5 pm EDT to 125 mph by 8:30 pm EDT. Between 5 pm and 11 pm, the pressure dropped from 962 millibars to 952 millibars.


***Due to updates, this one is no longer before the RI I mentioned, but is actually afterwords***
The next example is Hurricane Felicia in the Eastern Pacific (again using QuikSCAT).

This is an image of Hurricane Felicia at 14:36 UTC on August 5, 2009.



Same image as above, with the hook painted in
http://img651.imageshack.us/img651/8141 ... i1hook.png

NHC Discussion at 1500 UTC (8 am PDT) on Tue, Aug 4 (40 knots):
http://www.nhc.noaa.gov/archive/2009/ep ... .003.shtml?
Special NHC Discussion at 1700 UTC (10 am PDT) Tue, Aug 4 (60 knots):
http://www.nhc.noaa.gov/archive/2009/ep ... .004.shtml?
NHC Discussion at 2100 UTC (2 pm PDT) Tue, Aug 4 (70 knots):
http://www.nhc.noaa.gov/archive/2009/ep ... .005.shtml?
NHC Discussion at 0300 UTC (8 pm PDT Tue, Aug 4) Wed Aug 5 (85 knots):
http://www.nhc.noaa.gov/archive/2009/ep ... .006.shtml?

Main page: http://www.nhc.noaa.gov/archive/2009/FELICIA.shtml?

Hurricane Felicia strengthened from 80 mph at 2 pm PDT to 100 mph at 8 pm PDT. Between 2 pm and 8 pm, the pressure dropped from 985 millibars to 975 millibars. Please note that the image, due to QuikSCAT only taking so many in a day, was actually taken during a period of RI, versus before it...but I believe it still is worth adding on. (Due to updates...this images was taken the next day)
*** ***


The next system is Hurricane Fred, once again back in the Atlantic. I have two images for Hurricane Fred, and will post both.

The first image is from Sept 8 at 20:37 UTC.



Same as above, with the beginning of the hook painted
http://img186.imageshack.us/img186/448/fredri1hook.png

The second image is from Sept 9 at 07:42 UTC.


(Please see the painted image, apparently imageshack deleted the non-painted one)

Same as above, with the very sharp hook painted in (perhaps the sharper the hook, the longer/stronger the RI will be?)
http://img821.imageshack.us/img821/7291/fredri2hook.png

NHC Discussion at 1500 UTC (11 am EDT) Tue, Sept 8 (55 knots):
http://www.nhc.noaa.gov/archive/2009/al ... .004.shtml?
NHC Discussion at 2100 UTC (5 pm EDT) Tue, Sept 8 (60 knots):
http://www.nhc.noaa.gov/archive/2009/al ... .005.shtml?
NHC Discussion at 0300 UTC (11 pm EDT Tue, Sept 8) Wed, Sept 9 (65 knots):
http://www.nhc.noaa.gov/archive/2009/al ... .006.shtml?
NHC Discussion at 0900 UTC (5 am EDT) Wed, Sept 9 (90 knots):
http://www.nhc.noaa.gov/archive/2009/al ... .007.shtml?
NHC Discussion at 1500 UTC (11 am EDT) Wed, Sept 9 (105 knots):
http://www.nhc.noaa.gov/archive/2009/al ... .008.shtml?

Main page: http://www.nhc.noaa.gov/archive/2009/FRED.shtml?

The 12 hours between 11 pm EDT, Tue, Sept 8 and 11 am EDT, Wed, Sept 9 had an increase of wind speed from 75 mph to 120 mph, or 45 mph. The pressure also fell from 987 mb to 958 mb during those times as well.


This next one is from Hurricane Guillermo in the Eastern Pacific.

This is an image of Hurricane Guillermo on August 14, 2009 at 14:04 UTC.



Same image, with the hook painted in
http://img19.imageshack.us/img19/5628/g ... rihook.png

NHC Discussion at 0900 UTC (2 am PDT) Fri, Aug 14 (65 knots):
http://www.nhc.noaa.gov/archive/2009/ep ... .008.shtml?
NHC Discussion at 1500 UTC (8 am PDT) Fri, Aug 14 (70 knots):
http://www.nhc.noaa.gov/archive/2009/ep ... .009.shtml?
NHC Discussion at 2100 UTC (2 pm PDT) Fri, Aug 14 (85 knots):
http://www.nhc.noaa.gov/archive/2009/ep ... .010.shtml?
NHC Discussion at 0300 UTC (8 pm PDT Fri, Aug 14) Sat, Aug 15 (85 knots):
http://www.nhc.noaa.gov/archive/2009/ep ... .011.shtml?
NHC Discussion at 0530 UTC (10:30 pm PDT Fri, Aug 14) Sat, Aug 15 (100 knots):
http://www.nhc.noaa.gov/archive/2009/ep ... .012.shtml?

Main Page: http://www.nhc.noaa.gov/archive/2009/GUILLERMO.shtml?

Not even going to try to show wind speed increases and pressure decreases with this one...it is getting very late and will have to cut things short!


Now we move onto 2010, where Hurricane Igor went through a period of RI. This will be represented by a 24 hour loop from CIMSS' MIMIC. I will only paste the link to the loop here, please look in the SE quadrant of the hurricane for the "hook"...it is not as easy to spot as it is on QuikSCAT, though...but it appears around 0900 UTC (this is from Sept 12). I think it only lasts a couple hours, but I'm not sure that the hooks on QuikSCAT were present through the entire RI process, just the beginning.

http://cimss.ssec.wisc.edu/tropic/real- ... 2hr_05.gif

Main Igor MIMIC page:
http://cimss.ssec.wisc.edu/tropic/real- ... npage.html

NHC Discussion at 0900 UTC (5 am EDT - 70 knots):
http://www.nhc.noaa.gov/archive/2010/al ... .016.shtml?
NHC Discussion at 1500 UTC (11 am EDT - 90 knots):
http://www.nhc.noaa.gov/archive/2010/al ... .017.shtml?
Special NHC Discussion at 1830 UTC (1:30 pm EDT - 115 knots):
http://www.nhc.noaa.gov/archive/2010/al ... .018.shtml?

I think that last one should be either 2:30 pm EDT or 1730 UTC...since it is +4 from EDT...but wrote down exactly what the NHC Discussion reads.



I have not looked through Hurricane Julia, so I am not sure if she showed a hook or not. However, pending what happens tonight with Hurricane Karl, it appeared he also showed a hook, but this time on Infrared Satellite images. This is also a little difficult to show, but will try my best.

Image from 03:40 UTC Fri, Sept 17



Same image, painted with a somewhat of a "hook"
http://img337.imageshack.us/img337/7939/karlrihook.jpg

Okay...maybe once I get imageshack to work again, I can get the two images posted Edit: Woo, it worked!


In conclusion, those are just some things I noticed. I would of looked at prior years if I knew where old data was archived, but all I have are the images I saved and then happened to notice a "coincidence" in, and that all were prior to a RI episode. I would love feedback on this, it took a couple hours to prepare and type up!

[bvigal]

Wow! I'm impressed with the work you've done, even if I'm not a promet and thus can't speak to it's veracity in predicting RI!

[cyclonic chronic]

good work. u might be on to something. dont know if its true or not, but def. needs more scientific study to determine if u have the beginings of a reliable r.i. index/forecast.

[brunota2003]

I try on occasion Thanks you two, I wish I had more data than just those, especially since QuikSCAT is dead now, so I am not sure if IR/MIMIC would work out the same with identifying cases of RI several hours in advance. The thing that made me type it up is, if something similar is true, and RI could be predicted even just 6 to 12 hours in advance, there could be a lot of lives saved one day.

[curtadams]

What do similar strength storms that didn't RI look like? Do they lack hooks, or are hooks generally common in respectable TCs?

[brunota2003]

I do not have any QS images of storms of similar strength, but I do not recall seeing any that had such sharp hooks, most were more circular.

[Chacor]

Just FYI, the QUIKSCAT pass times are in purple above the text at the bottom. Nothing to do with the data buffer.

[brunota2003]

Oh...I thought you were supposed to use that to determine the time...so, what would be the date then? The date it was created? Unless it is one of the ones where the date created is like the 9th at 08 UTC, but the actual pass (in purple?) reads 21 UTC, so the date is really the 8th?

[Chacor]

Yes, the 21Z pass on an 08z image would refer to the previous 21z.

[JonathanBelles]

Hmmm. Interesting study. Personally it was a little hard to pick out your hooks even after looking at it being drawn in, but that being said, the theory could be right. I think that what might happen here is that there is a convective boost (Hot towers?) in the Rear Left Quadrant (RLQ), which is then brought around as it intensifies into the FRQ. MAYBE we can look to the RLQ for an explosion of convection to predict a RI.

Maybe you should try to turn this 20/20 vision in to a predictive measure. If you can use it to predict RI, then this theory can become practice. You should write a forecast or two when you see these 'hooks', and see if they end up rapidly intensifying.

I think you have a thesis, but now you have to test it. Good luck! I think you might be on to something.

[GCANE]

This is very interesting observations and great work.

I am not a pro-met, but have read a few papers.

If you don't mind, here is what I think may be happening.

I think the hook is a low-level jet of high theta-e air that could possibly be feeding a hot-tower.

Usually when an eye-wall hot-tower fires, RI is nearly guaranteed within 12 to 18 hrs, unless something drastically happens.

A recent paper I read suggests that the high theta-e air from the boundary layer is actually transported vertically by the hot-tower and deposited into the core.

The result is impulse-type of heating of the core which then kicks off the RI.

So, under the hot-tower, surface winds converge before they begin their ascent.

The SCAT seems to be showing an asymmetic infeed more like a jet rather than coming in from all angular directions.

This is very much like a tornado with an infeed jet.

That is why the hook signature with the sharp curve under the hot-tower.

[Ptarmigan]

I notice when before storms undergo rapid intensification, they tend to get more circular and often happens at night.

[brunota2003]

Okay...I went through and edited the times up above with the QS images, to reflect the real times that they are from (please feel free to correct me as necessary). There is also one more example from recently I would like to add, and it is an infrared image:



This is Tropical Storm Paula (soon to be Category 2 Hurricane Paula) at 00:15 UTC Oct 12, 2010. I am not going to highlight anything with this image, I think it is pretty visible.

0300 UTC Oct 12 (11 pm EDT) Advisory (60 knots):
http://www.nhc.noaa.gov/archive/2010/al ... .002.shtml?

0900 UTC Oct 12 (5 am EDT) Advisory (65 knots):
http://www.nhc.noaa.gov/archive/2010/al ... .003.shtml?

1500 UTC Oct 12 (11 am EDT) Advisory (65 knots):
http://www.nhc.noaa.gov/archive/2010/al ... .004.shtml?

1745 UTC Oct 12 (1245 pm EDT) Advisory (85 knots):
http://www.nhc.noaa.gov/archive/2010/al ... .005.shtml?

I was actually able to predict the RI after I saw that image. I stated (late on the 11th, after seeing that image) that I believed Paula was going to go through a phase of rapid intensification the morning of the 12th, and that it would be a Cat 2 or low Cat 3 by noon, but did not see it going beyond that (at least through RI) due to proximity to land. When recon flew into Paula, they found a Cat 2 with winds of 100 mph...and Paula did not strengthen beyond 100 mph.