After failing to receive any feedback from PF members, I received from one CharlesChandler of Baltimore, MD on another site the following rather comprehensive explanation of the Fort Worth "light show". Pursuant to PF policy, I have omitted links to his paper and embedded photos and video. This is the best video I've seen of this phenomenon -- thanks for posting this! These are called corona discharges, which occur when the electrostatic potential exceeds 100 kV/m.
Normally, a corona discharge in the air would produce violet light, and would be called St. Elmo's fire. As this phenomenon is most common at the end of a thunderstorm, sailors named it after their patron saint, believing that St. Elmo had once again delivered them from the perils of a storm at sea. There are a few violet-colored discharges, but St. Elmo's fire is generally a sustained discharge, and the "fingers" are rarely more than a couple of meters long. So this is something different.
The key to understanding this is the color. Violet means an electron avalanche passing through neutrally-charged air, where the arriving and departing electrons simply disturb existing electrons in the outermost shell, producing higher-frequency photons. The other colors, lower in frequency, indicate that the air was ionized, and the atoms were accepting and relinquishing electrons in full orbit (not just p-state changes). The most prevalent color is blue, which is ionized nitrogen or oxygen. Also visible is a blue-green color, which is ionized water molecules. The orange-red discharges are from ionized oxygen.
Meteorologists always say that this phenomenon is just the sky getting lit up by transformers blowing up. But that doesn't explain why they kept blowing up, nor why transformers all up and down the line kept blowing up (you'd think that once the line failed, the whole circuit would go dead, or the sub-station relay would have kicked out), nor why the city lights kept shining as the neighborhood transformers got knocked out. It also does not explain why some of the discharges are slightly directional and/or filamentary. These, in fact, are discharges between the ground and the air, and while power line towers are likely discharge points, they don't necessarily affect the power lines, and the sub-station doesn't sense the discharges, so it's not going to open the circuit. If the power goes out, it's because of a structural failure in the tower that was carrying the lines.
For more photos of similar events, see the Blue & Orange Flashes section of my paper on tornadic supercell thunderstorms. My personal favorite is the last image (and the associated video), where lamp-posts "caught fire" and burned an orange-red color just after a tornado passed by.
Chattanooga, TN, 2010-10-28
Interestingly, the street lights kept shining, and never failed, even though they were "on fire" for tens of seconds. I didn't think that there was anything flammable in a lamp-post...Respectfully submitted,
Steve
