NOAA Technical Memorandum NWS SR-193

Section 7 Table of Contents:

7. Time of Day Variations
A. National
B. Regional
C. 6-Hour Periods
D. Seasons

7. TIME OF DAY VARIATIONS

A. NATIONAL

The time of day when casualties and damages due to lightning were reported in the US shows a strong dominance of afternoon reports (Figure 30 and Table 27). Two-thirds of the casualties occur between noon and 6 p.m. Local Standard Time (LST). Casualties show a steady increase toward a maximum at 1600 LST, followed by a slightly slower decrease after the maximum. Therefore, most of the lightning-vulnerable activities engaged by people occur during the day.

However, damage reports are more broadly distributed through the 24-hour day. They show a steeper increase toward a maximum at 1700 LST than the decline after the maximum toward midnight. There is another slight increase in damages reports between one and two hours after midnight.

Differences between the two time series are in lower Figure 30. During the afternoon, there are up to 6% more casualties than damage reports. But at night, some hours have more than 3% less casualties than damage reports.

The difference at night is apparently due to the immobile nature of buildings and other objects. In contrast, people are less prone to be involved in lightning-sensitive activities at night, and also tend to avoid brightly-illuminated nighttime storms. There is a type of nocturnal thunderstorm complex, the mesoscale convective system, that occurs over the plains states during the summer, and produces tremendous amounts of lightning. These mesoscale convective systems, sometimes covering areas the size of Iowa, can produce rates over 3000 ground strikes per hour (Goodman and MacGorman, 1986; Holle et al., 1994).

Table 27 shows the numbers that were used to construct Figure 30. Not all Storm Data entries have times reported. Of the 13,057 casualties in Table 3, Table 27 shows that 8681 (66%) have times, while 61% of the damage reports have times associated with the report.

Figure 30: Top: Time of day of lightning casualty and damage reports for the United States from 1959 to 1994. Bottom: Percent casualties minus percent damage reports by hour. Time label for end of hour.

Maximum lightning impacts during the afternoon have been widely documented. For a small dataset from 1968 to 1985, deaths in North Carolina peaked at 1800 LST (Duclos and Sanderson, 1990). However, the most common maxima are from 1400 to 1600 LST:

• Deaths in Singapore from 1922 to 1979 (Pakiam et al., 1981).
• Deaths in Florida from 1978 to 1987 (Duclos et al., 1990).
• Deaths, injuries, and damage reports in Michigan from 1959 to 1987 (Ferrett and Ojala, 1992).
• Casualties in the US from 1959 to 1990 (López and Holle, 1995).
• Casualties in Florida during 1996 (Paxton and Morales, 1997).

On the time scale of individual storms, Holle et al. (1992) studied summer storms in central Florida. People had a tendency to be a casualty of lightning more often toward the start and end of thunderstorms, while objects on the ground were struck more uniformly through the lifetime of a storm. As on the daily scale, immobile objects were less sensitive to the storm than people who took actions against the threat.

B. REGIONAL

The times of casualties and damage reports by region are in Figures 31 and 32. Highest frequencies of casualties are typically within one hour of 1500 LST. Only the west coast has a broad afternoon maximum from 1200 to 1900 LST for a small sample size (100 casualties). Damage reports usually peak an hour or two later than casualties (Figure 30). In the plains and Midwest, many nighttime damage reports may come from the nocturnal thunderstorms mentioned earlier.

Figure 31 shows differences between plains and Midwest states, compared to the northeast, southeast, and Rockies. The main departures are in the afternoon, when the Rockies and eastern states have 13% more casualties and 8% more damages than on the plains and in the Midwest Relatively narrow distributions are apparent in the Rockies, southeast, and northeast. Time series in the plains and Midwest are broader.

Figure 31: Differences between hourly percentages of lightning casualties (solid line) and damage reports (dashed) in plains and midwest regions minus Rockies, northeast, and southeast regions.

Figure 32: Time of day of lightning casualties (solid line) and damage reports (dashed) by region in %.

C. 6-HOUR PERIODS

Casualties and damage reports by 6-hour periods of the day are shown by maps in Figure 33. These results can be compared to Figure 2 for casualties and Figure 9 for damage reports during all hours.

Night (0000-0559 LST)
Casualties and damage reports during the nighttime are most frequent in the plains and upper Midwest states, and a few populous eastern states. Compared to maps in Figures 2 and 9 for the entire sample, there is a shift in the highest-ranking states to the north and east from Texas, through Oklahoma and the plains states, to Wisconsin. Only the first 27 ranks of casualties are shown, since all states ranked lower than 27th have one or zero casualties. There were 62 deaths (3% of the day's fatalities with known times), 135 injuries (2%), and 1753 damage reports (15%) in Storm Data during these hours.

Of particular interest is how a lightning casualty can occur during these night to early morning hours. There were 29 deaths in 19 events between midnight and 0559 LST since 1980 in Storm Data. Of this total, 59% of the deaths and 58% of the events occurred when people were in a house set on fire by lightning. The next most common situation occurred when people were camping in tents (21% of the deaths and 16% of the events). Remaining incidents since 1980 involved a backpacker, an airman in an Air Force hangar, a driver losing control of his truck when it was struck by lightning while crossing a bridge, a night watchman at a factory, and people walking in a field.

Morning (0600-1159 LST)
Casualties during the morning are spread widely across the country. Damage reports are most common in the plains states in a similar pattern to the map for nighttime hours. There were 237 deaths (11%), 851 injuries (13%), and 1218 damage events (10%) in Storm Data during these morning hours.

Afternoon (1200-1759 LST)
Casualties during the afternoon resemble Figure 2 for the entire day. This result is to be expected since these are the most frequent hours for deaths and injuries. There were 1397 deaths (67%) and 4169 injuries (63%) in Storm Data during these hours. Damage reports are not as closely related to the 24-hour map in Figure 9, since the 4952 damage reports are only 41% of the Storm Data reports during these hours.

Evening (1800-2359 LST)
Casualties during the evening have the same shift in highest-ranking states to the north and east from Texas as noted for nighttime hours. Damage reports during the evening are similar to the whole-day map in Figure 9 and the afternoon map in Figure 30. There were 374 deaths (18%), 1457 injuries (22%), and 4064 damage reports (34%) in Storm Data during these hours.

Afternoon maps of both casualties and damage reports have the largest 6-hour sample sizes of the day and dominate Figures 2 and 9 for the whole day. There is a substantial number of high-ranking states for both categories in the southeast and eastern states. However, during other time periods, the maps show a tendency for events to occur away from the east and southeast, and greater activity on the plains and eastward to the Midwest Rasmusson (1971) found a similar pattern in the diurnal distribution of thunderstorms based on audible thunder reports.

Other studies have shown afternoon maxima in lightning impacts according to three- or six-hour periods:

*Deaths, injuries, and damage reports in central Florida from 1983 to 1990 (Holle et al., 1992).
*Fatalities in Australia from 1824 to 1991 (Coates et al., 1993).
*Casualties and damage reports in Colorado from 1950 to 1991 (López et al., 1995).

Figure 33: Rank of lightning casualties and damage reports by 6-hour periods and state of the US from 1959 to 1994.

D. SEASONS