Abstract
Cloud-to-ground lightning flash data collected by the National Lightning Detection Network were analysed in and around 16 central U.S. cities for the period 1989–92. Lightning data are well suited to study storm activity in and around large urban areas since their continuity and coverage in space and time is superior to historical, spatially limited records of thunderstorm activity. Frequency of cloud-to-ground lightning flashes (of negative and positive polarity) in the area immediately upwind, within, and immediately downwind of the cities were compared. An enhancement of lightning frequency on the order of 40%–85% was found over and downwind of many of these cities.
A number of possible urban-related causal factors were examined including effects of increased urban concentrations of cloud condensation nuclei, urban population and size, and the presence of distinct topographic features in and around the cities. Various factors, physical and anthropogenic, appeared to interact in diverse ways to account for changes in lightning flash frequency. The enhancement of lightning activity was largest during the afternoon hours when the urban–rural temperature differences are usually smallest, but when the atmosphere is generally the most unstable and when there is often a maximum in convective activity. The spatial distribution of the first 50 lightning flashes from each storm suggested that the urban area did not initiate new lightning storms. Thus, the overall results suggested that existing thunderstorms were the most strongly affected.
