The July-August 1999 STORMTRACK features the MAY 3, 1999 OKLAHOMA OUTBREAK
ITíS HOWIE AGAIN AND AGAIN AND AGAIN AND AGAIN AND AGAIN
I. COMMENTARY by Tim Marshall
Last year, the term "chaser convergence" was elevated to a new level when I met Dr. Howie Bluestein on top of Mt. Audubon (about 12,000 ft MSL) in the Colorado Rockies. I climb that mountain about once each decade and he climbs it once each year. Some statistician can figure out the odds of two people (knowing each other) meeting in such a remote place, but the odds are probably greater than one in a million. I figured this was just one of those things in life that was just happenstance. Then came 1999.
This year, I met Howie five times -more than in any other year. It also is the most number of times I have ever met any other chaser in a given year. Our first encounter this year was on May 16th on a crumbly paved road south of Seiling, Oklahoma. Storms fired along a dryline from western Kansas into northeast Texas. Howie and I just happen to pick the same forecast target and storm. We also just happend to pick the same rural road -as we both like to avoid the traffic on the main roads. Four days later, I met Howie again on a tornadic storm north of McLean, Texas -again on a rural road -and again the same forecast target. A few weeks later, dryline storms fired from western Kansas to southwest Texas. Numerous supercells developed. This time, I played storms further south -and so did Howie. We met again on a rural road northwest of Aspermont. The storm evolved into an LP supercell and moved slowly southeast . Tim and Howie on a remote road near Aspermont, Tx.
Our most bizarre encounter was in Amarillo, Texas on June 14. After a long, frustrating chase in the Texas panhandle, Carson and I decided to get a steak dinner in Amarillo. It was either going to be the Country Barn or Big Texan, both are on the east side of Amarillo. When we got there- the power was out due to an intense lightning storm. Thus, we decided to stop at a BBQ place on the west side of town. We arrived at the restaurant ten minutes before they closed. Five minutes later, Howie and his crew walked in the door. "Hi, Howie", I said. "Tim Marshall again", said a laughing Howie. Then, he told me they initially wanted a steak dinner and went to the Big Texan, but the power was out, so they decided this BBQ place. Wow! -same story -same time meeting. Gee, I never have met any of my neighbors or work colleagues even at the local Wal-Mart or restaurant, but yet, I keep meeting Howie in remote places all over the country!
The last encounter of the year was two days later in Norton, Kansas. Carson and I stopped for lunch at a McDonalds. We rarely stop at a McDonalds -but it was the only thing around. Five minutes later Howie and his crew drive up. (He rarely stops at a McDonalds too). As Howie approached the door, I opened it and said "Welcome to McDonalds Howie!". "This is incredible!", Howie said. "I hope when I go home, you donít open up the door!".
II. CHASER NEWS
The August 1999 issue of Weather and Forecasting is entirely comprised of articles on the history of tornado forecasting. It is an excellent compilation of the work of Fawbush and Miller by authors Maddox and Crisp. Other articles are about the birth of the Storm Prediction Center, storm spotting, and storm chasing by authors you know -like Doswell, Moller, Brooks, Ostby, Corfidi, Bradford, and Bluestein. The single issue price is $15 for AMS members and $35 for nonmembers, however, you can subscribe for an entire year for $35 if you become an AMS member. Just write to the AMS (American Meteorological Society) at 45 Beacon Street, Boston, MA 02108-3693.
80th AMS Annual Meeting - The annual meeting of the American Meteorological Society will be held at the Long Beach Convention Center in Long Beach, California from January 9-14, 2000. There will be many symposiums, however, symposiums of interest to chasers will be: The Mystery of Severe Storms: A Tribute to the Work of T. Theodore Fujita and Project Vortex: What We Have Learned and Where We Must Go. For registration, hotel, and general information contact the AMS or visit their web site at http://www.ametsoc.org/AMS, or phone: 617-227-2426 x 226, or x227, or x305).
ENVIRONMENTAL CONDITIONS LEADING UP TO THE 3 MAY 1999 TORNADO OUTBREAK by Rich Thompson - SPC forecaster
During the late afternoon and evening hours of 3 May 1999, a series of devastating tornadoes ripped much of central and northern OK, as well as southern KS. In excess of 60 tornadoes were reported in OK and KS, including large/violent tornadoes that struck Moore, Dover, and Mulhall in OK, as well as the south side of Wichita, KS. In this paper, the evolution of environmental conditions prior to the 3 May outbreak are discussed and compared to 26 April 1991, which is considered a "prototypical" tornado outbreak in the Plains, and important considerations in forecasting Plains tornado outbreaks are summarized.
2. Environmental evolution from 00-23z on 3 May 1999
a. Temperature and moisture profiles
Surface data from 00z 3 May (Fig. 1a) show that low-level moisture return was underway from central and north Texas into central OK, with mid 60 dewpoints available in TX. By 12z (Fig. 1b) on the morning of the 3rd, 60 dewpoints had spread as far north as the Wichita, KS area, and mid 60s covered areas from central OK southward. As of 18z (Fig. 1c) on the 3rd, upper 60 dewpoints were present from southwest OK to central TX, with mid 60s as far north as the Wichita area. Along with the moisture increase, surface temperatures warmed to the upper 70s in KS and low-mid 80s in southwest OK and northwest TX.
Figure 2 shows an overlay of the 12z 3 May and 00z 4 May soundings from Norman, OK. The 12z Norman sounding revealed a moist boundary layer about 1 km deep, with an elevated mixed layer from 825-600 mb located on top of the moist layer. The 00z 4 May sounding from Norman showed substantial warming from 600 mb to the surface, as well as some increase in the depth and magnitude of boundary layer moisture. The combination of steep midlevel lapse rates and relatively high boundary layer temperatures and dewpoints supported surface-based cape values in excess of 3500 Jkg-1 , when the 00z OUN sounding is modified for 80/68 F in proximity to the developing supercells near Lawton. (Note the loss of data above 300 mb at 00z which results in a loss of at least 500 Jkg-1 of CAPE. The moist profile above 500 mb denotes the anvil of the Bridge Creek/Moore tornadic supercell).
b. Vertical shear profiles
In conjunction with destabilization over KS/OK/TX during the day, a mid/upper level shortwave trough was moving eastward from AZ toward the southern Plains (see Fig. 3). The large-scale pattern was somewhat similar to 26 April 1991, though the midlevel trough was more amplified and farther east in the 1991 case. The 12z soundings from 3 May 1999 case did not appear to sample a mid-upper level speed max over central and eastern NM. However, time series of profiler winds at Tucumcari, NM from 10-15z on the 3rd (Fig. 4a), and later plots from Jayton, TX (Fig. 4b), Vici, OK (Fig. 4c), and Purcell, OK (Fig. 4d), all show dramatic strengthening of the flow above 5 km from mid morning over eastern NM to late afternoon over western and central OK. Deep layer vertical shear supported supercells by early afternoon in the warm sector, with 0-6 km shears ranging from 45 kt at 23z to 60 kt by 02z. In addition to the strengthening mid-upper level flow, low-level winds strengthened and backed at Purcell by late afternoon. Estimated 0-3 km SRH values at Purcell (east of the dryline) increased from 50 m2s-2 18z to roughly 250 m2s-2 by 23z. Through the evening hours, estimated SRH values remained in the range of 250-350 m2s-2 for the observed storm motions to the northeast at 25 kt.
The evolution of the troposphere followed a typical sequence of events preceding a tornado outbreak: Steep lapse rates overspread the area, boundary layer moisture increased, daytime heating further increased instability, and deep layer vertical shear simultaneously increased. In these regards, 3 May 1999 resembled a "prototypical" tornado outbreak. To further illustrate this point, the soundings and hodographs from 3 May 1999 and 26 April 1991 are compared in Figure 5a and 5b. Each sounding shows a deep layer of strong buoyancy, and the structures of the hodographs are remarkably similar. Surface analyses revealed a dryline across northwest TX and western OK by mid-afternoon on 3 May 1999, with a deep low to the northwest of OK. In contrast to 26 April 1991, surface convergence along the 3 May dryline was ill-defined (Figs 1c,d and Figs. 5c,d), and the initial supercell developed about 75 miles east of the surface dryline position. Also, a large plume of cirrus developed in the lee of the Rockies over eastern NM by mid-morning on the 3rd. The cirrus overspread much of the TX Panhandle and western OK by early afternoon, raising questions about continued heating and mixing in the boundary layer through mid afternoon. The lack of convergence on the dryline and widespread high clouds made the prospects for convective initiation less certain than for 26 April 1991.
In contrast, the 00z Norman sounding (modified for surface observations at Lawton) showed little convective inhibition and a level of free convection within 2 km of the ground. This thermodynamic profile suggests that relatively weak lift on the mesoscale could initiate deep convection. Additionally, a break in the high clouds existed over northwest TX and southwest OK through mid afternoon. The first small cumulonimbus clouds formed over northwest TX around 20z and quickly dissipated, while the Moore tornadic supercell developed to the southwest of Lawton around 2030-2045z. There was likely ambient large-scale ascent over western OK in association with the approaching mid-upper speed max from west TX, and the cirrus hole appeared to be crucial in allowing continued surface heating/mixing to maintain weak capping.
Finally, visible satellite imagery revealed a series of billows ("wave clouds") oriented north-south over north TX and south central OK. These clouds denoted a relatively shallow, more strongly capped boundary layer with more backed surface winds compared to northwest TX and southwest OK, where deep convection initially developed. As the southwest OK storms approached central OK, the storms encountered the confluent zone and greater storm-relative helicity (in the region where the billows were located prior to being obscured by high clouds). The greater low-level shear over central OK, where surface-based CAPE was still in excess of 3000 Jkg-1, appeared to coincide with the most significant tornadoes.
The 3 May 1999 tornado outbreak in OK showed many large-scale characteristics of historical tornado outbreaks. A mean trough was located over the Four Corners area, with a deep high Plains surface low. An embedded mid/upper level shortwave trough moved northeastward from the mean trough during the afternoon of the outbreak. The 00z Norman soundings for 3 May 1999 and 26 April 1991 were similar in terms of vertical shear and CAPE. However, convergence on the dryline was stronger in the 26 April 1991 case, and there were no large areas of high clouds over the warm sector on 26 April 1991.
There are a few important points to be learned from the 3 May outbreak. First, the most intense tornadic supercells are not necessarily associated with the most extreme CAPE/shear combinations. Many May-June events on the Plains with strong and violent tornadoes are characterized by large CAPE values (3000-5000+ Jkg-1) and sufficient vertical shear for supercells based on standard observations (VORTEX type observations are not normally available!). Second, the mode of convective initiation and the number of storms are critical. The initial storms evolved into tornadic supercells that lasted 3-6 hours in each outbreak, with no early transition to a squall line or other convective mode. The predominance of supercells and lack of a squall line on 3 May 1999 may be attributable to the lack of strong convergence ("forcing") on the dryline - it is conceivable that the outbreak would not have materialized had numerous storms formed simultaneously on the dryline! Storm motions were such that the supercells remained in an environment of favorable vertical shear and instability for many hours, allowing the storms to produce a large number of tornadoes. Third, the lack of other well-defined surface boundaries does not preclude tornadoes!
Combining these points with other aspects of these cases serves to illustrate a "recipe" for a Plains tornado outbreak:
1. Steep lapse rates with an elevated mixed layer overspread the warm sector, prior to the arrival of the maritime tropical boundary layer moisture from the Gulf. The combination of low-level moisture and steep midlevel lapse rates results in large CAPE values of roughly 2500-3000 Jkg-1 or larger. Maritime tropical air in the boundary layer (generally upper 60+ dewpoints) should be available around the northwest Gulf region at least a day before the outbreak, while soundings with deep, surface-based mixed layers can usually be found in upstream soundings over the Rockies the day before. Differential advection from moisture and lapse rate source regions results in "loaded gun" profiles.
2. Various storm-relative parameters appear favorable for "classic" tornadic supercells This usually means 0-3 km SRH of 250 m2s-2 or larger, midlevel storm-relative winds of 15-30 kt, and upper-level storm-relative winds of 40-70 kt. Be sure to remember that these are NOT magical thresholds!
3. A deep, moist boundary layer with only weak convective inhibition such that widely scattered storms can develop in the absence of strong low-level lift ("forcing"). On the flip side, a stronger cap may be desirable if low-level lift is stronger/more focused. The idea is to get several supercells, but not so many storms that storm mergers change the dominant convective mode.
4. Deep layer vertical shear increases to values supportive of supercells prior to the intiation of surface-based storms, such that the initial storms become supercells. If storms form before vertical shear is strong enough, then other modes of convection can dominate.
5. High-level storm-relative winds that are not parallel to the boundary where storms initiate, especially in cases of strong convergence along much of the boundary. If many storms initiate on a boundary and the storm anvils overlap a great deal, the storms may tend to coalesce into a squall line.
By carefully monitoring source regions for moisture, instability, lift, and vertical shear, forecasters can anticipate the range of possibilities for a given severe weather scenario. Such anticipation can then help forecasters judge the quality of numerical model guidance, and avoid falling victim to poor model forecasts.
THE ULTIMATE TORNADO CHASE by Chuck Doswell
The events on this day simply exceeded my wildest imagination. I've chased storms for 27 years and have seen many tornadoes. But I've never seen anything like the atmospheric spectacle of 03 May 1999. I had a chance to ride with the tornado that eventually devastated the Oklahoma City metroplex (hereafter referred to as the "OKC tornado") from its initial development just north of Chickasha, all the way into the western part of the city.
I left my home about 5:30 p.m. after coming from work and getting my cameras, drove west from Norman on Highway 9 to the intersection with Highway 62. From there, I went north on Highway 62 to I-44, north of Newcastle. Then, southwest along I-44 to just north of Chickasha then a U-turn and back along I-44. When I arrived at a location on I-44 just north of Chickasha, a tornado was already in progress under a spectacular storm. This tornado was a few miles west of me was moving northeastward. Shortly after I got set up on a tripod, this tornado dissipated. However, it was clear that the storm was not finished, with continuing cloud base rotation ... a the new tornado developed relatively quickly (at about 6:30pm). My path along I-44 ended up being the ideal intercept route, with no turns, no stop signs, no cross traffic, no railroad tracks ... nothing to keep me from riding along nearly parallel to the storm's track. Shortly after it formed, it was doing damage in Amber, OK (the power flash in the image). Mile after mile we traveled together, increasingly accompanied by a host of storm chasers, including two of the Doppler on Wheels trucks. The tornado hit the small hamlet of Bridge Creek (as seen via a host of power flashes in the video) in what I thought (erroneously) was uninhabited country. I wasn't really thinking about the threat to Oklahoma City right away, but then I began to notice that the tornado and its parent storm had reached a steadiness that I've never seen before. Normally, tornadoes fizzle after 15-20 minutes, but this one just kept on grinding away, like a giant mulching mower ... taking deadly aim on Oklahoma City.
Soon, driving through the rear-flank downdraft (RFD) a good part of the way, I began to notice debris falling from the air onto my car as I drove along, including some pink fiberglass insulation. Even though it was still in more or less open country, someone was being hammered! Then, finally, as we (the tornado and I ... and the caravan of storm chasers) neared the city, there was talk on the radio of a "Tornado Emergency" ... perhaps "Warning" is no longer effective, since we've overused it so much. In any case, that description certainly captured the essence of the moment and it apparently got everyone's attention. Power flashes began to occur as it entered the outskirts of the city. At the time, I thought the tornado had turned to the right, but in reality, the road had turned left! It crossed I-44 not very far ahead of me. No longer a turbulent "wedge", the tornado was changing its form into a wide cone-shaped funnel, dark gray against an inky black background. The cone then evolved into a stubby cylinder, with wild cloud base rotation above it. Power flashes became more frequent, and the dark gray sides of the funnel were streaked with brown ... something I have come to interpret as dirt and debris from human structures. Suddenly, to the left of the cylinder, I could see a second narrow funnel, which evolved into a multi-vortex tornado in its own right, as it rotated around to the other side of the cylindrical funnel. My batteries were dying, and the last view of the storm I captured on video was a stubby cylinder. Then, the pair seemed to merge into another giant wedge, as the storm and I continued to part company. I got off on 149th Street Southwest, and tried to go east, but I was blocked, first by state troopers and then by tornado debris from where the tornado crossed the street. My chase was over ... my camcorder batteries were used up, and I was suddenly exhausted. The tornado, however, still had a long way to go, through Moore, then Del City and Midwest City, crossing I-35 and then I-240. The worst of its destructive path was yet to come when I had to break off, but chasing in cities is problematic, especially when you're behind the tornado and encountering its track (as well as police roadblocks, etc.).
I returned home more or less the way I came. Everything at home was normal, but the contrast with the "world" I had left only minutes earlier was so utterly abnormal, I was dazed and a bit hyper at the same time. I reviewed my video as I made a VHS duplicate. My video had certainly captured faithfully what was in my viewfinder, but it couldn't possibly show anyone that compelling, utterly astounding ride I had. For about 30 miles and roughly 45 minutes, I was in a wild world, where all my 27 years of storm chasing had not prepared me for what I was going to witness. I came home pretty certain I had seen my first F-5 tornado (I had, of course), and I was also certain I'd seen my second killer tornado. Pleased about the former, I'm NOT pleased about the latter. I would have preferred the ride without the devastation.
TORNADOES ABOUND: GO WEST NOW! by Robert Satkus
What a devastating day. Normally I enjoy writing and submitting these summaries but this is a tough one. Bobby Payne and I went west on I-40 to the Calumet exit to await development and were somewhat depressed by the extensive cirrus shield spreading overhead. A few small towers tried to build to our southwest, but couldn't quite get their act together. After and hour or so, we could see better towers in the same area and dropped south from Hinton as storms rapidly developed near Altus and Lawton. As we neared Chickasha, we could see a left moving cell from a storm split that looked spectacular, however we knew the better stuff would be further south, and sure enough within minutes a tornado warning was issued for Grady and Caddo counties. A brief tornado was reported near Fletcher.
We dropped south of Chickasha, then west-southwest towards Cement. We got one inch hail in town and a large wall cloud was visible to the west. As we cleared a hill we could see a small cone shaped funnel and quickly, a debris cloud. At first, the tornado was more of a wispy funnel with a small debris cloud, but it eventually grew into a small cone with a vigorous debris whirl. A new circulation formed to the southeast of the tornado and a funnel came nearly to the ground, however, we couldn't confirm a touchdown. We pulled north of Cyril as the cone shaped tornado continued to our northwest. A large wall cloud passed nearly over us with a possible brief touchdown in the hills to our north.
A new circulation formed to our northeast, and we could see the beginning of a multiple vortex tornado, which I believe was the eventual Oklahoma City tornado. As we chase for a local TV station, we were told to head to new storms near Anadarko as there were numerous crews form our station on this storm. Reluctantly, we turned away from the tornado, racing north to get on a storm rapidly approaching Calumet. We watched the coverage of the monster to our south in awe at its size. Our storm was looking good though, with a large wall cloud to our east. We headed east towards Okarche as the wall cloud really wrapped up and a cone shaped funnel lowered to the ground. It was about 200 yards wide and just barely missed Okarche to the west. It moved north to just west-southwest of Kingfisher. It was one of the more picturesque tornadoes I have filmed. It slowly roped out, leaving a detached funnel just below cloud base.
We thought that was it and turned our attention to a rotating wall cloud over Kingfisher. I glanced back west and was amazed to see the tornado still on the ground, with a white debris plume and a hint of condensation above it, similar to the Union City OK tornado of 1973 as it crossed the river. As it finally dissipated, the storm weakened rapidly. By now we were hearing of the destruction in the Oklahoma City area and weren't sure whether to attempt to intercept that storm, or head back to the storm producing the Minco tornado. We turned towards the Minco storm as it was heading towards Bobbyís house in Yukon. However we could see a new storm exploding to our west near Hinton. It looked great and we felt the Minco storm would pass north of Bobbyís house, so we decided to head back west.
At first, the TV station wanted us to head away from the Minco storm, because this storm was only level 3 on radar. But other chasers on and I informed them this storm was a LP (low-precipitation) supercell. They switched over to velocity mode and about had a heart attack! "Go west now", they said! We did and as we neared Geary from the east a tornado formed rapidly, starting out as a rather sinuous, ropy vortex, becoming a larger multiple vortex and quickly evolved into a wedge. It moved rapidly northeast, and even though it was now dark, the lightning illuminated it nicely. It maintained its wedge shape for along time. At one time there was a debris whirl southwest of the tornado and later we saw a second, small cone tornado south of the wedge. This process occurred another time soon after. We followed the tornado to west of Kingfisher on Highway 33, but we were stopped by debris.
A new storm was rapidly approaching from the southwest, moving towards Loyal. We stopped south of Loyal near Omega and through the darkness we saw a small cone tornado, that roped out after about two minutes, just east of Omega. We followed this storm, again approaching Dover. It spawned another brief tornado just northwest of Dover. Just west of town we saw a new tornado form, northeast of Dover. It quickly grew into a large cone, but we were again stopped by damage, this time at Dover. Damage was extensive in town. This effectively ended the chase.
PLAYING TAG WITH A TORNADO by Blair Kooistra
One look at the weather data, and I knew it was a chase day. Instabilities were extremely high, Lifted Indices were at -10C and CAPES (convective available potential energy) edged over 3000 joules/kg along a projected dryline boundary in western north Texas and west Oklahoma. Upper air dynamics were also forecast to be in place late in the afternoon to produce rotating supercell thunderstorms--possibly tornadic. I left Fort Worth, TX around 1130 CDT, targeting a Childress-to-Altus line. I stopped for fuel, new storm data, and lunch at Wichita Falls, TX. Based on information that the dryline was being pushed eastward, I decided to head west to Vernon TX, nearer the dryline, and wait for storms to go up. I made it about half-way stopping to photograph the bluebonnet display in the 1890-era Harrold, Texas Cemetery.
By 1530 CDT, cumulus were beginning to gain height and wider bases. A tower with transient anvil quickly exploded to the west in Foard county and died. Two small cells to my north, however, were gaining strength quickly, and were near a projected moisture convergence bulls-eye. I headed west to Oklaunion, watched the cells intensify for about 10 minutes, then headed north into Oklahoma on US Highway 183. By this time, the SPC (Storms Prediction Center) had increased the risk assessment to a "High", and spotters in the Lawton area prepared for the arrival of these cells. I played catch up, going east on US Highway 70 to Grandfield, then north on Highway 36 through Chattanooga to Faxon (1633CDT), where I watched the southernmost of the two cells continue to launch updrafts, which the shear quickly toppled over.
I debated playing "tag" with the first storm, now approaching Lawton, in favor of dropping west toward the anvil developing some 60 miles west in Altus. However, spotters soon reported a wall cloud and developing tornado from the first cell, now located over Lake Ellsworth. I switched on the portable TV to Channel 7, showing a live view from a "towercam" of the wall cloud and lowering funnel though it was 20 miles to my northeast and a city away, I decided to go for this cell. By 1722 CST, I was just north of Fletcher on a farm road watching this storm wildly rotate and put down a long-lived cone-shaped ("classic") tornado. It was a beaut, but after 15 minutes of watching it, it had gained considerable distance on me. I took off north, zigzagging on a series of dirt and gravel roads, slick with red mud that proved treacherous to at least one chase group
The tornado lifted northwest of Cyril. By the time I finally figured out where I was and made it back to Cyril, the storm was again putting down a tornado; this time a large "stovepipe." I was finally able to see the tornado again to the north of Rorge, and again as I entered Chickasha from the south on Highway 92 (1815CDT). Tornado and police sirens added to the surreal nature of the scene. I was on a hill on the south side of town watching the large tornado chew through the northwest side of Chickasha and across Highway 81. The tornado then headed out (1837CDT) into open country between I44 and the village of Amber, gaining intensity. I made the mistake of getting behind a state trooper on I44 and took my place in a moving roadblock about 4-5 miles west of the wedge. Small satellite funnels repeatedly formed on an updraft trailing the main mesocyclone, and were all but ignored by the state trooper and those caught in the traffic jam. One funnel made an attempt earthward, but cut off a little less than halfway down. With the congested traffic and urban area ahead, I decided to pull off this storm and head west toward Anadarko, where the earlier Altus storm had become tornadic and was generating a large multi-vortex tornado.
Back to Chickasha (1915CDT), west to Verden for fuel, then north towards Lake Chickasha on an intercept course, I joined a four vehicle chaser caravan. We stopped briefly six miles north of Verden to watch the wall cloud generate a short-lived multi-vortex tornado. Several miles further north, the tornado returned to view (1950CDT), as a large, white "cigar," in the process of roping out. Pavement was regained (thank god!) at Cogar on Oklahoma Highway 37/152. Near the Grady county line, yet another multi-vortex tornado churned away, about 1/2 mile south of the highway. The tornado crossed the road in front of me, then set out northeast toward the Canadian River and Union City, where the DOW (Doppler On Wheels) and Dr. Howie ("Cb") Bluestein were scanning the storm. I made it across the Canadian River ahead of the police roadblock and continued to follow the storm in the deepening dusk as it passed over the little grain elevator town of Banner and finally northwest of Piedmont before darkness overtook us entirely. Spotters and TV chasers reported several tornadoes down in series during this period; power flashes on the outskirts of Piedmont lit up one funnel in a spectacular red burst. These were by far the most vigorous storms I had chased. They were moderately fast moving, too, about 30-35mph, which made keeping up with them difficult to someone not familiar with the territory. Most of my guidance once the storms passed Lawton was gathered by amateur radio storm spotters and Oklahoma City TV stations.
MUD, DOGS, AND ROADBLOCKS: OH MY! by Tim Vasquez
Near noon, the forecast for severe weather looked vague and SPC's (Storms Prediction Center) outlooks could only narrow the afternoon's action down to north Texas, Oklahoma, and Kansas. Oddly, the models had backed the winds behind the dryline, leaving no favored convergent areas. After studying the maps and models for about half an hour, I noticed a subtle boundary-layer convergence feature forecast in the 21-00Z RUC (Rapid Update Cycle) model panels near Wichita Falls. With 17Z westerlies at Guadalupe Pass, TX raging to nearly 50 knots, I suspected that the jet streak and associated dynamics were actually heading for Texas and Oklahoma rather than Kansas. With steep lapse rates supporting ducting of momentum towards this region, I suspected we'd see a response in the wind field in line with what the RUC was hinting at. I settled on a target region between Wichita Falls TX, Childress TX, and Altus OK. Right around that time my usual chase partner Gene Rhoden called me on his cell phone -- he was in Ardmore OK making his way north from Dallas and would arrive in about an hour and a half.
I met Gene at his house at 1pm and we looked at data. Veteran chaser Jim Leonard, arrived back from lunch about the same time. After about ten minutes of looking at data and discussing the situation, we agreed that southwest Oklahoma was the place to go. We were concerned about a broad chunk of cirrus moving out of west Texas, which threatened to shut down heating and keep storms from developing. However, since the mid-level cap was not strong, we had some hope that the cirrus would keep any convective initiation isolated. Gene and I packed up his Ford Explorer chase vehicle and left, while Jim continued to look at data.Driving south on I-44 in Gene's Ford Explorer, around 3pm, we could see a thick shroud of cirrus making its way into Oklahoma. We drove to Lawton, OK just in time to see towering cumulus already bubbling southwest of town. With the dryline still in Childress, we were not sure whether to abandon this cluster and move west of the moisture axis, but with nothing visible in that direction we decided to stick with what we had and move west if necessary. East of Cache, OK at 1545 CDT we pulled into a mobile home sales lot (quite ironic) and watched the towers mature and produce a weak rain shaft. Gene began shooting video of nearby towers were leaning over and corkscrewing. I had not seen shear this evident in towers since the 5/26/97 outbreak in east Oklahoma. A thunderstorm formed and moved northeast, producing a weak left-split as well as hail. We followed the right split north and east through the Wichita Mountains. At 1647 CDT, the updraft base had grown, became bowl-shaped, and produced a wall cloud with rapid motion. We parked along I-44 near Elgin, OK and by 1651 CDT a weak tornado touched down about 2 miles to our west. Debris was on the ground, but the condensation funnel wasn't connected to it, although it showed very tight, spectacular rotation. The tornado dissipated by 1653 CDT. Cumulus to tornado in only 90 minutes!This circulation occluded and a new, much larger updraft base began organizing. We got off I-44 at Elgin and moved north along US 277 to Cyril, where we parked and got tripoded video of the circulation. At 1720 CDT, a new tornado touched down about 4 miles to our west. This continued until 1731 CDT, widening into a beautiful cone backlit in shades of gray. As the storm base occluded, we noticed that a new mesocyclone was developing about two miles to the south-southwest, putting us in a dangerous spot. We abandoned our photography of the beautiful Cyril -Apache tornado and scrambled northbound on 267th Rd, which was NOT paved as was shown in "Roads of Oklahoma". The mud became treacherous and we skidded off the road and nearly dropped into a 15-foot ravine --with the real possibility of a new tornado descending on us in 30 seconds!As the vehicle came to rest, Gene and I mentally picked out spots to take cover and sized up the situation. Another look at the wall cloud (nearly overhead) showed that the strong rotation had subsided and we were no longer at risk. Now our main priority was to get out of the mess and get back on the road. I tripoded my video camera and aimed it east, where an ominous wall cloud was now receding then I did what I could to help Gene (and got quite muddy). Unfortunately the right-rear wheel had no weight on it and wanted to spin. Increasing friction with branches and digging out the mud gave us no traction -- an exercise in futility.Chaser Blair Kooistra drove by, getting out of his vehicle and braving mud to offer help (a big tip of the hat to you, Blair!), but Gene still wanted to work the truck out. As Blair left, a Cyril resident with his family and a pickup truck drove up, and after assessing the situation he felt he could tow Gene's truck out. Gene's nylon tow rope broke instantly, so the resident got a chain out the back. This proved successful in freeing Gene's truck, and we were back in business. Yet another testament proving the incredible hospitality of folks on the Great Plains, something I've seen and heard of often in my 15 years of chasing!
The mud had cost us 25 minutes, and with a new tornado nearly 15 miles east and heading for Chickasha, we made a futile attempt to get in position with it, heading north from Verden and east to Amber. The large cone-shaped tornado was clearly visible, but the poor contrast and our being on the move kept us from documenting it. As it was getting into more populated areas, making it increasingly difficult and dangerous to chase, we aborted the chase at 1836 CDT and headed west to another cell which was now moving towards Anadarko and based on two-meter amateur reports showed promise. This cell was much more LP-ish, with very little precipitation, and we had a fascinating view of its structure as we closed in on it. The storm was highly striated with a large, vaulted updraft and colorful backlighting from the sun. Simply beautiful!At 1850 CDT, admiring the view, we drove south towards Verden then slightly west, dodging two aggressive, stupid dogs who ran for our moving truck and nearly got under the wheels. We parked on a ridge north of Verden overlooking Lake Chickasha at 1900 CDT, watching for 15 minutes as the storm rotated and organized to our west. We backtracked past the stupid dogs, who once again charged our truck, then edged several miles north and west, meeting up with chaser Dave Floyd just in time to catch a violently-rotating, wispy tornado cross the road half a mile to our west at 1930 CDT. The tornado moved northeast and grew under a large wall-cloud. Unfortunately, with eight miles of mud to our north, we opted to head east on a paved road to Pocassett and go north on US Highway 81. We were sad to see the growing tornado disappear in our rear view mirror, but we learned our lesson well about dirt roads.
Around this time we were hearing of the tremendous damage in Oklahoma City on the 2-meter radios and viewing the coverage on our monitor on KWTV-9. It was sobering, and for awhile we were quiet. Finally meeting up with the storm 15 miles further in Minco with tornado sirens wailing, we encountered a large roadblock on the Canadian River, which proved to be a major choke point requiring a 30-mile detour. Here, the chase ground to a halt for nearly 30 chasers -- a smorgasbord of Who's Who in Chasing lined up at the roadblock where only a single TV news van (probably KWTV-9) was able to get through. Although I considered this to be the luck of the draw, Gene was livid at this ironic end to an otherwise great chase. After five minutes of considering road options, we opted west 15 miles then north, hoping the new activity to our northwest would organize. A half an hour later at dusk, we got on I-40 eastbound, and took the Country Club exit at El Reno where we fueled up. KWTV anchor Gary England was on the radio in the gas station, and the cashiers were absorbed in what was happening. After getting their attention to pay our bill, we then headed north on US Highway 81 to intercept the Okarche/Kingfisher storm. Night had fallen, and from this point on, we relied on lightning flashes and our 15 years of experience with storm structure to guide us. Nearing the new storms, Gene asked me how long until we reached Kingfisher. I glanced at the GPS (global positioning system) and said "two miles". I had to do a double-take, as I didn't see anything ahead. A few minutes later, I realized why. We coasted into a dark town was bathed in eerie blackness and lightning flashes -- the power grid had been taken out by a tornado. We quickly worked through the lack of traffic lights and continued five miles up the road to Dover. Around this time we learned on the TV monitor that a new storm had developed near El Reno and a tornado had touched down near I-40 and Country Club Road -- where we fueled up 45 minutes ago! This tornado went on to destroy parts of the city, including a hospital.Dover was a disaster area, having been hit on 10/4/98 and again just an hour earlier. The town was in blackness but busy with police and residents combing through the debris, littered extensively across lawns, sidewalks, and the street, and a new storm was already bearing down to the west. North of town, the road was partially blocked by power lines and a steel girder blocked the northbound lanes. We were able to drive around these obstacles with no problems. We scurried back south, by which time a deputy sheriff had arrived. He waved us over the lines, screaming, "Tornado!!! Tornado!!! Go south!!! Fast!!!!!!" After clearing the lines we looked again, and in just 15 seconds a new funnel had widened into a ghostly tornado about a mile to our west (2151 CDT). We moved gradually through town, knowing that the tornado would skirt us and the town to the north, and as the circulation passed north, the tornado widened into a beautiful, ominous stovepipe shape, departing northeast into open fields. The sight of this whitish receding stovepipe shape, illuminated by flickers of lightning, is one of the most haunting, eerie sights I have ever seen.Gene and I edged back north through Dover and across the downed lines. The tornado was now many miles away from us, and we decided not to pursue it. We could still see the stovepipe tornado on the eastern horizon, illuminated by lightning like a phantom. A roadblock was being set up for people not to cross the downed power lines. We decided to call it quits and got back to Norman after 1 am, muddy and exhausted.
A CHASE THAT ALMOST LEFT ME STRANDED by Shane Adams
This chase started off in the worst possible way. I got hung up at work way late, and was not on the road until around 5:15pm. I was west bound on Highway 9, when my cell phone rang. It was fellow chaser Dwain Warner in Ohio, calling me with information. Just as I was beginning my conversation with him, I reached for my shifter to gear-down for a stop and it came off in my hand. I managed to hang in third gear long enough to pull over and stop then desperately tried to repair the shifter. It wasn't gonna happen. So, I called my roommate and occasional chase partner Jeff Johncox to come and pick me up. Upon hearing I was stranded at the Highway 62-9 junction, he came to my aid and was there in twenty minutes. We quickly disassembled all the equipment from my vehicle and got on the road, heading to Chickasha. We were getting excellent scanner reports of the Chickasha tornado, and decided to turn north onto Highway 92 and intercept it from the south. We caught sight of the wall cloud as we turned north on 92, then headed rapidly north towards Amber.
After about a mile or so, we heard reports that a new tornado had formed, straight ahead of us. We looked and after another half mile or so, we saw the wedge, moving northeast along I-44. There were several chasers on this tornado, both land AND air, so we made our way to a good spot on the side of the road and shot video. The tornado began ripping through the Bridge Creek subdivision, although at the time we had no idea people were losing their lives. We watched the mile-wide wedge as power flashes from transformers blew in sequence, on the left side of the rotation. About two minutes later, we decided to continue north, still running the video. As we drove north towards Amber, we got sensational structure video, a huge white bell updraft, with rotation clearly visible even from as far as ten miles away, with the wedge looming underneath. This was the most fascinating, and most awe-inspiring moment of my chase career to date. I will never forget that tornado.
We lost visibility eventually, partly due to rain, partly due to the fact that the storm was moving away. I looked over my shoulder to the southwest and saw the day's second wedge, about 15-20 miles away, underneath a beautiful LP (low-precipitation) supercell. I wheeled the car around to get a better filming position. We sat and watched and filmed this tornado for over ten minutes, watching go from multi-vortex to wedge and then back again. I'm still amazed we had such a view of it from as far away as Highway 92 north of Amber.
This LP supercell produced what I believe to be two more tornadoes in the next twenty minutes, but our video of these was not very impressive, long-distance and sketchy. We decided to head north to Minco. As we arrived in Minco, we saw (but were hit by only one or two) baseball size hail stones littering the road. We stopped for gas there, and as I was coming out of the store, the sirens blew. We jumped in the car. Eventually, we turned west on Highway 152 and heard a spotter reporting a tornado crossing Highway 152, which would have put it about 5-7 miles straight west of us. We pulled over and looked for this tornado, and finally saw it ahead, crossing the road. We headed towards it, trying to get closer. After we had gone about a mile further west, I glanced to the southwest and saw another tornado coming fast. I shouted to Jeff about the new twister, so he could point the camera towards it. So here we were, driving west on 152 with two tornadoes right in front of us. The furthest west tornado disappeared beyond the trees to our northwest. Meanwhile, the southwest tornado was fast approaching. We lost sight of it as we topped a tree-lined hill, and when we saw it again it had moved to within about a quarter-mile away. Jeff screamed for me to turn around, and I slowed, trying to calm him down. I began making my six-point turn (152 has no shoulders there) and as I looked back to the southwest before turning back east, I saw multiple vortices tearing up grass and wheat in the field just outside my window. It was my first up-close encounter with a twister! We drove away from the tornado, while we filmed it out the window as it crossed the road (in the very spot where we had turned around) less then a mile away from us. Once we were in a safer position, we stopped the car, got out, and filmed. The tornado went through every phase of development from wedge, then multi-vortex, then classic tornado funnel, then stovepipe, then cone, then rope.
We filmed the tornado for over seven minutes before it roped out. The same wall cloud redeveloped and dropped a quick, brief tornado about three miles north of Minco, just as we pulled over across from Gene Rhoden and Carson Eads, among others. We then got caught in the huge roadblock on at the Canadian River on Highway 81, but finally just ran it after waiting a few minutes. Good thing we did, because we might not have seen the Union City tornado that touched down just a mile or so northeast of downtown. I saw Howard Bluestein for the first time in the field just as we were entering Union City. We finally made it back home around 5:30 the next morning, after a long night of dodging debris, traffic, and showing video to news stations.
TWO STORMS, ONE GREAT DAY by Aaron Blaser
I had looked at surface data and 700MB temperatures for the past two hours. There was bulls-eye of -11 C lifted index over Lawton where my target was. Only one small concern this day was the significant cirrus shield that limited the surface heating. But it was already 80-82F in the target area, so cutting off the surface temperatures would not hurt much. I left the television station about 2pm and headed west on I-40 to El Reno. Early turkey towers were trying to develop near Weatherford. I went south for a while on Highway 81 to near Pocasset (about 10 miles north of Chickasha) and pulled into a field, watching the cirrus shield moving in from southwest and watching the interaction with an area of cumulus congestus to the southwest.
I was too far north to see the base of newly developed severe storm in Commanche County. I turned on an AM station on the radio and noticed CG (cloud-to-ground lightning) static. I did get a rare FM station with good weather updates from Altus, OK. The storm was heading right for me. I opted to not risk a dangerous core punch, so I decided to get to Chickasha and head west about five miles to my original target town of Verden. I got there at 3:30pm, just as the storm went tornadic 20 miles southwest of Verden. The storm projection was north-northeast at 40mph, .so I knew that Verden would see this storm in 30-40 minutes. I also knew that this town would likely see large hail, so I scouted out a car wash, told people in 2 convenience stores about the upcoming danger, and went to a 100 foot cliff/high spot on south side of Verden. Cloud ceilings began to lower, then a rain shaft came in and lightning really picked up. I was now listening to simulcast weather coverage from KWTV stormtrackers on KXY 96.1FM. The constant reports from KWTV stormtrackers were a huge help.
As hail began falling, I decided to get off my perch and head to the car wash to wait out the hail. I heard of a tornado about two miles south of town and started to get worried. All of a sudden, the large hail stopped, and I headed southwest out of town viewing this huge wedge tornado. I turned around on the highway to find a good video stop, but had to re-enter the town to get good viewing area. I was able to get back to my original cliff location and shoot some video and still pictures from there. By this time, the wedge was approximately three miles to my east. I had limited time to shoot photographs, due to the rain wrapping around me on west side of mesocyclone. I followed the tornado east on Highway 62 to Chickasha. Due to heavy traffic from sightseers, I did not get any closer to video the tornado. I ended the chase for that tornado on north side of Chickasha.
I decided to head west to another storm. A factor in this decision was a large wall cloud with inflow tails on the storm which I could see clearly 20 miles west of my location. I also saw this was an LP (low precipitation) storm. So, I headed back west on Highway 62 through Verden. About eight miles later, the lowering dropped and a beautiful backlit tornado appeared in the front portion of updraft base while a secondary rotating wall cloud was churning in the back of the storm fed by the inflow tails. This tornado was below a corkscrew updraft base and stayed on the ground, moving slowly east-northeast for about ten minutes changing from a trunk, to hose, to rope. I got some great video, close-ups and good still pictures. As this tornado dissipated, the wall cloud at the back of the updraft base began to increase spin and size. I was certain it would produce large tornado, but did not until it moved over rural areas northeast of Gracemont and near Union City, OK. But by then, the local hills and setting sun made it difficult to see and video, but I saw two more separate tornadoes from this storm during its track from Anadarko to north of Union City, OK. My chase ended with catching a glimpse of another large wedge tornado from I-40 -through the lightning north of Yukon, OK from a distance of 5-8 miles.
CHASE ACCOUNT by David Wooten
On a business trip in Wichita Falls, Texas my pager alerted me to an extremely rapidly developing supercell just north of Lawton, OK. The cell was moving north-northeast at only 35mph. I did the math and figured I could catch up with the storm. I quickly departed Wichita Falls, TX northbound on I-44 and caught up with the storm just south of Chickasha, OK. The main towering cell was not hard to spot. It's top had grown rapidly and was bellowing upward with each mile marker as I approached the southwest band of clouds.
At the Chickasha , I got in front of the storm base. Within a few minutes, a huge wall cloud lowered and the entire cloud base began to rotate. I quickly found the local Ham repeater. An emergency weather net was up and running. I gave my report and pulled over to find my exact location on the map and observe the cell located atop the next hilltop on I-44 just south of Chickasha. A few seconds later, a large 1/4 mile wide funnel dropped slowly toward the ground. I reported this on the radio identifying my call-sign. There was a bright blue pop of a transformer then another and debris came up from the ground rotating with the funnel.
My wind gage went off the scale and the hail started falling along with sheets of red clay mud. This was the muddiest tornado I have ever been in. It was pouring down mud within seconds of the touchdown. The interstate was so covered by red clay and grass it was like trying to drive on top of a wedding cake. The slick roads slowed my departure allowing the cell to overtake me, worst of all cutting off my only escape route as I was unable to get to full speed. Several smaller funnels lowered and were dancing around the tornado.
The large tornado was on the ground paralleling I-44 from just north of Chickasha, OK. The tornado seemed to suddenly slow up and speed up again. It grew to over one mile wide and slowed in forward movement about 28 miles southwest of Oklahoma, OK. Hail up to golfball-size fell. I followed the tornado up I-44 for about forty minutes and entered the south edge of Oklahoma City. The tornado crossed over a bridge at Newcastle, OK at I-44. I continued north across the bridge over the Canadian River. With a State Trooper behind me, I made it across the Bridge before he slammed on his brakes and blocked the entrance to the bridge. State Troopers were on both sides of the highway and they shut it down in both directions. I stopped to call National Weather Service in Norman one last time and reported a tornado of unbelievable size now entering a heavily populated housing development. Hail up to softball-size began falling and I was deluged with a massive downpour of mud and debris. I drove out of this and saw the first group of survivors wandering in the streets where there homes had just been leveled to the ground by the tornado. The thrill of the chase was over.
MAY 3, 1999 OKLAHOMA HISTORICAL OUTBREAK by Jeff Piotrowski
While I did not expect the magnitude of this historical day, severe weather parameters were coming together over Oklahoma signaling a higher than usual tornado threat. Threat was probably an understatement of the events later that day. The questions remaining to be answered late that morning and early afternoon were where would supercells develop and when would the supercells finally break the cap. The parameters for a severe weather and tornado outbreak were starting to become more apparent with time. A strong mid-level wind max was expected to blast across Oklahoma by evening. A very strong low-level jet was positioned across Oklahoma during the day. The flow aloft was very diffluent (splitting) and the low-level winds were becoming more backed with time. The crossing of low- and mid-level jets was projected to be in the central Oklahoma area, and this area is typically a region of intense tornadoes. A deepening surface low was over northeast Colorado and a sharp dry line extended from the low to Dodge City, KS to Childress, TX to near Abilene, TX. Mid- to upper-60 dew points were working northward into Oklahoma and the air mass was very unstable, particularly over southwest and central Oklahoma. A special afternoon sounding from Norman, OK revealed extreme low-level shear, and the severe storm outlook was immediately upgraded to a high risk. Significant and very damaging tornadoes were possible over the high risk area, and it was becoming more apparent that central Oklahoma would be in the highest risk.
I monitored the forecast models and the RUC (Rapid Update Cycle) model is my model of choice. The RUC model progged explosive supercell development over southwest Oklahoma between 3-4 pm. This area matched up well with the extreme instability and favorable surface patterns setting up by early afternoon. I decided that southwest Oklahoma was where I would place my target zone. One feature which I noticed and Brian Stertz also noticed was the presence of a secondary boundary/warm front extending across southern Oklahoma. We both felt that this boundary would act to increase low-level shear even more with time, and would probably be the major player in tornadic development by late afternoon. This was the very same boundary we had chased the day before in north Texas. Deep moisture would also tend to be pooled in the vicinity of this boundary. It was also apparent that the boundary would lift northward as the surface low deepened further and the low-level jet roared over the top. Low-level shear was definitely going to be concentrated today, so the potential for violent tornadoes was very high. Brian Stertz concurred on my choice of the target area, and he planned to head that direction after work from Tulsa.
I left Tulsa/Broken Arrow around noon after finishing my recap of weather data. Southwest Oklahoma remained capped; but instability was building up such that in the next 2-3 hours, explosive supercell development looked to be likely. I passed through Oklahoma City, and continued on to the southwest on the H.E. Bailey Turnpike (I-44) towards the Lawton area. Little did I know, I was passing through areas that would be devastated by a historical F-5 tornado. By early afternoon, the RUC model bulls-eyed southwest Oklahoma even more, indicating that a supercell would rapidly develop to the south of Lawton, OK. More and more factors would come together over the next three hours to make this one of Oklahoma's worst and most damaging outbreaks ever. The cap restrained convective development early in the afternoon. Meanwhile, extreme instability was building up over southwest Oklahoma and northwest Texas. This was not a good combination of factors for the safety and well-being of resident's of this area. By mid-afternoon, surface LI's (Lifted Indices) were in the -10 to -12 degree C range, and low-level winds started to back over southwest and central Oklahoma. I was very concerned that damaging tornadoes were imminent. I reached Lawton just after 2pm, and continued on to the southwest to near the town of Faxon, OK. I decided to pull over and setup my tripod for video. The cap was fairly strong; I watched several towers bubble up and attempt to break through the cap without success.
With time, the cumulus clouds started to change appearance. Vertical growth was increasing, and oddly enough, several of the small towering clouds had a twist as they developed in the unstable air. Between 3:45pm and 4pm, a small but organized towering cumulus started to break through the cap. The cloud struggled for the first few minutes, but erupted with fury shortly before 4pm. I was almost certain that this was the supercell the RUC model hinted would form south of Lawton. I was also concerned that this storm would rapidly become severe and tornadic, given the degree of instability and low-level shear. I called into the southwest Oklahoma repeater to report this to the Norman NWS (National Weather Service) that a storm was breaking through the cap. An accessory cloud on the back side of the now exploding cumulonimbus showed incredible spin as it was absorbed into the organizing updraft.
The chase begins at 4:10pm. I decided the storm had organized to severe levels and left my position near Faxon, OK on Highway-36. To my north, the storm went through the usual transition phase of lone thunderstorm to supercell, only today the process happened at an accelerated rate! Almost immediately, the storm had obvious signs of rotation. The storm tower/updraft was becoming striated in the lower levels, while up above the updraft knuckled back. A clear slot started to form on the western portion of the storm and it was apparent that an RFD (rear flank downdraft) already was developing. Inflow bands appeared out of nowhere. A supercell was now becoming full-fledged. The supercell gained intensity as it tracked north-northeast towards Lawton. At 4:28pm, I was approaching the storm from the south as it was very near Ft. Sill Army base. A pronounced hook echo started to develop on radar as storm rotation increased rapidly. Also at this time, the flat rain-free base dipped lower, as a bowl-shaped wall cloud to my west about 3-4 miles away. Data from my laptop computer showed that low-level shear was rapidly increasing as well. I knew a tornado was about to happen so I called in my report again to the Norman NWS. Not more than a minute later, a tornado warning was issued for Comanche, Caddo, and Grady counties. The developing tornado location was listed as three miles east of Medicine Park in Comanche Co. I was at the intersection of I-44 and Highway 49 at this time (4:50pm). I started to notice rapid condensation under the wall cloud at this time. Between 4:50 and 4:55pm, I watched a pronounced funnel cloud work down from the wall cloud. The funnel was not more than two miles to my west-northwest. The tornado would soon follow!!
The funnel appeared to be very close to touching down as it moved north-northeast near the town of Elgin, OK. I was on US Highway 277 southwest of Elgin when I could see a small tornado touch down to my east. My road options were limited and hampered my luck at getting closer to the tornado. I decided to head up towards Cyril, OK (southeast Caddo Co.) to get into position to view the tornado which remained on the ground to my east about eight miles away. The tornado remained small as it passed very near the town of Cyril at 5:20pm. The tornado crossed Highway 19 before I was able to get to Cyril. I pulled over (as numerous other chasers, spotters, and news crews did) to watch the tornado slowly increase in width and intensity to the northwest of Cyril. At this time, I was also watching radar and an even larger supercell was exploding over the Altus, OK area. For a split second, I had thoughts of heading west to that very large supercell. Hail to the size of softballs had been reported over northern Jackson Co. I decided to stick with this storm as it was tornadic already, and certainly had a favorable environment to grow and sustain rotation. I watched the tornado slowly weaken at 5:30pm as it moved northward to near Cement, OK. I did notice that two circulationís were now starting to split from the parent updraft. Circulation #1 was heading north-northeast towards the Cement/Laverty area. Circulation #2 had an even more northerly course as it was heading for Anadarko. I decided to follow circulation #2 as my road options were a little better to stay even with it. At 5:35pm, I followed the rotation north into Anadarko without much success. However, radar was indicating that circulation #1 was rapidly intensifying near Laverty (west Grady Co.). Spotters soon reported a large multiple vortex tornado in that vicinity. My attention was now directed solely on that tornado. I plotted an intercept somewhere near or east of Verden, OK. I left Anadarko on Highway 9 to intercept this increasing tornado now reported to be heading for the large city of Chickasha. I gained quickly on the tornado following an Oklahoma trooper in escort fashion east of Verden. The time was now 6:02pm and the storm base was very dark to my east and southeast.
My visual observation was severely reduced by the rain and hail on the northwest side of the large hook echo I saw on radar at my location east of Verden. I continued on knowing a large tornado (reported to be 1/4-1/2 mile wide at 6pm) was just southeast of me about five miles! I broke out of the rain and hail quickly as I approached Chickasha. There was a very dark area that soon revealed the large tornado that was tracking almost due north. The width of the tornado looked to be in the 1/4 mile wide range as it approached Highway 9 at 30-35mph. I watched the tornado slowly wind down about 1-2 miles south of Highway 9 at 6:10pm. Vortex breakdown was starting to occur. The solid cone tornado dissipated into weaker multiple vortices and eventually a weak dust swirl. At 6:12pm, the tornado had dissipated but there was still strong rotation at cloud base. I decided to head east into town thinking that the tornado might re-form further north. How wrong I was. My van was rocked by extreme wind gusts on Highway 9 about three miles west of Chickasha. I knew I was in trouble. The inflow/outflow couplet had just passed the road prior to my scary experience. Almost as fast as the tornado dissipated, the tornado re-formed not more than a 1/2 mile north of me. The time of the new tornado touchdown was 6:14pm. Rotation became very violent just to the west of town. The tornado quickly transformed from multiple vortex structure to a very large condensated cone. I reached town (still shaking from my close call) and reached Highway 281. The tornado was going to barely miss the north section of town, but I could see that the airport was in the direct path. I called in my report and made my way slowly north about 2-3 miles south of the large and dangerous vortex.
The Chickasha Municipal Airport took a direct hit from the tornado. I crossed the damage path on US Highway 281 and could confirm F-3 damage in the area as the 1/2 mile wide tornado continued northeast at 6:20pm. The tornado was also joined by an accessory tornado which traveled in tandem with the larger F-3 tornado. This smaller tornado (F0-F1 intensity) remained on the ground between 6:20pm and 6:24pm. The twin tornadoes were an impressive sight as they moved across open country just to the west of I-44/H.E. Bailey Turnpike. The tornado remained very large for much of its track which ended as it approached Amber, OK (Grady Co.) at 6:28pm. Almost instantly, an even larger (and visually stronger) tornado was on the ground southeast of Amber, and developed from the same mesocyclone. The tornado was well over 1/2 mile wide as it passed east of Amber, or 4-5 miles to my east. Numerous chasers formed into a caravan as I took County Rd E-1280 east from Amber. Fortunately, Amber would be spared from the severe damage which would be soon inflicted on communities further to the northeast along I-44. At 6:38pm, the tornado was approaching one mile in width as it moved up the turnpike three miles northeast of Amber, or five miles to my north. The forward speed seemed to slow between 6:40 and 6:50pm. Unfortunately the tornado also increased in intensity. The tornado ripped across the unincorporated town of Bridge Creek, resulting in complete devastation and many fatalities. It is also in this time frame that the DOW ( Doppler on Wheels) measured a 318 mph wind velocity just above the surface in the maxi tornado. My road options were not good but I had to continue on to get back even with the now F-5 tornado. My distance from the tornado did allow me stunning views of the incredible rotation and RFD on the backside of it.
I knew that the slower movement would allow me a better chance to catch up to the tornado. The tornado's position at 6:50pm was near the Grady/McClain Co. line on the northwest side of the Newcastle city limits. I finally reached Highway-76 (a paved road) and quickly made my way north towards the mile wide F-5 tornado. The tornado passed about a mile to my north at 6:54 pm as I approached the intersection of Highway 76 and Highway 130. I tried to get out on my cell phone, but all circuits were jammed, and I am almost certain that several cell towers were wiped out. I followed the tornado along Highway 130 into the town of Newcastle. The tornado fortunately passed northwest of town about 2-3 miles. However, the tornado nearly followed I-44 and several motorists were injured or killed by the advancing 1/2 mile wide tornado. I made my way through town on US Highway 62/277 and caught up to the tornado once again as it crossed I-44 less than a mile to my north. The tornado remained very intense, but the structure was more of a stove-pipe appearance (1/4 mile wide) at 7:08pm. It was now that I realized that the tornado was tracking right for the city of Moore and possibly the south sections of Oklahoma City. I had a very sick feeling in my stomach that this was going to get even worse. I continued video of the tornado as it moved northeast from I-44 towards Moore.
At 7:10pm, the tornado weakened slightly as it crossed the south Canadian River in rural sections of extreme northwest sections of Cleveland County. Once again an accessory tornado formed north of the main tornado. This small tornado soon rotated around to the southwest side of the larger tornado. Eventually, it became absorbed into main tornado. At 7:12pm, I reached OK Highway 37 (SW 134th St.) and the road which goes into the City of Moore. The intensity of the tornado weakened to strong F2- weak F3 as it approached Highway 37. I was hoping that the weakening trend would continue. One bad thing I noticed was that today's tornadoes never went into the rope stage. They either dissipated completely or re-generated into an even worse tornado. The tornado did the latter as it moved towards SW 134th St. At the intersection of SW 134th and May Ave., the tornado started to max out again in intensity. I had managed to get ahead of the now 1/2 mile wide tornado as it crossed SW 134th. I was on the northeast side of the tornado looking southwest from a subdivision area (Meadow Ridge Estates). I called in my report to the Norman NWS to give them a detailed account of streets and sub-divisions in the path of the intensifying tornado. I was on Pennsylvania Avenue just south of SW 119th when major debris, power flashes, explosions, and a load roar were observed. I had a real hard time trying to keep composed as I knew hundreds of homes were about to be leveled. My winds bordered on the extreme to scary as the tornado passed just to my west. Tons of debris started falling from the sky around me so I moved east on 119th St. A thud on my roof (either from debris or one huge hailstone) signaled a time to move. Swirling debris, mud, and 80-100mph winds blasted me at the intersection of 119th and Western at 7:22 pm. I continued on 119th to I-35 as the tornado was passing to the northwest about a mile away from me. Reports were now coming in that many houses in northwest and north sections of Moore had been leveled or destroyed. I watched in shock/amazement as the 1/2 mile wide tornado approached I-35. There was tremendous debris swirling in and around the tornado. Millions of pieces of debris! A horizontal vortex tube also formed on the southeast side of the large tornado as it neared I-35. This will mark the second time I have seen an accessory tornado. The last time was on April 26, 1991 and it was over Noble Co. Oklahoma. The roar that the tornado made as it moved through Moore will never be forgotten!
The rain curtains swirled around the south side of the tornado as it crossed I-35. My view was blocked so I headed east into the northeast section of Moore. The tornado continued but was partially obscured from view by the wrapping rain curtains. I reached Eastern Ave. at 7:30 pm and headed north. The tornado was approaching the intersection of Eastern Ave. and NE 23rd St. I made my way cautiously north on Eastern and pulled over when I reached NE 23rd St. and watched the tornado pass just 1/4 of a mile to my north. I was in almost continuous contact with the Norman NWS since the tornado moved into the west section of the city of Moore. I gave them detailed location information as my positions never varied from 1-2 miles from the devastating tornado. While pulled over, power poles quickly were toppled like dominoes by the tornado. I had to move! The power pole 100 feet from my van broke off, and the edge of 100 mph winds! The time was now 7:33pm.
The tornado's appearance was becoming even more shrouded in rain as it moved towards I-240 and Del City. I drove over to Bryant Rd to head north since there was likely major debris/power poles across Eastern. At 7:38pm, the 1/4-1/2 mile wide tornado continued to head northeast towards Sunnylane Rd and I-240. The tornado was about 1-2 miles to my north-northwest at this time. Very intense damage was still being reported as it was now moving into Del City. I wanted to avoid damage on Sunnylane so I headed over to Sooner Rd near I-240. The tornado actually started it's northerly track as I headed north on Sooner Rd, immediately behind the tornado. Between 7:40 and 7:48pm, the tornado struck the western portion of Tinker AFB along Sooner Rd, and then continued along and just east of Sooner Rd. F3-F4 damage continued at this time. I saw a horse on the road that was wandering around dazed and injured. Several vehicles and a tractor trailer were badly damaged on the side of the road. I called in for emergency vehicles to respond to this area at Sooner Rd and NE 29th. I had several incredibly horrifying views of the damage path. For several miles I looked up the 1/4-1/2 mile path and very little was left standing. The tornado approached I-40 and Sooner Rd at 7:46pm. Midwest City was next in line for the tornado's path. As the tornado continued north, it started to weaken and narrow shortly before 8pm. I could see yet another circulation developing well east of the occluding/dissipating tornado. This new circulation was going to be very near Choctaw. I stopped at exit 157 on I-40 and saw numerous cars that were ripped apart and rolled. I decided to stop chasing this storm at this very nasty scene. I had to regain my composure and reflect on the scope of what just happened. Yet another tornadic supercell was near Yukon to the west of Oklahoma City. This storm also had a very large hook echo on radar. I started west to continue my chase of more extremely violent tornadoes in NE Canadian and Logan Counties.
My chase led me through the highs and lows of storm chasing. I prepared myself for months in advance for the storm chase season, but I was still not fully prepared to deal with what happened. I would just hope that my reports made a difference today. Destruction, devastation, death, and despair are the 4 D's that a chaser must realize on the sad side of storm chasing. Today was a mixture of extreme emotions in this seasoned storm chaser!
CHASE STRATEGY: MAY 3, 1999 by Tim Marshall
I did not realize a tornado outbreak was going to occur today. Being a workday with two weeks before my annual chase vacation, I glanced at the morning (13z) SPC (Storms Prediction Center) forecast then headed off to work. They issued a slight risk of severe storms over a large portion of the central U.S. A long-wave trough had moved into the inter-mountain region but it looked like the system was ill-timed with surface features and upper dynamics coming together after dusk. Given the late show possibilities, I figured to work until noon before checking the weather situation again. When I came back to the office, my secretary said that Carson Eads called. I called him after lunch and was informed that SPC had upgraded Oklahoma to a moderate risk at 1630z. It appeared a short wave was approaching the area and convection was likely along the dryline. "Violent tornadic supercells" was mentioned in the discussion. That did it.
I canceled my afternoon appointments and headed home to plot a map and gather my chase equipment. Carson was not able to chase until later in the day, thus, I was on my own. However, knowing it was safer to chase with two people, I asked my wife Kay if she would like to have dinner in Wichita Falls, TX. I proceeded to plot a 2pm surface map. Positive features included a surface low in northeast Colorado, good convergence along the dryline in far western Oklahoma, and ample surface moisture. Negative features included persistent cirrus cloud cover which would inhibit surface heating. I placed my forecast area in southwest Oklahoma along an axis from Kingfisher, OK to Wichita Falls, TX. My forecast area was placed between the temperature axis (Clinton, OK to Childress, TX) and the dewpoint axis (Oklahoma City to Ardmore, OK) at 15z. Around 2:45pm, Kay and I departed for Wichita Falls. At 4:30pm, we saw a thunderstorm to the northwest at the north end of a hole in the cirrus. The storm looked mushy with no backsheared anvil. Other storms were beginning to fire to the west near Quanah, TX and Altus, OK. We stopped briefly in Wichita Falls and topped off the gas tank. I decided we should head north and monitor the Lawton storm. However, my main interest was still to intercept the other storms moving off the dryline to the west since the Lawton storm wasnít on any noticeable boundary. A tornado warning was soon issued for the Lawton storm which changed my interest. As we approached Lawton, we already could see a tornado northeast of town.
Around 5:30pm, we passed Lawton and saw the elephant trunk tornado to the northeast on the west (back) side of the storm. We turned northeast on I-44 and watched the tornado slowly shrink then dissipate around 5:35pm to our north. Then, we noticed a large, lowered cloud base off to our northeast. Inflow bands extended from the tapered cloud base to the southeast. I also noticed another storm was approaching Lawton to our west and had a large rain free base. I decided to continue chasing the first storm northeast on I-44 since the road paralleled the storm and it already proved itself to be a tornado producer. Soon, a large wall cloud formed and a truncated-cone tornado developed at 5:46pm. We stopped at a toll booth southwest of Chickasha and filmed the tornado. A large debris cloud formed as the tornado lumbered across open country to the northeast. Around 6:07pm, a ropy satellite tornado developed a few miles northeast of the cone tornado. The ropy tornado moved southwest and rotated counterclockwise around the cone-shaped tornado. The ropy tornado reached itís maximum size southwest of the cone-shaped tornado then dissipated at 6:10pm. The cone-shaped tornado continued for a few more minutes before suddenly evaporating at 6:14pm just northwest of Chickasha, OK.
At 6:19pm, a large, wedge-shaped tornado formed northwest of Chickasha and moved northeast paralleling I-44. Overhead, helicopters from Oklahoma television stations hovered in the strong inflow. We filmed this tornado for about ten minutes before racing ahead of it to Newcastle, OK. Around 7pm, we stopped on a bridge atop a hill just north of Newcastle. This was the perfect location from which to view the tornado. We were about ten miles ahead of the tornado and watched it slowly emerge from the precipitation. I soon realized we were in the path -but I wanted to stay as long as I could to film the tornado from this great vantage point. My wife was not in agreement with this decision. Mud and leaves began falling around 7:10pm. Then, the cone-shaped tornado turned 3-D with cloud elements wrapping around the front side and trees being spun around its base. The tornado roar followed and the vehicle began to shake as pulses of inflow wind howled by. The tornado emerged over a field right in front of us. Transformers exploded with the sparks accelerating horizontally into the base of the tornado. Unfortunately, we had to leave the bridge location and traveled about 1/2 mile to the south. The tornado took on a cylindrical shape before crossing over the bridge where we had just been. As the tornado moved east, a satellite tornado spun around the south side of the larger tornado and they headed across the Canadian River into Moore, OK around 7:18pm. We knew that many people were now in its path.
Just after the tornado crossed the Canadian River, it began to re-form into a wedge again. We saw the tornado cross a line of steel towers with high-tensioned power lines. Five large explosions occurred simulatneously as five towers crumbled. We followed the tornado east on Rt. 37 and watched the vortex fill with dirt and debris. The tornado was frequently illuminated by exploding transformers. Eventually, a policeman blocked our way so we headed back to I-44 then drove east on I-240 then south on I-35. The tornado had just passed leaving incredible devastation in its wake.
We knew that chasing this storm was over, however, we could see another tornadic supercell approaching from the west. So, we headed west on I-40 and intercepted the supercell at Yukon. A multi-vortex tornado formed just as it crossed the interstate. We continued north on some back roads following the sub-Vortex vehicles. It was dark by now and occasional lightning flashes would illuminate the tornado. Later, near Cashion, we saw a large cone tornado form and move northward toward Piedmont. We decided that was enough for one day and headed back to Dallas.
MOORE, OKLAHOMA DAMAGE SURVEY by Tim Marshall
On May 4th, I volunteered to be of service in a disaster survey of Moore, Oklahoma. I spoke with Dr. Kishor Mehta at the Institute for Disaster Research at Texas Tech University (my former alma mater) and he asked me to travel to Oklahoma City that evening. I met a number of students and faculty the next morning and we proceeded to plan are survey strategy. I was asked to lead one team and survey the damage from southwest of Newcastle through Moore and up to I-240. Another team was to survey north of I-240 through Dell City and Midwest City. The third team was in search of residence shelters. The sub-tasks of our team was to: 1) map out the damage path and assign an F-scale rating to each damaged house F-1 and above by doing both ground and aerial surveys, 2) document the performance of housing by noting variations in construction, anchorage, etc., 3) and document the types of projectiles.
A. DAMAGE PATH
We began our ground survey three miles southwest of Newcastle in a rural area northeast of Bridge Creek. The tornado damage path was about 1/2 mile wide. Most branches were removed from the sturdy oaks. One homeowner we spoke with had a below ground tornado shelter where he and his wife hid during the storm. Their house sustained F-3 damage. Our first F-5 rating was found just northwest of Newcastle, about one mile west of the bridge where Kay and I had filmed a tornado. The house, which was anchor bolted to a concrete perimeter grade beam, was completely
removed and disintegrated. Only the steel bathtub and part of the washer were found in an open field about 100 feet northeast of the foundation. The occupant drove away from the house just before the tornado struck.
The next area surveyed was in extreme southern Oklahoma City south of 149th Street. Homes in this rural area sustained up to F-3 damage. The tornado damage path narrowed to about 1/4 mile wide but the F-scale ratings increased to F-5 at Country Place subdivision south of Rt. 37 (134th St). Many homes were just a pile of debris (F-4).
Interestingly, several homeowners survived by seeking shelter in a bathtub or interior hallway, however, there were many injuries and a few fatalities. Everyone we spoke with knew the tornado was coming thanks to warnings from the media and relatives phoning.
The tornado damage path remained consistently 1/4 of a mile wide through the Briarhollow subdivision. These were tract homes with attached two-car garages. We found a mix of homes that were either nailed, strapped and nailed, or anchor bolted to their concrete foundations. The West Moore High School was located on the north side of the damage path and sustained some damage to the roof and metal cladding. An awards ceremony was being held at the time the tornado struck and hundreds of people were in attendance. The people were evacuated to interior portions of the school and all survived, however, their vehicles in the parking lot were tossed around like toys. Some of the automobiles ended up in the houses south of the High School.
The tornado crossed Western Ave. and struck the Emerald Springs Apartments. Some of the two-story wooden structures were reduced to one-story or less. Several injuries and fatalities were reported here. A metal building just north of the apartment complex contained a church and was crumpled. The tornado churned through the Greenleaf subdivision reducing many townhomes to rubble. Numerous F-4ís and one F-5 townhome was found. The tornado completely eliminated another two-story metal building which contained a church, then crossed Santa Fe Avenue entering the heart of Moore. The damage path extended diagonally from NW 12th to NW 19th Street through smaller wooden-homes with attached one-car garages. A total of 93 homes had F-4 damage and 7 had F-5 damage in this one subdivision alone.
The tornado bulls-eyed the Kelley Elementary School traveling right over/through it. The steel frame buckled and the entire school had to be bulldozed. I hate to think what would have happened if school was in session. There were few areas that offered occupant protection in this school. I hope that if they build the school back, that someone will think of "hardened" places within the school to put the kids.
The tornado continued through another subdivision inflicting up to F-4 damage before crossing I-35 at Shields Blvd. A number of fatalities and severe injuries occurred by people trying to seek shelter under the bridge there. However, the bridge was a concrete slab with no girders -like many bridges in Oklahoma. The tornado crossed I-35 and struck an apartment complex and Best Western Hotel. Roofs were removed and some of the two-story buildings were
reduced one-level. The tornado then traveled through the Ridgewood subdivision causing up to F-4 damage before traveling out of town over rural areas. The tornado turned more northerly at I-240 and headed on to Dell City and Midwest City. After assembling all of our information, we found 284 F-1 homes, 405 F-2 homes, 558 F-3 homes, 317 F-4 homes and 17 F-5 homes. Homes F-3 and greater were bull-dozed.
B. PERFORMANCE OF HOUSING
Houses were conventionally constructed. Wooden bottom plates were either nailed, strapped, or bolted to their foundations. Most of the houses had bottom plates nailed to their concrete foundations. Nails were two inches long, concrete type, and penetrated into the foundation about 1/2 inch. In many instances, failure of the home occurred at this connection. The wooden bottom plate either pulled through the nail (leaving the nail in the concrete foundation) or the wooden bottom plate and nail were removed. Stronger straps and anchor bolts held the wooden bottom plates in place, however, failure occurred where the wall studs were attached to the bottom plate. Most wall studs were straight-nailed (two nails) to the wooden bottom plate and toe-nailing was observed in a few instances. The straight and toe-nailed connections would be weaker in than the straps or anchor bolts, thus, the nails failed first. There was some diagonal bracing of the walls with wooden let-ins.
Typical house failure initiated as the garage doors buckled inward and windows broke allowing the wind to enter and pressurize the buildings. The roofs were then uplifted and the walls were removed. As expected, none of the houses were designed or built for an F-5 tornado. However, failure of the homes initiated at wind speeds closer to code values which are in the 70 to 80 mph range. It should be noted that building codes represent the minimum requirements.
The tornado generated thousands and thousands of projectiles. These projectiles impacted and/or went through houses and vehicles. Most of the projectiles were wood. Wooden boards from houses and tree parts were most common. The largest projectile found was a 12 foot diameter, 14 foot tall steel oil tank which tumbled end-over-end 906 feet just west of the Newcastle bridge. The tank was not anchored. Several vehicles traveled up to 1000 feet and ended up in ravines or culverts. Some parts of mobile homes (i.e. steel beams up to 35 feet long) traveled similar distances.
The most unusual projectiles were as follows: 1) the leg of a steel lawn chair penetrated a solid 5 x 5 wooden post, 2) an eight foot long wooden 2 x 6 entered a hole in the roof of a house and penetrated the refrigerator freezer, 3) an eight foot long wooden fence post went through a window and lodged in an interior wall, 4) a 6 foot long steel pipe, which weighed about 40 pounds, went through the front door and came to rest in an interior hallway. Also, a bathtub was moved 330 feet in which two people seeking shelter survived.
The tornado disaster at Moore, Oklahoma provided an opportunity to learn about building response as well as human response. As expected, there was significant variability in house construction and performance. Houses with F-4 and F-5 damage provided little shelter to the occupants. Portions of houses need to be "hardened" to provide occupant safety. This is especially true for schools. About 100 people were interviewed and ALL said they had sufficient warning to seek shelter thanks to the media (#1) or relatives/friends calling (#2). People sought shelter in bathtubs (#1), closets (#2), and hallways (#3). A few people actually got in their cars and drove out of the path. A number of projectiles were documented including a few never before seen items.
NEW VIDEO - THE MAY 3, 1999 OKLAHOMA TORNADO OUTBREAK- This video is a collection of tornado footage from the historic outbreak which ravaged central Oklahoma. Ride along with Tim Marshall up the Bailey Turnpike witnessing tornado after tornado including a close look at the F-5 Moore tornado from the top of a bridge. Travel with other chasers too as they witness tornado after tornado with a second storm to the west. Youíll also take part in the damage survey following the event. 90 minutes, VHS only, color with sound. $25 U.S., $30 Canada, $45 overseas (U.S. Currency please) Send check or money order to: Tim Marshall, 4041 Bordeaux Circle, Flower Mound, Texas 75022-7050.
REBUILDING AFTER THE STORM: WHAT HAVE WE LEARNED?
by Tim Marshall
I've been back to Oklahoma City several times on assignment to determine the extent of tornado damage to several structures. This has given me the opportunity to monitor the rebuilding efforts. A total of 40 buildings were examined in Moore and southern Oklahoma City to see if building practice has changed in the wake of the tornado. Out of the 40 new homes I inspected (all were on concrete slab), five homes had bottom plates bolted to their foundations, six homes had bottom plates strapped to their foundations, and 29 had bottom plates nailed to their foundations. Tapered concrete nails were used and spaced from 12 to 52 inches apart. Nails had a maximum of 1/2 inch penetration into the concrete. In general, construction was no better in quality after the tornado than before, and in some cases, the construction quality was worse now.
WORST HOUSE: One home in southern Oklahoma City had the bottom plate nailed to the concrete foundation every 52 inches, foundation straps were provided -however- they were bent outside the finished wall so they did not engage the bottom plate. Wall studs were cut for let-in braces, however, they did NOT contain the let-in brace. Also, ceiling joists were toe-nailed into the top plates and were driven too close to the edge of the plate -splitting the wooden plate.
BEST HOUSE: One house just south of the West Moore High School had 2 x 6 perimeter framing, a bolted bottom plate to the foundation, and an 8 x 8 x 8 concrete safe room with steel door. However, rafters and studs had standard toe-nailed connections.
GOOD NEWS: Six of the 40 homes inspected had concrete safe rooms. These rooms had six inch thick walls of steel-reinforced concrete; the reinforcement extended into the concrete slab foundation. The concrete rooms varied from 4 x 8 feet to 9 x 9 feet and were 7.5 feet tall. Builders told me they charge $3,500 to $5,500 per safe room, depending on room size and the design.
STATISTICS OF THE MAY 3, 1999 OKLAHOMA OUTBREAK
The tornado traveled 42 miles and the damage path reached one mile wide.
MAY 3, 1999 WEB SITES
ACKNOWLEDGEMENTS FOR THIS ISSUE: Thanks to Dr. Chang, Carson Eads, Mark Martinez, Dr. Kishor Mehta, NWSFO- Norman, Robert Prentice, Jim Purpura, Greg Stumpf and all the contributors.