July-August 1998 STORMTRACK features the Nashville, TN Tornado, Chasing in Illinois, Joe Galway Remembered, and Storm Chasing in Israel?


I. COMMENTARY by Tim Marshall

There have been a number of discussions about what is a tornado so, Iíll take a stab at it again before the new Glossary of Meteorology is published by the American Meteorological Society next year. The old glossary definition (circa 1959) was: "A violently rotating column of air, pendant from a Cumulonimbus cloud, and nearly always observable as a funnel cloud or tuba." This definition is of course outdated as there have been many advancements since the late 50ís in the study of tornadoes. I am naturally curious to see how the official definition will change.

To me, the term "tornado" should be broad enough to include landspouts, "tornadic" gustnadoes, and waterspouts which come ashore, but should exclude dust devils even if they extend to cloud base. I proposed an updated definition in 1995: "A violently rotating column of air in contact with the ground, usually associated with Cumulonimbus; condensation to the ground and/or a funnel cloud is not required." My proposed changes are in italics. The old definition did not mention that the circulation had to be in contact with the ground which is quite important. Also, we have learned that tornadoes do not have to be associated with Cumulonimbus clouds as in the case of the "landspout", a tornado which typically forms from lines of towering cumulus in higher terrain, a term coined by Dr. Howie Bluestein. We also know that such tornadoes donít always require a visible funnel. See, now wasnít that simple?


Friend and fellow chaser Beverly Bishop passed away on May 24, 1998 at her home in Chattanooga, TN after a courageous battle with cancer. She was 48. "She loved tornado chasing", said chase partner Steve Tabb who took her on numerous chases, sometimes traveling all night from Tennessee only to end up near Big Spring, Texas by the next evening. Steve, I and others will miss her bright smile and good company.

Beverly in a light-hearted moment at a Memphis, Texas fast food restaurant. She chased with Carson and myself for two seasons. Although she never saw a tornado with us, she never expressed dissapointment. For her, like many of us, just chasing storms and being out on the plains was fun enough. Photograph by Steve Tabb.


Rich Smrz, rsmrz@mail.state.tn.us, sent in this historical, I mean hysterical story: In 1842, a researcher named Elias Loomis performed an experiment related to tornadoes and chickens. From Loomisís own words: "Several live, featherless chickens attracted much attention after a recent tornado. In order to determine the wind velocity needed to pull feathers from the chicken we experimented as follows. A six-pound cannon was loaded with five ounces of powder and a freshly killed chicken, with its feathers. The gun was pointed upward and fired. My conclusions are that a chicken forced through the air with this velocity is torn entirely to pieces; so tornadoes likely posses wind speeds of less than the measured chicken speed of 341 miles per hour." But still, 135 years later, Professor Bernard Vonnegut in Albany, New York was asked to look again at chicken plucking as a measure of tornado wind speed, but not the same way. Vonnegut quickly learned about flight molt, something that makes feathers loosen in the face of danger. Its like tear away feathers, a defense mechanism so that a chicken starved predator ends up with a mouthful of feathers instead of a mouthful of giblets. Vonnegut figured that tornadoes scare chickens and flight molt kicks in, so that even a tiny tornado could blow the feathers off. Too bad there are no Ďchickenometersí to record tornado windspeeds. Vonnegut did have a distinguished career studying foul weather, but not fowl weather."


Mike Zweifel sent in the incredible cover shot in this issue: "I was out at the Modine-Benstead Observatory for our general meeting around 7pm on July 20, 1998. I saw some clouds coming in from the west, turned on the television, and saw there was a severe thunderstorm warning for southeast Wisconsin until 8:30 pm. The storm struck around 7:30 pm and was accompanied by winds to 65 mph along with heavy rain. We started getting a lot of cloud-to-ground lightning about fifteen minutes later, so I went outside with my camera to photograph it. Just as I was opening the shutter on my camera, three lightning bolts struck, the closest being 1/4 mile away. The thunder was so loud that my ears were ringing for about an hour."


JOSEPH G. GALWAY by Chuck Doswell

Joe Galway passed away at the end of June. He was born in 1922, in Boston, MA. For some aspects of his forecasting life, see Lewis (1996) for a more detailed biographical sketch than I can provide here. He was a forecaster with the Severe Local Storms Forecasting Unit (SELS) for 27 years ... from 1952 to 1965 and again from 1972 to 1984. The break in his forecasting service came when he served for several years as the "Principal Assistant" (or PA ... what would now be called a "Deputy Director") of the National Severe Storms Forecast Center (NSSFC). He was on duty during the infamous 08 June 1966 tornado in Topeka, KS (Galway 1966), and issued the successful tornado watch for that event.

In addition to his severe storm forecasting duties, Joe was a severe storms researcher who contributed many useful ideas and tools to the operational SELS forecaster, arguably the most famous of which is the Lifted Index (Galway 1956). It is of some interest that this famous article appeared in the "Correspondence" section of the AMS Bulletin because of some stupid bureaucrats who were preventing its publication. Joe was so determined to see his ideas published that he sent his paper in as correspondence rather than an article, thereby sidestepping the "coneheads" trying to block its appearance in the literature. He also did ground-breaking work with tornado outbreak climatology (Galway 1977; Galway and Pearson 1981).

As if all of this (and I have by no means exhausted the list of his scientific publications) wasn't enough, Joe was a dedicated historian of severe storms forecasting and research. His historical works, formally published only after his retirement, set a high standard by which all such works will be judged in the future. His biography of J.P. Finley (Galway 1985) is superb in its scholarly content, in the value of its personal insight, and its readability. To me, it's obvious that Joe recognized a kindred spirit in the determined Finley, who fought numerous battles with ignorant "superiors" because he cared deeply about what he was doing. The next two papers (Galway 1989; Galway 1992) are superb summaries of the history that only someone like Joe, who was there virtually from the beginning, could write. The last of these was completed as his health was declining ... I'm terribly grateful he finished it and saw it published ... it's a masterpiece.

Personally, I first met Joe in the summer of 1967, when he had assumed the duties of PA. I had been a student trainee in Madison, WI during the previous two summers. The magic of the names in SELS was not lost on me. When those watches came clattering over the "RAWARC" teletype (Lord, has it been that long?), sending the office into a flurry of activity, each one had a name on it: Galway, Wood, Magor, Crumrine, Sanders ... . Those names had mysterious, God-like qualities to a would-be meteorologist. I was seized by hero worship immediately.

The last year I was at the University of Wisconsin (1967), Dave Barber (then a graduate student), encouraged me to apply for a student trainee job at SELS after I graduated. By some profound miracle, I was accepted and gratefully traveled to Kansas City for the first time, arriving in a tornado watch, appropriately enough. I met Joe and he turned out to be warm and helpful and not at all intimidating, in spite of my being awestruck at meeting him. I had two and a half summers at NSSFC as a student trainee. I thought I'd died and gone to heaven, and Joe's presence was a big factor in my joy over being there.

Subsequent events, too numerous to detail, only reinforced my respect for Joe. Time passed, and after various misadventures, circumstances brought me back to SELS, this time with the ink still wet on my Ph.D., Joe had been "demoted" back to the forecast desk ... nothing like a demotion in reality, but almost certainly an inappropriate wounding to his pride. Nevertheless, to my delight, I was now in a position to work forecast shifts throughout the year with Joe. Working with him on shift wasn't always pure joy. He could be intense and demanding, and had his own ideas about what were appropriate things to do at certain times. There were times when it got downright unpleasant, for a time; Joe was not a saint and had some vices that he began occasionally to bring to work with him. It's an important measure of the man that these difficulties he eventually overcame, like most other obstacles in his life, by the time he retired.

It was my privilege to be able to provide some small measure of assistance as he began to publish more often ... the bureaucracy by which the SELS forecasters found their publications being suppressed had lasted into the mid-1970s. When the Techniques Development Unit arrived in 1976, we became the de facto publication judges of the SELS papers, and we did everything we could to push their ideas into publication, rather than trying to suppress them. It's a transformation to which I'm pleased to say I contributed. Joe's papers were a joy to read, even in their first draft form. If I made any contributions to them, they were trivial ... he was a role model in that way, as well as his forecasting.

I made a special point to travel to Joe's retirement party in 1984. I took the opportunity to thank him for all he had meant to me ... I'm happy I said those things then, so there could be no doubt that he knew them. I don't have to feel bad about not having said them to him, after his passing. I know I'm not alone in saying that Joe was a significant influence on me and my career. He has been an inspiration to many of us and his contributions will survive for many years to come. He's a hero unlike most of our youthful heroes ... he lived up to my expectations, and more! He's already missed. cdoswell@telepath.com


Galway, J.G., 1956: The lifted index as a predictor of latent instability. Bull. Amer. Meteor. Soc., 37, 528-529.

______, 1966: The Topeka tornado of 8 June 1966. Weatherwise, 19, 144-149.

______, 1977: Some climatological aspects of tornado outbreaks. Mon. Wea. Rev., 105, 477-484.

______, 1985: J.P. Finley: The first severe storms forecaster. Bull. Amer. Meteor. Soc,66, 1389-1395.

______, 1989: The evolution of severe thunderstorm criteria with the weather service. Wea. Forecasting, 4, 585-592.

______, 1992: Early severe thunderstorm forecasting and research by the United States Weather Bureau. Wea. Forecasting,7, 564-587.

______, and A. Pearson, 1981: Winter tornado outbreaks. Mon.Wea. Rev.,109, 1072-1080.

Lewis, J.M., 1996: Joseph G. Galway. Wea. Forecasting, 11, 263-268.



Israel is "a bridge between continents" - Africa, Europe and Asia. The variety of weather patterns and systems affecting this region is large, and Israel lies in a small area that combines all of them - European rain, Asian cold and African heat. This is the reason for the weather that Israel experiences, some of which is severe. It is common to divide the severe weather in Israel into two categories: winter weather, and spring/autumn weather.

Severe winter weather may occur from December through March, when low pressure frontal systems from Europe pass through Israel, but it is rare usually forming as local events. The cold fronts are very active, as the air ahead of it is coming from the deserts surrounding Israel, and the air behind it comes from northern Europe. The most severe thunderstorms in winter are usually along the cold front, and they are accompanied by heavy hail, high winds and heavy rains. Supercells are rare, although a few may develop. Tornadoes can be found under the most severe thunderstorms, but they are weak. Unfortunately, with very few chasers and spotters and almost no radar data, tornado sightings are rare. But we do know they develop, and even more frequently than what was thought before.

The winter thunderstorms are most frequent, but usually the least severe. The most severe thunderstorms, which causes the loss of human lives and property every year, develop during the hot days of autumn and spring. During those months (October-November, April-May) the Sudan low, which affects Equatorial Africa all year round, sometimes centers itself between Ethiopia, Yemen and Saudi Arabia. A trough begins to move northwards from that low, along the Red Sea, and is called the Red Sea Trough (RST). The trough may reach Israel and even all the way north to Turkey. When the RST starts to move northwards, the weather in Israel is usually sunny and hot. As the trough develops, it becomes even hotter, when winds blow from the E or SE, and temperatures can reach up to 48 degrees C. At this point thunderstorms develop almost daily along the Red Sea coast of Arabia and Sudan. Because of the lack of meteorological information from those fairly remote areas the nature of the storms is not known, although it is obvious some are severe. When the RST reaches Israel another trough in 500mb comes down from northern Europe. This has two main effects: 1) Israel is at the point of both the 500mb and the surface trough, which encourages cloud growth. 2) The southerly flow in 500mb brings very humid and unstable air masses from tropical Africa, while the surface remains hot and very dry.

The instability under these conditions is huge. But when the humid unstable air masses move north from the tropics, they go through a drastic change. As the air masses pass over the desert regions of Africa and the Middle-East they lose their moisture, and with it their ability to produce active clouds, yet the huge instability remains. Every tiny change in their moisture content may determine the fate of a potential cumulonimbus. In many cases not enough moisture is supplied, and an alto-cumulonimbus cloud forms. A severe feature of this cloud is intense lightning, most of which are cloud-to-ground, about every second or so. The precipitation from this cloud is very weak, and most rain evaporates before reaching the ground. But this causes one more severe feature: A dry downburst. Great damages to property and life in the Middle-East have occurred as a result of dry downbursts. The surface air under the alto-Cb is extremely hot and dry (humidity less than 10%). The rain falling from the thunderstorm under the downdraft evaporates as a result of the dry air. Evaporation causes the air to cool, which increases the speed of the downburst. Sustained winds can reach 160 km/h (100 mph) or more, for a few minutes.

In Israel, the dry downbursts usually occur over the desert regions, in the south and east, when the humidity is very low. But the most severe and destructive weather in Israel occurs when surface humidity rises. This can happen either by southerly winds, bringing moisture from the Red Sea to central Israel, or by westerly winds, bringing moisture from the Mediterranean. Southerly winds are rare, but westerly winds are common, when the RST reaches the Mediterranean Sea, and causes the formation of a shallow low pressure area that drives relatively humid air from the sea into Israel. As said above, the instability of the air masses that reach Israel from Africa is huge. Even a tiny rise in moisture can give the air mass back it's power. Supercells may form within minutes in different areas of Israel, with flash floods, hailstorms, and even tornadoes. These supercell outbreaks, which usually develop across southwest Jordan, central and southern Israel, and Sinai, cause more deaths and damages than any other weather event in this part of the Middle-East. They are hard to forecast, and the Israeli public doesn't always heed the warnings issued by the Israel Meteorological Service. A large outbreak occurs every 2 to 4 years, but smaller, local ones may occur twice or more every year.


On November 2, 1994, under a deep RST, southerly winds in Eilat (southern Israel) caused a sharp rise in humidity. Around 13:00, within minutes, a supercell developed above Eilat and Sinai, and two large waterspouts dropped side by side, tossing dozens of boats and ships on the Red Sea and causing great property damage, but luckily no deaths were reported. Hundreds of people died in Egypt that day, mostly from floods, but also from wind damage (not enough data from the damaged areas...). Dozens of injuries were reported in Israel as a result of downbursts with winds in excess of 80 mph. On October 17-18, 1997, at least five supercells developed across Israel. Thirteen people died from the massive floods which occurred in the Dead Sea Valley. A hailstorm with baseball size hail which lasted several hours left three feet of accumulation, scores of injuries and millions of dollars of property damage. Eilat received a yearly average rainfall in just six hours. Hundreds of hikers, in spite of the warnings, were trapped in many areas of Israel.


Chasing in Israel is not so hard: It is a small country, and in eight hours one can drive from one end to another. The roads are modern, and there are plenty of them. Traffic can be a problem, but not everywhere. The two main problems with storm chasing in Israel are: 1) There are no mobile radars and weather stations that can provide instant soundings and data. 2) Weather forecasting for the RST is hard, because crucial information on the weather conditions in many places in the Middle-East is insufficient, or none at all. In addition, Israeli citizens cannot go to Saudi Arabia or Sudan, which is very unfortunate, because there must be some beautiful storms over there. Going through the border to Sinai from Israel may take an hour or more (customs, security), which will spoil any on-going chase. But once a good chase is on, spectacular events can be seen, such as intense lightning over the Dead Sea at dusk, flooding in the Judean desert creating huge waterfalls, and even the occasional tornado. For questions and comments, please e-mail me: davidsh@geocities.com or check my web-site, for photographs and chase reports: www.geocities.com/CapeCanaveral/Hangar/1911/


 April 15, 1998 RED BUD, IL CHASE by Jay Antle

On the morning of April 15, 1998, my chase partner (John Moser) and I headed toward St. Louis, MO. on I-70 from Lawrence, KS. Our initial target area was a St. Louis MO/Carbondale IL/Mt. Vernon IL. triangle. This was much further than we were used to trying on a one day chase. However, the dynamics looked fantastic with EHI's forecasted close to eight with helicities over 500 and CAPES near 3000. Moisture was not going to be a problem with dewpoints forecast well into the 60s. As a result, the SPC (Storm Prediction Center) issued a HIGH RISK for southeast MO. and southern IL.

Near Columbia MO at 11 am, we registered 70 degree temperatures and 60 degree dewpoints which boded well as we kept going towards St. Louis. By this point, we were already in the moderate risk zone as per the SPC. We guessed that a tornado watch would likely be issued for the St. Louis area by the time we arrived there (not exactly a bold prediction). We drove through broken cloudiness from Columbia to near St. Louis which caused us to worry about surface heating. However, nearing St. Louis (about 1:20 PM), the clouds gave way to VERY hazy sunshine. The visibility was poor but at least we had good surface heating. By 1:40 p.m., we were in a PDS (Particularly Dangerous Situation) Tornado Watch Box extending south and west from our location at St. Louis. Radio reports indicated that convection was beginning in south central Missouri (near the Lake of the Ozarks). We considered cutting south on I-44 towards Rolla (and even drove down it a few miles) but were faced with unfriendly chase territory (lots of hills and trees) and I knew chasing in southern Missouri was impossible for that reason. So we would have to wait until this stuff crossed into more friendly terrain in Illinois. But where to cross the Mississippi River? Once you get south of St. Louis, you have to travel almost fifty miles south to Chester in order to find the next bridge across the Mississippi. We decided (given what we were hearing about where the convection was building) to head south on I-55 and cross at Chester.


By 3:30 p.m., we stopped at a rest area on I-55 about twenty miles northwest of Chester and got data from Chris King and Mike Phelps who were both monitoring things for us. Given how terrible the visibility was -despite sunshine, their help was invaluable. A supercell was coming out of central Missouri. We could also see a line of low cumulus with a few congestus mixed in extending east to west. This was the infamous outflow boundary along which convection was to fire during the afternoon according to forecasts. Dewpoints on the south side of this boundary jumped from 62 to 68 F. We decided to head further north and cross the Mississippi at St. Louis (a poor decision) and then make any northerly or southerly adjustments on the Illinois side. We were also hoping that other convection might go up along the outflow boundary we had just left to our south. We assumed this supercell would still have a slight northerly component to its movement. As it turned out, it moved more east than north as it crossed the state. By this point, on a non-hazy day, we should have been able to see the anvil of our target supercell...but not with the poor visibility on this day.

By 4:10 p.m., we began to see nice anvil mammatus to the west and the skies darkened quickly as we booked north on I-55 trying to cross the river at St. Louis and then get south on the Illinois side. By 4:30 p.m., we crossed into Illinois and a tornado was reported on the ground near St. Clair Missouri moving in our direction. By now, inflow bands were clearly visible feeding into the dark skies to our west as we crossed the Mississippi River. At this point, poor maps betrayed us. Our strategy was to drive through the town of New Hanover and end up at Valmeyer. Illinois is one of the few states in chase country that neither one of us had good maps. To make a long story short, we were about three miles further north than we thought. Further, the storm had no real northerly movement to it. We were now in the flood plain of the Mississippi staring up at a supercell and we were too far NORTH. By 4:48 p.m., we were sitting on top of a flood levee, had several good escape routes, and were congratulating ourselves for our good placement as we had great views across the river and into Missouri. At this point, we discovered we were in trouble. By 5 p.m., the National Weather Service suggested that the possible tornado would cross into the Festus, MO area by 5:30 p.m., so we began to head south on county roads (which our map did not handle particularly well). Inflow into the storm was clearly obvious by watching flags and trees as we drove south. Great inflow bands were visible with lots of lightning to the west with the obligatory green skies. Clearly, this cell qualified as a BEAST!!!

By 5:15 p.m., the precipitation core was on us, marble to dime-sized hail pelted our car, and we decided to head northeast towards Waterloo on roads that were actually ON our poor map. Looking back at this, it seemed clear that the storm back built and the mesocyclone produced the tornado near St. Clair began to weaken while the second mesocyclone formed further south and crossed the river near Festus. In any case, we were on the north side of the precipitation core and maturing mesocyclone. So much for our self-congratulation about good placement. At Waterloo we decided to head south and east on Highway 154. The NWS reported a possible tornado 7 miles west of Crystal City MO. (On the west bank of the River near Festus) at 5:15 p.m.. We decided a true core punch in this situation was out of the question. By this point, tornado warnings were out for Monroe and Northern Randolph counties. (We were in Monroe County. The town of Red Bud was in Randolph County).

By nudging the core, we encountered no worse than marble-sized hail (we managed to avoid the reported baseball sized hail with this storm). We let the worst of the hail pass to the east and then crawled south and east on highway 154. As we got south of Waterloo, (roughly 5:40 or so) we heard the tornado sirens wailing. Along the way, I discovered a new chasing resource. We passed a bar with an open door and a satellite dish (PKG liquors). We stopped, I ran in and got some quick radar data from the local TV meteorologists who were on air covering the situation. Everyone was tossing down a brew apparently "underwhelmed" by the whole event. The big tornado on my T-shirt started the usual round of "TWISTER" questions but we were in a bit of a hurry.

Just north of Red Bud, we broke out of the core. The ground was covered with hail which gave off a great deal of fog and reduced visibility to near zero for about half a mile. Looking to our east we could see the core and an impressive lowering to our south but poor contrast hampered further identification. At about 5:55 p.m., however, just to our west, there was a large "scuddy" lowering which almost reached the ground and showed some signs of rotation. Trees blocked view of any ground contact. Further, the hail fog was blowing INTO it. We accelerated quickly to get out of the way. Frankly, I'm still not sure what to call this thing. Surface winds were not that strong near it and the rotation was fairly weak. We watched it meander off to the northeast and weaken. At about this time, John Farley was south of Red Bud and reported several large lowerings in his chase account for the day. After comparing notes with John Farley and reviewing the video, I believe the lowering/tornado he caught on video was the one we saw to our south as we broke out of the core.

At the same time (roughly 6 p..m.) a tornado was reported by spotters two miles west of Red Bud (perhaps our lowering?). We stopped to look at the situation again two miles NW of Red Bud on Highway 154. As we drove through Red Bud the sirens were going off there as well. Once when we pulled over to the side of the road south of town we were met by the local emergency management coordinator who was out spotting and confirmed to us that one of his spotters had reported a tornado. By this point, roughly 6:15p.m., we also unfortunately picked up two vehicles of locals who were looking for "TORNADERS" who followed us for almost twenty minutes.

We decided to leave this cell and pursue interesting lowerings on a cell near Ruma to the south of Red Bud on Highway 3 (ever in search of tail-end charlie). After watching these lowerings (which never did much from our vantage point for about half an hour) we headed back north to Red Bud and stopped at the local Hardeeís for a break before we headed back home (a LONG drive home ahead of us). We knew there were severe warnings up for Monroe County on a new storm just to our north but we didn't expect much from an already worked-over atmosphere. In Hardeeís we talked to locals about the storms (one kid reported seeing a funnel cloud) and answered the usual TWISTER questions.

At this point, the Hardee's effect took over. John and I have gone into Hardee's burger joints (which have lousy food) and come out to find convective nirvana. Well, it happened again. This was 7:10 p.m.. We came out of Hardeeís, looked up and our jaws dropped. Right above us was a extraordinary, tiered, shelf cloud which extended to the north and east. We turned on the weather radio and a tornado warning was out for Monroe County (sirens were going off in Waterloo.). By now, it was getting close to dark but we headed north to see what we could see. Once we crossed the shelf cloud, the winds really didn't do that much. (Stopped at another "weather" bar for a quick look at radar. These stops take no more than 40 seconds and are very helpful). By this point, tornado warnings were going out all over the place to the north and east of us. By 7:20 p.m., as we were just south of Waterloo, we could barely make out a sloping feature half hidden by rain curtains off to our northeast. This could have been the wall cloud from the tornadic storm moving out of Monroe County. By this point, is was getting dark and we called it a day. We made the long drive back and reached Lawrence, KS around 1a.m..



I conducted a ground damage survey of the tornadoes which occurred on April 16th. The damage track was about 70 miles long, through some pretty rough country, so I couldn't look at all of it. Based on the severity of the damage at several points along the track, I believe an F-5 rating is justifiable. Lawrence County EMA personnel asked me to accompany them and give opinions regarding a fair rating for the tornado. I met Barry Roberts, a UAH grad student and Lawrence County EMA spotter, in Loretto TN. We had to ride in a heavy duty pickup truck, because most of the roads accessing the damage are bone-jarringly rough and rutted, with lots of mud everywhere. We decided to start the survey near Brace TN in northern Lawrence County, and work our way upstream. The tornado also did severe damage extending east-northeast into Maury County.


The reader may assume that, unless otherwise indicated, all severe structure damage described here involved complete removal of all major superstructure components from the original foundation footprint. Such damage looked very much like the standard picture of F-5 damage that appears regularly in the back of Storm Data. The only issue was how well constructed the buildings were. It was clear that damage near Brace was severe. The damage swath was at least 0.5 miles wide, with major forest blowdowns in spots. The few houses that got hit sustained major damage, if not outright obliteration. Our first hint of F-5 damage came with a well-built, two-story brick and frame house (with basement opening to the backyard). The house was totally destroyed, with only a clean slab remaining. Tilted anchor bolts were clearly visible protruding from the foundation. The superstructure was spread out over 100-200 yards into a forest behind (N) of the house. Some debris fell into the basement, where the residents took cover. They survived, although one person is still in the hospital. The yard was now mud, with all the freshly planted sod having been stripped away by the tornado.

Several miles upstream, we found an area near a small pond where the tree blowdown rate was near 100 %. Interestingly, this tornado blew down trees in all directions about equally. There was little evidence of an along-track bias in blowdown direction. More houses were obliterated in this area, along with a high-voltage support tower.

Major forest damage was evident upstream for several miles, then there was what appears to be a small break in the damage swath. We cannot confirm the discontinuity from a ground survey alone, due to lack of roads and a patchwork of farm fields with few trees. There was then a gradual increase in damage again toward the west-southwest. By the time we got to Deerfield TN, especially the Piney Road area, damage was again colossal. Here a well-built stone house was destroyed by the tornado, with only a stub of a chimney left standing. A country grocery was also leveled, but much of its debris remained near or on the original building site. Along Piney Road, several more houses were annihilated. Their construction was of variable quality. One house had no apparent anchoring of the superstructure to the masonry foundation. The house was completely gone, and we could not see its remains from the vantage point of the clean slab. Just upstream, a house that was under construction (framing was up, but no wall panels) suffered one surprising bit of damage. The 3.5 inch thick solid concrete slab of the front porch was lifted up by the tornado and broken into pieces and scattered across the front yard.

The dirt road serving this and nearby houses was polished clean. Across the street, some of the fresh 2x4 framing from this house was driven into the soil almost full depth, with only an inch or two of wood protruding. Some of these wood members could not be budged, even with considerable manual effort. Their actual depth of penetration into the ground is not known, but it was not shallow. It was in the Piney Road area that a new and interesting piece of video was found by a local TV station Friday. A fellow calling himself a "storm chaser" got decent hand held video at a range of one mile or so of this phase of the tornado. It showed a Dimmitt-like funnel (but with a bit more tilt) ripping up houses. One house broke up and the debris jumped up nearly to ambient cloud base before being centrifuged out. A few seconds later something else spewed a spiral of black debris around the violently rotating funnel. This video showed just how impressive the rotation of this tornado was. It looked comparable to that of other F5 tornadoes I've seen on video. In the Piney Road area, it was apparent that the damage swath was wider than the funnel cloud envelope at the surface.

Further upstream in Wayne County, Police had the highways blocked off, but they let us in to see the damage. In a hollow near Highland, TN, forest and structure damage was again severe. In this area, several fatalities occurred. A grocery store-gas station building was turned to rubble, and ~200 yd W a house was completely destroyed. Several fatalities occurred in association with the disintegration of this house. One of the bodies landed in the aforementioned grocery store rubble. Forest blowdown nearby was again extensive, often more than 90% in the hills and hollows west-southwest. The swath width here was roughly 0.8 miles, but an aerial survey was needed to arrive at accurate numbers. Much of this part of Wayne County has been clear-cut by logging companies, with only a carpet of short young pines in much of the area, which limited the damage evidence in some places. In the still-forested hollows, which were mostly deciduous trees, damage was near complete. Ground crews working to get to a damaged area in the back-country reported they had to work for hours with chainsaws just to clear a driveable path on the roads. This was the area where the "mile-wide" wedge video was obtained by San Satterfield's field spotter/chaser. Photogrammetry should be done to determine the actual visual dimensions of the funnel. We have not had the chance to try to do this yet.

The tornado actually continued to produce reportedly severe damage on into southern Maury County TN before lifting. As I mentioned earlier, I intercepted this storm in Marshall County, but it was already much weaker there. Why did it weaken so quickly? And why was it so intense earlier? Inspection of the composite radar loop shows that shortly before this supercell came along, there was another storm that took a rather similar track, and also dissipated quickly after crossing I-65. Perhaps it left a fresh boundary for the Wayne County storm to work with. The radar loop also shows that the Wayne County storm inhaled a small cell on its S flank about the time it was near I-65. However, instead of invigorating this major storm, it appeared to contribute to disrupting and perhaps killing it. This kind of "death merger" is common around here; I've seen the same thing happen to some other supercells. The question is: Why? After all, I've seen many mergers lead to storm intensification.


I was watching the weather, the day before, because I expected the worst of it to come through with the warm front. I even had my VCR taping the local TV station, but no storms developed. The next morning, I wasn't very concerned when I got up at 6:30am and saw storm clouds. Listening to the radio while going to work, I quickly realized that this storm was a serious one, and I even considered going back home and setting the VCR up again. Once at work, I kept my head-phone radio on the various local news-talk and TV stations. I found out that the first dangerous wave of storms came through at about 7:00am; they had spawned tornadoes in Western TN. The next wave of storms (with a tornado warning) came through at 12:00pm. I listened to the TV station as they plotted the probable course of the storm; these storms were going straight across southern Davidson county, which was rare. They usually come across to the north or south of us. A tornado was reportedly sighted to the south-west of where I work but no damage occurred.

My workplace is located about a mile from the city, on top of a hill, where we can see the downtown buildings. I went outside at about 2:50 p.m. and noticed that the air was very hot: it was unusually hot for earlier in the day. At about 3:00pm, the TV reported that another tornado had been spotted, going in the same direction as the noon tornado. I walked around, looking out the windows. To the west and north of the city, the sky turned the most awful color of a storm I have ever seen: it was a very dark green-grey color. The TV station I was listening to is located to the west of the city, so it was the first to feel the effects of the tornado as it formed. I was listening to them as they started to sound more and more anxious. I heard one of them say to the other: "Did you feel that?" He had felt the room shake. He also asked if the door had moved. Soon after that, the station went off the air. I got up and went into the breakroom, where a TV was on. A different station was showing a camera view of downtown Nashville and it was showing just very dark grey sky. I went into where some women work and they were already up and going to the restroom for protection. One of them was saying, almost in a panic, that it was headed this way. I went to the large front window, which affords a perfect view of downtown, and saw that others were already there. This is what I saw: It was a low, grey mass of clouds, with a wedge-shape wall cloud west of the city. I seem to remember it being brownish at it's thickest. I thought that it was a wall cloud, but later found out that the whole thing had been the tornado. I should have know what it was, because the cloud went all the way to the ground. It was so huge! It was larger than all of downtown Nashville. It moved slowly over the city, dimming it out with darkness. The clouds moved very slow in a counterclockwise rotation: it moved so slowly that that was another reason why I couldn't imagine that I was seeing the tornado itself. I expected a much faster rotation. The clouds around the tornado were moving crazily: one puff of grey cloud was closer to us, and I watched as it moved in a random way: there was no definite direction to it. The main tornado slowly crept over the city and we began to see flashes in the city. I thought it was lightning, but found out later that they were transformers. We watched as the tornado crossed across our sight; it was very hard to distinguish anything once it was to our west: that was where the tornado did the most damage. I never got to see the suction vortices, but I DID get to see a wedge tornado! Sadly, I heard that a man that had been injured in the storm died this morning. He was the only victim of the Nashville tornado.

Let me say that I don't think that a city can be much of a handicap to a tornado. The one I saw just engulfed downtown: it was huge. Lot's of people said that they felt the buildings sway. Damage to buildings amounted to windows blown out and the corners of brick buildings toppled. Aluminum roofs got ripped off and old brick buildings had their fronts collapsed. Trees were down everywhere. The houses seemed to be spared: I think that the trees weakened the winds enough to save them. The most incredible sight I saw was a large billboard, next to the interstate: it had eight 18 ft tall iron I-beams a foot wide, and the whole sign was bent over at a 45 degree angle! The power must have been incredible! They say that this tornado was an F2 or F3. If it had been a major tornado, I think some buildings would have toppled. A city is NOT safe from a tornado. Merdoug@nashville.com


A tornado going through downtown is every emergency managers worst nightmare. On April 16, 1998, it happened in Nashville as an outbreak of tornadic thunderstorms struck central and southern Tennessee. Nashville just happened to be in the way. The supercell responsible for the tornado developed about one hundred miles to the west of the city during the early afternoon. The storm tracked eastward pretty much along Interstate 40. A small tornado developed just south of Dickson, Tennessee on Highway 46, a mile or so south of Interstate 40, and moved eastward for about 1.5 miles before lifting. The tornado damaged a number of mobile homes and pushed two unanchored pre-manufactured homes off their foundations. There was nothing left of one home. Although the damage appeared F-5, the intensity of this tornado was probably not more than F-2. I do not know why the storm didnít produce another tornado for almost an hour but this certainly gave the folks in Nashville time to prepare.

A large rotating wall cloud developed just west of downtown Nashville around 3:00 p.m. F-0 damage began about three miles west of downtown with occasional downed limbs and power lines as well as displaced roofing shingles. The damage increased to F-1 about one mile west of downtown along Charlotte Avenue when a metal building lost part of itís roof and one wall. Circulating winds beneath the wall cloud knocked down trees and flipped poorly attached roofs on itís way to the capital building. Then, like a scene right out of the movie Independence Day, the dark saucer-shaped object emerging out of the haze right over downtown. Office workers in the high rises suddenly had "front row seats" to witness what was about to happen. Fortunately, many workers fled into interior hallways, just as thousands of glass windows cracked or were shattered by flying debris. Winds of around 100 mph carried debris through the streets. Over 100 windows were blown out of the Tennessee Performing Arts Center (TPAC) building. Some old masonry buildings collapsed. In general, the damage throughout downtown was F-1. As the wall cloud continued east of downtown, it crossed the Cumberland River and encountered the Tennessee Oilers stadium, a concrete structure, which was under construction. Three of the six tower cranes toppled due to the high winds and flying debris. Workers had fled into part of the structure as they saw the storm cross the river. The local sheriff began shooting some video during this time as he was just southeast of downtown. The video shows the circulation passing over a power substation and blowing out a number of transformers which illuminated the bottom of the wall cloud. The wall cloud circulation was on the ground!

For an unknown reason, the circulation would not tighten and remained broad and relatively weak through the residential and light commercial neighborhoods in east Nashville. The damage remained F-1 with torn roof shingles and an occasional partial roof removal. Thousands of mostly shallow rooted trees were uprooted. Many of these trees fell on the houses creating most of the structural damage. There were only isolated spots of F-2 damage within the ten mile path primarily due to poor attachment of the roofs to the walls. Two steel transmission towers were downed. The greatest concentration of damage occurred at the Cornelia Airport where dozens of planes were destroyed. One plane was lofted into a pickup truck and a number of hangars were destroyed. The tornado continued eastward crossing the Cumberland River again and passed just south of the Opryland Amusement and Entertainment Complex traveling across the golf course uprooting a number of trees before dissipating. At this time, a second storm formed south of town and produced an F-1 tornado that traveled from just north of the Nashville airport through Hermitage, a suburb of Nashville.

Within a day of the disaster, I received a phone call from NOAA Headquarters asking me to conduct an aerial survey of the damage. Mr. Tim Troutman, from the National Weather Service in Nashville, Tennessee, accompanied me on the survey. Numerous damage photographs were taken. A review of the radar images indicated a pronounced "hook" and strong reflectivity gradient. Similar characteristics were noted with the storms in southern Tennessee which were F-5 intensity. Why the Nashville circulation didnít tighten will remain a mystery for me for now. Hopefully, this event will be studied in greater detail.