I am currently watching the big middle US blizzard move from Oklahoma City towards Chicago. My lightning data feed shows occasional ground strokes and cloud flashes well into the snow sector along the I-44 corridor. I have been plotting these events on Google and for the events that have a small location clustering (i.e., 2 km diameter), they plot right on top of a visible tower. I suspect that these are all upward lightning flashes that self-initiate from the tower. The ground strokes and cloud events recorded by the lightning location system are either subsequent recoil leaders that form on weak branches and connect with the main upward leader channel or dart leaders (i.e., recoil leaders) that form after current cutoff and connect with the tower tip; both of which produce an impulsive return stroke that is detected by the location system. In Rapid City, most of the upward flashes that I observe are triggered by a nearby preceding positive ground flash. I am not seeing any preceding flashes for these cases. This is likely due to the lower cloud base (i.e., lower charge region) and the intense electric field produced in heavier snow. Given the data shows negative events (negative charge lowered to ground), the lower charge region influencing the towers is likely negative resulting in an upward propagating positive leader from the tower.
So a fundamental question comes to mind. Are most of the lightning events recorded during strong snow storms upward lightning from tall objects? If so, what is the percentage? If indeed most "thundersnow" is caused by tall structures, are we in essence creating more lightning by building tall towers and buildings?
I plan to investigate and address this in the future. Stay tuned.