STORM TRACK: January 31, 1983 (Volume 6 Issue 2)

The ST Editor's 1982 storm chase report has been delayed by other more dramatic accounts (Leonard, Marshall, Moller), but can now finally be told. My chase covered May 23 to June 11 not a very active storm period last year, despite something over 12,000 miles of driving (just two weeks earlier, more tornadoes were recorded at that time and in that part of the country than at any other time in history). Vacation was originally scheduled two weeks earlier, but a last minute requirement for emergency leave fram one of the staff precluded my absence at that, time. Hopefully, this year will turn out differently, as I plan -again- to be in tornado country during mid-May (specifically May 8-28). However, I made the most of the time last, year, filming four small tornadoes and nine funnels in the Texas triangle bounded by Mineral Wells, Plainview and Midland. Three of these storms were unique in my experience and bear recounting here.

On May 27, 1982, I started from Wichita Falls, Texas at 11:00 AM with my own forecast calling for severe weather from there to Sweetwater to Floydada, and then to Erick, Oklahoma. Subsequently, Kansas City (the NWS National Severe Storms Forecast Center) issued a tornado box 80 miles either side of a line from 30 miles WSW of Hobbs to 10 miles NE of Childress. There being nothing in my forecast area, overlapping but at the east end of theirs, I assumed that later activity would move in from the west and headed west to intercept it. I entered Lubbock 45 minutes after the watch began (in the middle of the box) but was depressed to see a sky full of anvils as early as 1:30 PM (all times are CDT). The atmosphere was too primed for severe weather, and the storms started too soon, before maximum heating could occur. They lined out early, before becoming well organized, and one of the first big cells formed a ponderous shelf cloud 20-25 miles W to SW of Lubbock. The precip overcast was so heavy that an area beneath the shelf was almost a purple-black (2:30 PM). Proceeding southwest out of Lubbock, I followed US 82 to Farm Road (FR) 179 and turned south to FM 211. The shelf cloud was lining out now in a N-S orientation, curving back to the NE at the northern end. Strong surface to cloud outflow dust was boiling up ahead of it. I began driving east, on FM 211, passing a string of 10-15 farm trucks with concerned families, moving away from the storm.

Enroute, I stopped twice (8.0 and 9.6 miles east of the FM 179/211 intersection 2:42 and 2:45 CDT respectively) to record ground dust swirls of suspicious nature. These were occurring well ahead of the swiftly advancing dust, wall (Illustration 1A). The second of these stops drew my attention to the SSW at two swirls beneath new convection ahead of the outflow lip (Illustrations 1A and 2). The new, small (N-S) convective line seemed detached (at lower levels) from the main storm to the N and NW and was building into the pervasive anvil overhead, from the old cell to the west. I charged on to US 87, and then south about 3-4 miles. At this point, my record keeping on picture locations/times failed completely under the pressure of "close in work," as the two swirls reappeared, in almost the same location as before. In swift succession, each rotation gave evidence of tornadic structure: First the northern one, with a narrow tube connecting the tight, small conical ground swirl to the turbulent cloud base; and then the southern one, with a laminar structure 5-10 seconds later (Illustration 3B). Both lasted 25-30 seconds, with the outflow dust wall still within 1/2 mile to the west of the rotation, but rapidly approaching. It subsequently overtook the new convection and swept on east.

Illustration 2

Illustration 3A

Illustration 3B

Illustration 4

Illustration 4 shows a possible explanation for the final "gustnadoes" in #3A/B by Tim Marshall. However, this doesn't explain the initial paired-rotation, well ahead of the gust front. In such a case, Dr. Charles A. Doswell III believes that, strong updraft can occasionally form tornadoes (and not necessarily just small ones) without a coincident "rear flank downdraft." Undoubtedly, the new convection was enhanced, through some interaction, by the gust front to the west.

At this point, I submit that the term "gustnado" needs to be redefined into tornadic (TGO) and nontornadic (NTGO) gustnadoes. I have no specific citation on use of the term "gust'- nado" in the literature but suspect that such applications have been more general than specific. I submit that a TGO is identifiable by (1) proximity to a gust front, (2) proximity to updraft convection, and (3) laminar structure. An NTGO develops directly at the interface between outdraft and inflow, or between outdraft and normal environmental winds, but without (2) or (3) above. The latter rotation is characteristically disorganised, fitful and translates with the moving outflow boundary. Tim Marshall and

Illustration 5

I spotted an example of a pair of these on June 11, near Snyder, Texas (Illustration 5). They lasted 10-l5 seconds and were poorly organized, moving rapidly to the SE along the leading edge of the gust front.

The only other gustnado examples from my early chase years (which also tend to confirm this TGO/NTGO dichotomy) were from 1964 and 1971 storms in Kansas and New Mexico. On May 29, 1971 (5:32 PM), a TGO occurred 2-3 miles east of Tucumcari, New Mexico (Illustration 6A). The ground tube was laminar for about 5-10 seconds along an outflow boundary and was immediately adjacent to strong updraft convection. The storm had only begun outflowing a minute or so earlier from a very turbulent, fractus charged base (tops to 44,000'). Another example of a possible NTGO (although not as easy to "call" as the others) is from a July, 1964 Kansas storm north of Junction City (Illustration 6B). This rotation developed right, on the outflow boundary, with a strongly defined SW wind spinning off the back side and lighter, northerly inflow in front (see inset). This rotation never became completely laminar, beneath a high level convective base (maybe 7-8,000'), but was strong and potentially damaging. This suggests that an NTGO could do severe, localized damage and -yet- be nontornadic (i.e. not be linked to convective rotation at or above cloud base). What do you think? Write Storm Track and share your experiences.

On June 3, 1982, I left Norman, Oklahoma at noon toward my forecast area for NW Texas, from about Childress to Fort Worth to Abilene. A severe storm developed at the eastern end of my area, just west of Springtown, and several small funnels were photographed between 7:56 and 8:00 PM. I then proceeded south of Springtown at 8:48 PM on State Road 51. At 9:00 PM and 3.3 miles south of town, I observed a remarkably rapid convective development to the SW -- near Peaster (Illustration 7). A small cumulus materialized quickly about 1-2,000' above ground, beneath a rain free base and just ahead of a southward moving rain curtain (suggesting a "rear flank downdraft"). I watched it build up at a rate of at least 100 ft per second (probably more) until it joined the rainfree base overhead. It then drew in its original base and became almost perfectly conical. The last picture in Illustration 7 was taken a few seconds afterward, still showing the general conical shape. The picture before that is reproduced from memory. I presumed that this was a funnel but had never before seen a formation grow like that. This small cumulus was the only one like it in the area, there being almost no fractus below cloud base anywhere nearby. Unfortunately, frequent lightning at the time prevented any radio reports on local warnings or possible damage from this event.

Illustration 7

On June 7, 1982, I started from Wichita. Kansas at 10:00 AM, under northerly winds, overcast skies and temperatures in the 60s (how depressing). With my forecast for the southern Texas panhandle and SW Oklahoma (roughly from Lubbock to Plainview to Shamrock to Lawton), I drove down I-35 and broke through the NE-SW front south of Oklahoma City. The radio reported that Wichita Falls was baking slowly in 100 deg. At this, I turned west to Hobart, where the NWS had left a tornado watch, blocking most of the two panhandles. I drove to the SE corner of their box, where it overlapped my forecast, and photographed two small tornadoes including the F1 at Happy Union. The remarkable thing about this storm was that the vortex developed (1) not only a mile or more south of the turbulent wall but (2) did so from one of the smoothest cell bases I have ever seen. The vortex, never clearly seen from my vantage, appeared as a rising dust column, from ground to cloud base (Illustration 8). Tim Marshall was surrounded by dust in Happy Union but did see a small tube from the cloud. Shortly afterwards, I encountered a local chaser on Farm Road 1914 near Happy Union, who was routing traffic around a downed "60,000 volt" power line pole. He confirmed hearing the roar and spotting three separate rotating dust swirls. Lubbock radar reported a hook on radar, but I am unsure if it was looking at, Happy Union or the wall cloud to its north. One interesting visual reference to the Happy Union tornadic cell: Along with the unnaturally smooth base was a narrow band of cumulus "teeth" (downdraft?) along the western edge.

Illustration 8