Cyclone Road


Saturday, April 29, 2006


I chased with Scott Eubanks today and we saw a few semi-interesting storms, the first near Snyder and the second at Sweetwater. The Snyder storm exhibited a large wall cloud but I never discerned serious rotation. The later storm also had lowerings in the vicinity of various mesos, but never threatened. When the first sensible RFD from our Snyder storm reminded me of the walk-in freezer from my high school Dairy Queen job, I wondered if we might be in trouble.

Scott and I took a wild ride through northeastern Scurry County and we tried to find the secret passage northeast of Snyder across the creek in the northern part of the county. Street Atlas was dead wrong about not one but two rural roads and we lost our storm because of a missing road (who took the road?) and a bridge that didn't exist (Bring Your Own Boat?). We through canyons, over dried creek beds, into open cattle rangeland, and on what looked like an access road astride someone's field of winter wheat. We might as well have beein chasing in the Caprock Canyon. Meanwhile our storm was spinning like a top, with MTM displaying higher shear values than ever. This is when the sheriff reported the brief funnel in Stonewall County near Peacock.

"Bob's Roads" are always a gamble but we assumed today's storms would have a limited period before they lined out or joined larger convective clusters and we wanted to stay as close as possible for as long as we could. Plus the map indicated there was a way to do it.

Our assumptions about today's mode and evolution were wrong. Despite the removal of CIN (per SPC objective analysis and hinted by MAF 12z sounding), when the lift arrived, the convection was not immediately widespread and messy. Our first storm was a discrete cell for nearly two hours and the second one for about half that time, before a line filled in and the squall was on.

I had an early dinner with a friend and decided to return to Denton where I met the night's weirdest tornadic supercell. This storm raised reports of power flashes near Alliance Airport and lowerings galore (none of which was easy to confirm due to the lack of lightning). They blew the sirens in town and that was probably the right thing to do with the power flash reports. I found myself under a wall cloud or a relatively compact updraft base after I couldn't find the intersection of 156 and 114 to flee east from the approaching meso.

Monday, April 24, 2006

I intend to chase whatever target remains available after I teach this afternoon. That confines me to southern Oklahoma or north central Texas, unfortunately. Yesterday I chased in Kansas and here’s the report I wrote on the drive home last night. Given my teaching schedule, today’s chase, and a busy work week ahead, pictures may be several days in the offing.

Report for April 23, 2006:

Eric Nguyen, Robert Hall, and I intercepted a supercell west of Salina, Kansas between 0z and 0130z. The storm produced three interesting wall clouds plus a large rain foot and several midlevel funnels. One of the wall clouds featured a tail cloud that hung from the elevated base to very near the ground, the longest example of this feature I've seen. The storm exhibited rotation for much of the time we chased it, but never threatened a tornado as it was too elevated.

Our initial target was the nose of what we expected would be a vigorous dryline bulge. We hoped storms would fire ahead of this lift and traverse the warm front along or just north of I-70. Like everyone else, we were disappointed by the quality of boundary layer moisture and by the apparent weakness of the impulse that had tempted us to chase under a ridge in the first place. Early RUC forecasts for midlevel winds were discouraging, but we checked profilers throughout the day and never gave up entirely on the idea that something along the boundary could see improved SR flow if it anchored to the front.

We wandered about and encountered Scott Eubanks on his first day of chase vacation, and enjoyed a lazy dinner in Wakeeney before turning back east with the intention of reaching I-35 and our route home. The lack of cumulus and consistent forty degree dewpoint depressions in central Kansas had convinced us it was time to turn south. Along the way, all three of us were either sending text messages or engaged in cell conversations about the bust when a rounded updraft base appeared north of the highway.

The base was elevated but compact and over the next twenty minutes a core developed then split away. We met Ryan McGinnis and Darren Addy at this point and watched the slowly-developing storm spit intermittent CG’s that ran us back into our vehicles. Afterwards, the remaining updraft base looked much like it had earlier near the highway, a smooth underside with a signs of circular organization. Our inflow picked up substantially and the storm developed a collar cloud that turned while scud condensed and attached to the center of the elevated meso. Within ten minutes, the storm produced a large, blocky wall cloud that rotated lazily while two elevated funnels formed around the periphery of the collar cloud.

After the first occlusion, we followed the storm a few miles south (mean forward motion was probably around five knots!) while looking in amazement at a new, larger wall cloud and associated tail cloud near a dark and well-defined rain foot. When we stopped, the tail cloud grew very low. Our best guess is that this feature was 1-1.5 km long and very vertical, nearly reaching the ground. While it's important to emphasize that this was not tornadic or even rotating with any vigor, it was nevertheless interesting for illustrating the elevated nature of the storm and how much trouble it went through to tap into boundary layer moisture and higher RH air along the stationary front.

Soon the outflow-dominated fate of this cell befell our last attempt at organization and the final wall cloud was much more elevated and broad. By 0130z, the storm suffered both from a lack of midlevel winds and contaminated inflow courtesy of convection to the south. With pictures and video we hadn't anticipated capturing, we turned for home around 0200z. A clear-sky bust became a fun and unexpected chase with a great looking supercell.

Eric and I will both have video and pictures on our websites within a few days.

Friday, April 21, 2006

The Oklahoma City City Arts Center is hosting a display of work from two southern plains artists, Deanna Wood of Denton, Texas and George Wilson of Oklahoma City, which focuses on severe storms, tornadoes, and the relationship between these phenomena and the people who share the plains with them.

I have seen Deanna's work and believe that all chasers would find some value in this show, particularly those who are interested in the compositional elements of storm and tornado photography and imagery.

For chasers on vacation looking for "down day" activities, this is cool event to consider. Here's the full press release:

OKLAHOMA CITY, OK- Seeking Shelter: The Art of Deanna Wood and George Wilson features two dynamic artists who passionately capture and explore the forceful nature of tornadoes. Approaching their subject with different intentions, the two artists will provide visitors with a stirring experience that captures the destructiveness and energy of tornadoes as well as the nearly mythic presence these storms possess. These artists grapple with the majesty of nature’s force and examine the concept and perception of shelter as it relates to these great storms. Seeking Shelter: The Art of Deanna Wood and George Wilson will be on display in the Eleanor Kirkpatrick Gallery at City Arts Center from May 4, 2006 – May 20, 2006.

There will be an opening reception on Thursday, May 4, 2006 at City Arts Center from 5:30 p.m. – 7:30 p.m., free and open to the public. Music will be provided by a trio of percussionists from the University of Central Oklahoma. The Eleanor Kirkpatrick Gallery is located in City Arts Center at State Fair Park, 3000 General Pershing Blvd., OKC, OK. Gallery hours are Monday - Thursday 9:00 a.m. - 10:00 p.m. and Friday - Saturday 9:00 a.m. - 5:00 p.m. Concluding the exhibit, Deanna Wood and George Wilson will discuss their work and artistic process on Saturday, May 20th, 2006 with a Conversation with the Artists. For more information regarding the exhibit or Conversation with the Artists call 405.951.0000 or log onto

Exploring the imposing, nearly omnipotent, presence of tornadoes as well as the humility they bring to their spectators and victims, Deanna Wood and George Wilson provide a beautiful and emotive homage to the destructive power of these storms. Seeking Shelter: The Art of Deanna Wood and George Wilson examines the interplay between an uncontrollable natural force and the people who seek refuge from its path. “Visitors to this exhibition will be confronted with installations, sculptures, and paintings that will challenge them to face their own thoughts, beliefs, and memories regarding the most fearsome and awe-inspiring natural force faced by Oklahomans,” explains Artistic Director Clint Stone.

Texan artist Deanna Wood’s work is inspired by recurring dreams about tornadoes. In her dreams these tornadoes loom in the distance threatening to cause injury and destruction. Wood likes viewers of her art to be unsure if the tornadoes in her images will remain in the distance or become an approaching danger. Wood uses some common images symbolically. Telephone poles and houses represent the safety of shelter. Trees and limbs represent the destruction and chaos of strong winds. Through all of these images, Wood illustrates the fragility of life. Speaking of her work, Wood states, “By focusing on the human need for order in the face of chaos, it serves as a reminder that life is fleeting.” Visitors to Seeking Shelter can expect to find Wood’s work crafted with a varied assortment of media including clay, wood, and encaustic technique. In the encaustic process, Wood melts beeswax and natural resin to create a wax medium. She then applies several layers of hot wax and fuses them to the surface of her work with a heat gun. Then alternating between scratching the surface and rubbing in oil paint, Wood creates painting that is rich in color and texture yet features an interesting play of opacity and transparency.

George Wilson, who has lived and worked in Oklahoma for the past 34 years, takes a different approach to his subject. Attempting to create art that depicts the transcendental nature of the human imagination, Wilson addresses the spiritual and mythical nature of both humans and tornadoes. He focuses on the action of the tornado and the counter reaction of humans to the tornado. He explores the creation and destruction cycle that accompanies tornado. He investigates the human response to tornadoes which includes fear, awe and mystery; feelings which according to the artist result in humor. Wilson states, “I make jokes about the possible permutations of tornadoes in my imaginary world.” Wilson is sculptor, metalsmith, and jeweler. His long artistic career has offered him the opportunity to work with other sculptors including Ken Little and Jesus Morales. During his career, Wilson’s style has evolved from realistic sculpture to its current state of abstraction laced with realistic elements.

This exhibit is supported by our season sponsors: Allied Arts, Classical KCSC 90.1FM, Kirkpatrick Family Fund, Oklahoma City Housing Authority and Southwestern Publishing.

Wednesday, April 19, 2006

I wrote last year about streamlining my internet time during chase season and while I made some progress, I want to make more this year. My idea for this season is that I'll post chase reports here after I process pictures or video grabs and polish the text. That might mean the next day or even the next non-chase, "down" day. I'll drop the quick updates that make it seem like I'm some kind of news organization trying to break a story. It's not important to me that I'm first with images or text, and that's a good thing because I'm never first anyway. Heh.

I'll attempt to provide some forecast rationale to explain, post-event, how I decided on a particular target. This removes the "pressure" of always posting a rushed forecast before a chase. I want to shorten these posts too. Perhaps a paragraph or two on the chase followed by a paragraph or two on the forecast, depending on the magnitude of the event. I don't think anybody enjoys a 1000 word chase report and probably nobody reads them either. I don't anymore. But then I read a lot of old chase reports over the winter and I might be burned out.

What this means is that posts here won't be as timely, but should have a more consistent quality.

Monday, April 17, 2006

Exciting news today that my mentor and friend Lee Martin was one of two finalists for the 2006 Pulitzer Prize in Fiction for his novel, The Bright Forever.

Lee was my thesis director when I was a student at UNT but more importantly he was the person who showed me why fiction writing was a worthwhile and rewarding vocation for reasons other than money or fame. He's a great writer and a remarkable teacher who has deserved this sort of recognition a long time.

Video aired on CNN over the weekend showing a chaser driving near the outer circulation of a tornado and hooting about it while his windows blow out and debris whips over the road. The anchors had a good laugh and characterized all stormchasers as thrillseeking fools who put themselves in harm's way. The trouble is that this isn't entirely false. There are plenty now who want to put themselves as close to a tornado as possible and some who want to "get inside." I wouldn't be surprised if a handful might even welcome injury if they thought the video payout would be worth the suffering. I'm not joking.

In all but a few cases, this has nothing to do with any love of atmospheric phenomena, aesthetic or scientific. Instead it's all about drawing attention to themselves, promoting their name, website, or affiliated media. Someone found a post written by that same death-defying CNN chaser where he said that he had nothing else to live for. Too bad he can't find a suicide method that doesn't impune the rest of us.

It doesn't matter to me why people chase. Motivations for any human behavior are incredibly complicated. But it's unfortunate when chasers are represented by those who chase for reasons so far from why most of us go out each spring, and when all the beauty and benefit of chasing is ignored in favor of violence and obvious stupidity.

Saturday, April 15, 2006

It looks like many of the reasons I didn't chase verified, especially that the dryline in Kansas did not initiate until it was in northeastern parts of the state, where I wasn't willing to go. Storms in southern Nebraska produced, around Beatrice, but I had ruled that out for a solo chase.

All in all, however, I suppose I regret not chasing, but that's easy to say when I don't know how difficult the chase was given terrain and speed (we haven't seen any first-hand reports yet) or how photogenic the tornadoes were. Still, the storms in eastern Kansas right now don't seem to have ungodly motions and I could have stopped chasing around KC and had a relatively easy seven hour drive home on all interstate highway. The regret is not particularly painful. May is near.

I decided to pass on this setup for several reasons. Most importantly, I don't believe sufficient moisture will be in place until the lift is further east and north than I want to go this weekend. So it's a distance thing. However I also had zero confidence in the NAM for either moisture (that's a no-brainer) and the placement of surface features, another area I believe it has performed poorly in recent events. The dryline will mix east faster than progged and one reason is that the air east of the boundary isn't as moist as the NAM believed it would be. So the movement of the dryline and the mixing ratio east of the dryline are related.

Another relationship that can't be ignored is the moisture content and the ability to erode CIN. People are citing optimistic CAPE values and CIN evolution in one breath and then admitting NAM is over-forecasting Tds by 5-7F in the other. Well, those are directly related. Those derived parameters are derived from base values of temperature and moisture content.

I do believe that supercells will form today, but they will do so in initially very high dewpoint depressions. LCLs do matter. Yet we know from recent systems that if a storm can find a boundary, all bets are off. I expect to see one or two tornado pics from later today, probably from the southern end of the MDT or even the dryline if it fires very late in central or southern Kansas.

Friday, April 14, 2006

Well, what can I say? I'll probably sit this chase out. I have no confidence in the model solutions in not one but TWO critical areas: thermodynamics and the position of the surface low. That's no big deal if you can be in the general area early to make adjustments which for me would mean leaving right now and spending the night in Salina.

I imagined a little while ago finding myself in western Iowa Saturday night with 1000 miles between my front bumper and Denton and thought, 'fuck that.' May is two weeks away and I'll go anywhere I please then.

I'll post an excerpt from a private email I sent to some friends earlier. I'm a reluctant forecaster today because there are striking differences in our model solutions and because we know that the NAM's dewpoint forecasts are cause for real skepticism. What this means is that the best I could do is compose two forecasts, one based on the NAM and one based on the GFS. Or I could say very little today and try and make a reasonable forecast tomorrow based on the actual atmosphere.

The problem with having extremely low confidence in any solution at this hour is that the target region is far away. I will have to make a decision rapidly about when to leave and where to go. If I chase this at all, I am committed again to leaving this afternoon or early evening and at least making Wichita, Kansas. There's some chance I'll back out of the whole thing. We'll see. Here's the excerpt:

"Last night I felt like the dryline bulge in nc Kansas was the best play for Saturday. I still think the bulge is the right play, where midlevel flow more normal to the dryline and a stronger cap should promote discrete, rotating storms. But I’m getting more worried about where that bulge may appear, with the GFS continuing to insist on a faster track for the system and the NAM developing the surface low in southwestern NE and moving to central and east central portions of the state between 18z and 0z. The strongly backed flow along and east of the low in southern Nebraska around Hebron to Lincoln looks hugely attractive to me. And that’s on the NAM.

If we buy into the GFS solutions over the last several runs, then the dryline bulge surges from nc KS into extreme se NE and even sw IA between 18z and 0z, while surface flow in ne Kansas veers rapidly in the same timeframe, decreasing low level shear values. Given the CIN forecast for the region, a decrease in convergence resulting from parallel wind fields and/or dryline passage before max heating could keep the area capped to convection until very late. Of course if the dryline sweeps through as early as GFS shows, convection would likely fire in those above mentioned regions closer to early evening when the CIN reduces somewhat.

Rich Thompson just posted on ST about a possible cold core play in Nebraska tomorrow also. I noticed on COD how the dryline occludes nicely on the 12z NAM run (GFS too), just like what... were hoping for a few weeks ago. Actually, I think people have looked for this evolution several times this year and it never seems to happen quite like that. I don’t trust the NAM for a process so “specialized” but it does show strong cold air advection and the vortmax moving into the area of the DL wrap."

Headed out to run a dozen errands this afternoon and must cut yet another entry short.

I expect to chase tomorrow somewhere in north central Kansas ahead of the dryline bulge. Some friends and I will most likely leave very early tomorrow morning for the target. I don't want to post a forecast until I can see this morning's model runs.

Thursday, April 13, 2006

I'm rushed as usual for a weekday morning. Saturday remained on life support as a potential chase day until last night's NAM and now again this morning the model that has vexed chasers all spring suggests that we should keep our weekend plans flexible. I don't trust the NAM enough to even bother with details, but the 0z EC from last night offers a similar solution for the timing of the main shortwave trough. Here we go again? More detail tonight and tomorrow morning.

Sunday, April 09, 2006

Struggling with a short story this morning and decided to waste an hour looking over weather data. 12z GFS sending a faint signal for possible convection on Saturday somewhere in Oklahoma or Kansas. The current solution is capped like crazy but that's not such a bad thing six days out. Our questions about moisture quality will arise again although we managed several healthy and photogenic tornadoes on the last go-round.

Speaking of April 6, my pal Mike Hollingshead captured some fantastic images as usual, posted on his page here. From that same storm, veteran chaser and forecaster Al Pietrycha enjoyed this remarkable view.

As for how my forecast performed, I don't consider it my best effort. I failed to properly discount model forecasts for the evolution of the surface low, which in turn wrecked my conception of how the warm front would progress. At 9:00 AM I mentioned Columbus to O'Neil, Nebraska. While supercells formed in this area during the event (and moved north or even northwest a la May 17, 2000, Brady tornado day), they did not produce tornadoes. Two hours later I mentioned a second area of south central Nebraska to north central and northeast Kansas. Obviously this was the better play as the storm reports map from SPC and the fine images linked above demonstrate.

I have come to believe that the early season promotes bad forecasts from me not only because I'm rusty, but because I have too much time to examine literally dozens of model runs leading up to a single event. The effect of this cumulative numerical brainwashing is that on Day 1, I tend to try and align observations with the expectations created by all those graphics in the previous several days. When the Day 1 data clearly shows something different evolving, it's difficult to dispense with preconceived ideas about what was supposed to happen. Bad habit.

During the real chase season (21 days away), this isn't possible. An active period offers no time for gazing four and five days down the road. A chaser is forced to deal with Day 1 data only and let tomorrow be damned until the end of the day when you scramble for an internet connection and the SPC Day 2 outlook to know which direction to drive until bedtime. As a result, you encounter the morning's forecast with a tabula rasa. The chaser also benefits from staying "in the weather story" each day, aware of trends with that particular airmass and how things have unfolded in the prior days. This effect is most acute when forecasting consecutive days under a similar regime, such as the first ten days of May in 2004.

Last note today is on a fine New York Times article that should interest chasers, particular in the northern plains. No, nothing political here; it's a piece about the emptying out of northern and northwestern North Dakota, and how the increasingly dry conditions and economic changes in the agrarian business model have drained the population so much that many worry the area will be completely uninhabited soon. I don't know if that's such a bad thing. I'm fascinated with the rise and fall of these communities and the ghostowns that remain, but I don't find the phenomena awfully tragic. It's the way things go in this species. Here's the story.

Thursday, April 06, 2006

A quick update. The surface low is deepening in northern Nebraska and winds are backed north and northeast of the feature, up into southern South Dakota! Sheesh. However a surface trough extends southward from the low into north central Kansas and perhaps the double-barrel low structure suggested by last night's NAM is revealing itself. This could serve to create two areas of favorable low level shear, the earlier area north and then another area south as the low level jet cranks up. By south I mean south central Nebraska and north central to northeastern Kansas. The best strategy today would be to pick a target region either north or south and then fly by the seat of your pants watching the evolution of surface features with great vigilance.

One advantage the southern area has that I haven't mentioned is the possibililty that later initiation will allow for greater moisture return and pooling. I'm not a fan of "new moisture" which might be on the menu for the Nebraska portions of the outlooks area.

With instability values suggested by insolation already underway and the powerful presentation of the upper level shortwave trough on satellite imagery, today has the potential for dozens of tornado reports from 20z until 6z tomorrow across eastern Nebraska, eastern Kansas, northeast Oklahoma, Missouri, Arkansas, and other locations. Even extreme northeast Texas could see supercell storms by 0200z.

Chasers from Nebraska to North Texas will be scrambling around later today waving their printouts of the latest SPC high risk or recently downloaded Jon Davies' cold-core papers while trying to decode a complicated surface map. It's complicated enough that I've started and deleted two separate posts already. I'll summarize my scattered and useless thoughts thus far.

The surface low developing in north central Nebraska currently is forecast to slide south into extreme north central Kansas near the border. A dryline should extend south of the low with a pseudo warm front stretching eastward. As the dryline pushes west toward the warm front, storms should fire on the boundary and move north northeast into the backed flow along the front. At least that's what would happen if this were a textbook and not the fucked up situation we're dealing with currently.

Here's what I really think. The shear in northeast and north central Nebraska will dwindle rapidly but sometime between 20z and 23z, adequate shear in the vicinity of the pseudo warm front will exist for low-topped supercells from around Columbus to O'Neil, Nebraska. This is dependent on the evolution of the low and whether or not it deepens and glides south to align our surface features as needed. It appears that the current RUC solution is on track for the position of the low thus far.

I don't like the storm motions in the southern targets and would stick with the Nebraska play if I were chasing. As for the southern end of the dryline? While 60 knot motions in the Jungle might be an adventure, I'll be having drinks with some friends at 5:00 PM in Sweetwater's patio bar downtown. Good luck to everybody chasing today.

Wednesday, April 05, 2006

About that never going to Nebraska in April thing? Heh. Let me take that back. No, I'm not going, but not because I don't want to go. I can't. But anybody who can beg, borrow, or steal their way out of obligations tomorrow should haul ass north starting ten minutes ago. This morning's ETA coming into UCAR right now shows a blockbuster system with a deep surface low (and trending deeper) holding over central and western portions of northeastern Nebraska and wrapping significant moisture around its eastern and northeastern sides. Highly backed and vigorous surface flow will create excellent low level shear profiles along the bulging dryline upon which storms will fire and traverse the warm front east of the low. Helicties and impressive lapse rates beneath a rapidly cooling midlevel temps (yes this is a closed upper low, so we can certainly call it "cold core") should support numerous low-topped supercells with tornadoes.

I would not be surprised if SPC doesn't upgrade their categorical risk on the 1730z SWODY2, though it is unusual to do so on a Day 2 with some questions about moisture. Yet in this case I think it's justified. In fact, given the extreme instability forecast on this morning's ETA (~3000 j/kg in place over much of northeastern NE by 18z), some of these tornadoes could be violent. What's more, while upper level winds are out of the south or even southeast, storm motions should be favorable for chasing, in the thirty to thirty-five knot range to the immediate north and northeast of the surface low.

This is about as perfect a tornado forecast as you'll find on a model, I believe. Whether or not those sorts of instabilities will truly develop is a function of the moisture return, insolation, and other factors that climatology leans against this time of year---but the model output is quite consistent and, perhaps more importantly, trending in the more favorable direction from run to run.

If I were able to leave, I would head for North Platte as a staging area, looking to refine my target tomorrow within the region of Thedford to North Platte to Columbus.

Saturday, April 01, 2006

Updating from inside the truck while Eric and Rob load their gear. We're heading northwest on 287 looking to keep pace with the boundary moving more slowly than progged earlier. I suppose our target is the double point (or triple if we're lucky), so today is a day of analysis and waiting. My expectation is that we'll find our first storm in the vicinity of Interstate 40 corridor in the panhandle somewhere from Pampa to Plainview.

Blog Front Page