Elaborate Tornado Avoidance Techniques
Southern Kansas to east-central OK
19 May 13
SHORT: Intercepted 4 supercells in 3 tornado warnings and saw no tornadoes: 1. NW of Ark. City KS, 2. near Parkland OK (middle storm of OK triplets), 3. ENE of Prague, 4. Near Okemah and Okmulgee. Somehow managed to miss all the tornadoes two days in a row.
LONG:
Setting up “the day after the day before the day”
Having intercepted clearly the most unfavorable of several Kansas storms the previous day, the “day before the day”, Rich T and I were determined to get on a tornadic supercell on “the day”.
A classical setup set up for the corridor from southern KS to central OK, with rich low-level moisture return, a strong dryline, a lack of antecedent precip to mess up the boundary layer, large low-level hodographs, and strengthening deep-shear vectors with a decent component across the dryline. In short, this was a prime southern-plains tornadic-supercell regime, the only potential hitches being mode certain initiation with storm-mode concerns up north (near a cold front) and initiation questions farther south.
Our strategy, therefore, was to get in between the two ends and be in prime position to strike like a predator at the best developing storms in a corridor from OKC-ICT. We arrived early at a truck stop N of Perry and awaited development. Early storms near ICT looked interesting, even supercellular and potentially tornadic, (one did produce there), but also, potentially tangled up quickly in a messy cluster near the front. Fateful decision #1 had been made: blow off the early, tornadic ICT storm for more discrete and ostensibly longer-lasting activity to form closer, later.
Two areas of convection began to develop almost at once: just W of us, moving NNE toward Wellington, and about 75 miles SSW of us, on the NW fringes of the OKC metro area. Either was easily reachable. We had much better visibility on the closer activity, which seemed likely to remain in a favorable environment well into mature stages. Fateful decision #2 was sealed: check out the closer storms, which would take us away from central OK. We had seen this picture play out several times before–jump on middle or northern convection only to have the southern storm near OKC produce tubes (whether or not we would see one with our convection). Still, we risked missing something closer to home for something closer to us.
(Somewhat) northern play
Fully aware of that dubious history, we headed N to near Blackwell and took a look at the growing Cb, already showing some supercellular characteristics on radar–as did an initially separate, trailing storm behind the cold front and to our storm’s WSW. This is how it looked before the base came into view. We hopped N of the border, zigzagging E and W a little across I-35 N of Wellington to get in viewing position. The storm turned out to be a messy HP with poor contrast. Only a brief, surging attempt to wrap a mesocyclonic occlusion was visible, with a funky, spiked, ground-tickling tail cloud, before it buried itself deeply in rain again.
Optimism wasn’t the word here. The storm had 1) accepted a cascade of percip from the dying remnants of the post-frontal supercell, 2) dealt with big towers and showers forming and merging into its rear flank, and 3) interacted with a new, quasi-linear frontal segment nearby. Meanwhile, radar showed the storms near OKC coalescing with frightening speed into a big, honking supercell near Edmond. This is when we started getting that sinking feeling…here we go again.
Back to central Oklahoma
Fateful decision #3 actually didn’t take long: let’s not linger with this HP rubbish, and try our best to bust SSE ahead of the now violently tornadic convection just NE of Edmond (or anything that might form farther upshear). We didn’t dither on this, which made an intercept from that distance at least marginally feasible. Somehow, the at HP rubbish, and accessory convection building into its flank, produced a small but photogenic tornado near the South Haven exit of the Kansas Turnpike…not long after we left!
Unaware yet of that morale-crushing development, we made decent time southward on US-77/177 through PNC to the Cimarron Turnpike, then SE on the speedy slab. All the while, we could see the booming, huge flanking towers rolling into the back side of the leading storm, now near Stroud, with a known history of atmospheric violence. However, as we closed in for a potential end-around the back side, potentially in the Bristow/Kellyville area, the convection was softening, the mesocyclone weakening, the storm moving into somewhat cooler, lower-theta-e air.
Resigned to missing all that storm had to offer, we saw two newer supercells on radar, and fuzzily from the sunny N side–a small one to our near SSW, near Carney and trailing behind the first. The westernmost storm was organizing fast…over Norman! That was the cell we really wanted to see, with its unimpeded inflow and longer prospective resident time in juicy air…but the middle storm (which was sucking some outflow from the formerly tornadic lead supercell) was smack in the way.
Since the middle storm was considerably smaller than the others, with lower VIL and MESH hail indicators and a thinner forward-flank core, we made fateful decision #4, hedged a bet against the risk that could strengthen its hail production, and headed S through its forward-flank core toward the Agra/Parkland area. True to its algorithmic evolution, that core indeed contained only marginally severe hail and was surprisingly translucent, allowing us to pop out of precip a safe distance of 4-5 miles ENE of the storm’s mesocyclone. We could check this out briefly while plotting a course to get ahead of the Norman storm, by now producing a tornado over Lake Thunderbird–just a few miles from both my house and Rich’s. That sinking feeling came back again.
Yet we weren’t obligated to leave this storm just yet, not with us in an unusually good vantage point for this hilly, forested part of the central Oklahoma crosstimbers, and a tightening mesocyclone approaching rapidly on track toward a spot on the highway immediately to our north. From a distance, the main cloud area was so low that it might have appeared tornadic, but we certainly were close enough to confirm otherwise. As it crossed less than a mile away, the persistent, deep, scuddy lowering was rotating, but not especially fast, and had no debris or dust beneath. Inflow felt rainy and somewhat cool the whole time, which seems to have been more related to the earlier storm than this one. Nonetheless, it tried hard (deeply enhanced crop)…just not hard enough.
Having spent just a few minutes examining that supercell, we made our expected run for the former Norman storm, now approaching Prague, offering a tremendous radar signature, and yielding terrible media-broadcast tales of tornadic destruction from the Dale-Shawnee area. By the time we got to I-44 and US-377, it was obvious we would have to either: 1) make a big zigzag to take a safer approach, or 2) penetrate this storm’s forward-flank core too, but come out perhaps even closer to what was known to be a dangerously tornadic mesocyclone on 377.
No, thanks…fateful decision #5–live to chase another day! I don’t know what would have happened had we turned stupid and tried to core-punch that sucker, but I’m glad not to know. Even though the NWS tornado mapping for the event shows a break at 377, the risk was too great…and of course gorilla hail can take out a windshield and end a storm intercept quickly. Notice also the tornado path NE of Prague, between there and Welty (the W road E0960 off OK-48, halfway from Bristow to I-40).
Zooming safely ENE on I-44 then S on OK-48 toward Welty, we passed instead through far-forward-flank precip that was dense but only rain. We passed through hilly terrain near Welty and hunted for the first relatively open view of the big supercell that wasn’t right in the path of the meso. By the time we saw the storm, looking W on side road E0980 S of Welty, it was a deep, dark, low-contrast, HP drum. The storm definitely was producing a tornado at the time based on official mapping, but one we just couldn’t see with our eyeballs because of extremely dense wrapping precip. At the time of the last shot, the tornado would have been directly down the road, in the distance (deeply enhanced crop).
Shortly after I walked a little way down the road to try a different angle (deeply enhanced crop), a small car pulled up. The unknown chap inside noticed I was shooting the storm, and started an impressive routine of big-timing, offering instruction about the supercell’s structure and what was going on where. His descriptions were fairly accurate in a general sense, but this presumptuous little game was wasting both of our times and just had to stop. I’ve seen and studied a supercell or two myself over 28 years of storm observing and professional meteorology, so it was hard to resist temptation to bust out laughing. Instead I decided to have mercy and let him off the hook relatively painlessly. Once I could get a word in edgewise, I succinctly discussed the likely pattern of vorticity lines around the wrapping, precip-filled occlusion downdraft and its interface with the forward-flank gust front of a supercell of this configuration and morphological stage. [Thank you V.O.R.T.EX.] Quickly my interlocutor quieted down and turned his attention back to the storm. 🙂
Duly amused and a bit grateful at not being recognized, and not wanting his vehicle (or its glare) to take up too much of the shot, I wished the dude good luck, and jogged back up to the vehicle. Rich (who had stayed by the vehicle) and I watched hard for any evidence of the ongoing tornado somewhere in there (deeply enhanced crop). The rotation of the precip cage and surrounding cloud deck was impressive, but we still couldn’t see the tornado by eye. Even the deep enhancement leaves room for doubt.
After that, the visual appearance got even messier, the mesocyclone shifted NW then NNW of us, and RFD precip began to hit. We bailed S, glancing back at the remnant circulation crossing behind us (by then nontornadic, as it turned out) expecting that to be our last view of the storm in the fading daylight. An awful lot of rain was falling well S of the hook of our previous supercell, and once we got out of a cell-phone hole and saw radar again, it was obvious why. Yet another supercell had gone up farther SW, playing a game of mutual hindrance with the former Norman-Shawnee-Prague storm. They were aligned in ideal geometry to dump rain and/or outflow into each other’s inflow regions!
Enough was enough
Despite its ragged and somewhat high-based appearance (looking NNE from near Castle), we followed the last supercell N a few miles out of Okmulgee before twilight and unimpressive structure mercifully ended the intercept. Somehow we had witnessed four supercells and no obvious tornadoes on a big-event day, and aside from that, had seen very little in the way of photogenic structure. On top of our chase failures the previous day, this dealt a most demoralizing blow; and that wasn’t the worst of it.
At some point in there, I learned that the beloved cat Iniki, who suddenly had fallen ill a couple of days before, had to be put to sleep by an emergency veterinarian the same evening, while I was out driving hundreds of miles and failing to see anything of note. I never got to hold her and say goodbye. This day officially sucked. How could doing an activity I loved so much get so stinking miserable?
We ran into Bobby Prentice and Scott Fitzgerald at an Okmulgee gas station, hearing of I-40 being closed on our way home due to tornado damage (indeed, it was). We didn’t need more bad news. Yet, to make matters worse, we then learned that the Shawnee tornado had been a killer, and also, crossed the power-supply lines N of Rich’s house and cut electricity thereto for many hours ongoing. His food was unloaded into my fridge and big freezer, and he stayed at Elke’s and my place that night.
Our wish for the 20th was to turn the heretofore painful storm-intercept fate of the previous two days around on the third and final day of this weather system, despite fighting the urge not to even bother…