The Cheyenne Tornado
July 16 1979

Jonathan D. Finch

On July 16, 1979, the most damaging tornado in Wyoming history touched down 3 miles west-northwest of the Cheyenne airport. A severe thunderstorm watch was issued by SELS well before the tornado occurred. The tornado occurred close to the severe thunderstorm watch area. This strong tornado moved east or east-southeast across the northern part of Cheyenne, causing $22 million in damage and 1 fatality. 140 houses and 17 trailers were destroyed. 325 other houses were damaged.  Four C-130 aircraft and National Guard equipment sustained $12 million damage. Municipal hangars and buildings suffered another $10 million damage. There were no hail reports with this storm or any other storm that day in Wyoming. A woman was killed instantly by lightning after climbing to the top of Pingora Peak (11884 ft) about 28 miles west of Lander.

On July 14 a strong shortwave trough and associated jet streak was moving across the northern plains as shown by the 00 UTC July 15 500mb and 250mb charts. Behind this feature across Wyoming, a cold front was already beginning to surge down the plains as of 21 UTC. Severe weather actually developed by early afternoon in western SD and Nebraska and in the evening in Colorado. The warmest part of the 700mb warm plume at 00 UTC July 15 extended from central NM to northern Colorado into southern Utah and eastern NV. Late on July 14th a cold surge developed primarily in the lee of the Rockies. By the end of July 14th (06 UTC July 15th), the cold front had already surged into northern Colorado. A front that was enhanced by outflow was located across southern Nebraska into central Missouri.

By 12 UTC July 15, the surface to 700mb cold surge was apparent. The cool air was actually spilling over the Rockies into southwest Wyoming and central CO. The shortwave trough (500mb)  and associated jet streak that moved across the northern high plains the previous evening was located over Minnesota and Iowa. The front that was draped across southern Nebraska and Missouri at 00 UTC July 16 was surging into central Kansas by 12UTC.

By 00 UTC July 16, low-level moisture was beginning to surge upslope (to the southwest) from northwest Kansas into eastern Colorado. The surface front was stalling out across central New Mexico. An outflow boundary stretched from near Dodge City to near Tulsa. Another cluster of convection in the Texas Panhandle resulted in ain cooled air at Amarillo. The front in New Mexico was about to undergo frontolysis since it had already moved to near the mean ridge position at 500mb. The northern plains frontal boundary tightened considerably on July 15 as the elevated mixed layer hot plume shifted north and the low level cool dome shifted south. 700mb temps on the 15th ranged from about 3 C around Sheridan to around 15-16ºC at Jackson Hole. With  no upper air data between Boise and Lander, one may ask how I analyzed +15ºC 700mb temps across western Wyoming. Well, I was able to make judicious use of surface data to fill in details where upper air data were lacking. This wedge of warm air was analyzed based on the surface tempertatures at Jackson Hole and Yellowstone. Afternoon surface temperatures at Jackson Hole and  West Yellowstone  were lifted up to 700mb to yield the 700mb temperatures above these stations. Sometimes the low-levels become superadiabatic, yielding a 700mb temperature that is too warm. So the best method is to use the  mixout temperature, whether this occurs at local noon (in windy conditions) or sometime in the early afternoon in lighter wind conditions. In most of my front range tornado cases, surface winds under the mid level hot plume are breezy to windy and afternoon dewpoint depressions are fairly high. Surface winds on July 16, 1979 were not as strong. Also notice the 700 mb baroclinic zone across the midwest and northern plains. There is upper level support for a front across this area, but not further south in New Mexico. By the end of July 15 (06 UTC July 16), the front over New Mexico is about to undergo frontolysis. The surface winds over eatsern Colorado were becoming more southeasterly, resulting in strong moisture transport onto the front range. The surface dewpoints at Denver and Akron were up to 59F and 60F respectively.

By 12 UTC July 16, the front over New Mexico was undergoing frontolysis. A 700 mb baroclinic zone was located across Iowa. The surface front associated with this baroclinic zone was further south as one would expect in Iowa Illinois and Missouri. A shortwave trough was evident from the 500 mb chart over Montana and northwest Wyoming. The tail end of the upper jet streak was located over eastern Montana. The surface dewpoints at Sidney, Akron, and Cheyenne were up to 60F, 61F and 56F respectively. Note again that an adjustment of 7F was made to the Sidney dewpoint. Strong surface pressure rises were occurring at Dodge City and about to occur at Garden City at 12 UTC but I am not sure why. Accas was reported at Cheyenne on the morning of the 16th after the low clouds burned off. Therefore, high lapse rates were likely present along the eastern extremity of the warm plume(elevated since low levelcooling occurred at Cheyenne overnight).

From 17 to 20 UTC July 16, an area of thunderstorms moved east-southeast across Nebraska, leaving an outflow boundary trailing back to the west into southeast Wyoming. Frontogenesis was also occurring across southern Wyoming and northeast Colorado. By 20 UTC, the leading edge of a strong pressure gradient marked the location of the new surface frontal boundary. This makes sense based on the position of the upper jet at 12 UTC. I was able to plot the location of the outflow boundary at around 15 UTC using time series of surface observations from Scottsbluff, Alliance, Mullen and Ainsworth, NE. 17 UTC   18 UTC   19 UTC  20 UTC. It is possible that this boundary sagged into the Cheyenne area around the time of the tornado. A satellite loop of this event is here.

By the afternoon of  July 16, a well defined frontal boundary extended from east of Yellowstone to east of Jackson Hole to west of Lander to near Rawlins, to near Laramie to near Cheyenne. Note that Lander was north of the front all day on July 16 since the winds were light and the dewpoints were high (uper 40s F). The boundary was undoubtedly situated west of Lander where the woman was struck and killed by lightning. The observations at Rock Springs show the frontal boundary to the north of them in the afternoon. Then the wind shifted to the northeast at 00 UTC signifying a frontal passage.

The mid level warm plume shifted east from July 15 to July 16, with 700 mb temps warming to +18 to 19ºC across parts of southwest and south central Wyoming and adjacent northern Colorado. A strong sfc-700 mb baroclinic frontal zone was apparent across Wyoming. The 00 UTC 17th Denver sounding showed +14C at 700 mb. However, this sounding was contaminated by convection. Since Denver mixed out completely by 20 UTC with a surface temperature of 93ºF, the 700 mb temperature was surely 17C just before the cool outflow arrived. The 700mb temperature at Grand Junction at 00 UTC July 17 was 17C. The maximum temperature was 97ºF. If you lift a parcel dry adiabatically from the surface (856 mb) to 700 mb you get 18ºC. Therefore it is typically better to subtract 3ºF from the high temperature before lifting to 700 mb when determining 700 mb temperatures, especially in light wind situations. In strong wind and very dry situations at elevations above 6000 ft, the local noon temperature would give a closer approximation. In this case I used the maximum temperature (ºF) or maximum temerature (ºF -3) depending on the station) to augment the 700 mb chart.  700 mb temperatures in the hot plume on the morning (12 UTC) of the 15th and 16th were about 2 to 3ºC cooler than at 00 UTC. This is because the diurnal cooled layer extends up beyong 700 mb at stations above 5000 ft. This is obviously not the case at high plains stations such as Amarillo and Dodge City. At these locations, the hot plume sometimes actually moves east away from the rockies between 00 and 12 UTC, yielding very warm 12 UTC 700 mb temperatures.

The maximum temperature at Jackson Hole was 87ºF. Lifting 87ºF at 805 mb up to 700 mb yields a 700 mb temperature of 18.5ºC. However, you can see that Jackson Hole had not "mixed out" by local noon since they were stuck in a fairly fairly light wind regime. Therefore, I decided to use MaxT (F) -3 = 84ºF as the mixed out temperature. This yields a 700 mb temperature of 16.5ºC. The same procedure was used for Yellowstone.  Vernal, UT as well as much of northern Utah is lower in elevation. I used the maximum temperature there to find the 700 mb temperature.

Theta-e Comparison

The first indication of a tornado was 3.5 miles west-northwest of the Cheyenne airport at 335 pm MDT (2135 UTC). The 22 UTC surface chart shows the front just south of Laramie. The T/TD were 71ºF/56ºF. At this time the tornado was probably just finishing its rampage across  northern Cheyenne.  By mid-summer standards in the central or eastern USA this would be a very cool temperature and a low dewpoint such as would be found behind a strong cold front. However, the elevation of Laramie is 7270 ft. The author of this page believes that the theta-e was similar near Cheyenne and Laramie on the immediate cool side of the boundary. Despite the T/TD being 12ºF/16ºF lower at Laramie than Topeka, the theta-e was actually slightly higher at Laramie than at Topeka.

Since the storm was near the boundary, surface based CAPE estimations were made on both sides of the boundary. To estimate the CAPE on the cool side of the boundary I used the 22 UTC surface observation at Laramie. This was the only high elevation location in the most air immediately north of the boundary at 22 UTC. The T/TD were 71ºF/56ºF with a  surface pressure of 788 mb. I used this information along with nearby soundings from North Platte, Lander and Rapid City, Dodge City and Denver to create an approximate sounding for the "cool" side of the front. The surface based CAPE was around 2800 j/kg at 22 UTC. A T/Td of 71ºF/56ºF at Laramie would give the same theta-e as 77ºF/57ºF near Cheyenne. At Cheyenne it was 81F in the early afternoon. But immedialtely north of Cheyenne it was surely cooler, with much higher dewpoints. In fact, the dewpoint jumped to 56ºF at Cheyenne by 22 UTC, but the temperature was cooler due to nearby thunderstorms.  Just south of the outflow boundary or front, surface based CAPE values were lower.  The T/TD at the Cheyenne airport at 20-21 UTC were 80ºF/50Fº and 81ºF/49ºF. Using the same thermal profile aloft(500-200 mb), the surface based CAPE at Cheyenne was 1500 j/kg. Keep in mind that the Cheyenne tornado moved just north of the airport, possibly along the surface boundary. So the storm could have been ingesting air from both sides of the boundary.

I do not have enough information to determine if the storm was realizing the higher CAPE just north of the boundary, or if the storm was mainly ingesting air from the warm side. If the former is the case then the storm would have been lower based and the vertical wind shear would have been stronger given the easterly winds and higher dewpoints in the cooler air. I cannot reliably compute the shear profile or storm relative helicity. On the front range and mountain region, it is very difficult to approximate the 1 to 2 km AGL wind profile. While the surface wind and winds at and above 500 mb (~4 km) might be known in this case, wind speed and direction approximations from just above the surface to 3 km are not very accurate.