Recognizing Side Lobes Versus Rotation Associated With a Tornado

[Mike Smith]

Hi Everyone,

We had a classic side-lobe situation this past Friday but, unlike some, it turned into a tornado in Texas. That posting is here: MSE CREATIVE CONSULTING BLOG

Some information for chasers that I didn't include in the blog posting:

• Side lobes can indicate extreme winds, even more than 100 mph. It can be very difficult not to want to pull the trigger when looking at winds of that velocity.
• Side lobes do occur with/near circulations on the leading edge of squall lines (QLCS), but I have not had enough experience to provide quality guidance.
• As with everything in meteorology, there are (rare) exceptions. The April 29, 2022, Wichita-Andover Tornado had such a high base (see radar imagery below) that it appeared the tornado's circulation was detached from the supercell. It could have been misinterpreted as a side lobe. But, these are rare.

Happy and safe chasing in 2024!!

 

[Jeff Duda]

Thanks for bringing this up, Mike. Always important to consider the potential for sidelobe contamination!

A recent paper by Boettcher and Bentley (2022) in Weather and Forecasting provides an excellent introduction and discussion on how to recognize and evaluate sidelobe contamination, including that coming from the vertical. The final figure, Figure 16, is particularly useful.

 

[Mike Smith]

Jeff, I had not seen that, thank you.

We had a recent example where people were awakened to a tornado warning at midnight due to side lobe misinterpretation.
About Thursday Night's Tornado Warning False Alarm in Little Rock

My post was not meant to be as comprehensive as Boettcher and Bentley (2022), which I will read in its entirety, but a "heads up" so reasonably obvious situations do not trip us up.

Thanks again, Jeff.

 

[Ben Holcomb]

Steve Piltz WCM at Tulsa NWS Office did a decent video on the SPC Lecture series about issuing warnings. There is a good portion starting about 20 minutes in on Sidelobe

 

[gdlewen]

Hi Everyone,

We had a classic side-lobe situation this past Friday but, unlike some, it turned into a tornado in Texas. That posting is here: MSE CREATIVE CONSULTING BLOG

Some information for chasers that I didn't include in the blog posting:

• Side lobes can indicate extreme winds, even more than 100 mph. It can be very difficult not to want to pull the trigger when looking at winds of that velocity.
• Side lobes do occur with/near circulations on the leading edge of squall lines (QLCS), but I have not had enough experience to provide quality guidance.
• As with everything in meteorology, there are (rare) exceptions. The April 29, 2022, Wichita-Andover Tornado had such a high base (see radar imagery below) that it appeared the tornado's circulation was detached from the supercell. It could have been misinterpreted as a side lobe. But, these are rare.

Happy and safe chasing in 2024!!

@Mike Smith I have a couple of questions to hopefully clarify my understanding of things:

1) In the sequence, Right-Mover--Hook Echo--Rotation with a Hook, what additional cues led you to decide to issue a tornado warning? It seems like there are lots of situations where these occur and a tornado does not form.

2). Regarding the 4/29/2022 Storm, I have been identifying reflectivity/velocity patterns like that as "Low-Precipitation Supercell". Especially using an elevation that likely samples the storm and not the layer below it, like in the image below (same storm, but using level 3, with a beam height of about 2kft at the center of rotation.)

So, is that a correct classification? LP Supercell?



If not, then if someone can point me to some resources, that would be great. I have been through some early papers on LP-supercells, like Bluestein (1993), but that doesn't mean I understood them.

 

[Mike Smith]

Hi Everyone, I'm back to answer Geoff's questions (above) about the process of accurate tornado warnings.

I have a checklist. The first two are environmental. [In black] Was it present on 2-2-24 in northwest Texas?

• STP (any of the three formulations) ≥0.5? Yes
• Does the radar pattern resemble any of the "pattern recognition" configurations we taught? Yes, isolated supercell ahead of a line (see radar image).

If the above two above are met, strongly consider tornado warning when any of the two below are met. Tornado warning mandatory if three are met. These are in no order.

• Hook echo.
• Right-mover.
• Couplet.
• Lightning polarity suddenly flips to positive.
• Genuine TDS

I would gladly add some of the newer criterion such as the arc but I can't find evidence it contributes to more accurate or timely warnings beyond what I have written above.

I've mentioned elsewhere the upbraiding I received from the Midwest WCM about my warning techniques and how hook echoes are "so 1980's."

It depends, I suppose, on your goals. Are we there to warn the public or experiment on them? Nearby, you will find the NWS's own figures pertaining to their tornado warning accuracy. If these newer methods are truly superior, how does one account for the cratering of PoD and lead-time in NWS warnings the last dozen years?

I use this example to illustrate my point. Two people died in the Gaylord, MI Tornado of May 20, 2022. While an advance tornado warning was issued, it provided just 8 minutes of lead-time to the mobile home park where they were killed running for their car to drive to shelter. Please keep in mind the NWS's published tornado warning goal is 13 minutes. It was routinely achieved from 2000-2011. That is no longer true.

Gaylord Sequence of Events: The first image is at 1:42pm CDT (local time was Eastern). You see a broken line of thunderstorms moving east off Lake Michigan. There was a severe thunderstorm watch out but no tornado watch.

Second image from 2:18, shows a transition to a supercell well underway. The storm took an abrupt 23° turn to the right with evidence of an appendage.

2:26pm, we now have an isolated supercell ahead of a line.

Simultaneously, 2:26pm, we have an absolutely classic hook echo (zoomed image) but no discernible couplet.

It was at this point I issued a tornado warning for my Twitter followers. Using my criteria above, the following were present:

• STP above 0.5.
• Isolated cell ahead of a line.
• Right-mover.
• Absolutely classic hook.

The first time a couplet appeared was at 2:35pm, which is when the NWS issued its tornado warning (close-up image).

As the tornado reached the western city limits (2:46pm), it still did not have any sort of consistent adequate couplet for an EF-3 intensity as viewed from the perspective of a storm warning meteorologist.

I use this example as one where today's oft-use techniques were inadequate in a major, fatal tornado.

When I was training meteorologists at WeatherData, Inc. and then AccuWeather Enterprise Solutions, there was -- occasionally -- a significant pushback from a new meteorologist(s). Can't tell you how many times I heard, "that's not how we did it at [name of last job or college from which the meteorologist matriculated]." Again, "What is the goal?" Our clients absolutely depended on us for safety and multi-million dollar decisions and they loved us. Our verification stats the decade before I retired (2018) were outstanding.

Any questions? I'd be happy to answer them.

Best wishes, everyone, and GO Chiefs!!

 

[gdlewen]

@Mike Smith thank you very much. Very clear and comprehensive. As always. (I suppose this could be a DM but it seems ungrateful not to publicly acknowledge the time you—and others—take to help those of us who are learning to understand severe weather.)

 

[Mike Smith]

I was watching Convective Chronicles (

) new piece on diagnosing the surprise F-5 Plainville, Illinois, Tornado. They have an illustration that is germaine to what we have discussed (above) about tornado warnings. As some of you may know, this F-5 was a complete surprise with no tornado watch and the tornado warning was not issued until after the tornado dissipated.

The nearby illustration shows a lightning polarity flip from 95% positive (very rare) to negative at the time of tornado genesis. Usually, the flip -- when there is one -- is the other way around, But, the main point is that if there is a sudden flip in lightning polarity, that is one of my five criterion for a tornado warning.

It is important to note that "5" tornadoes form and occur without a flip like JLN. Still, this is a good indicator, especially in radar-sparse areas.