*Note that this website is an expanded version of the paper Stuart et al. (1998) in the Pre-print volume of the 19th Conference on Severe Local Storms, to be presented in Minneapolis, Minnesota during the week of September 14-18, 1998.

Here are the two lightning images that prompted our study, courtesy of Global Atmospherics.

Note the maxima around Hampton Roads for both time periods (click on images to enlarge).

• Note the units of Flashes per square kilometer annually.
• A maximum of 6+ flashes per square kilometer is apparent in both images in approximately the same area of Hampton Roads.
• These consistent lightning maxima suggested to us that something significant occurred in Hampton Roads during these periods of record.

Before we look at some individual cases, there are some important points that must be considered: (click to enlarge)

Additional notes to Important Considerations

• This study is based solely on the concept, that the process of collision and coalescence of frozen particles in thunderstorms, is the primary mechanism for the charge separation that produces lightning in thunderstorms (Dye 1990).
• Furthermore, our study is based on the concept that enhanced thunderstorm updrafts are most likely to penetrate the freezing level, generally increasing the production of frozen particles, consequently increasing the collision and coalescence of frozen particles.
• Through the summer of 1998, we expanded our study from events of >5000 lightning strikes to >2000 lightning strikes, again an arbitrary number, which may be further adjusted.
• Many events in 1994 occurred before the Wakefield, VA WSR-88D (KAKQ) was accepted.
• We will be including as many 1994 events as possible as this study progresses.
• One flash event possibly being composed of multiple strikes is important to consider as multiple strikes can be much more hazardous than just a single stroke, and some of the lightning events in this study may have been characterized by a significant number of multiple strikes.

Another important consideration:

What if there are more thunderstorms annually in Norfolk, compared to, say, Richmond, then wouldn't that account for some of the increased lightning activity in Hampton Roads? Hopefully, these 2 graphs will help answer that question (click to enlarge):

• Warm Season Thunderstorm Days - Warm season is defined as May through September, when greater than 90% of the total annual lightning occurs.
• Richmond and Norfolk essentially had the same number of warm season thunderstorms for the 1990-1997 period of record.
• The number of thunderstorms from October through April is negligable for both locations.
• Warm Season Lightning Flashes - Warm season is defined as May through September, when greater than 90% of the total annual lightning occurs.
• Norfolk had significantly more lightning during the 1990-1997 period of record than Richmond, particularly before the radar period of record, but Norfolk had significantly more lightning in 1996, which included many of our studied events.
• Lightning events from October through April are negligable, in fact, only one March event was included in the study and it was nearly last on our list of events of greater than 2000 strikes.

We evaluated KAKQ VAD Wind Profile (VWP) data in order to categorize the lightning events by mean wind flow. The three main categories are:

• West to Northwest dynamic flow - mean wind >15 knots through a deep layer
• Weak deep flow - winds of varying direction through a deep layer, <15 knots
• Southwest dynamic flow - mean wind >15 knots through a deep layer

• An important point to consider is that low-level flow may be weak in dynamic west to northwest and southwest deep flow events, since winds above the boundary layer (above 4,000 to 5,000 feet) often determined the overall mean wind.
• Note that most of the events occurred in west to northwest mean flow.
• Weak low-level winds are important, as will be illustrated in the case studies and conclusions.

Now, here are a few cases (click on images to enlarge).

June 24, 1996 - This was our number 1 lightning event with over 25,000 lightning strikes in roughly 18 hours. Important points:

• Base Reflectivity
• Scattered pre-frontal convection occurring over Hampton Roads, where the 4 and 7 year lightingin maxima are.
• Maximum reflectivities at the 0.5 degree elevation angle are 61 dBZ.
• Composite Reflectivity
• Pre-frontal convection near Norfolk has 65+ dBZ and is tracking through the area of the 4 and 7 year lightning maxima.
• Maximum reflectivities were 66 dBZ in this image, which means that these high reflectivities were above the 0.5 degree elevation angle.
• Higher reflectivities aloft suggests the presence of hail in the thunderstorms, and in fact widespread severe hail occurred during this outbreak.
• VAD Wind Profile
• Low-level winds through 4000 feet were south to southwest.
• However, the mean wind through 12,000 feet was west to northwest, since a deeper layer was west to northwest and at higher speeds.

July 18, 1997- Over 3,600 lightning strikes occurred in roughly 5 hours. Important points:

• Base Reflectivity
• Quite a bit of ground clutter around the radar site, into parts of Hampton Roads.
• Careful examination of the area just east and northeast of the clutter shows a weak boundary extending northward out of Hampton Roads.
• More importantly, there is an indication of two boundaries intersecting over Hampton Raods near the area of the 4 and 7 year lightning maxima.
• Composite Reflectivity
• Even though the ground clutter is more apparent, the weak, intersecting boundaries are a little better depicted.
• Convection did develop later along this intersection, but the boundaries were not apparent at that time.
• VAD Wind Profile
• The flow was very weak through a deep layer, and was generally north to northwest above 5000 feet.

• resulting in higher reflectivities aloft
• increasing the size and number of frozen particles in thunderstorms
• increasing the charge separation in thunderstorms
• increasing the frequency of lightning

May 30, 1998 - Not a part of the 1994-1997 data set but best depicted the local boundaries often present during weak flow regimes

• Base Reflectivity
• Note one prominent boundary nearly parallel to the James River, and the other just northeast of the James River.
• Note the boundaries intersecting in the area of the 4 and 7 year lightning maxima.
• VAD Wind Profile
• The flow was very weak through a deep layer, around 5 knots through 12,000 feet, and quite variable in direction.

• resulting in higher reflectivities aloft
• increasing the size and number of frozen particles in thunderstorms
• increasing the charge separation in thunderstorms
• increasing the frequency of lightning

July 31, 1996 - Over 10,000 lightning strikes occurred in roughly 34 hours. Important points:

• Base Reflectivity
• Parallel lines or "streets" of targets, possibly moisture, are present around the radar site, oriented parallel to the mean wind in the VAD wind profile (Hardy 1968).
• Some convection has formed within the "streets" through the process of interaction of horizontal rolls (Hardy 1968).
• Main area of convection associated with the surface front was well north and west of Hampton Roads, and produced some outflow boundaries.
• Maximum reflectivities at the 0.5 degree elevation angle are 58 dBZ.
• Composite Reflectivity
• Convection within all the convection exhibited reflectivities as high as 68 dBZ, 5 to 10 dBZ higher than at the 0.5 degree elevation angle.
• This pre-frontal convection within the moisture streets was tracking toward Hampton Roads.
• Higher reflectivities aloft suggests the presence of hail in the thunderstorms, and in fact hail did occur during this outbreak.
• VAD Wind Profile
• Southwest flow of 20 knots or greater through a deep layer.

Southwest flow events were usually characterized by pre-frontal lines of convection, but only the dynamic southwest flow events produced enough lightning to be included in our study. The multiple lines of convection likely contributed to the overall lightning frequency in any area during any particular southwest flow event.

So what did we learn, what benefits do we expect from this study, and what are our prospects for future research?

From left to right: Preliminary Conclusions, Benefits to NWS Wakefield, Benefits to Virginia Power, and Future Research (click to enlarge)

• Additional notes on Preliminary Conclusions
• As stated earlier, in dynamic west to northwest flow or southwest flow events, winds below the boundary layer in some cases were less than 15 knots.
• When surface winds were 10 knots or less, the KAKQ WSR-88D was most likely to depict the intersecting boundaries.
• Once convection developed and increased in areal coverage, outflow boundaries and local "James River" or Chesapeake Bay Boundaries were not well depicted, if at all.
• Additional notes on Benefits to NWS Wakefield
• Much of this summer lacked in widespread or severe convection, but in some cases, enhanced wording for lightning has been included in local statements from NWS Wakefield.
• Once we have investigated other meteorological parameters and determine a more precise relationship between these parameters and lightning production/frequency, NWS Wakefield and Virginia Power will embark on an organized public education campaign.
• Emergency Managers and local law enforcement will be notified of increased lightning threats.
• Perhaps a special product could be created to describe daily lightning threats, similar to what NWS Melbourne, FL is currently doing (Sharp et al. 1998).
• Additional notes on Benefits to Virginia Power
• Many of these benefits are related to more effective use of repair crews and reducing overtime.
• Unexpected events cost Virginia Power significant amounts of money, but effective use of resources before an expected event can save significant amounts of money.
• Greater strength arresters are basically systems that are less likely to be damaged by acts of nature. In fact, some stronger strength arresters have already been placed in some areas due to ice storm and winter weather threats.
• Additional notes on Future Research (some parameters are based on Anderson and Charlton 1990)
• Many of the parameters we will be investigating are related to enhanced updrafts and their relationship to production of frozen particles such as hail.
• Production of frozen particles in thunderstorms doesn't necessarily result in hail reaching the ground. In fact, hail does not reach the ground in many thunderstorms due to melting.
• Advection of frozen particles from upstream convection (upper-level winds advecting ice crystals from upstream storms) will be difficult to evaluate, but we will determine how to approach this issue.
• Affects of condensation nuclei such as salt particles near the Bay and Ocean, or pollutants near urban centers, may be addressed far in the future, but for now, no means exists to determine if condensation nuclei contribute to frozen particle production or charge separation in thunderstorms (Dye 1990).
• Lower freezing levels (or wet-bulb zero (WBZ) levels) can result in increased production of frozen particles, as updrafts are more likely to penetrate the freezing level (or WBZ level).
• Advection of colder or dryer air at mid levels can lower the freezing level (or WBZ level), thus possibly increasing production of frozen particles. Hence, mean wind flow is a starting point in determining temperature advection at mid levels.
• Study of Convective Available Potential Energy (CAPE) and surface/layer-based Lifted Indices can indicate potential strength of updrafts and combined with freezing level (or WBZ level) information, can indicate potential for production of frozen particles.
• Once we have analyzed meteorological parameters that we have determined to contribute to production and frequency of lightning, we will formulate a "Lightning Index".
• A "Lightning Index" would likely be based on specific thresholds of meteorological parameters, and offer some form of a prediction of potential for production and frequency of lightning on a daily basis.
• A probablistic approach may be necessary for the "Lightning Index", meaning that a range of possibilities is offered in the form of percent possibilities for certain frequencies of lightning, obviously this concept is in its infancy stage.

References

Anderson, Kerry, and Robert Charlton, 1990: Predicting Lightning Occurrence and Frequency from Upper Air Soundings
over Stony Plain, Alberta. Preprints, 16th Conference on Severe Local Storms, Amer. Meteor. Soc., Kananaskis

Park Alberta, Canada, J40-J45.

Bluestein, Howard B., 1993: Synoptic-Dynamic Meteorology in Midlatitudes Vol. II, Observations and Theory of Weather
Systems. Oxford University Press, Oxford, 594 pp.

Dye, James E., 1990: Cloud Physics and Cloud Electrification What are the Connections? Preprints, Conf. On Atmos.
Electricity, Amer. Meteor. Soc., Kananaskis Park Alberta, Canada, 687-691.
Hardy, Kenneth R., 1968: CPS-9-Radar Investigation of Clear Air Convection. Preprints, 13th Radar Meteorology
Conference, Amer. Meteor. Soc., Montreal, Canada, 236-241.
Sharp, David W., 1998: Implications Involving the Issuance of Public Lightning Advisories for Florida. Preprints, 19th
Conference on Severe Local Storms, Amer. Meteor. Soc., Minneapolis, Minnesota, In Press
Stuart, Neil A., Hugh D. Cobb, Wayne F. Albright, A. Todd Anderson, James Browder, 1998: Correlating Thunderstorm
Lightning Patterns with WSR-88D Signatures and Resultant Benefits for Utility Companies: A Preliminary

Investigation of the Hampton Roads "Hot Spot". Preprints, 19th Conference on Severe Local Storms, Amer.

Meteor. Soc., Minneapolis, Minnesota, In Press

Comments and suggestions please e-mail Neil Stuart

Link to imprtant privacy statement