Interesting Findings from Two Fall HSLC Cases (23 Nov. 2014 & 18 Nov. 2015)

CSTAR folks,

We wanted to post a review of several HSLC cases that had some significant operational utility here at FFC to garner further insight, comments, or discussion you may have (included some discussion points at the end).  Both happen to be November events, with the first (23 Nov. 2014) providing a unique outlier to some previous local research, and the second more recent case (18 Nov. 2015) lending to a situation where the SHERBS3 may have been the heaviest hitter of the available predictive severe parameters.


23 Nov. 2014 QLCS Tornadic Event: Persistent Regeneration of Weak Tornadoes with Pronounced Tornado Debris Signatures

  • Poster presentation at 2016 AMS Annual Meeting
  • Coordinated topics/previous research with Jason Deese (FFC), John Banghoff (Ohio State Univ.), Steve Nelson (FFC), and Dr. Gary Lackmann (NC State Univ.)

The tornadic development across central Georgia on the afternoon of 23 Nov. 2014 was not the typical case one would expect in the Southeast U.S. for two reasons:

  1. Persistent northern bowing segment or “broken-S” QLCS convective mode resulting in six separate tornadoes
  2. Pronounced tornadic debris signatures (TDS) seen with five of the six tornadoes, some of which lofted debris to a significant height above the ground more than previously documented with weak tornadoes (Banghoff and Nelson, 2014)

Figure 1. Surveyed event tornadoes and damage points.

The synoptic setup consisted of a negatively titled upper shortwave trough tracking northeastward across the southeast CONUS (Fig. 2) with an attendant low pressure system drawing strong low level Gulf moisture advection (Fig. 3).


Figure 2: 500 mb analysis (2000z)


Figure 3. 850 mb analysis (2000z)

A parent surface high pressure system situated off the mid-Atlantic coast had previously resulted in hybrid cold-air damming (CAD) along the eastern slopes of the Appalachians, and allowed for the periphery of the wedge of cold air or “wedge front” to be present across central Georgia (Fig. 4).


Figure 4. Surface analysis (2000z)

The combination of these influencing features provided a HSLC environment for convective development (Fig. 5).  A special 18z upper air release also came out of TAE and indicated HSLC parameters even away from the wedge influence (Fig. 6).


Figure 5: Instability and shear parameters (2000z)


Figure 6: Special 18z TLH raob instability and shear parameters

Tornadogenesis was observed to consistently occur along the wedge front as it rapidly retreated northeastward across central Georgia ahead of the aforementioned system.



Figure 7: Radar snapshots of 6 tors at time of genesis

Dual-polarization radar data were analyzed for each of the tornadic cells to assess strength of rotation (Vr and NROT) and max TDS heights (used GR2Analyst).


Figure 8: Method of using dual-pol radar data to analyze TDS height


Figure 9: Resultant TDS heights during 6 tornado lifespans


Figure 10: Resultant Vr and NROT values during 6 tornado lifespans

Much of this data were also analyzed during the event in real-time to assist in enhanced wording of the tornado warnings (prior to implementation of Impact-Based Warning (IBW) wording).


Figure 11: Examples of enhanced wording used in real-time warnings

The analyzed TDS heights primarily stayed in a 6-11 kft range, which is more common to the significant EF2 category observed with previous research (Fig. 12).  While the surveyed tornadoes in this event mainly fit in the weak EF0-EF1 categories, it is proposed that such anomalously high TDS heights were due to the presence of abundant fall foliage and lofted leaf debris combined with subsequent tornadic updraft regeneration.


Figure 12: Annotated from Entremont and Lamb, 2013

It is also proposed that the wedge front provided a nearly steady source of low level streamwise vorticity available for tilting into the vertical within the convective updraft as subsequent downdrafts instigated persistent tornadogenesis by bringing vorticity to the surface.  Presence of the front thus compensated for the lack of surface based instability in the HSLC environment and helped focus tornadic development. This serves as an extension to previous research on wedge front influence in conversely low shear high CAPE environments (Fig. 13).


Figure 13: Annotated from Baker and Lackmann, 2009

Trends in observed radar data and associated near-storm environment from this particular case provide unique utility in operations.  The findings not only extend the proposed effect of wedge front influence on convection in HSLC environments, but also present an upper bound of TDS height correlation to tornado strength during the fall season.  This provides aid to awareness and enhanced wording in warning decisions. Warning operators could justify a seasonal adjustment to the threshold for tornado damage threat tags with the newly implemented IBW structure (Fig. 14).


Figure 14: Decision Aid utilizing previous TDS research from Banghoff and Nelson

Since the 2016 AMS Annual Meeting, Keith Sherburn provided some archived SHERBS3 and SHERBE plots near the time of initial tornadogenesis for the case (Fig. 15).


Figure 15: SHERB parameters during tornado lifespans (2000-2200z)

Interestingly, the SHERBS3 parameter captured the threat better than the SHERBE as critical values nosed farther north and east near the storm locations.  The SHERBE actually had an opposite trend in central Georgia as it weakened and diminished in spatial coverage. While SHERBE calculations set the effective shear magnitude to zero where CAPE is absent, it is noted this could miss events where CAPE is under-forecast/diagnosed.  In a rapidly retreating wedge situation, this is more likely to occur, especially if guidance struggles to resolve the strength, timing, and extent of the wedge to begin with.  It’s hypothesized the SHERBS3 has potential to more likely illustrate the favorable environment interacting with the wedge periphery in an HSLC environment.



18 Nov. 2015 QLCS Tornadic Event

A strong upper low pressure system and associated surface front brought a squall line of storms across north and central Georgia during the late afternoon and evening hours of 18 Nov. 2015.  Three tornadoes formed in the northern part of a pronounced bowing segment along the squall line as it tracked across portions of Coweta, Fulton, and DeKalb counties (Fig. 2). The first two tornadoes resulted in EF-1 rated damage, with one near Palmetto and another near Fairburn. The final tornado was a very brief EF-0 in DeKalb County near Tucker (specifics listed below). Some observed TDSs in radar imagery allowed for enhanced wording of the warnings.


Figure 1: Surface analysis


Figure 2: Radar imagery of northern QLCS bowing segment

There were also indications of a lingering wedge front/retreating warm front near the northern notch of the bowing segment, where there were still some ageostrophic roots to a 1030+mb parent high off the NE coast, and perhaps some in-situ reinforcement occurring from upstream solar sheltering seen in the 2000z obs plot below (Fig. 3).  This boundary could have been interacting with the near-storm environment, providing a more steady source of low level streamwise vorticity necessary for tornadogenesis and subsequent regenerations.


Figure 3: Surface analysis and obs


First tornado (Coweta Co. 448 PM EST):

Rating: EF-1
Peak Wind: 105 MPH
Path Length: 0.2 MILES
Path Width: 200 YARDS

Figure 4: Radar imagery of first tor


Figure 5: Surveyed path of first tor

Second tornado (Fulton Co. 459 PM EST):

Rating: EF-1
Peak Wind: 95 MPH
Path Length: 2.6 MILES
Path Width: 200 YARDS

Figure 6: Radar imagery of second tor


Figure 7: Surveyed path of second tor

Third tornado (DeKalb Co. 547 PM EST):

Rating: EF-0
Peak Wind: 75 MPH
Path Length: 0.12 MILES
Path Width: 100 YARDS

Figure 8: Radar imagery of third tor


Figure 9: Surveyed path of third tor

The environmental parameters also fit the bill for HSLC.  Below are reanalysis plots for 0-6 km bulk shear, 0-1km SRH, SBCAPE, SHERBE, and SHERBS3 compared to the SPC storm reports at times near the initial tornado occurrence (2000z and 2100z in Fig. 10 and 11 respectively).


Figure 10: Instability and shear parameters (2000z)


Figure 11: Instability and shear parameters (2100z)

Again, both SHERBE and SHERBS3 have critical threshold values north of the instability axis near the areas of tornadogenesis, and the SHERBS3 had better coverage than the SHERBE.  This still is in support of the SHERBS3 parameter potentially being a greater predictive parameter in retreating wedge front situations with limited CAPE.




Initial discussion thoughts from both case analyses…

  • Did any other office have some notable observations or analysis from either of these past two events (more likely from 23 Nov. 2014 since 18 Nov. 2015 was quite localized)?  Perhaps another event with a similar setup or outcome?
  • Any further insight into the following proposed operational applications:
    • Raising the Fall seasonal threshold for TDS height and enhanced wording correlation for Impact-Based Warnings
    • Influence of wedge front given HSLC environments
    • Potential for SHERBS3 to capture severe threat better than SHERBE in rapidly retreating wedge cases


This entry was posted in CIMMSE, Convection, CSTAR, General Information, High Shear Low Cape Severe Wx. Bookmark the permalink.

3 Responses to Interesting Findings from Two Fall HSLC Cases (23 Nov. 2014 & 18 Nov. 2015)

  1. Jonathan Blaes @ WFO RAH says:


    These are both very interesting cases and the summaries are very well down and result from what appears to be a tremendous amount of work already. Thanks! Now to pass along a couple of comments to your questions.

    Both of these November events did not extend into central NC and the RAH CWA from what I can tell from the SPC reports database ( and Our local case study database includes some notes and imagery relation to the demise of the CAD for the November 2014 event though.

    Your comment about raising the fall seasonal threshold for TDS height and enhanced wording correlation for IBW is interesting. We have not had as much experience with TDS ‘ as you or many other offices but a consideration of what debris is available to be lofted is warranted. While your TDS heights were on the higher end of your box and whisker distribution, I’m not sure they were anomalous or extreme outliers, definately on the higher end though. Did you get any video of these tornadoes, were there more leaves visible then in the summer months? If the sample size is large enough, it might be interesting to examine or compare TDS heights in the fall vs spring/summer.

    The wedge front strikes again! From your summary, it appears the wedge front was retreating, is that correct? If so, this may point to results similar to what Jessica King noted regarding support for HSLC severe events in locations with rapid changes in surface temperature and moisture. In your previous work with Dr Lackmann you noted that CAD makes the strongest contribution to severe convection in shear-limited events with sufficient instability. Here you have noted that in high-shear events but with limited instability the CAD or wedge front compensated by providing additional vorticity to the surface. I assume that the limited instability was complemented by localized forcing for ascent near the wedge front. This is a really cool case.

    Great write up Adam and thanks for sharing, Jonathan

    • akbaker84 says:


      Regarding the TDS and the actual debris – we did not have videos, though we did extensive surveys along the almost 120 mile stretch where it had the 6 broken paths including aerial views for the SW majority via helicopter, and the vast majority of damage occurred in largely wooded areas near a local peak in foliage. So we have good reason to believe most of it consisted of pine needles and leaves. Good point about sample size – if I remember right, I don’t think the previous research from John Banghoff and Steve Nelson had enough signal in the results to distinguish seasonal differences, but this was on the upper end of their overall range.

      As far as the role of the wedge, it was indeed retreating (quite rapidly in the 2014 case) and overall does seem to be the “tipping point” in the CAPE/shear balance for favorable environments. Dr. Lackmann and I hadn’t simulated a HSLC case in the past work, but it would be interesting to see if it had similar contributions to stronger convection like it did in the LSHC environment. Of course as I’m writing this, we have a wedge front in a marginal shear and low CAPE setup… :^) Thanks for the thoughts!

  2. mdparker says:

    The debate about the SHERBS3 rages on. There is a contingent that doesn’t like the SHERBS3 because it can be large even when storms are impossible (i.e. CAPE is literally zero). The problem with the mesoanalyzed SHERBE is that it necessarily goes to zero as the SFCOA CAPE goes to zero, *even when those SFCOA CAPE values are in error*. I am beginning to suspect that the scenarios where CAPE develops “just in time” (as shown in Jessica King’s model runs and Keith Sherburn’s composites) are really not very well handled by the SFCOA on many days. So, although the physical reasoning underpinning the SHERBE is “better”, for situational awareness it may well be more important to keep tabs on the SHERBS3.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s