sign in sign up
<<
Home , Supercell

Supercell by Eyre to the effect that evidence of 'twisting' in the damage can be used to imply something Gharles A. Doswell, lll and Harold E. of the character of the high . The impli- Brooks cation in Eyre's remarks is that some of the National Severe Laboratory, Norman, damage may have been tornadic, as evidenced pylons. This by the twisted trees and electricity implication apparently is based on the mis- 'We read with interest the account by L. A. conception that tornadic winds would show Eyre (1992) oftwo violent events rotation on scales of a few metres. In virtually in , and have some comments we all tornadoes, the scale of rotation is at least one would like to share with the readers. \tr7e believe and perhaps two orders of magnitude larger that the events described might belong to a class than the scale oftrees and electric power pylons. of severe thunderstorms that has been relatively Therefore, on the scale of the damaged obiects, unrecognised heretofore, but which occasion- the rotation will not be at all evident. For all ally can be characterised by the sort of extreme practical purposes) the damaging winds in a and events associated with these two are 'straight' winds on a scale of a few examples. Because the information about the metres. Twisted damage is a consequence of environments available to us, either from the way structures on that scale are affected by Eyre's note or from other sources) is rather the wind. Based on the evidence in Eyre's note, limited, our suggestions cannot be given a rig- it appears that the twisting is simply the result orous test. However, the descriptions of the of what he calls 'local eddies' produced by events and our experience with similar storms the complex interaction of effectively straight in allow us to- speculate on winds and objects like trees and buildings. If the character of the storms producing these the rotation is on the scale of the convective devastating events. storm (several kilometres), as would be the case The storm apparently occurred in an in a without a tornado, the argument environment with strong vertical , is even stronger that twisting damage on the according to Eyre. Eyre calls the storm 'multi- scale of individual trees was not the direct result cellular' although he does not present any of that rotation. specific evidence in support of that description. \ile think that the storm may have been a It has been our experience that virtually all form of supercell with relatively strong storm- convective storms lasting more than 30 minutes scale rotation in mid-levels (say, higher than have at least some multi-cellular aspects and so 3km above the surface), but with only weak describing a storm in that way conveys little or storm-scale rotation near the surface. Such no real information about the character of the storms often take the form of what has been stonn. It is not hard to imagine that the Sydney called a'high-' (or un) supercell by storm was, in fact, of the supercell variety, Moller et aI. (1990). Based on our observations although it may not have had an intense meso- of such storms in the USA and elsewhere, they at low levels. I7e have found that many can be accompanied by extensive swaths of people still cling to several of the old-fashioned damaging winds (and thus may fall into the descriptions of 'supercell characteristicsr' one class of events called '' by Johns and of which is the unicellular character of the Hirt (1987)), as well as large hail and torrential convective storm. Modern definitions of rainfalls. focus on the presence of a deep, Incidentally, although it has not been persistent (see, e.g., Doswell and explained satisfactorily, the green colour noted Burgess 1993). Therefore, the multi-cellular by Eyre and the eyewitnesses has been associ- character ofthe storm cited by Eyre is not really ated with hailstorms for many years in the USA. relevant in its classification; based on the events I0flhile we do not have any suggestions for its produced by the storm, we are fairly confident origin, our experience has confumed repeatedly that it was a supercell. the association with hailstorms. \7e also note with some dismay the comments The Arnhem Land event is also very inter-

209 esting. Supercells generally are considered to or . It is interesting to note that be midlatitude phenomena and so it might a recent windstorm in (see Cummine el appear that we could rule out the possibility of al. 1992) produced a 'blowdown' of trees over supercells in such an obviously tropical an area of roughly 2000km2; the damage area location. However, recent theory, observations, of comparable intensity for Hurricane Andrant and numerical modelling results all sug- in Florida is about the same size, so if the event gest that the critical element in convection in Canada had occurred over a populated area becoming supercellular is the vertical wind it is possible that such a severe thunderstorm shear profile. Therefore, it appears that there (which we believe may have included one or is nothing inherent in the tropical more HP supercells) could have been destructive that precludes supercells. The fact that the on a scale comparable to a maior hurricane. generally have relatively litde vertical wind shear most of the time would explain the References relative rarity of supercells. However, in the P. Leduc, M. (1992) presence of the necessary vertical wind shear, Cummine, J., McCarthy, & Blowdown over northwestern . A a convective storm is as likely to be a iust event - July 18, 1991. In: Preprints, 4th AESI supercell as its mid-latitude counterpart. Again, CMOS \Vorkshop on Opnational Meteorolog, the description of the environment by Eyre Vltistlu, British C olumDlo, Atmospheric Environ- does not allow us to determine even crudely ment Services Canada, Meteorological and pp. 1-317 whether or not sufficient vertical wind shear of Oceanographic Society, 3l Doswell, C. A., III and Burgess, D. \Uf' (1993) was present on the day of the appropriate type Tornadoes and tornadic storlns: A revbw ofcon- dre event, but the eyewitness accounts of an ceptual models. In: Church, C. (Ed') Proceedings, eyeJike structure suggest the presence ofa rear- Tomado Symposium .I//, American Geophysical flank downdraught (see Lemon and Doswell Union, In lress (1992) How severe can a 'severe thun- 1979) and clear siot associated with a rotating Eyre, L. A. derstorm' be?: An investigation into two violent supercell. It may well be that the presence of elecuical storms in Australia. Veather, 47, PP. the intertropical convergence zone somehow 37+383 created a supercell-favourable local environ- Johns, R. H. and Hirt, uil. D. (1987) Derechos: detailed proximity Widespread convectively induced windstorms. ment, but in the absence of 'Vea. sounding data, it is difficult to be certain. Forecasting,2, pp. 3249 Lemon, L. R. and Doswell, C. A.,I[ (1979) Severe description makes it plausible Nevertheless, the thuriderstorm evolution and mesocyclone struc- also 'Wea. to speculate that the Arnhem Land storm ture as related to . Mon. Reo., was a supercell. 107, pp. 118,t-1197 lWhilst the implications of global warming Moller, A. R., Doswell, C. A., III and Przybylinski, and its effects on the scale of individual con- R. (1990) High-precipitation supercells: A con- ceptual model and documentation. In:. Prepints, vective storms make for interesting speculation, I 6th Conference on S eoere Local S torms, Kananaskis we prefer not to indulge ourselves in that par- Park, , American Meteorological Society, ticular aspect of Eyre's note. If we cannot be pp.52-57 certain whether or not the events were supercells, it seems quite beyond our capability to extrapolate to global implications. However, in answer to Eyre's nominal question about "Supercell how severe a 'severe thunderstorm' can get, it Reply to: appears that peak 'straight' winds approaching thunderstorms" 75 ms-1, accompanied by hailstones up to l0cm in diameter, are possible over areas of l500km2 I L. Alan Eyre or more, with hail and winds exceeding 25 m s- University of the West lndies, Jamaica lasting perhaps as long as 30-45 minutes at any one point. Anyone experiencing such an event certainly would have to be impressed, perhaps I would like to thank Charles Doswell, III and as impressed as he or she might be by a tornado, Harold Brooks of the US National Severe

2r0 connection is cer- Storms Laboratory at Norman for their auth- would emphasise that the oritative cornment on the two Australian tainly beyond the realms of interesting specu- throughout the world storms. Theirs was exacdy the kind of response lation'. activity signal, the Schu- that I was hoping my article would elicit' Only produces an electromagnetic two of their five bibliographic references to mann Resonance, that is scalable and can be has been done in the supercells are readily available to non-American calibrated. In fact, this storm: "The and especially tropical departments, and the very area of the Arnhem Land more crucial contribution of the researchers and number of lightning strikes multiplied average monthly tem- scholars at Norman to the genesis and structure than 100-fold when the jumped by 2'C. . . SR is of is not as widely known and perature in Darwin temperature" appreciated as it should be. related to change in tropical in this instance, the suggested identification (Nadis 1993). of both storms as supercells, their clear expla- It is certainly high time for the mesocyclone, (whatever want to nation of the character of the wind damage at 'derecho', or supercell we Ku-ring-gai, and particularly their recognition call it) to appear in standard and of the eye-like structure at Jabiru as a 'clear physical geography texts in all its maiesty and slot' initiated by rear-flank downdraught in a power. As a Caribbean national who has,studied per- rotating supercell are all most helpful to our tropical storms for 30 years, and having pub- understanding of these spectacular and dam- sonally experienced and described for storm systems aging storm systems. lication three of the most severe the The Sydney storm was described to me as on record in the Western Hemisphere, far and away 'multi-cellular'by a radar analyst of the Bureau Arnhem Land storm of 1990 was spectacle it of Meteorology immediately after the event' the most awesome meteorological The personal communication was verbal: no has ever been my privilege to witness. I can hold the supporting data were given to me at the time' assure Doswell and Brooks that I now Close examination of the synoptic situation and supercell in even greater respect. the satellite imagery at a later date would seem to lend support to this view. \trfhile Doswell and Brooks say that they pre- Reference fer not to comment on my suggested relation- Nadis, S. J. (1993) A flashy global thermometer. ship between storms and global warming, I Technol. Rets.,12 Jan, pp. lO-11

patterns are similar to 1859 when Vic- What the paPers said torians had a miserable and a bleak .t' an to Except where indicated, the items below come l0 Noztember. "Cautious hopes of end came when from the Daily Telegrapft. Our thanks to David South Africa's crippling drought Naylor for forwarding some of the material. storms, followed by hours of soft, soaking , yesterday. The 19 October. (Dqily Express) "Bitterly cold fell over much of the country Orange Free weather has turned October into December , over Cape Province, the Transvaal across Britain and worse is to come. Tem- State and the western and central - land, peratures have been as much as six degrees regions, revitalized parched gtazing which below normal with blanketing parts of started to fill dams and caused rivers year flow . . . the north . . . Yorkshire weather diviner Bill dried up more than a ago, to surrounding Foggitt, hired to predict weather for the English Flooding in Johannesburg and Tourist Board, says all the signs point to a suburbs caused damage estimated at several areas, turned to bad winter. And freelance Frank million rands, while in farming Iilfalker, of Girvan, south-west Scotland says dustbowls by the drought, storm waters washed 211