March 5, 1999   E.Oklahoma and W.Arkansas Chase          

My third chase of the year was surprisingly successful. By late morning, hopes
for chasing were slim as moisture was limited, and a strong cap prevailed over
E.Oklahoma and W. Arkansas. The Storm Prediction Center was indicating that
a few supercells were possible by late afternoon and outlooked the area with a
slight risk of severe storms. The forecast models did not catch on to a "hidden"
vortmax over N.New Mexico during the morning hours. As a result, the models
downplayed the convective threat over Oklahoma and Arkansas. By afternoon,
the vortmax was expected to move eastward fairly rapidly. The cooling aloft and
the strong lift was forecasted to break the cap, particularly over E.Oklahoma and
W.Arkansas. A surface low was expected to track from NC Oklahoma towards
SW Missouri by evening. As a result, the dry line was progged by all models to
reach the Oklahoma/Arkansas border by early evening. The dry line was not a
strong boundary, but was expected to focus t-storm and isolated supercells by
evening. A strong mid-level jet max (associated with the vigorous shortwave)
was also expected to sweep eastward across Oklahoma, thus aiding in strong
divergence aloft and mid-level drying. The limiting factors were a lack of deep
low-level moisture and a fairly strong cap. As the expected shortwave moved
closer to the narrow instability axis, these factors would result in a dramatic and
sudden change to the atmospheric profile.

Brian Stertz and I waited patiently and reviewed the weather data exhaustively
before committing to a chase or particular chase area. The RUC (Rapid Update
Cycle) model indicated the area between Joplin Mo and Miami Ok was the area
primed for convective development. I noticed this was present in all updates of
the model. This area would be enhanced by the approach of the surface low as
well as the approach of the shortwave. Convergence would be strong in this area.
We both also noticed the increase in instability/heating over S.Oklahoma and
N.Texas. A pronounced thermal ridge was setting up by early afternoon from the
Dallas/Fort Worth area up towards McAlester Ok/SE Oklahoma. Moisture was
also increasing quickly in this area. Today looked to be more of a mesoscale set
up than synoptic. Brian projected that the thermal axis was going to "poke" up
towards the Fort Smith Arkansas area by late afternoon. I was skeptical that the
cap could be overcome there until after dark. I targeted the NE Oklahoma and
SW Missouri area for supercells, since this area would have strongly backed
surface winds and strong convergence. Brian stuck with his target area of Fort
Smith, which would be on the northeast "nose" of the low-level thermal axis.
After further review of data, I started leaning more towards Brian's forecast. The
temperatures over SC/SE Oklahoma and N Texas had warmed into the lower
80's and dew points rose to 58-62 in this area. All of the unstable air would be
focusing towards Fort Smith. Brian and I reached agreement , and we packed
up quickly and headed out of Tulsa for Fort Smith.

We headed down the Muskogee Turnpike for I-40 around 2pm. We started to
notice a marked change in the "look" of the atmosphere as we headed southeast.
The cloud bases were much lower in the Muskogee area, and scattered showers
(not elevated) were popping up around Muskogee. The showers were cellular
in nature and looked to be intensifying possibly toward t-storm status. Radar
confirmed our observations, but it also indicated stronger t-storms were firing
over NE Oklahoma between Tulsa and Joplin. A few of the storms were on the
verge of becoming severe with large hail. The storms quickly intensified, but
other storms were starting to organize northeast of Muskogee. More importantly,
new cells were trying to break the cap well to our south near Stigler Oklahoma.
Apparently, the cold air aloft was starting to impede on the instability axis. We
were concerned about the storms over NE Oklahoma since these were already
severe. The cap was definitely stronger further on south into SE/EC Oklahoma.
The question of the day was- will the cap break before sunset??

We continued on to I-40 and then headed east towards our target area- Ft.Smith.
The cap was still holding tight as we headed east on I-40 around Sallisaw. The
storm to our southwest struggled to break the cap, but was still not completely
surface based. I called the local Ft.Smith station to check on the progress of the
storm to our west about 35 miles. He indicated the storm was trying to organize
and consolidate, but still looked capped. Time was still very much on our side
so we pulled over near Roland (west side of Fort Smith) for a radar update, and
a refreshment break. Radar still showed the storm was capped, but the storm
was growing in size over Haskell Co. Oklahoma. The structure was transforming
quickly, and this caught our attention. Before the storm looked disorganized and
multi-cellular. By 3:15pm, we started to notice that a dominant updraft tower was
taking over on the south flank. Radar was even indicating some hail in E. Haskell
Co.- to the east of Stigler. At 3:20 pm, the storm started to look like it was trying
to take on supercell characteristics. The core started to become banana-shaped
and a pronounced inflow region was seen on radar. We scurried back to I-40
and headed west to intercept the storm. A severe t-storm warning was surely
imminent!!

We were fortunate because the storm's track was headed right for us , and along
I-40. The only problem was the storm was 20 miles to our west-southwest, and
rapidly becoming severe. A severe t-storm warning was issued for Sequoyah Co.
at 3:30pm. Hail up to quarter sized was expected as the storm tracked directly
for Sallisaw. By the time the warning was issued, we had already passed the
Muldrow exit. Sallisaw was now just 12 miles west of us. The updraft tower was
about 260 degrees from our location, and was very prominent. The anvil was
also becoming prominent/"razor" sharp. The storm was still southwest of the town
of Sallisaw, but the core looked to be closing in quickly. We decided to pull over
about 2 miles east of Sallisaw to avoid the core. The updraft region advanced on
us quickly, and was definitely showing signs of rotation. Soon the cloud base
started to "scud bomb", and a wall cloud rapidly developed about 2 miles to our
southwest. The rotation steadily increased very near Sallisaw, and large hail
reports soon came across amateur radio from Sallisaw. The wall cloud was
tightening up rapidly as it headed northeast from Sallisaw. Brief funnel clouds
signaled that this storm meant business, but it still looked like it had quite a bit
of organization to do before it would produce a tornado. Inflow bands jutted
into the updraft area as the storm continued into the hills northeast of Sallisaw.
Not a great place to chase, so we decided to stick with I-40 and try to get ahead
of the circulation. The severe t-storm warning was quickly upgraded to a tornado
warning for E. Sequoyah Co. Oklahoma and for Crawford Co. Arkansas at 3:52
pm. The towns of Figure Five, Van Buren, Rudy, and Alma were listed as being
in the path of the circulation.

We continued on through Van Buren and on to Alma. The circulation was deep
in the hills, and the only way to intercept the circulation was to get far enough
ahead of it. Very large and damaging hail soon pounded the Cedarville and
Rudy areas in Crawford Co. The hook echo on this storm was immense and the
hail was also bordering on the ridiculous. Baseball to a times softball hail and
a very strong circulation were both being reported in the Rudy area. Alma was
most definitely the next in line for the circulation which was now heading e-se.
We thought originally that the meso was going to head east so we headed for
the north side of Alma. When we reached that location, we struggled to make
out the wall cloud. Numerous inflow bands and very hazy skies limited our view
to the west. A new tornado warning was issued for Franklin Co., and the bulletin
advised the location of the tornado was 5 miles w-nw of Alma moving southeast.

We headed back south into town, and found a good overlook on the hill north of
the business district. The overlook was okay except for the power lines. Brian
started video right away, and I called in to the NWS to give them a fix on the wall
cloud which was bearing down on Alma. At 4:26pm, the wall cloud was spinning
like mad between Rudy and Alma. Very strong rotation was noted on radar, and
the overall meso was starting to tighten. At 4:30 pm, the wall cloud was in warp
factor rotation and inflow jet bands were pulling in from the northeast. A large
and pronounced "cigar" cloud also was seen to the north of the wall cloud. This
marked the edge of the huge and very damaging hail. The circulation turned to
almost a 330 degree heading and was heading directly for the I-40/ U.S. 71
intersection, and the Alma business district. Rotating rain curtains swirled to our
west about 3 miles, and we could see cloud elements connecting to the  ground
at 4:35pm. A tornado was on the ground briefly about 2 miles west of Alma. We
did not want to get caught in the huge hail core that was bearing down on Alma,
so we headed south to I-40. The circulation crossed I-40 about 1 mile west of
town and continued on to the southeast. We pulled over (very briefly) to watch
the circulation spin like a top to our south. The rotating rain curtains were still
present to the south of Alma. We had to be very careful from this point on. The
large hail core rapidly advanced on us, and we barely made it out of there !!
Extremely large hail pounded Alma and intermittent hail (larger than golfball)
started to fall on I-40. The hail core ominously hung up about 1/2 mile north of
the highway as we were trying to get out of it's path. We safely made it past the
menacing hail core, and we both let out a collective breath of relief!!

The circulation was blasting east-southeast at speeds approaching 50 mph.
Our plan was to get ahead of the circulation near the town of Ozark, and let it
head for us again. The circulation seemed to cycle between Mulberry and Ozark
just before 5pm. The original circulation occluded and headed southeast for
Ozark. A new and very strong circulation organized rapidly along I-40 west of
Clarksville. A tornado warning was issued for Johnson Co. for both circulations.
We had already committed to the Ozark area, so it was starting to look like the
Clarksville circulation was going to leave us behind. We pulled over south of
Ozark only to watch the circulation gust completely out. Meanwhile, the mean
circulation bore down on Clarksville. First with golfball to baseball sized hail,
and then later a F-2 tornado which tracked from 6 miles sw of Clarksville to the
south side of town. We were also watching the storms to our south in Polk Co.
which were also tornadic. Since there was no way we could catch up with the
Clarksville storm, we thought we might be able to catch the Polk Co. supercell
before dark. Topography was not going to allow this to happen!! The Ouachita
Mountains were a huge obstacle to overcome, so we dropped that idea real
quick. We pulled over to re-group, but realistically our chase was through. The
supercell near Clarksville cycled again near the town of Lamar. Yet another
tornado touched down right on I-40 just before dark. The scanner traffic was
incredible as this supercell veered a little more to the northeast from the Lamar
area. More tornado warnings were posted for Pope and Van Buren Co. after
dark, and a strong F-3 tornado raked the area between Scotland and Formosa
in Van Buren Co.

We passed through the south side of Clarksville and saw lots of legitimate F-1,
and possibly weak F-2 damage. We continued east on I-40 seeing if we might
find the tornado damage around Lamar. An interesting event occurred as we
neared the Lamar area. The strong smell of cedar prevailed as we headed east
on I-40. We checked both sides of the road and did not see any visible tornado
damage. The smell got stronger, and soon damage was found on both sides of      interstate just west of the Lamar exit. We knew right away that this was fresh
tornado damage. The path on the south side of the road was very narrow, about
20-40 yards wide. However, on the north side of I-40, the damage widened to
70-100 yards wide. The damage was not particularly intense (F0- weak F1), but
this most certainly indicated a tornado path. We decided to call it a chase and
pulled over for dinner at the Russellville Cracker Barrel. Lots of nervous travelers
approached us there asking us if it was safe to continue on to the west. I guess
a storm chaser's work is never over- storm chaser turned travel consultant!! We
did advise them the worst storms were heading east and passing south. Not a
bad day for these tired and hungry chasers. We learned that Alma had been
pounded severely by the baseball hail. On the way back to Tulsa, we stopped
by Alma and saw lots of nasty damage. In fact, there was still hail up to golfball
sized in the ditches nearly 4 hours after the storm had passed !!