After a record number of tornadoes in 1995 and the recent media blitz about tornadoes, Amarillo has experienced a rash of tornadoes being reported in 1996. Many of these reports, both from the public and from our regular storm chasers, have left us mystified as we try to reconcile these reports with what we see on the radar.
Video tapes brought to the office leave us scratching our heads and asking, "Is it a tornado?" There were three occasions this year when one chase group came into the office after hearing about a tornado report and offered, "I was right under that cell; and let me tell you, there was no tornado." On two of those occasions the reporting team came into the office and shared their video of the tornado they reported. One of these events occurred with reputable, experienced chase teams. How could they have such a different perspective?
The National Weather Service (NWS), defines tornado in WSOM C-40 as "...a violently rotating column of air, usually pendant to a cumulonimbus, with circulation reaching the ground." Is this a definition we want to continue to use? Should any vortex near a convective cloud be defined as a tornado? We have in the past listed a gustnado in Storm Data as a weak tornado and described its actual character in the narrative section. We even issue tornado warnings at times based on such sightings. Yet, it would be difficult to say it was pendant to the thunderstorm. What about a vortex not associated with deep convection?
Two examples from spotters will help illustrate some of our problems. On June 5 we received reports of a tornado in one of our counties. Spotters and public reports alike gave a classic description of a gustnado, except at the end of its life cycle it became attached to a rain-free base and then dissipated from the surface upward. This event lasted about ten minutes. The same 40dBZ/30VIL storms then went on to produce another gustnado (seen by the county Emergency Manager), which was said to last over 20 minutes--a temporal scale exceeding anything I have heard of for a gustnado.
On July 12, about 25 mi southeast of the WSR-88D in Amarillo a report of a tornado came in from one of the local TV storm chase crews. This storm produced a deep meso from 3,000 to 20,000 ft, with a strong rear flank downdraft (RFD) feature on the radar (currently under research by our forecasters). Review of the video shows strong inflow with an associated dust signature which appears to interact with the RFD and cause a brief (about 30 seconds) spin-up. Does a dust swirl lasting for 30 seconds in an updraft downdraft region deserve to be classified as a tornado?
Two events I personally saw this year help confuse matters even more. On April 4, during my return from a school presentation I observed what appeared to be a dust devil approaching the highway ahead of me. Off to the north I observed virga from a high based and fully glaciated cell. As the dust devil approached the road and was about one-half mile ahead, the van suddenly swerved due to the interaction with the unexpected outflow boundary. Should this four- to five-minute duration dust devil now be called a gustnado?
On July 8 I observed from my back porch what at first appeared to be a very large dust devil which I estimated to be 15 mi to my south southeast. Viewed through binoculars, I could make out a double concentric ring and a tube extending above the dust devil going up about 90 per cent of the way to the cloud base. On the synoptic scale, we had an easterly wave moving over the region which developed a line of storms moving to the northwest. This so-called dust devil was under the south end of a feeder band. The rain-free base was about 6,000 ft above the surface with a vertical development I estimated at 5,000 to 6,000 ft over the tube feature. This feature was well developed when first observed and lasted 8 to 9 minutes before dissipating. Was this originally a low level convergence vortex or dust devil which developed due to interaction with the convective process? Should it be classified a tornado?
We are rapidly expanding our understanding of the processes or combination of processes leading to tornadogenesis. The classical mid-level mesocyclone associated tornadoes generate little doubt about what is being observed by either the radar operator or the storm spotter. However, great confusion arises with weaker, short duration vortices.
Landspout tornadoes in this area cause confusion from event to event. We have a history in our area of producing multiple landspouts. In 1995, six were captured on video at one time. At times they appear to be pendant from the rain-free base and develop downward; and at other times they appear as a surface vortex which develops upward. This causes problems with spotters who are dedicated to understanding what they are seeing. Additional problems come into play for the radar operator, since these events are often displaced spacially from the core of the cell. Another event causing confusion is the gustnado. Where should gustnadoes fit into the scheme of things? And finally, how do we deal with the myriad of unexplainable vortices being reported by the hoards of neophyte storm chasers with a camcorder at hand?
I have three operational concerns evolving from these marginal type events. First and most importantly is how does the radar operator reconcile these marginal events with what he/she is observing? If the event is not evident on the radar, many valuable minutes are lost trying to understand a reported event and its likelihood of existence and continuation. Life and death decisions have to be made, and made quickly.
Second, I am concerned about the climatological records created by me and fellow WCMs. What we put in Storm Data determines the number, intensity, and distribution of tornadoes in the United States. Secondary to the climatology (maybe equally important to some) is the effect our selection of events has on the verification program. Forecasters definitely want to know which tornado reports are valid and which are discounted.
Given our current resources, should we make an effort to list each known vortex, describing it and its environment? The July 12 example might be listed, "A short duration (30-second), shallow, surface-based rotation developed on the south end of northwest moving line. This unusual movement was due to an easterly wave. The parent cell had a radar identified mesocyclone from 3,000 to 20,000 ft." The video offers little more than a shallow dust swirl, yet the radar supports calling it a tornado. What would you call it? Since Storm Data is the only tornado database, we probably should make some effort to include more detail about these events. Operational and research meteorologists deserve a more detailed database for research purposes if we expect to expand our knowledge of these events!
Finally, there is the need for more detailed training to our core spotter groups. We cannot keep providing the same categorical definitions to these groups. We must try to explain our latest understanding of tornadoes and the varieties they are likely to encounter in the field. We need to develop better communication between the office and the spotters on these marginal events. Our forecasters need more training and field experience to understand the confusion the spotter is having.
Before we can deal with the issues above, we need to reach some consensus on how we define these various vortices. Do we list vortices not associated with the updraft/downdraft interface? Do we need a minimum time frame for existence? Do we need damage, even if only a surface swirl over open terrain? What about damaging vortices not associated with thunderstorms or moist convection? Do we need research on repeatability of these weak events and their probability to increase in strength? The forecaster needs to know if that event will lead to a risk of life and property. We need to reach some consistency on which vortices will be called a tornado and which ones will fall on the other side of the line and simply be a notation in the call log. With video and increasing numbers of spotters (from the professional to the thrill seekers) we will see events that were easily dismissed when only supported as an eyewitness description.
As you can tell, I have many more questions than answers; and I am sure many of you have experienced some of the same frustrations and have similar questions. The purpose of this essay is to share some of our problems and spark discussions to deal with the growth we will see in reported tornadoes (valid or not). However, I also caution forecasters not to misjudge those unexplainable reports. It takes but a few days in the field to realize there are many small-scale events that do not fit the classroom theories or the pixel display of our remote sensors.
I wish to thank MIC Jose Garcia for his thoughtful review and insights on this essay. However, any remaining flaws are my responsibility.
Brady, R.H., and E.J. Szoke, 1989: A case study of nonmesocyclone tornado development in northeast Colorado: Similarities to waterspout formation. Mon. Wea. Rev., 117, 843-856.
Davies-Jones, R.P., 1986: Tornado Dynamics. Thunderstorm Morphology and Dynamics 2nd ed., E. Kessler, Ed., University of Oklahoma Press, 197-236.
Doswell, C.A. III, 1985: The operational meteorology of convective weather. Vol.II: Storm scale analysis. NOAA Tech. Memo. ERL MSG-15 (NTIS Accession No. PB83-162321), 158pp.
Doswell, C.A. III, and D.W. Burgess, 1993: Tornadoes and tornadic storms: A review of conceptual models. In The Tornado: Its Structure, Dynamics, Prediction and Hazards (C. Church, D. Burgess, C. Doswell, R. Davies-Jones, Eds.) Geophys. Monogr. 79, Amer. Geophys. Union, 161-172.
Forbes, G.S., R. Wakimoto, 1983: A concentrated outbreak of tornadoes, downbursts, and microbursts, and implications regarding vortex classification. Mon. Wea. Rev., 111, 220-235.
Lewellen, W.S., 1993: Tornado vortex theory. In The Tornado: Its Structure, Dynamics, Prediction, and Hazards (C. Church, D. Burgess, C. Doswell, R. Davies-Jones, Eds.), Geophys. Monogr. 79, Amer, Geophys. Union, 19-39.
Trapp, R.J., and R. Davies-Jones, 1996: The dynamic pipe effect and its role in tornadogenesis. Preprints, 18th Conf. Severe Local Storms (San Francisco, CA), Amer. Meteor. Soc., 387-391.