The Cheyenne Ridge Tornado
April 23 1960

Jonathan D. Finch

                                     
rough draft form

Historical Tornado Cases for the Boulder Warning Area

Historical Tornado Cases for the Cheyenne Warning Area

Historical Tornado Cases for the United States

May 7-8  1965 Front Range Tornado

June 14-17  1965 Front Range Superstorm

Elevated Mixed Layer

Elevated Heating

High plains and front range topo maps



Brief Overview


                                        On April 23, 1960 , a thunderstorm developed near Fort Collins, Colorado. This storm moved north-northwest and
                                        and became severe, producing tornado(es) in western Laramie county in Wyoming.  The tornado touched down
                                        southwest of Granite, Wyoming and moved northwest or north-northwest into eastern Albany county (elevation
                                        7600-7900ft). The tornado was observed for about 25 to 30 miles. The tornado(es) was in very rural areas for most
                                        of its life cycle, doing most of the damage near Interstate 80. The tornado did no documented damage as it moved
                                        into the Medicine Bow National Forest (elev. 7700-8200ft) in eastern Albany county. Due to the lack of 
                                        documentation, I didn't draw the tornado path this far north and west.
                                                       

Synoptic setup leading up to the Tornado


                                               
                                        The 00z April 23 1960 500mb chart showed 2 branches of the westerlies, one from Old Mexico into the southern
                                        plains, and another from southern California into western Utah. Moisture was streaming north from the Gulf of
                                        Mexico with dewpoints in the 50sF as far north as Nebraska. Moisture was present north of the surface
                                        front across the nothern plains with 750mb dewpoints/mixing ratios around > 5C / 7.5 g/kg).  I used surface
                                        observations to augment the upper air observations. For example, mid afternoon, mean T/TD values of 73F/47F and
                                        and 75/41 were recorded at Alliance, NE (elev. 4000ft) and Scottsbluff, NE (elev. 4000ft) respectively. The
                                        moist layer probably extended up to 750 or even 700mb. In the case of Alliance, the mixing height would be
                                        5000 ft, assuming the dewpoint was correct. This would mean that the top of the moist layer at Alliance would be
                                        around 9000ft (~725 mb) with a 750mb dewpoint of about 43F(mixing ratio 8 g/kg). The surface observation sheet
                                        for Alliance indicated building cumulus clouds in the remarks. The top of the moist later at Scottsbluff would be
                                        675mb, with a 750mb dewpoint of 37F(mixing ratio 6.2 g/kg).

                                        At 00 UTC April 23,  a weak cold front was about to push through Cheyenne where the dewpoint jumped into
                                        the upper 30s by 01 UTC, along with a wind shift to the northeast.  The airmass north of this boundary in the Nebraska
                                        panhandle was actuallly moderately unstable in the afternoon, but was becoming less unstable by 00 UTC.
                                        Dewpoints were in the 40sF in the Nebraska panhandle.  A sharper frontal zone was located from southern Montana
                                        into northern SD. The surface dryline was rather ill-defined in Colorado but was more pronounced in west Texas,
                                        buldging east of  Amarillo. A cold surge was apparent west of the Big Horns. A pacific cold front was located over
                                       western Utah.

                                        By 03 UTC April 23, convective outflow was apparent in western Kansas, eastern Colorado and southwest
                                        Nebraska.

                                        By 06 UTC, the weak cold front had apparently pushed through Denver as the sea level pressure jumped and
                                        the dewpoint jumped into the upper 30sF. The northern cold front was progressing south into southwest SD.
                                        The northern front moved through Chadron after 3 UTC (22 MST). Note the abrupt wind increase, pressure rise
                                        and drop in dewpoint.   

                                        By 09 UTC, the northern cold front was approaching Scottsbluff. Drier air had filtered into Chadron and Rapid
                                        City. This front was overtaking the southern front. Drier air had filtered into Scottsbluff by 09 UTC. Notice the
                                        drier air, pressure rises and dewpoint drop between 02 MST and 06 MST. The dry air intrustion may have been
                                        fairly shallow.  The area of outflow over northwest Kansas and southwest Nebraska eroded from the south overnight
                                        with southerly low level winds shoving the rain cooled air further north.

                                        The northern cold front passing Cheyenne around 12 UTC was accompanied by low clouds, fog and strong
                                        pressure rises between 0500 and 0600 MST.  
                                       
                                        By 12 UTC April 23, two branches of the westerlies were still evident. A deep 500mb trough was still parked in
                                        the western US. The flow over the Rockies was meridional with the eastern edge of the strong flow from central AZ
                                        into western Colorado and central Wyoming. Ample elevated moisture was still in place north of the surface front, with
                                        700mb/750mb  dewpoints from 3 to 7 C (6 to 8.5 g/kg).

                                        The 15 UTC surface chart showed north winds in eastern Wyoming and western Nebraska. But surface dewpoints
                                        in the cool air were still 40F at Cheyenne(mixing ratio 6.5 g/kg) and 42F (mixing ratio 6.5 g/kg) at Sidney, NE.
                                        Dewpoints were in the mid 40s to lower 50sF immediately north of the surface front in northeast Colorado and 
                                        southwest Nebraska. For example, Imperial, NE and Akron, CO had  dewpoints of  53F and 46F respectively
                                        (mixing ratios 9.2 g/kg and 8.3 g/kg. Thus, given that the mixing ratios at the surface and 750mb were between 7.5
                                        and 9 g/kg east of Cheyenne  in the morning, strong upslope flow would tend to advect this higher theta-e air onto the
                                        front range later in the day. The surface pressure at Cheyenne was about 800mb, so the lowest 50mb of moisture
                                        should have an average mixing ratio of 8 g/kg later in the day.  

                                        By 18 UTC, the surface winds were beginning to turn upslope in western Nebraska, northeast Colorado and southeast
                                        Wyoming. By 21 UTC  the dewpoint at Cheyenne had dropped to 38F due to vertical mixing. However, dewpoints
                                        just southeast and east of Cheyenne still ranged from 49 to 53F.  I do not know if the surface observation at Fraser,
                                        Colorado is reliable, but Fraser seems to be north of the front at 23 UTC. I am suspicious of the dewpoint temperature.
                                        Here is the 22 UTC surface chart.

                                        At 23 UTC, there was a narrow corridor of higher dewpoints from southwest Nebraska and the southern Nebraska
                                        panhandle into northeast Colorado. The dewpoints at Sterling, CO, Imperial, NE and Sidney, NE were 54F,
                                        51F and 49F respectively. The surface pacific cold front was still west of Vernal, UT and Rock Springs, WY and
                                        just west of Grand Junction, CO. I chose to draw the dryline just east of Grand Island, NE since the moisture
                                        was mixing out immediately ahead of the front. This is not the only solution. The surface cold front in Nebraska was
                                        now surging south. In fact the western end of the frontal push was around North Platte. Another baroclinic zone
                                        appeared to be entering the northern Nebraska panhandle.




Lifted Index Technique Without Upper Air Data



                                        I roughly estimated the 500mb temperatures across much of Wyoming and Colorado at local noon based on
                                        a technique that I have been using for several years. This technique was partially based on Dr. Toby Carlson's
                                        pioneering work with the elevated mixed layer(EML) in the mid and late 1960s. Carlson was to first to document
                                        the existence of the EML. He used local noon surface temperatures over the dry and well mixed western US to
                                        calculate the dry adiabat that the well mixed temperature fell on. Hence, if you know the local noon surface
                                        temperature, you can estimate the potential temperature and 700mb temperature since the low to mid levels are
                                        dry adiabatic. In fact, I found that I was able to estimate the 500mb temperatures in very deep mixing situations.
                                        The surface dewpoint depression should be at least 50F and preferably 60F and the station must be in the warm
                                        sector in windy conditions, with a station elevation > 6000ft. Stations such as Casper(5300ft), Rawlins(6813ft),
                                        Laramie(7270ft), Rock Springs(6760ft), Gunnison(7700ft) and Alamosa(7600ft) typically mix out to 500mb by
                                        noon in windy, warm sector situations. Stations such  as Denver(5300ft), Colorado Springs, Albuquerque and
                                        Las Vegas occasionally mix to 500mb, but it generally takes until later in the day. I used the noon local temperature
                                        for the stations that usually mix out quicker. In these cases, the high temperature is typically 1 to 4 degrees higher
                                        than the noon temperature, leading to slighly superadiabatic low levels in the late afternoon. But in windy conditions
                                        the super is usually much less pronounced since heat cannot build near the ground. In the absence of strong cold
                                        advection, these 500mb temperature estimations are still valid for that evening. In the later case(sites that mix
                                        out later in the afternoon) I use a variation of this technique. I subtracted about 3 degrees from the maximum hourly
                                        temperature in the afteroonoon and considered this to be the mixout  temperature. If a station experienced a warm
                                        frontal passage in the afternoon, I use the mixed out temperature following the frontal passage. Of course, once
                                        you know the 500mb temperatures in the region of interest, and the 500mb flow pattern, you can estimate the
                                        500mb temperatures for locations immediately downstream.  In well mixed situations discussed here, the wind is
                                        typically unidirectional from the surface to 500mb.




Lifted Index Calculation Without Upper Air Data

                   1. Using noon surface observations

                                        The 19 UTC surface observations at Rawlins, Gunnison, Rifle, Montrose and Colorado Springs were used
                                        in estimating lifted index in the tornado affected area at 02 UTC.  The following table shows the elevation, station
                                        pressure, T/Td, mixing ratio, surface potential temperature. The 500mb temperature was calculated from these
                                        and are shown in the last column. Alamosa may not have completely mixed out by 19 UTC since max temperature
                                        was 67. So the 500mb temperature approximation may have been too cold there.      

                                        Surface winds in the warm sector (well west of the tornado affected area) were mainly from the south-southwest
                                        and afternoon temperatures did not fall. If 500mb temperatures had been falling, then sfc temps would also be
                                        falling since the vertical temperature profiles were dry adiabatic. There was definitely no cold front aloft as far east
                                        as Cheyenne or the tornado affected area by 01-02 UTC. In the later case(sites that mix out later in the afternoon)
                                        I used a variation of this technique. I subtracted about 3 degrees from the maximum hourly temperature in the
                                        afternoon and considered this to be the mixout temperature.   

                                        The station pressures in millibars were calculated by multiplying the station pressures in inches of HG by 33.86.
                                        The hourly station pressures are available on page B of the surface observation forms.

19 UTC Elev(ft) Pres.(mb) T(F)/Td(F) Max T theta(F) 500mb T
Rawlins 6813 782 66/16 69 104.2 -16
Gunnison 7678 760 63/16 63 105.5 -16
Rifle 5540 820 73/25 104.0 -16.5
Montrose 5759 815 70/25 101.8 -17.5
Alamosa 7539 765 62/10 67 103.4 -16.5
Colo. Springs 6140 803 72/11 75 106.4 -15.5

                                                                            

                                        The surface T/Td/P at Sterling, CO (3900ft), Sidney, NE, Akron and Cheyenne, WY at 19 UTC were
                                        67F/51F/870mb, 65/48/860mb, 66/49/851mb and 66F/38F/804mb. Given the estimate of 500mb temperature
                                        of -15C at Colorado Springs at 19 UTC, an estimate of the lifted indices at Sterling, Sidney and Akron are
                                        -4.5, -3.5, -5 and -4.                                


                   
                    2. Using only the 00 UTC surface observation from Denver

                                        In the case of April 23, 1960, the surface observations at Denver around 00 UTC were used to estimate the
                                        01-02 UTC lifted index at Cheyenne. In this case, Denver mixed out late in the afternoon after a warm front moved
                                        north of the station.


00 UTC Elev(ft) Pres.(mb) T(F)/Td(F) theta(F) 500mb T
Denver 5360 822 77/7 108.3 -14.5


                                        The surface T/Td/P at Cheyenne at 01 UTC were 64F/46F/800mb. Obviously, there was a strong push of most
                                        air into the Cheyenne area. The surface observations from Cheyenne indeed confirm this. Note the windshift to the
                                        east-southeast, along with a moistening and slight cooling between 22Z and 01Z. Using a 500mb temperature of
                                        -14.5C yields a surface based lifted index of  -7.
              
                                                                                                  

                  
                3. Using only the 00 UTC Denver sounding

                                         The 00 UTC Denver sounding(text format) confirmed the above finding that the 500mb temperature was around
                                        -14.5C. Of course, using this sounding with the 01-02 surface observation from Cheyenne gives a lifted index
                                        of -7.
                                       

Elevated Heating on April 23 1960


                                        
                                        As experienced storm chasers are well aware, severe thunderstorms occur on the high plains with much lower
                                        surface dewpoint temperatures than at low elevations. For example, supercell thunderstorms occur in spring and
                                        summer with dewpoints only from 3 to 15C(37 to 59F). Exactly why can we get by with lower dewpoints and still
                                        get explosive storms? The answer is twofold. In answering this question I would like to correct a common
                                        misunderstanding concerning dewpoint temperature. I have often heard something like, "wow Cheyenne has a 60F
                                        dewpoint, thats like a 80 dewpoint at our elevation". This is incorrect. The dewpoint lapse rate is only about 1F /1000ft.
                                        This means that a 50F dewpoint at CYS has the same moisture as roughly a 56F dewpoint at Houston. This
                                        corresponds to a mixing ratio of  9.6g/kg. In June, if the dewpoint at Houston is 56F, then the CAPE is probably
                                       close to zero for any reasonable 500mb temperature. But at Cheyenne, if the wind is from the east or southeast
                                        and the dewpoint is 50F(same amount of moisture), then interesting things can happen. Obviously it is not the
                                        high moisture content that is responsible for a threat of storms at Cheyenne.



                    1. Sounding Comparison--Cheyenne vs Fort Worth                   

                       
                                        First, let's compare the 00 UTC Fort Worth sounding to an approximate 01 UTC sounding for Cheyenne, WY.
                                        I used 01 UTC since the rich moisture arrived in Cheyenne between 00 and 01 UTC.  It is interesting to note
                                        that the Cheyenne sounding starts where the moist layer ends at Forth Worth (800mb).  The boundary layer at
                                        Cheyenne is at the nose of the cap at Fort Worth. The level of free convection(LFC) is actually higher in the Fort
                                        Worth sounding since the cap is so strong there. The LFC is fairly low at Cheyenne owing to high 0-5 km lapse rates.
                                        The surface and boundary layer mixing ratios are lower on the Cheyenne sounding than on the Fort Worth sounding.
                                        However, the surface is potentially much warmer at Cheyenne. When judged on a level playing field (comparing
                                        potential temperatures), one can easily see that the low-level thermal profile is much warmer at Cheyenne. Now
                                        take the surface T/TD at both locations up dry adiabats to the LCL and then trace the parcel curves up moist
                                        adiabats. You will see that the parcel curves are nearly identical. This means necessarily that the theta-e is nearly
                                        identical at the two stations.


                    2. Comparison of Surface Observations


                                        I calculated the mixing ratio, potential temperature and equivalent potential temperature(theta-e) for several stations
                                        at 02 UTC. The closest surface observation to the storm was Cheyenne(elev. 6140ft), located about 23 miles to the
                                        east of the tornado path. Luckily, Cheyenne was close to the theta-e ridge and on the inflow side of the tornadic
                                        storm.
                                        
                                        Since no observation was available at higher elevations closer to the tornado, I decided to make an approximation of
                                        the T/TD near the location of the tornadic storm. The tornadic storm first did damage south of the interstate between
                                        Twin Mountains and Granite and then moved to the north-northwest. The tornado moved into uninhabited areas to the
                                        east of Green Mountain near the Albany county line. Along the path of the storm the elevation was mainly 7600-7800 ft.
                                        In uplsope scenarios, it is often potentially warmer on the higher terrain. However, since low clouds were absent on
                                        the lower terrain, we assumed similar surface potential temperatures at Cheyenne and west of Granite. Since the
                                        dewpoint lapse rate is 1 deg F/1000ft, and since the elevation of the tornado and Cheyenne are 7700ft and 6140 ft,                          
                                        the dewpoint would have been 1.5F lower at 7700ft, assuming the mixing ratio was conserved in the upslope. In my
                                        experience, mixing ratio is not conserved in upslope flow since some mixing occurs from mid levels on the higher terrain.
                                        I am estimating that the dewpoint was around 42F in the inflow of the tornado affected areas. So the T/Td were probably
                                        about 56F/42F with mixing ratio 7.5 g/kg.

                                        Now let's compare the mixing ration, potential temperature and theta-e at selected locations at 02 UTC.  Keep in mind
                                        that the surface observation at Granite 3W was approximated.

                                
02 UTC Elev(ft) Pres.(mb) SLP(mb) T(F) Td(F) MR(g/kg) theta(F) theta-e(K)
Granite 3W  7700 757 997.0 56 42 7.6 99.0
 334.0
Cheyenne 6140 800 998.4 65 46 8.3 99.5 336.6
Dallas 487 994.5 1011.9 78 61 11.7 78.8 333.5
Galveston 6 1016.6 1016.1 72 67 14.1 69.5 334.4

                                                

                                        The dewpoint jumped to 46F (mixing ratio 8.3 g/kg) at Cheyenne between 00 and 01 UTC and remained there
                                        at 02 UTC.  With breezy east-southeast to southeast surface winds (and surely even stronger boundary layer winds),
                                        this moisture had time to make it onto the higher terrain west of Cheyenne by 0210 UTC (time of the tornado). The
                                        station pressure at Cheyenne at 02 UTC was 800 mb. Meanwhile, at Dallas, TX the temperature and dewpoint were
                                        78F/61F, with a station pressure of 995mb. But the surface theta-e was slightly higher at Cheyenne compared with
                                        Dallas.  The potential temperature and mixing ratio at Dallas/Cheyenne were 78.8F/99.5F and 11.7 /8.3 g/kg
                                        respectively. So despite the mixing ratio being 41% higher at Dallas and the temperature being 19F higher, the
                                        theta-e was actually higher at Cheyenne(336.6K vs 333.5K). The theta-e was also slightly higher at Cheyenne
                                       
                                        The theta-e was also slightly higher at Cheyenne than at Galveston at 02 UTC. This is despite the mixing ratio being
                                        70% higher at Galveston(14.1 vs 8.3 g/kg). Again, the much higher potential temperature due to elevated  heating
                                        compensated for the marginal moisture.                                  
                    
                                        There is even a larger contrast  between the low elevation stations and Granite 3W. The mixing ratio was 86% higher
                                        at Galveston than Granite 3W. But the potential temperature was 99F at Granite 3W and only 69.6F at Galveston.
                                        Keeep in mind that a 42F dewpoint at Granite has the same moisture as a 50.5F dewpoint at Galveston.
                                                                        
                               


Lifted Index,  CAPE and Shear Approximations using 00 UTC UA data


                    1. 00 UTC and 12 UTC  Upper Air Analyses
                      

                                        The 00 UTC upper air charts showed the leading edge of the strong 500, 400mb, 300mb and 200mb flow
                                        somewhere between Denver and Grand Junction. My 700mb chart shows a tight thermal gradient from
                                        southwest Wyoming to western Colorado and along the NM/AZ line.  A large area of strong elevated heating
                                        was  located over the Rockies with the warmest plume from Alamosa, CO to Rawlins, WY. (+10C). The
                                        northeastern edge of this warm plume was located from Laramie to Denver and was beginning to be
                                        eroded in the Cheyenne area due to moist upslope flow that developed in the late afternoon. How did
                                        I add such detail in regions where upper air data were lacking?  I simply took the midday surface temperatures
                                        at stations such as Rawlins, Colorado Springs, Alamosa, and Gunnison up the dry adiabat to 700mb and read
                                        the resultant temperature off of a skew-T. Since there were no frontal passages between midday and 00 UTC,
                                        I feel that this is a legitimate use of surface data. Simply analyzing the 700mb chart with no discretionary use of
                                        surface data would yield unsatisfactory results.  

                                        The 12 UTC upper air charts showed very little cooling at 500mb , 400mb and 300mb over the region of 
                                        of interest since 00 UTC.

                                      


                    2. 02 UTC  Thermal and Wind Profile Over Southeast Wyoming


                                        The thermal and wind profile above Cheyenne at 02 UTC (just before the tornado) was estimated using the 00 UTC
                                        Denver sounding, surface data from Cheyenne, and the 00 UTC and 12 UTC soundings and upper air charts.
                                        Since the Denver sounding was taken at 00 UTC (2 hours before the tornado) and since Cheyenne is directly
                                        downstream of Denver, the 600-200mb Denver thermal profile was probably a good proxy for Cheyenne's 
                                        thermal  profile. However, some very weak cold advection was occurring after 00 UTC. Also, Granite 3 W is
                                        25 miles west-southwest of Cheyenne. In addition, the 500mb temperatures that I calculated from selected
                                        surface observations all indicated cooler 500mb temperatures compared to Denver's 00 UTC sounding.
                                        However, sounding comparisons from Denver (00 and 12 UTC), Glasgow (00 and 12 UTC) and Rapid City
                                        (00 and 12 UTC) indicate that only slight cooling occurred during the 00 to 12 UTC time frame. Mid to high
                                        level(  500, 400 and 300mb ) 00 UTC vs 12 UTC upper air charts also indicate that any 500-30mb cooling
                                        was weak. 
               
                                        The surface front was well west of the surface location at 00 UTC.  In fact the cold front just moved through
                                        Grand Junction between 00 and 01 UTC,  Montrose between 01 and 02 UTC, Rawlins between 03 and 04 UTC
                                        and Eagle between 03 and 04 UTC.
        
                                        Given the dry adiabatic profiles (that are implied from the surface to 500mb in the dry air at Rawlins), 500mb
                                        cold advection would have required surface cooling as well.  So there was little in the way of mid level cold
                                        advection prior to the tornado, especially above 700mb. However, as the upslope flow deepened in the late
                                        afternoon and evening, the 700mb layer was cooled from Cheyenne west to the lee slopes of the Laramie range.
                                        Prior to the tornado, the 700mb level cooled down to about 7-8C.

                                        As already discussed, 700mb temperatures were known with fairly high accuracy. Of course, surface dewpoint,
                                        surface temperature and surface pressure at Cheyenne were known quantities.  

 
                                       
                    4. Surface Based CAPE at Cheyenne


                                        In the worst case scenario for Cheyenne, I used the unmodified sounding from Denver and the surface from
                                        Cheyenne and arrived at 2200 j/kg surface based CAPE. Then in a best case scenario, I allowed for some slight
                                        high level cooling (500-300mb) from the southwest (about 1C) and arrived at close to 2500 j/kg. It appears that
                                       moderate instability was in place at Cheyenne with CAPE values possibly exceeding 2000 j/kg.


                    5. Surface Based CAPE and Vertical Wind Shear in the Tornado Affected Area


                                        Cheyenne was located in the inflow of the tornadic storm, and the dewpoint jumped from 41F at 00 UTC to 46 at
                                        01 UTC. So the rich moisture had 1 hour 10 minutes and possible even 2 hours to make it upslope to west of 
                                        Granite by 0210 UTC as the tornado passed 24 miles west southwest of Cheyenne.

                                        In the section on elevated heating, we already estimated the T/TD at Granite 3W at 02 UTC. Given that the 500-300mb
                                        layer was no more than 1C cooler than at Denver at 00 UTC or Cheyenne at 02 UTC,  that the potential temperature
                                        at Granite and Cheyenne at 02 UTC were similar, and that the surface dewpoint was likely close to 42F, the surface
                                        based CAPE was likely around 1900 j/kg, or slightly less than at Cheyenne. Please note that the CAPE estimation for
                                        Cheyenne is based on reliable surface data while the surface conditions at Granite 3W were approximated. It is a shame
                                        that surface observations are so scarce on the Cheyenne ridge. The vertical wind shear profile was excellent, with 20kt
                                        east-southeast winds at the surface, and south winds around 35kts at 500mb(3km), south winds at 50kts at 400mb(5km)
                                        and south-southwest winds at 60 kts at 300mb (7km).


    Upper air charts


UA

12 UTC 23

 700        500        

00 UTC 24

 700        500        400        300        250        200

12 UTC 24

700        500        400        300        250        200

00 UTC 24  vs 12 UTC 24 Comparisons

500       400         300

Surface

00        03        06        09        12        15        18        21        22        23        01        02