FORECASTING SUPERCELL TYPE
by Richard Thompson and Roger Edwards
© Copyright Richard Thompson and Roger Edwards
How many times can you recall cruising out in some "meatwagon," feeling good about the day's potential, only to be baffled by storm behavior later that afteroon? We've lived that sequence on dozens of our 200+ Great Plains chases. Storm structure and motion can vary wildly, even from county to county. To help remove some of the mystery from chase day forecasting, we've compiled a table summarizing recent research findings, and how you can apply them to your forecasts. Much of this information is adapted from the suggested reference reading, with some interpretation based on experience in the field. To keep this table relatively simple, it has only the main features; it is absolutely not an all-inclusive sounding guide. Check the references at the end of this article for more comprehensive coverage of background topics.
LP supercells: The key is, how well (efficiently) does the storm produce and dump precipitation? LP stomms are inefficient due to either dry-air entrainment with large CAPE, or strong shear with small CAPE. In the former, dry entrainment into the narrow updraft combines with already low condensation rate. If any significant precip is produced at all, it tends to be large hail. Small CAPE/high shear LPs pump much of their water into the anvil, at the expense of precip. LP storms can be spectacular, with laminar striations and a towering, sculpted, "barber pole" updraft. Many are non-tornadic, but LPs with strong SR winds (say, >30 kt) and high helicity can yield a dream chase - a series of highly visible tornadoes for several hours. LP example include Wheeler TX from 18 May 80 (non tornadic; see Eric Evenson's cover shot on the March 31, 1991 Storm Track), Boise City OK/Baca County CO from 11 May 91 (at least 4 small tornadoes), and Laverne OK from 15 May 91 (one large, long-lived tornado). Remember, LPs may be "inefficient," but low-precip does not mean no-precip! They can still destroy your car with giant hail.
"Classic" supercells: These resemble the idealized artwork found in spotters' guides and meteorology texts. [In the real world, there is a lot of disagreement over what "classic" means.] Moderate precip efficiency results from a combination of large CAPE and large SR shear (indicated by helicity), which causes abundant precip to be dumped away from the tilted, rotating updraft. Some begin LP in appearance, like the Red Rock OK storm (26 Apr 91), but don't act that way for long. Hail is very common, but its location and size can vary greatly within a storm, and over time. "Classic" supercells usually develop well-defined wall clouds, inflow regions, and rear-flank downdrafts (RFD). They may take an hour or more to form and mature, then produce tornadoes in cyclic fashion (from several successive mesocyclones). They then dissipate (usually after sunset), turn into HP supercells, or become part of a multicell complex such as an MCC or squall line. Other excellent examples are Union City OK (24 May 73), Binger OK (22 May 81), and Wichita/Andover KS (26 Apr 91).
HP supercells: These are obviously very efficient precip producers, and are often (not always!) slow-moving. Warm cloud depth >2 km (on a sounding, CAPE below the "freezing level" of 0 deg C) results in a large amount of low level condensation. SR winds can be weak, but CAPE is high, and HPs in the plains almost always produce huge destructive hail (baseball to grapefruit size). HP rotation is often enhanced by interaction with, and movement along, a boundary that is oriented near the direction of the mean flow. Such storm motion can result in major mesocyclones despite marginal helicity. Plainfield IL (28 Aug 90) is an extreme example of this. More commonly (like the 18 Apr 92 storm in SW OK), precip falls near or into the updraft due to weak shear, and tornadoes are either weak or nonexistent. In HPs with somewhat stronger shear (higher helicity), the precip load still isn't advected away from the updraft faster than it wraps around, resulting in "bears cage" mesocyclones with hidden tornadoes. Beloit KS (15 Jun 92) and Tulsa (24 Apr 93) are a couple recent examples.
Remember, morning soundings do not represent afternoon conditions! Always modify the sounding using winds, temperatures, and dew points you expect to occur, at the surface and each upper level possible. Only then should you analyze the sounding for shear and instability. When analyzing soundings or interpreting SELS outlooks, refer to our table and discussion. This information can help you recognize and forecast storm potential on a particular chase, and the different possibilities within a region on the same day. Keep in mind that there is no distinct boundary separating different supercell types, only roughly defined parts of a continuous spectrum. "Borderline" situations and poorly analyzed soundings can still produce unexpected results. On the road, think of possible viewing angles and intercept strategy based on expected storm type. When you approach the storm, that knowledge and careful visual observation will enable you to note changes in storm structure and motion, and plan your route accordingly. Hopefully, we will all encounter fewer nasty surprises, and more productive chasing. See y'all in the inflow!
Branick, M.L., and C.A. Doswell, 1992: An observation of the relationship between supercell structure and lightning ground-strike polarity. Weather and Forecasting, vol. 7, Mar 1992, 143-149.
Davies, J., 1994: A look at hodographs, helicity, and supercells. Storm Track, vol. 17, Jan-Feb 1994.
Doswell, C.A., 1992: A review for forecasters on the application of hodographs to forecasting severe thunderstorms. National Weather Digest, vol. 16, Feb 1991, 15pp.
Johns, R.H., and C.A. Doswell, 1992: Severe local storms forecasting. Weather and Forecasting, vol. 7, Dec 1992, 588-612.