Revisiting the Hodograph

[Donald Giuliano]

Those of you who know John Esterheld or myself probably know that we've been working on an independent research project for some time now. We are finally at a point where we can present the results to the public, and will be doing so Aug. 29th in the auditorium at the National Weather Center in Norman, OK. We are seeking feedback from fellow stormchasers and/or meteorologists before submitting our results for peer review, and are certain that at least some Stormtrack members in the area will be interested in attending.

So without further ado, allow me to present the title and abstract for our talk later this month:

Revisiting the Hodograph: ​

A New Examination of Low-Level Shear Variation ​

Between Storm Classes​

John Esterheld​

Donald Giuliano​

​

ABSTRACT​

​

After examining the hodographs found in many significant tornado environments, the presenters hypothesize that environments characterized by two features strongly favor significant tornadoes. These features consist of a strong straight line shear vector in the 10-500 m layer and storm-relative 10 m inflow normal to the base of the shear vector (thus yielding purely streamwise 10 m inflow).


An analysis of these hodograph features was performed on 67 surface-based supercells in Oklahoma between 1997 and 2004 that were classified as non-tornadic, weak-tornadic (F0-F1), and significant tornadic (F2-F5). By quantitatively analyzing the 10-500 m storm-relative helicity (SRH) as determined by the 10-500 m shear vector and storm-relative 10 m inflow, variation in 10-500 m SRH values between all three storm classes was found. The magnitude of the 10-500 m shear vector itself was also noted to vary significantly between storm classes, although to a lesser degree. Further, histograms of the angle defined by the storm-relative 10 m inflow and 10-500 m shear vectors reveal that tornadic storms, and particularly significant tornadic storms, tend to be characterized by angles near ninety degrees, whereas non-tornadic storms do not.

 

[rdale]

Is this similar to the "hockey stick" soundings that have been talked about over the last year by some operationalists?

 

[Donald Giuliano]

Yes, it is related. This feature is often referred to as a "kink" or "sickle-shaped hodograph", and was first mentioned by Thompson and Edwards (2000):

http://www.spc.noaa.gov/publications/thompson/3may99/waf.htm

While preparing the presentation, we also came across (just a few minutes ago) a recent paper by Dan Miller that explored this a little further and shares some of our ideas:

http://ams.confex.com/ams/pdfpapers/115403.pdf

--Don

 

[rdale]

Thanks for finding Dan's - I saw his presentation within the last year (forgot where it came from) and that got my curiosity going, I'd be very interested to see what your side comes up with...

 

[Gabe Garfield]

Don gave me a sneak peek of his research today, and I can tell you, it is very compelling. Surprisingly little research has been dedicated to the hodograph "kink," so I'm quite glad that Don and John took the bull by the horns. This feature is nearly ubiquitous to proximity soundings for major tornadic events (provided adjustments are made for observed surface winds and storm motions), so it certainly makes sense to find a way to quantitatively represent this feature.

If you're in the Norman area, I would highly recommend that you attend Don's talk!

Gabe

 

[Matt Ziebell]

Just curious if your research used strictly observed data (e.g. RAOBs and wind profilers) or were model proximity soundings also incorporated?

 

[Donald Giuliano]

Just curious if your research used strictly observed data (e.g. RAOBs and wind profilers) or were model proximity soundings also incorporated?

We used strictly observed data, as in our experience model proximity soundings cannot resolve these features and are essentially worthless from a quantitative point of view. Even RAOBs were of little use due to poor temporal/spatial resolution, so we wound up excluding them as a result. It should be noted, however, that it was through looking at hundreds, actually more like thousands, of RAOBs that led to the development of our ideas.

 

[rdale]

If you dropped RAOBs what did you use?

 

[Donald Giuliano]

If you dropped RAOBs what did you use?

We used wind profilers. VAD profiles would likely work as well, but we elected to stick with one data source for all observations AGL. To get the surface wind, we then supplemented this with 10 m winds from the Oklahoma Mesonet.

 

[Patrick Marsh]

One of the key pieces of Dan's presentation was that it is more than just the sickle-shaped hodograph. You needed to have a very moist sounding as well. If I recall correctly, none of the soundings in Dan's presentation included the typical loaded gun sounding and were more "tropical" in nature in the lowest levels.

With that said, did / can you analyze the sounding structure associated with the hodograph for each event?

 

[Donald Giuliano]

One of the key pieces of Dan's presentation was that it is more than just the sickle-shaped hodograph. You needed to have a very moist sounding as well. If I recall correctly, none of the soundings in Dan's presentation included the typical loaded gun sounding and were more "tropical" in nature in the lowest levels.

With that said, did / can you analyze the sounding structure associated with the hodograph for each event?

We did not, and instead chose to focus solely on the nature of the shear. The poorly-mixed aspect of the low-level thermodynamics pointed out by Dan is something we also came across while performing this research and believe to be important, but decided it was beyond the scope of what we were doing, and did not pursue it. Part of the reason for this is that we think it is quitely likely the two features are significantly related to one another, as it is not physically realistic to have intense boundary layer shear when the BL is well-mixed.

**EDIT

I think it would be possible to analyze the thermodynamic structure of the soundings, perhaps by using RUC analyses. However, you might start to get into the same issues there as you do with wind speeds, where the RUC is unable to properly resolve the near-surface thermodynamics. It would certainly be interesting to investigate, however.

 

[Patrick Marsh]

I agree that it is quite likely that the sickle shaped hodograph and poorly mixed boundary layer are inherently linked somehow. I was just hoping that a larger dataset existed than Dan's for comparison.

 

[afischer]

I think your study sounds fascinating... particularly in that you looked at direct observations via mesonet and wind profilers, rather than hodographs from RUC proximity soundings or modified adjacent RAOBs.

 

[Glen Romine]

These features consist of a strong straight line shear vector in the 10-500 m layer.

We used wind profilers. VAD profiles would likely work as well, but we elected to stick with one data source for all observations AGL. To get the surface wind, we then supplemented this with 10 m winds from the Oklahoma Mesonet.

Maybe I'm missing something here, but if you have a wind ob at 10 m, from a mesonet site, which you are transplanting and merging with the lowest available wind measurement from a wind profiler (which is 500 m AGL), don't you have to have a straight line hodograph (2 pts = line)? That said, is it fair to assume the profile would look straight if you had wind measurements between these two heights? Since the shear in this layer being normal to the shear above sounds important for your conclusions, this is something I'd want to look into shoring up. Also, be careful of site elevations of mesonet sites and wind profilers. Good luck with your project.

 

[Donald Giuliano]

Maybe I'm missing something here, but if you have a wind ob at 10 m, from a mesonet site, which you are transplanting and merging with the lowest available wind measurement from a wind profiler (which is 500 m AGL), don't you have to have a straight line hodograph (2 pts = line)? That said, is it fair to assume the profile would look straight if you had wind measurements between these two heights? Since the shear in this layer being normal to the shear above sounds important for your conclusions, this is something I'd want to look into shoring up. Also, be careful of site elevations of mesonet sites and wind profilers. Good luck with your project.

You're absolutely right that using 2 points implies a straight line. That part of our hypothesis remains unproven as we did not quantitatively address it in this study, though we contend there is enough observational evidence to suggest this is the case (enough to proceed with our study at any rate), and that our results are difficult to explain unless it is a common (though not necessarily ubiquitous) feature. There's plenty more to say, but I do not want to steal our thunder before we have had a chance to present. I'll be posting a link to the power point at a later date for interested parties who do not live near Norman.