National Weather Service United States Department of Commerce

Storm Electricity Aspects of the
Blackwell/Udall Storm of 25 May 1955


Don Burgess
University of Oklahoma (CIMMS)
Retired from the National Severe Storms Laboratory


One interesting aspect of the Blackwell/Udall storm was its unusually vigorous electrical activity and the research/scientific controversy associated with analysis of electrical data from the storm.  A number of publications that analyzed, reanalyzed, the tried to explain the data appeared in scientific literature for over 20 years.  The Blackwell/Udall electrical activity was so intense that it led to an electrical heating theory for tornado formation.  That theory remained viable until better data was obtained in the 1970s…again in Oklahoma.

Early in its life the Blackwell/Udall storm featured frequent, nearly continuous, intracloud lightning that was observed by this writer (then an 8-year-old boy), his family and many other families that gathered to watch the storm at the west edge of Stillwater, OK.  Surprisingly, there was not much observed cloud-to-ground lightning at that time.  Later, as the nighttime Blackwell tornado formed south-southeast of Tonkawa, it was made visible by frequent lightning flashes.  At Blackwell, very frequent cloud-to-ground lightning was observed ahead of the tornado, and unusual electrical activity was seen in and around the tornado.[1][2][3]  Very bright electrical discharges were seen within the funnel and ground-originating corona current (also known as St. Elmo’s Fire) was seen just ahead of the tornado.  The Udall tornado was also made visible to observers all along its path to locations northeast of Udall by the occurrence of frequent lightning.[3][4]

There were two types of electrical measurements taken on the storm, both cutting edge research for the mid-1950s.  Sferics and radar measurements were taken by the Electrical Engineering Department at Oklahoma A&M College (now Oklahoma State University) in Stillwater.[5][6]  Advancing sections of the developing lightning bolt/channel produce bursts of electromagnetic radiation known as “sferics.”  Multiple sferics bursts occur with a single lightning flash.  The U.S. Weather Bureau recorded electric field mill observations from a network of 10 stations that stretched east-west across far southern Kansas and included the Oxford, KS area, close to the path of the Udall tornado.[7]  Electric field mills record charge buildup and lightning discharge above the sensors.  Both electrical measurement systems recorded rather extreme rates of lightning discharges associated with the Blackwell/Udall storm.  The electric field mill data indicated storm flashing rates of 10 to 20 per minute, but the sferics data indicated flashing rates of somewhere between 60 and 1500 flashes per minute…a flashing rate which at the time was the highest ever recorded.  Therein lay the scientific debate.  If the flashing rate was as high as indicated by the sferics measurements and concentrated at the exact azimuth of the tornado (as apparently measured), enough heat might be generated to create tornado-scale pressure drops and vertical motions.  In fact, the visible signature of the extreme intracloud flashing rates seen in the convective tower before the tornado was dubbed the Tornado Pulse Generator [6] and was mentioned to storm spotters to use as a tornado precursor.  Also, since sferics are observed in radio and TV reception bands, some emphasis was placed on use of radio/TV interference as tornado precursors.

During the 1960’s, numerous papers were published, both supporting and showing exception to lightning-induced tornadogenesis and sferics warning systems.  Finally, at the end of the decade, the Wave Propagation Laboratory (Boulder, CO) and the National Severe Storms Laboratory (Norman, OK) began a careful, systematic data collection program (with newly-developed and better sensors) in Oklahoma.  Central Oklahoma tornadic storms of 29-30 April 1970 were studied, with results suggesting that high flashing rates occurred from the whole storm, not just the tornado area.[8]  A later paper in 1975, reporting on many storms, both tornadic and non-tornadic,[9] found that many tornadic storms produced high lightning flash rates, but some didn’t, and some non-tornadic storms also produced high flash rates.  The scientific debate was completely put to rest by analyses of newly developed Doppler radar and electrical measurements from the Union City, OK tornado of 24 May 1973.  The Union City measurements showed conclusively that the enhanced lightning activity originated with the parent storm as a whole, not with the tornado itself.[10]

Today, we know that charge separation leading to lightning flashes occurs as a result of collisions between different types of precipitation particles.  These collisions are increased as updraft speed increases.  Therefore, lightning flash rates are related to updraft size, speed, and duration.  Up to the current time, unique signatures for tornadoes have not been found in lightning data.  However, we do know that intracloud lightning dominates severe/tornadic thunderstorms, sometimes numbering as many as 90% of the total flashes.[11]  New, experimental lightning mapping arrays that measure total (intracloud and cloud-to-ground) lightning have been developed and installed at a few locations in the United States.[12]  One such array has been deployed by the National Severe Storms Laboratory and the University of Oklahoma and covers a large portion of the middle of the state.  It is hoped that data from this array can be used to augment Doppler radar and other observations, leading to even more reliable tornado warnings than exist now.


[1] Montgomery, F.C., 1955: Tornadoes at Blackwell, OK, May 25, 1955.  Monthly Weather Review, 83, Pg 109.

[2] Montgomery, F.C., 1956: Some observations of the tornado at Blackwell, Oklahoma 25 May 1955.  Weatherwise Magazine, June, Pg 97, 101.

[3] Phillips, V.V., J.G. Galway, and D.M. Hanson, 1955: Tornadoes of Blackwell, OK – Udall, KS, May 25, 1955.  Monthly Weather Review, 83, Pg 224, 238.

[4] Staats, W.F., and C.M. Turrentine, 1956: Some observations and radar pictures of the Blackwell and Udall tornadoes of May 25, 1955.  Bulletin of the American Meteorological Society, 37, 495-505.

[5] Jones, H.L., 1958: The identification of lightning discharges by sferics characteristics.  Proceedings of the 2nd Conference on Atmospheric Electricity, Portsmouth, NH, Pergamon Press, Pg 543-556.

[6] Jones, H.L., 1965: The tornado pulse generator.  Weatherwise Magazine, April, Pg 78-79,85.

[7] Gunn, R., 1956: Electric field intensity at the ground under active thunderstorms and tornadoes.  Journal of Meteorology, 13, Pg 269-273.

[8] Taylor, W.L., 1973: Electromagnetic radiation from severe storms in Oklahoma during April 29-30, 1970.  Journal of Geophysical Research, 78, Pg 8761-8777.

[9] Taylor, W.L., 1975: Detecting tornadic storms by the burst rate nature of electromagnetic signals they produce.  Preprints of the 9th Conference on Severe Local Storms, Norman, OK, American Meteorological Society, Pg 311-316.

[10] Brown, R.A., and H.G. Hughes, 1978: Directional VLF sferics from the Union City, Oklahoma, tornadic storm.  Journal of Geophysical Research, 83, Pg 3571-3574.

[11] MacGorman, D.R., and W.D. Rust, 1998: Severe, winter, and tropical storm systems.  The Electrical Nature of Storms, Oxford Press, New York, Pg 235-301.

[12] Krehbiel, P.R., R.J. Thomas, W. Rison, T. Hamlin, J. Harlin, and M. Davis, 2000: GPS-based mapping system reveals lightning inside clouds.  EOS, Transactions American Geophysical Union, 81, Pg 21-25.