Director - Severe Storms Research Center
Georgia Tech Research Institute
The Severe Storms Research Center (SSRC) at the Georgia Tech Research Institute (GTRI) has developed and deployed the North Georgia Lightning Mapping Array (NGLMA). The NGLMA uses twelve very high frequency (VHF) sensors, distributed about the metropolitan Atlanta region, to measure “total lightning” in electrical storms.
Conventional lightning mapping, such as is typically seen on television news shows, depicts the strike points of lightning flashes which originate in the parent thunderstorm and reach the ground. Currently, only these cloud to ground (CG) lightning strokes are accurately measured over the North Georgia area. These CG strokes also only account for about 10% of all the electrical activity in severe local storms.
Figure 1. A comparison between what a cloud-to-ground network observes in a lightning flash (left) versus what a total lightning network will observe in a lightning flash (right). Note how the cloud-to-ground network only provides a single point of information. Also, the cloud-to-ground network would observe nothing if the flash were solely intra-cloud. (Illustration source: NASA SPoRT)
The deployment of a VHF LMA in North Georgia allows the real time detection and location of the three dimensional paths of Cloud to Ground (CG), Intracloud (IC) and Cloud to Cloud (CC) lightning. The IC and CC strokes are much more prevalent than CG strokes and have been found to be useful gauges of convective growth within storm cells. The intensity of intracloud electrical activity may also indicate storms which may (or may not) pose a threat to aircraft operations or may develop shortly into CG lightning threats for personnel and operations on the ground, giving an extended lead time of up to five minutes before the first CG strike.
Figure 2. A sample of 31 thunderstorms observed by the Kennedy Space Center Lightning Detection and Ranging network showing the number of cloud-to-ground strikes versus total lightning observed in each storm. Notice how the intra-cloud component dominates the total lightning observed in each storm. It is also interesting to note that two storms had no cloud-to-ground strikes at all, yet were still very electrically active. (Illustration courtesy NASA SPoRT)
Research on sudden increases of total lightning (the combination of CG, CC and IC lightning), known as ‘lightning jumps’, have been shown to be related to intensifying updrafts in thunderstorms and often precede the development of severe weather events, including large hail and tornadoes (figures 3 and 4). Using this information in real-time can help forecasters in warning operations decide which storms have the potential to become severe and issue warnings sooner.
The use of the NGLMA will also allow the use of new data types, such as flash extent density, which measures the spatial extent of lightning segments within CG, IC and CC flashes and how these distributions evolve over time.
Additionally, total lightning data will be made available on a faster update rate than NEXRAD radar data, with only a minute or two elapsing between successive updates. This rapid update rate will allow more precise tracking and timely warnings by the NWS than are possible on the basis of radar data alone.
Figure 3. A screen capture from AWIPS II showing the total lightning flash extent density (colored contours) versus the cloud-to-ground strike locations (negative and plus signs). Notice how the total lightning indicates that lightning flashes are covering a wide area whereas the cloud-to-ground observations only show single locations. (Illustration source: NASA SPoRT)
Figure 4. A time trend plot (top) of a storm that had two separate lightning jumps at 1906 and 1920 UTC that led to the issuance of a tornado warning at 1920 UTC ahead of the touchdown of an EF-1 tornado. The bottom two images show the AWIPS display before (left) and during (right) the lightning jump. (Illustration source: NASA SPoRT)
The VHF sensors being deployed in the NGLMA detect radio frequency (RF) emissions from short sections of the lightning path as the discharge advances both within the cloud and as the flash extends towards the ground. Each element within the array of VHF sensors looks for impulsive radiation within short (80 us) windows in an unused TV band (82 to 88 MHz, channel 6) and notes the time of the maximum impulse. The time of this maximum signal within the measurement windows are temporally aligned through the use of GPS signals. Figure 5 shows a typical impulsive signal within a portion of the sampling window.
Figure 5. RF Impulse from Lightning Segment (Illustration source: New Mexico Tech)
By accurately determining the time of arrival (TOA) of these emissions at several ground stations at precisely known locations and times, a unique three dimensional position may be determined for the source, as demonstrated in Figure 6.
Figure 6. Time of Arrival used to locate lightning segments (Illustration source: New Mexico Tech)
A more complete description of the mathematics behind the operation of the LMA can be found at: http://lightning.nmt.edu/nmt_lms/descrip.html
By properly associating these distinct emissions in both time and space, the three dimensional path of the lightning's path within the thunderstorm may be plotted. An example of an exceptionally long lightning stroke, detected by the LMA in Northern Alabama, is shown in Figure 7. This particular lightning extended over 45 km in length. Four ground stroke points associated with this cloud flash are depicted as “+” or “-” symbols on the lower map and labelled “1” through “4” on the time-height plot in the upper part of the illustration. It is easy to see that the electrically active portion of the thunderstorm has a much greater extent than the ground stroke points would indicate.
Figure 7. Depiction of a large lightning flash above the North Alabama Lightning Mapping Array (Illustration source: NASA MSFC)
The SSRC currently operates eight VHF sensors that are centered on downtown Atlanta (figure 8):
The full deployment of the network allows the NGLMA to detect and plot electrical activity over an area within 150 to 250 km of the center of the array.
Figure 8. Locations of active lightning sensors. The array center is downtown Atlanta.
The equipment deployed at the SSRC at GTRI, Newnan and Queen of Angels locations consists of an electronics package contained in an electronically cooled enclosure and an antenna tuned to an unused TV frequency. This equipment is illustrated below in Figure 9.
Figure 9. Original LMA Hardware in Electronic Cooler (Illustration source: New Mexico Tech)
While three of the NGLMA sensor units are of the electronic cooler variety, the rest are improved self-contained LMA units (figures 10-12). These units consist of three main components: a VHF and a GPS antenna, a main electronics box (containing the processing computer, solid state hard drive, GPS receiver and filters), and a power conditioning box. The modern design of these components allows them to function without the electronic cooler. The boxes are also somewhat smaller than the earlier cooler-based units. Functionally, however, both sensor unit designs are equivalent.
Figure 10. The main electronics enclosure.
Figure 11. The electronics and power supply fit in the two small boxes on the lower shelf, with the backup battery on the floor beneath them.
Figure 12. The antenna mounted and attached to a pallet on the Queen of Angels roof.
With twelve sensors on line, the NGLMA is now producing total lightning data that is available for examination by outside partners. Data from the sensors is updated every one minute and posted in real-time on http://nglma.gtri.gatech.edu. An example of the web based data is shown in Figure 13.
It should be emphasized that this data is experimental, however it can also prove useful to forecasters and emergency managers. The SSRC worked with NASA SPoRT to include the NGLMA data into the NWS LDM feed. Through this mechanism, the data is now available in real-time to the NWS office in Peachtree City for ingestion into their AWIPS-II system. Forecasters have been trained on how to use this data in in warning operations. We will continue collaborating with the NWS to analyze data for case studies to learn how the data can be used optimally.
Figure 13. Data from the NGLMA as thunderstorms that produced two EF-1 tornadoes moved through Atlanta on November 18, 2015