Guest Post: The science of Sandy's transition

By: Angela Fritz , 9:36 PM GMT on November 05, 2012

Share this Blog
12
+

The following is a guest blog from Lee Grenci, a senior lecturer and senior forecaster at the Department of Meteorology at Penn State. Lee will be blogging with Wunderground more often beginning in December, and we're looking forward to having him on board. --Angela

The debate about whether Sandy was a hurricane or hybrid system at landfall has generated much discussion, particularly as it relates to the National Hurricane Center passing the responsibility for issuing warnings to local offices of the National Weather Service. I cannot accept the notion that Sandy was purebred hurricane at landfall. Indeed, I'm convinced that Sandy was in the process of transitioning to a post-tropical low-pressure system as it approached landfall (just as NHC had anticipated). Let's investigate.

Looks Like a Duck, but…

As Sandy bore down on the New Jersey Coast late Monday afternoon (October 29, 2012), I looked at the 2047Z AMSU-A cross section through the storm (Advanced Microwave Sounding Unit; read more). This cross section (below) displays the temperature difference (in Kelvins) between the core of the storm and the storm's environment at each pressure altitude from 1000 mb to 100 mb. Right off the bat, you might jump to the conclusion that Sandy was a purebred tropical system...warm core from the earth's surface all the way to the lower stratosphere.


The AMSU-A cross section through Sandy at 2047Z on October 29, 2012. At first glance, it looks as though Sandy was a warm-core and purebred tropical system. As you will soon discover, Sandy was not warm core and was on the cusp of transitioning to a post-tropical low-pressure system. Credit: Courtesy of the Cooperative Institute for Meteorological Satellite Studies.

As it turned out, what looked like a duck and quacked like a duck was really not a duck. In other words, we need to "read between the lines" on the AMSU-A cross section. But I'm getting a little ahead of myself on the details of this cross section, so I'll get down to brass tacks and show that Sandy was in the process of transitioning to a post-tropical storm as it bore down on the New Jersey Coast.

Transitioning to Post-Tropical

Once Sandy completed its sharp turn toward the west off the Middle Atlantic Seaboard, the hurricane started to interact with the surrounding baroclinic environment. To see what I mean, check out (below) the 00Z NAM model analysis on October 30 (8 P.M. EDT on October 29) of 700-mb heights (in meters), 700-mb wind barbs, and 700-mb temperatures (color-filled in degrees Celsius). Here's a larger image to aid in the analysis. I annotated a portion of Sandy's circulation (on the image below) where cold-air advection was taking place (yes, there was cold advection produced by southerly winds).


The 00Z NAM model analysis of 700-mb heights (in meters), 700-mb wind barbs, and 700-mb temperatures (color-filled in degrees Celsius) on October 30 (8 P.M. EDT on October 29). The white circle marks a portion of Sandy's circulation where cold-air advection was occurring. Credit: Penn State University.

The bottom line here is that cold air had already begun to wrap into the circulation of Sandy, indicating the storm's transition to a post-tropical low-pressure system.

This pattern of cold advection also prevailed at 850 mb (00Z NAM model analysis of 850-mb heights, wind barbs, and temperatures). Also, please note the warm-air advection to the north of Sandy. I circled a portion of Sandy's circulation where there was warm advection on the annotated version above (yes, there was warm advection produced by northeasterly winds).

To gain another perspective on Sandy's evolving post-tropical circulation, focus your attention on the 850-mb temperatures measured by radiosondes at 00Z (station plots below). Note the rather large temperature gradient south of Sandy's center (3 degrees Celsius at Wallops Island, -3 degrees at Roanoke / Blacksburg, and 2 degrees Celsius at Washington's Dulles International Airport compared to 15 degrees Celsius near the core of Sandy just off the New Jersey Coast (Sandy's 850-mb ob is not shown on the station plots; see 2355Z aircraft reconnaissance). Moreover, 850-mb temperatures near 0 degrees Celsius so close to the center of Sandy is hardly compatible with a purely tropical system. Without reservation, Sandy was transitioning to a post-tropical low-pressure system (consistent with NHC's forecast discussion issued at 5 P.M. on October 27).


The 700-mb station plots based on data measured by radiosondes at 00Z on October 30, 2012. Larger image. Credit: Penn State University

Okay, with the transition to a post-tropical storm taking place as Sandy approached New Jersey, let's tackle the misleading message sent by 2047Z AMSU-A cross section on October 29. At the time, it sure looked like Sandy was warm core (and tropical) as it bore down on the New Jersey Coast. As it turned out, I didn't read between the lines.

Tricky AMSU-A Cross Section

Having convinced myself that Sandy was not a hurricane at landfall (it was more of a hybrid during its transition to a post-tropical low), I was faced with resolving what I was apparently missing on the 2047Z AMSU-A cross section. What was I missing on the AMSU-A cross section?

In a nutshell, the vertical and horizontal resolution of the AMSU instrument is pretty coarse, with several kilometers between weighting functions in the vertical and 50 kilometers at nadir in the horizontal. With regard to the vertical resolution, any cooling in the upper troposphere associated with the approach of the negatively tilted short-wave trough "might have fallen through the cracks" on the AMSU image (as a reminder of the 500-mb short wave that played a pivotal role in Sandy's evolution, check out the 18Z NAM model analysis of 500-mb heights and absolute vorticity on October 29).

With these "cracks" in mind, I turned to the Special Sensor Microwave Imager / Sounder (read more about the SSMIS), whose weighting functions are slightly different from the AMSU instrument and whose horizontal resolution is higher. Perusing the SSMIS data at the 1333Z data on October 29, we can see the footprints of Sandy's warm core on Channel 3 (350 mb) and Channel 4 (250 mb). For geographical reference, note that the East Coast of the United States is outlined in white, so both channels contain upper tropospheric data over the eastern United States and offshore Atlantic waters.

There's a different story on Channel 5 (approximately 150 mb; see image below). Indeed, the footprint of Sandy's warm core isn't apparent at 150 mb (there aren't any closed, "warm isotherms" in the vicinity of the storm). That's probably because of wind shear (15Z analysis of deep-layer shear) associated with the approach of the short-wave trough. The point I'm trying to make here is that as early as 1333Z on October 29, the footprint of Sandy's warm core had, at the very least, weakened at 150 mb (roughly 14000 meters). In other words, the first dominos were falling, setting into motion Sandy's gradual transition from tropical to post-tropical.


SSMIS Channel 5 data (at approximately 150 mb) over the eastern United States and offshore Atlantic waters at 1333Z on October 29, 2012 (the East Coast is outlined in white). Note that Sandy's warm core wasn't apparent at 150 mb at this time, although 150-mb temperatures indicated that there was still relatively warm air over the region. Credit: Courtesy of the Cooperative Institute for Meteorological Satellite Studies

Although the warm core of Sandy isn't apparent on the SSMIS Channel 5 image, it was still relatively warm air at 150 mb over the eastern United States and offshore Atlantic waters at this time. If my theory is correct (that Sandy was starting to transition to a post-tropical low), how is that possible?

Well, heights were falling as the short-wave trough approached. To see what I mean, check out the 12Z station models at 500 mb on October 29 and compare them to the 00Z station models at 500 mb on October 30. Note that 500-mb heights fell 210 meters at Wallops Island, Virginia, and 170 meters at Upton, New York during the 12-hour period (both upper-air stations are circled in white).


The 12Z NAM model sounding over Sandy's Center at 12Z on October 29, 2012. Credit: NOAA's Air Resources Laboratory.

What were the ramifications of these height falls? To answer this question, check out, above, the NAM model sounding at 38.4 degrees North and 74.4 degrees West at 12Z on October 29 (the lat-long coordinates of Sandy's center at this time). Clearly, with falling heights associated with the short-wave trough,150 mb now lay in the lower stratosphere (the tropopause began around 200 mb on this NAM sounding). So the footprint of relatively warm stratospheric air was present above 200 mb mb on the AMSU-A cross section (150 mb, 100 mb, etc.), lending the false impression that Sandy was purely warm core from the sea surface to 100 mb.

I should point out that, in the deep Tropics, 150 mb typically lies in the upper troposphere, a fact that might confuse any purely tropical interpretation of the AMSU-A cross section. In reality, the mere fact that the tropopause over Sandy lay at 200 mb at 12Z on October 29 speaks volumes about the transitioning character of the storm (not to mention the developing baroclinic circulation I talked about earlier on this page). Eventually, Sandy's warm core at lower altitudes eventually weakened and dissipated as the transition to a post-tropical storm was complete.

You might argue that I'm talking about nuances here, but, in my opinion, Sandy was not a hurricane (not purely tropical) at the time of landfall. Rather, it was in the process of transitioning to a post-tropical low-pressure system. Yes, to be fair, Sandy was not completely post-tropical either. Needless to say, the true nature of Sandy at the time of landfall will be debated in the scientific community. One thing's for certain...it was one of the most atypical storms that I can remember in my 35+ years as a forecaster.

Many thanks to Dr. Derrick Herndon (University of Wisconsin) for his input.

Reader Comments

Comments will take a few seconds to appear.

Post Your Comments

Please sign in to post comments.

or Join

Not only will you be able to leave comments on this blog, but you'll also have the ability to upload and share your photos in our Wunder Photos section.

Display: 0, 50, 100, 200 Sort: Newest First - Order Posted

Viewing: 34 - 1

Page: 1 — Blog Index

34. drosihn
9:21 PM GMT on November 12, 2012
FWIW, I think it was called "Frankenstorm" because it was arriving on Halloween...
Member Since: January 18, 2002 Posts: 0 Comments: 0
33. georgevandenberghe
2:29 PM GMT on November 08, 2012

Posted the following on another blog in response to hyperbolic
wording about the strangeness and unprecedented energetics
of Sandy being outside of current theroretical knowledge. It isn't
although it was certainly a rare event. By the way living in the DC
area I consider myself very lucky our Sandy impact wasn't much worse.
Perhaps the June Derecho cleaned out a lot of trees that would
otherwise have fallen with Sandy.
For what it's worth, I dislike terms like "Frankenstorm" which suggest an odd event outside of science.


Would like to see other comments on this, particularly from a more recently or deeply educated dynamic meteorologist (My MS is 25 years old and I've not used it thoroughly so I'm rusty)

********** ORIGINAL POST 10/26 *************

Most hurricanes have a length scale of a few hundred kilometers.
Eddies this size in a baroclinic region damp out because of
stratification (the circulations created by the geostrophic
adjustments are thermally indirect and convert eddy kinetic
energy to potential energy intensifying the temperature gradient
and killing the eddy.. This energy is eventually released in a
larger scale more classic baroclinic disturbance some time later.
)

If stratification is less or the disturbance is larger though, the
eddy can extract energy from the baroclinicity and grow. This
is common with large areal extent hurricanes that encounter
temperature gradients and transition into large baroclinic waves, which are generally intense (maintain gale strength). The tropical cyclone does not have
to be as large in low stratification (weak static stability) situations.

The energetics of hybrid storms is beyond my current knowledge.
Member Since: February 1, 2012 Posts: 17 Comments: 1620
32. Barefootontherocks
9:55 PM GMT on November 07, 2012
Quoting 24hourprof:
Please ignore this post. Is there a way to delete it?
Quoting 24hourprof:
Please ignore this post...I must be doing something incorrectly because when I modify a post, I seem to get a new one that's exactly like the original. Oh well. I'll learn eventually.
Hi Lee. In case you have not got this answered yet... When you modify comment, you have to leave at least a single character there. A comment can be removed, as in gone, by the blog owner. Since you are in Angela's blog, you cannot remove one here.

If you are adding to or changing a comment, and not removing it, be sure and click the "submit" option when you finish with the modify comment box, and the changes should show up.

PS
When you modify a blog entry, you have to click "Update entry" or the changes will not take.
Member Since: April 29, 2006 Posts: 151 Comments: 18360
31. WunderAlertBot (Admin)
9:09 PM GMT on November 07, 2012
angelafritz has created a new entry.
30. 24hourprof
7:31 PM GMT on November 07, 2012
Please ignore this post...I must be doing something incorrectly because when I modify a post, I seem to get a new one that's exactly like the original. Oh well. I'll learn eventually.
Member Since: October 24, 2012 Posts: 90 Comments: 798
29. 24hourprof
7:26 PM GMT on November 07, 2012
Quoting gippgig:
#26: I wasn't asking for a full explanation, just a figure legend (blue=pressure (obvious), red=temp (I think), green= (I don't know),...).


Here you go:

Annotated 12Z skew-T this morning
Member Since: October 24, 2012 Posts: 90 Comments: 798
28. gippgig
6:36 PM GMT on November 07, 2012
#26: I wasn't asking for a full explanation, just a figure legend (blue=pressure (obvious), red=temp (I think), green= (I don't know),...).
Member Since: December 5, 2005 Posts: 0 Comments: 64
27. 24hourprof
2:19 PM GMT on November 07, 2012
Please ignore this post. Is there a way to delete it?
Member Since: October 24, 2012 Posts: 90 Comments: 798
26. 24hourprof
2:18 PM GMT on November 07, 2012
Quoting gippgig:
The final figure should indicate what the various numbers represent; it's confusing as shown.


This is a skew-T, a basic tool that meteorologists use every day. I didn't want to explain it because, to do so, is a blog (or blogS) in and of itself. Those are the kind of issues I can tackle in the future, but it would have been way off base to take the time to explain what amounts to a basic, every-day tool that weather forecasters use.
Member Since: October 24, 2012 Posts: 90 Comments: 798
25. 24hourprof
2:13 PM GMT on November 07, 2012
Quoting whitewabit:


Lee, I am assuming you will soon have your own blog ?


Yes, eventually. A big thanks to Angela. She's gone the extra mile for me.
Member Since: October 24, 2012 Posts: 90 Comments: 798
24. 24hourprof
2:08 PM GMT on November 07, 2012
Quoting carlakid50:
Lee very interesting even though most of it went over my head. Hopefully; I wiil continue to learn with your new blog as I have with Dr. Masters as well as all those who post. The question of storm classification has come up frequently on the upper Texas gulf coast especially with the occasional storm that pops up at the last minute in the gulf, but without the particular cold air front aspect. I will be curious to see if we have more of the hybrid storms forming in the coming seasons. Thank you.


Hopefully, my mini-lectures will help you to learn. Many thanks for your kind words.
Member Since: October 24, 2012 Posts: 90 Comments: 798
23. 24hourprof
2:07 PM GMT on November 07, 2012
Quoting cicadaknot:
It's great to see you again Lee! I took your Meteo 101/361 classes almost 10 years ago and i keep telling everyone how great those classes were. I learned more in those two courses than anything I've taken since. I look forward to your unique writing style and informative pieces in the future. Everyone here is in for a treat!


You're too nice. Much appreciated, though.
Member Since: October 24, 2012 Posts: 90 Comments: 798
22. carlakid50
3:45 AM GMT on November 07, 2012
Lee very interesting even though most of it went over my head. Hopefully; I wiil continue to learn with your new blog as I have with Dr. Masters as well as all those who post. The question of storm classification has come up frequently on the upper Texas gulf coast especially with the occasional storm that pops up at the last minute in the gulf, but without the particular cold air front aspect. I will be curious to see if we have more of the hybrid storms forming in the coming seasons. Thank you.
Member Since: September 24, 2008 Posts: 0 Comments: 20
21. cicadaknot
1:33 AM GMT on November 07, 2012
It's great to see you again Lee! I took your Meteo 101/361 classes almost 10 years ago and i keep telling everyone how great those classes were. I learned more in those two courses than anything I've taken since. I look forward to your unique writing style and informative pieces in the future. Everyone here is in for a treat!
Member Since: July 17, 2004 Posts: 0 Comments: 74
20. 24hourprof
12:19 AM GMT on November 07, 2012
Quoting gippgig:
The mini-lectures idea sounds great - I hope Jeff will mention in his blog when they appear.
Clicking on the 700-mb larger image gives a DIFFERENT image (black is -25 rather than -10, the heights are 5xxx rather than 2xxx, etc.)
The caption of the Oct. 29 AMSU-A cross section reads Oct. 20 and the caption of the 850-mb station plots reads 700-mb.
The final figure should indicate what the various numbers represent; it's confusing as shown.


I wish I could fix these, but Angela graciously put me on her blog. I should have done a better job proofing it, but I was in a hurry to get it posted.

Thanks everyone for the welcome and being so gracious.

Try this version on my Web site:
Link

Better?
Member Since: October 24, 2012 Posts: 90 Comments: 798
19. gippgig
10:30 PM GMT on November 06, 2012
The weighting functions and 50 km at nadir links also need to be fixed.
Member Since: December 5, 2005 Posts: 0 Comments: 64
18. whitewabit (Mod)
10:21 PM GMT on November 06, 2012
Quoting 24hourprof:
Many thanks.

Everybody...please just call me Lee.


Lee, I am assuming you will soon have your own blog ?
Member Since: August 17, 2005 Posts: 361 Comments: 31207
17. gippgig
10:10 PM GMT on November 06, 2012
The mini-lectures idea sounds great - I hope Jeff will mention in his blog when they appear.
Clicking on the 700-mb larger image gives a DIFFERENT image (black is -25 rather than -10, the heights are 5xxx rather than 2xxx, etc.)
The caption of the Oct. 29 AMSU-A cross section reads Oct. 20 and the caption of the 850-mb station plots reads 700-mb.
The final figure should indicate what the various numbers represent; it's confusing as shown.
Member Since: December 5, 2005 Posts: 0 Comments: 64
16. TampaCat5
10:07 PM GMT on November 06, 2012
Agreed, wonderful technical analysis. Well written and put some of the harder-to-understand weather products (and concepts!) in context for me. Thanks, Lee.
Member Since: June 11, 2006 Posts: 4 Comments: 445
15. seminolesfan
8:09 PM GMT on November 06, 2012
Great write up and analysis!

Cheers...
Member Since: June 14, 2006 Posts: 0 Comments: 2108
14. fireitup
7:07 PM GMT on November 06, 2012
Fantastic post, Lee! Any sort of mini lectures you would be willing to offer would be appreciated by me and many others I am sure.

Looking forward to future readings.
Member Since: April 20, 2009 Posts: 0 Comments: 23
13. Mousseline545
4:46 PM GMT on November 06, 2012
Salut !

Je ne suis pas très forte en anglais, c'est pour cela que je vous envoie ce message en français.

J'aime beaucoup venir ici où je fais tout pour essayer de tout comprendre de moi-même.
Ce qui fait que de temps en temps, vous parlez de choses importantes qui intéresseraient certainement des internautes de France dont des américains, mais comme je ne veux pas faire trop d'erreurs, je préfère vous écrire en français.
Je vais faire connaitre votre blog, mais pour l'instant, j'essaie d'en profiter jalousement.
En tout, vous êtes très forts. MERCI A TOUTE L'EQUIPE.

De votre compréhension, je vous remercie beaucoup. Merci pour tout.
Member Since: June 8, 2012 Posts: 0 Comments: 0
12. 24hourprof
1:22 PM GMT on November 06, 2012
By the way, I'm retired (yes, I'm an old guy) now from Penn State after 28 years of teaching. I figured I'd keep on teaching and Weather Underground graciously offered me the opportunity.
Member Since: October 24, 2012 Posts: 90 Comments: 798
11. 24hourprof
1:19 PM GMT on November 06, 2012
Quoting AySz88:
FYI, some of the "bigger image" links are pointing at wunderground.com/worldofweather/... instead of www.e-education.psu.edu/worldofweather/... .

[edit] Also, I think you're misisng a link in "I circled a portion of Sandy's circulation where there was warm advection on this annotated version "?



Yes, I'm keeping all my images in house at A World of Weather. I think Angela assumed they were on Weather Underground. They're all fixed now. Thanks again, Angela.
Member Since: October 24, 2012 Posts: 90 Comments: 798
10. 24hourprof
1:17 PM GMT on November 06, 2012
Quoting Skyepony:
Lee~ Excellent blog..& not just because I was in the not a hurricane at landfall camp in the war that went on over this in Masters':)

Welcome to Wunderground.

Your links didn't work.


Many thanks. Looks like Angela fixed them. Much obliged.
Member Since: October 24, 2012 Posts: 90 Comments: 798
9. Angela Fritz , Atmospheric Scientist (Admin)
2:30 AM GMT on November 06, 2012
Links were my fault -- bad editing! Should be fixed now, thanks everyone!
8. Skyepony (Mod)
12:59 AM GMT on November 06, 2012
Lee~ Excellent blog..& not just because I was in the not a hurricane at landfall camp in the war that went on over this in Masters':)

Welcome to Wunderground.

Your links didn't work.
Member Since: August 10, 2005 Posts: 161 Comments: 37337
7. AySz88
12:20 AM GMT on November 06, 2012
FYI, some of the "bigger image" links are pointing at wunderground.com/worldofweather/... instead of www.e-education.psu.edu/worldofweather/... .

[edit] Also, I think you're misisng a link in "I circled a portion of Sandy's circulation where there was warm advection on this annotated version "?
Member Since: August 25, 2005 Posts: 0 Comments: 8
6. 24hourprof
12:09 AM GMT on November 06, 2012
Many thanks.

Everybody...please just call me Lee.
Member Since: October 24, 2012 Posts: 90 Comments: 798
5. Barefootontherocks
11:29 PM GMT on November 05, 2012
Thanks also to AngelaFritz who seems to bear some responsibility for this blog coming forth. Ps. And to Dr. Herndon at Wisconsin.
Member Since: April 29, 2006 Posts: 151 Comments: 18360
4. Barefootontherocks
11:28 PM GMT on November 05, 2012
Wow, Lee Grenci. From a water vapor loop junkie, thank you so much. Don't know what your blog all means yet, maybe never will, but this is really neat and I am sure I will learn from it. Thank you.
Member Since: April 29, 2006 Posts: 151 Comments: 18360
3. 24hourprof
11:07 PM GMT on November 05, 2012
Many thanks for the welcome. When I start regular blogging, I hope to explain basics that will allow readers to learn and understand. Sort of a series of mini-lectures that will build toward a specified scientific concept.

How does this idea sound?

Again, many thanks for welcoming me.
Member Since: October 24, 2012 Posts: 90 Comments: 798
2. whitewabit (Mod)
10:20 PM GMT on November 05, 2012
Thanks Lee Grenci .. I didn't understand everything in your article and need to do some more reading, but you have helped me understand how Sandy was transitioning.

Happy to see you coming on board.
Member Since: August 17, 2005 Posts: 361 Comments: 31207
1. HawksNestPoint
9:50 PM GMT on November 05, 2012
Thanks for scientific explanation that left this 1960's educated engineer in the dust. As much as I appreciate the analysis, it is not of interest to weather nerds alone. Most of the insurance companies have higher deductibles for hurricane damage, and the fact that it was not will hurt the underwriters and help the claims beneficiaries. It is interesting to note that several state insurance departments have already made the determination that Sandy was not a hurricane, and this will reduce the arguments with insurers over deductibles.
Member Since: October 6, 2002 Posts: 0 Comments: 8

Viewing: 34 - 1

Page: 1 — Blog Index

Top of Page

About angelafritz

Atmospheric Scientist here at Weather Underground, with serious nerd love for tropical cyclones and climate change. Twitter: @WunderAngela

Local Weather

Partly Cloudy
65 °F
Partly Cloudy