Lightning location in a Colorado thunderstorm
MacGorman, D. R.
Few, Arthur A.
Master of Science
The method of acoustically reconstructing lightning channels from thunder recorded with a small array of microphones (having a base line of about 5 m) is relatively slow (about 7 hours per lightning flash). We examine another method, which we have called thunder ranging, that does not provide as many points on the lightning channel but is considerably faster (about 2 hours per lightning flash). In this technique, the ranges of a source of a thunder pulse, .1-2 sec long, (also called a clap) to three, noncollinear microphones, separated by distances on the order of 1 km, are measured manually from an analogue oscillogram. From these ranges, the coordinates of the channel segment generating the pulse can be calculated. We discuss the sources and typical magnitudes of errors in this calculation. In the 1972 Colorado experiment, from which our data was taken, the error in the calculated location of a channel segment is typically within 15% of its range. We plot the location of twenty lightning flashes occurring between 18:7 MDT and 18:3 MDT in a multi-cell storm on 25 July 1972. These plots are superimposed on radar reflectivity contours in order to study ways in which lightning and storm structure are related. As in earlier acoustic studies (e,g Teer and Few,1974), the lightning channels are predominantly horizontal. They occur in a relatively thin layer; 4-5% of the calculated locations are between 4 km and 5 km above the ground; 7-8% are between 3.5 km and 5.5 km. The narrower layer is completely above the ° C isotherm in the cloud. Two other conclusions are tentative, since they are based on fewer observations, (1) The lightning channels seem to avoid regions of heavy rainfall 1.4 mm hr^-1, as determined by 1-cm radar) and often extend long distances into regions from which the echo is less than 3 dBZ. (2) Lightning is roughly aligned, having two possible correlations with the storm: the channels often parallel radar reflectivity contours, and they point from the region around the array spread toward regions into which a radar contour is expanding.