An 8 Year Lightning Climatology of North Carolina

This week is Lightning Safety Awareness Week across the United States.  North Carolina is no stranger to the dangers of lightning; “Storm Data” ranks North Carolina sixth in the U.S. for the number of lightning fatalities between 1995 and 2010. This week the Insurance Information Institute reported that North Carolina ranked third in the nation in the number of lightning-related insurance claims filed by property owners.  To further investigate this problem, NWS Raleigh constructed an eight year (2003-2010) cloud-to-ground (CG) lightning climatology using lightning data from the National Lightning Detection Network (NLDN) to explore the influences of the season, time of day, various geophysical features, and mesoscale processes on the spatial and temporal distribution of CG lightning across the state.

2003-2010 average annual flash density in flashes/km2.

2003-2010 average annual flash density in flashes/km2.

An 8-year average annual flash density analysis for North Carolina was created and it shows some interesting features.  The largest flash densities are located in the southern Coastal Plain, the coastal region, and the Sandhills where sea breeze boundaries, the Sandhills convergence zone, and the Piedmont trough provide a focus for convection. The greatest flash density across the state is located in the southern Coastal Plain with the lightning capital of NC located near Tabor City, located south of Lumberton, near the SC border. The elevated amounts of lightning across the southern coastal region likely results from a coastline orientation that promotes inland penetration and collisions of sea breeze boundaries. The maximum in the Sandhills region likely results from enhanced surface convergence along the clay-sand soil transition zone on the western perimeter of the region referred to as the Sandhills convergence zone (Wootten et. al 2010). In addition, the Piedmont trough (Koch and Ray, 1997) can be identified with the localized maximum that stretches from northeast to southeast from Raleigh to Charlotte with a local minimum just to the west.

Percent of annual flashes that occur in a month.

Percent of annual flashes that occur in a month.

An examination of monthly CG lightning statistics reveals that the greatest amount of lightning occurs in July, comprising 33% of the yearly amount, followed by 23% in August and 22% in June.  June, July, and August account for 78% of the annual statewide lightning while November, December, January, and February combine for only 1% of the annual total. The monthly percentage of all flashes loosely resembles a bell shaped distribution with a more gradual increase in flashes during the spring and a dramatic decline in flashes from August to September likely resulting from the climatologically drier fall.

Statistical point data was computed for eight selected cities using a 25 km2 grid box centered over the associated airport location (AVL, CLT, ECG, EWN, FAY, GSO, ILM and RDU). Examining the location specific daily data reveals that Wilmington has the greatest number of total strikes per year while Greensboro has the fewest.  The top three locations in number of strikes per year; Wilmington, Fayetteville, and New Bern are located in southeastern NC where climatologically there is the greatest instability and sea breeze boundaries, the Sandhills convergence zone, and the Piedmont trough provide local foci for convection. Asheville has the greatest number of days with lightning strikes (nearly 57 which is 5 more days than the next city, Wilmington). Interestingly, Asheville had the second fewest number of strikes per year with Wilmington having the most. All 8 locations experience days with excessive lightning with 50% of the total annual lightning occurring on just 4 to 6 days.

Chart of daily lightning statistics from 2003-2010 for 8 cities across NC that includes the total number of CG lightning strikes, average number of strikes per year, average number of days per year with strikes, as well as exceedance and daily threshold counts.

Chart of daily lightning statistics from 2003-2010 for 8 cities across NC that includes the total number of CG lightning strikes, average number of strikes per year, average number of days per year with strikes, as well as exceedance and daily threshold counts.

Percent of annual flashes that occur during a particular hour.

Percent of annual flashes that occur during a particular hour.

The statewide percent of all CG strikes per hour indicates that the average peak hour is 21 UTC or 5pm EDT when 12% of all CG strikes occur.  More than 59% of all strikes occur between 18-23 UTC while the fewest number of strikes occur during the 13 UTC hour.

Average flash density in flashes/km2 for a specific hour.

Average flash density in flashes/km2 for a specific hour.

Hourly flash density plots show an interesting evolution in the diurnal cycle of thunderstorms in NC. The hourly charts show that lightning typically develops across the mountains and the favored sea-breeze locations along the southern coast during the 17 and 18 UTC hours when local forcing for ascent from differential heating in the mountains and sea-breeze convergence along the coast helps initiate convection. During the next few hours, lightning likely connected to the sea-breeze moves inland slowly while lightning across the southern mountains near Asheville moves east toward Charlotte and Hickory. Eventually the signal becomes less clear as the lightning converges toward the central part of the state toward evening.

The diurnal evolution described above can also be seen in hourly lightning counts at Asheville, Raleigh and Wilmington. Not the very focused distribution of hourly lightning at Asheville and a somewhat similar but not as focused distribution at Wilmington with both locations showing a dramatic ramping up of lightning in the early afternoon.  In contrast, Raleigh shows a much flatter distribution and more general increase and decrease owing to the varied mechanisms for initiating and maintaining convection.

Average hourly lightning counts at Asheville, Raleigh and Wilmington.

Average hourly lightning counts at Asheville, Raleigh and Wilmington.

Examining the location specific hourly data reveals that the peak hour at Asheville and Wilmington occurs the earliest, at 19 UTC or 3pm EDT.  The three locations in central NC, Charlotte, Greensboro, and Raleigh had the latest peak time, averaging around 22 UTC or 6pm EDT.

Chart of hourly lightning statistics from 2003-2010 for 8 cities across NC that includes the hour with the most CG lightning strikes, the hour with the fewest strikes, the percent of strikes in the peak hour, and the percent of strikes during various 4 hour windows.

Chart of hourly lightning statistics from 2003-2010 for 8 cities across NC that includes the hour with the most CG lightning strikes, the hour with the fewest strikes, the percent of strikes in the peak hour, and the percent of strikes during various 4 hour windows.

Additional Information

A summary of the lightning climatology is available in this presentation:
http://www.erh.noaa.gov/rah/science/presentation.20120315.central.nc.ams.lightning.climo.ppt

Additional analysis images including statewide seasonal, monthly and hourly charts along with other imagery from the project are available at the following URL:
http://www.erh.noaa.gov/rah/lightning/

References

Koch, S. E., and C. A. Ray, 1997: Mesoanalysis of summertime convergence zones in central and eastern North Carolina. Wea. Forecasting, 12, 56–77.

Wootten, A., S. Raman, and A. Sims, 2010: Diurnal variation of precipitation over the Carolina Sandhills region.  J. Earth Syst. Sci. 119, No. 5, October 2010, pp. 579–596.

Advertisements
This entry was posted in Lightning. Bookmark the permalink.

2 Responses to An 8 Year Lightning Climatology of North Carolina

  1. A very interesting study, JB! Do you have any sense that there’s a bias in the NLDN over the mountains? There have been times this summer when I was reasonably certain that a strike had occurred and yet it did not register on the NLDN. Thank you for any insights you can provide!

  2. Nice work, thanks for sharing!

    There are many similarities, but a few noticeable differences, between this work and a reflectivity-based study performed by Parker and Ahijevych (2007: “Convective episodes in the east-central United States. Mon. Wea. Rev., 135, 3707-3727”). The PA07 results (e.g. their Figure 3) show a more pronounced minimum extending from GSP up toward Richmond, VA, and a more uniform zone of moderately-high frequency throughout the Sandhills and Coastal Plain. In response to Doug Miller’s question, above, the PA07 study also found a local minimum in storm occurrence over the Appalachians. Storm frequency there is very strongly diurnal, such that frequencies there in the 18-21Z period approach those found along the Atlantic Coast (e.g. PA07’s Figure 4). However, for the remainder of the 24 hour period, storm frequencies over the mountains appear to be quite low.

    As for any remaining discrepancies between the present work and PA07, I think there are a few possible explanations. One is that there may be some artifacts from varying NLDN detection capability (as mentioned by Doug above). Another possibility is that differences in radar calibration, plus the use of composite (i.e. vertical maximum) reflectivity, are masking the true distribution of convective storms in the PA07 study. Or, I suppose, a final possibility is that convective storms produce more or fewer CGs in the storms that typify different parts of the state (i.e. perhaps there is an extravagant meteorological explanation for the discrepancies).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s