Utilizing Total Lightning and the Tracking Meteogram Tool to Assist in the Warning Decision Making Process

Herein is an example of the Tracking Meteogram Tool, which was developed by NASA SPoRT, being used to track and create a time series plot of the total lightning associated with a thunderstorm at the National Weather Service forecast office in New Braunfels, TX (Austin/San Antonio – EWX). The information gleaned by the time series plot from the tracking meteogram tool assisted in the warning decision making process.

For full disclosure, I have a background in total lightning and its operational uses in severe weather operations. My Master’s thesis at the University of Alabama in Huntsville was on the utility of total lightning and the lightning jump to assist in the quasi-linear convective system (QLCS) tornado warning decision process. Also, as a CIMMS research associate at the NWS Warning Decision Training Division, I developed a four-part series on best practices for using total lightning to assist in storm interrogation for various convective modes and severe hazards. I have been an intern at the NWS forecast office in New Braunfels, TX since May 2016.

On the evening of November 1st, 2016, there were isolated thunderstorms in the forecast across the Interstate 35 corridor between San Antonio and Austin, but severe weather of any sort was not anticipated across our area. The Storm Prediction Center convective outlook highlighted the eastern half of our CWA for possible thunderstorms, but did not have even a marginal risk area outlined.


Storm Prediction Center (SPC) Convective Outlook product issued at 1z on November 2nd, 2016 (8 pm CDT on November 1st, 2016).

On this particular shift, I was working the public service desk, while my colleague Nick Hampshire, a lead forecaster at EWX, was working the short-term forecast desk. Given my background in total lightning, I typically overlay the one minute 5 km by 5 km Earth Networks Total Lightning Detection Network (ENTLN) total lightning product on top of reflectivity for situational awareness purposes. Isolated showers and thunderstorms began initiating across the region around 6-7 pm that evening. These showers and storms were, as expected, fairly mundane and short lived, only producing light to moderate rainfall before the updraft was cut off and the storm dissipated. When the showers did manage to produce lightning, the lightning frequency was low and short lived.

Around 7:40 pm, a shower initiated east of Seguin, moving northward toward the cities of San Marcos and Austin. By the time it reached San Marcos around 8:20 pm, the shower began producing lightning. As the storm progressed northward toward the city of Austin, the total lightning flash rates continued to increase. To monitor the time series trend of the total flash rate, I used the Tracking Meteogram Tool and configured it to display the sum of the values, thereby plotting all the lightning being produced by the storm at any one time. I noticed a steady increase in the lightning flash rate that coincided with and even slightly preceded the strengthening of the storm as determined by radar signatures. A quick interrogation using radar and the standard environmental package from LAPS of the storm at around 8:51 pm showed 50+ dBZ echoes up to beyond the -30 degree Celsius level (~30,000 feet).

20161101 EWX Storm NASA post.png

4-panel display of reflectivity at different tilts from KEWX radar at 8:51 pm CDT on November 1st, 2016 (0151 UTC on November 2nd, 2016).

The total flash rate at this time was 46 flashes per minute, and the flash rate had increased from 34 flashes per minute at 8:47 pm to a local maximum of 47 flashes per minute at 8:52 pm. Given the radar signatures as well as the rapid increasing trend in total flash rate, Mr. Hampshire and I decided that a Significant Weather Advisory was warranted. In the text product, we mentioned pea to nickel sized hail associated with this storm. The SPS was issued around 8:52 pm. We received a few reports of pea sized hail in southwest Austin on social media shortly after 9 pm (2z).


1 minute ENTLN total lightning 5 km grid with tracking meteogram tool (left) and time series plot of total lightning for the storm of interest from 0133 UTC (8:33 pm CDT) to 0204 UTC (9:04 pm CDT) on November 2nd, 2016 (November 1st, 2016)



Radar loop from KEWX from 0054 UTC (7:54 pm CDT) to 0210 UTC (9:10 pm CDT) on November 2nd, 2016 (November 1st, 2016)

This case demonstrated the value of total lightning and the tracking meteogram tool. Given the forecast and the atmospheric environment, severe weather was not anticipated. However, it was the large, rapid increase in total lightning that initially prompted my attention to this storm and caused me to delve further into interrogating the severe potential. Had I not had the total lightning information available to me, the Significant Weather Advisory almost certainly would have come out later and perhaps not at all. Granted, this storm did not meet severe criteria, but not having any product issued for pea sized hail when hail of any size was not in the forecast would not have been an ideal situation, and the value added from the total lightning was still noteworthy.


Tweet posted from the NWS Austin/San Antonio twitter account shortly after the storm had passed through Austin, dropping pea sized hail.


4 thoughts on “Utilizing Total Lightning and the Tracking Meteogram Tool to Assist in the Warning Decision Making Process

    • Brad,

      I appreciate the kind words! And I absolutely anticipate similar applications once GLM becomes operational. GLM has been a long time coming, and I am excited for the data to finally begin streaming.

  1. I’ll second here! Great post Brett and welcome to the SPoRT blog! We’re glad to have you aboard and thanks for taking the time to make this post. I know blog posts, especially ones that are this detailed can be time-consuming. So, thanks for your efforts here.

    At first, I was surprised by the high flash counts in this case. 1-minute flash counts around 50 or so are as high as anything I’ve ever observed here. But, then, I noticed that the Tracking Tool was probably set to sum all the the values (pixels) under the circle. Here at the Huntsville office, we’re used to looking at the interpolated source and flash extent density data from the North Alabama LMA, using the maximum rather than a summed value. At least that’s my understanding. There are of course other differences between the LMA and ENTLN data sets and the way the data are generated. Anyway, I’ll let the lightning experts chime in here.

    Thanks again for the great post and how the data and tracking tool are being used operationally at EWX!

    • It was my pleasure! I am grateful for the opportunity to post. Hopefully I’ll have more cases to publish posts on in the coming months.

      As far as the flash counts, I did indeed set the meteogram to the sum configuration. I was tracking the total flashes from the storm at any one time, or at least that’s how I was interpreting it (it was the one minute flash rate), but someone correct me if I am wrong. In my research and experience with lightning data, I’ve seen intense supercells produce as many as 150+ flashes per minute (see some of Sarah Stough’s (UAH Ph.D. candidate) research on lightning associated with tornadic supercells).

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com 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