Lightning Strike Frequency vs Severe Thunderstorm Frequency

[Frank Kienast]

I've been looking at maps of lightning strike frequency and comparing them to severe thunderstorm frequency, and am a bit surprised at what I see. For reference, here is a lightning strike density map for the US:
http://www.weather.gov/os/lightning/lightning_map.htm
And here is a severe thunderstorm watch frequency map for the US:
http://en.wikipedia.org/wiki/File:A...torm_watch_frequency_in_the_United_States.svg
Granted, this a map of severe thunderstorm watch frequency, but presumably there is a strong correlation between watch frequency and severe event frequency.

I'm surprised by the lack of correlation between lightning strike frequency and severe thunderstorm frequency.

I would think it is certainly a fair statement that severe thunderstorms tend to have tall tops. How often do you see a warning, for example, for a thunderstorm that is just 25 or 30 thousand feet tall? And I know lightning frequency is directly proportional to storm top height as well. So severe storms = tall storms = lots of lightning strikes.

But look at some examples on the above maps. Take New York state. Severe thunderstorm watches are most frequent in the eastern part of the state. But lightning frequency is the greatest in the western part of the state. Likewise, look how North Dakota has a very high frequency of severe thunderstorm watches, but a relatively low number of annual lightning strikes.

Can someone explain to me the lack of correlation here? I have an idea myself, but will wait to see what others say first.

Thanks,

Frank Kienast

 

[rdale]

Your assumption of a strong correlation between watch boxes and severe thunderstorms is probably in error. You might be better off at the least comparing to LSRs.

 

[dmckemy]

Florida and the Southeast/Gulf Coast states get a lot of pulse type thunderstorms during the summer -- the lack of shear and high freezing levels usually inhibits any sort of severe thunderstorm development.

Meanwhile, in the Plains and northern U.S., thunderstorms usually occur during the spring and summer months, and often times there is shear to help aid in the organization of severe thunderstorms (especially in the Northern sections of the U.S. during the summer months since they are closer to the jet).

As for why some sections of some states have higher severe t-storms watch box frequency than other parts of the state (like New York)...perhaps topography or the presence of the ocean could have an effect on severe storms? I grew up in Western North Carolina, and the Appalachian Mountains had sizable effect on squall lines and storms that tried to cross it...often, the storms would weaken and die after crossing the mountains.

 

[Jeff Duda]

A storm doesn't have to be tall to produce severe weather and a tall storm won't necessarily produce severe weather. As the tropopause is higher in the summer than winter, most thunderstorms are also taller in the summer than winter, regardless of severity. Also, many low-topped storms can produce severe wind (although I don't have any direct evidence to back up that claim at the moment, but I sense it is true).

You should expect to see some degree of correlation between lightning flash density and severe thunderstorm watch frequency, and to an extent you do. It's just not that strong is all. Just remember that lightning is not a criterion for a severe storm, so you shouldn't expect to see a perfect relationship.

 

[rbstokes]

I've seen many good lightning storms with non severe criteria. For lightning i would look into FZLVL/WB0, as well as PW. Also you need a good charge seperation for a lot of lighning, so a strong updraft. An ordinary cell can produce alot of lightning. Ive seen SVR cells produce occasional lightning, so it all depeneds.

 

[Frank Kienast]

I found a map showing tornado frequency in the US.
http://earthstorm.mesonet.org/materials/graphics/TornadoFreq.gif
Actually, this map does seem to show a lot better correlation to the number of lightning strikes than the severe thunderstorm watch map does. For example, you see lower numbers in North Dakota where lightning strikes are relatively low, and you see a maximum in Florida that corresponds (somewhat) to the lightning strike max in that state.

My theory based on observations from living in Virginia, Indiana, and western New York, was that the most severe storms tend to have their energy concentrated in a smaller area than marginally severe storms. Often when I lived in Indiana, there would be a tornado or larg hail relatively close by, but nearby locations would have nothing at all. In NY and VA, severe weather tends to be less strong but cover a larger area than in IN. While in western NY we do of course get garden variety storms that affect a small area, when there is a severe outbreak it seems to hit a much larger area. So that in say the Dakotas, a severe storm will be much stronger (including a high density of lightning strikes) than here in western NY, but will cover a smaller area. So while the area immediately south of the tornado receives a large number of strikes, surrounding areas get very little. In western NY, if storms reach severe levels, they tend to be in lines or clusters. At any one given location there is not an exceptionally large number of lightning strikes, but the large area covered means a large effect on the strike density for the region.

 

[Dan Robinson]

There are far more sub-severe storm days east and southeast of the Great Plains than severe storm days in the Plains, which explains why the Midwest and even parts of Appalachia have higher strike densities than much of the Plains. Sub-severe storms can produce CG activity equal to severe storms, many times more so. Supercells and many severe squall lines often will have relatively little CG activity compared to their higher-altitude in-cloud flashes.

 

[Frank Kienast]

I'm going to have to start watching lightning strike data more carefully next time supercells develop in the plains.

Back about 20 years ago, when online radar data was very course (think Compuserve "SD" radar text radar reports), lightning provided the most accurate information about storms. At the time I lived in Virginia, and I almost always observed a direct correlation between lightning strike count and storm intensity. In the mid 90's, I moved to Indiana. About that time, high resolution radar data became available, and this provided the most information about a particular storm. So I hardly ever looked at lightning strike data after that. I think I may have made some generalizations from my experience back in VA that are not valid nationwide.

I just found a free source that shows lightning strike density over the last hour and last 24 hours for anyplace in the us:
http://www.strikestarus.com/
So I think it will just be a question of observing for a while.

 

[MClarkson]

For any two storms in similar locations, there will be a decent correlation between more lightning and more severe storms. If you go out on your porch and observe a storm with more lightning than yesterday, it will probably have a higher chance of crossing the severe thresholds than yesterdays. Now there will be exceptions of course, for example a line of relatively weak convection mixing down strong winds aloft and creating widespread high wind reports with little lightning. Changes in freezing level and CCN concentration and water amounts will also create different charge separations, even if updraft strength is the same.

But then if you fly to a completely different location or climate, the triggering mechanism behind the storms can be vastly different. For obvious example summertime afternoon storms in Florida, which are very common, are triggered by smaller scale features often in environments with light winds at most levels, less gusts to mix down and little vertical shear to separate updraft from downdraft. These storms typically cause less severe weather, but due to their frequency large amounts of yearly lightning. Whereas storms in the plains(less frequent) are typically triggered by larger scale forcing in shear profiles much more conducive to severe weather, causing more severe weather with less total yearly lightning. Thus the correlation over widely separated areas between severe tstorms and lightning is not so good.

 

[Keith Wilson]

A good example of severe low top storms could be made with the outer bands of landfalling tropical systems, depending on what you are considering as low topped. Some of these are rotating supercells, and I witnessed this as recently as TS Lee over the Labor Day weekend in Panama City Beach, FL. I measured 70+ mph winds and witnessed rotating wall clouds with two separate storms. Lee wasn't a very impressive storm as they go, either. These storms did not produce lightning and for the most part stayed below 30K'.
An even better example would be Hurricane Ivan in 2004, which was a prolific tornado producer.

Of course, these are warm core storms.