Relationships between Thunderstorms and Cloud-to-Ground Lightning in the United States

Stanley A. Changnon Changnon Climatologist, Mahomet, Illinois

Search for other papers by Stanley A. Changnon in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

Data from networks of lightning sensors operated during 1986–89 were employed to perform climatic assessments of cloud-to-ground (CG) flashes, and of the relationship between CG flashes and thunder events, as reported at 62 first-order stations. The average annual CG flash pattern resembles the nation's thunderstorm pattern, with lightning being most frequent at points in the southeast United States (>18 000 flashes in Florida) and least frequent at West Cost stations (< 100 flashes). Flashes in the intermontane area are relatively less than thunderstorms, and greater than 50% of all thunder events there are due to intracloud lightning. An anomalous high in CG frequencies exists from Virginia to New York. Cluster analysis identified ten regions of similar storm activity with several reflecting localized influence (Great Lakes, Florida, the Piedmont, and Arizona).

Thunder events provide poor estimates of CG lightning incidence and durations. Cloud-to-ground flash data reveal that 20% (far West) and 50% (southeast United States) of all thunder events are missed at weather stations; 30%–60% of all thunder events have durations too short; 10% (North and West), 40% (mountains), and 25% (Southeast) of all CG flashes within 20 km of weather stations occur at times that are not reported as thunderstorms. The average annual point duration of thunderstorm activity, when based on thunder data, is underestimated by 23% (West Coast) and 44% (East Coast). A real variations in CG flashes reveal that audibility of thunder is better in the cooler northern latitudes, poorer in the mountains, and good on the West Coast. Regional values allow estimation of the average thunderstorm frequency and durations of point storm activity. Peak single-storm CG flashes vary from 50 at West Coast stations to greater than 2000 flashes in the Northeast and Florida, being between 2 (West) and 15 (East) times the point averages. The results allow for assessment of the risks associated with CG lightning throughout the United States. Findings also reveal that the use of historical thunder data, as a surrogate for lightning activity, is improper, and thunder values need to be adjusted with the relationships presented. A future shift to making thunderstorm observations using lightning data, as opposed to the thunder-heard method of the past and present, will bring sizable difference in the incidence and duration of storm activity, with increases ranging from 10% to 50% depending on the area of the United States.

Abstract

Data from networks of lightning sensors operated during 1986–89 were employed to perform climatic assessments of cloud-to-ground (CG) flashes, and of the relationship between CG flashes and thunder events, as reported at 62 first-order stations. The average annual CG flash pattern resembles the nation's thunderstorm pattern, with lightning being most frequent at points in the southeast United States (>18 000 flashes in Florida) and least frequent at West Cost stations (< 100 flashes). Flashes in the intermontane area are relatively less than thunderstorms, and greater than 50% of all thunder events there are due to intracloud lightning. An anomalous high in CG frequencies exists from Virginia to New York. Cluster analysis identified ten regions of similar storm activity with several reflecting localized influence (Great Lakes, Florida, the Piedmont, and Arizona).

Thunder events provide poor estimates of CG lightning incidence and durations. Cloud-to-ground flash data reveal that 20% (far West) and 50% (southeast United States) of all thunder events are missed at weather stations; 30%–60% of all thunder events have durations too short; 10% (North and West), 40% (mountains), and 25% (Southeast) of all CG flashes within 20 km of weather stations occur at times that are not reported as thunderstorms. The average annual point duration of thunderstorm activity, when based on thunder data, is underestimated by 23% (West Coast) and 44% (East Coast). A real variations in CG flashes reveal that audibility of thunder is better in the cooler northern latitudes, poorer in the mountains, and good on the West Coast. Regional values allow estimation of the average thunderstorm frequency and durations of point storm activity. Peak single-storm CG flashes vary from 50 at West Coast stations to greater than 2000 flashes in the Northeast and Florida, being between 2 (West) and 15 (East) times the point averages. The results allow for assessment of the risks associated with CG lightning throughout the United States. Findings also reveal that the use of historical thunder data, as a surrogate for lightning activity, is improper, and thunder values need to be adjusted with the relationships presented. A future shift to making thunderstorm observations using lightning data, as opposed to the thunder-heard method of the past and present, will bring sizable difference in the incidence and duration of storm activity, with increases ranging from 10% to 50% depending on the area of the United States.

Save