February 28, 2017

A new year brought with it a new storm chasing season and a new vehicle. The van was dead. The transmission blew in Great Falls, Montana the previous year, and that’s where we left it. I was looking for a good replacement vehicle. I wanted another van, but I also didn’t want to cut another hole in the roof for my camera mounts. The van leaked terribly. So instead I was looking for a vehicle with a sunroof. The Subaru Forester looked like it was the best fit in the end, with its oversized moonroof, seats that folded down flat providing enough room to camp, all wheel drive, and decent fuel economy. I rode a train all the way up to Chicago to get a used, low mileage 2015 that was turbocharged. Known for being reliable, long lasting, and popular with outdoorsy types, I thought I had found the perfect car. I didn’t know it would wind up being the bane of my existence due to unresolved stalling issues that Subaru either didn’t know how or was too lazy to fix. But for now it was still new, filled with potential and possibilities, but also completely untested out in the field. I had yet to do any major work on it or configure it for chasing. So I was going into 2017 with a stock Forester that had yet to go on a chase. Without my screen or camera mounts I’d be chasing with one hand tied behind my back.

It was way too early in the season to be thinking about chasing, but the GFS was advertising a potential significant severe weather event in the Midwest at the end of February. Unseasonably warm and moist air was forecast to push up into central Illinois with dews as high as 60 F.

Early season setups are often marked by excessive wind shear yet marginal instability. The GFS was advertising 90+ knots of midlevel pushing into Illinois. Storms would be flying. I don’t like chasing low topped rockets. They’re impossible to keep up with and storm mode often isn’t the best on these kinds of setups, but these are the kinds of parameters you never want to turn your back on.

The setup was trending favorably for a significant severe weather event as February 28 approached with the 3km NAM showing more than 2000 J/kg over Illinois and 60 knots of bulk shear. That’s an impressive combination for February.

In the wake of an expected morning MCS, the low level jet was forecast to be cranking hard out of the south. The directional shear was causing the tornado parameters to spike with values that suggested strong or even potentially violent tornadoes were possible. After years of being burned on overhyped setups, I’m still a pessimist, and wasn’t convinced this was going to be a good storm chasing event. Storms looked like they’d be stuck on or behind a cold front to the northwest, removed from the best tornado parameters, or firing after dark to the south. The maxed out parameters in the center of the warm sector across central Illinois could go unused.

The Storm Prediction Center was on the ball and gung ho for the setup, issuing a moderate risk for tornadoes across the Ohio River Valley with the possibility for significant tornado coverage across the Midwest.

I have a couple of go to tornado forecasting parameters that I extensively use and on which I heavily rely. I call them my secret weapons. They’re the 0-3km CAPE and the lid strength index. I don’t see them utilized much by forecasters, but I think they’re excellent tools for forecasting environments that support surface based storms with robust low level updrafts. They’re great for honing down a large target area and homing right in on a specific storm chase target, and also for discriminating environments that will and will not support a tornado threat. The 0-3km CAPE shows the instabilityin the lowest layers of the atmosphere and is a necessary ingredient for storms with robust low level updrafts. I generally want to see about 100 J/kg or more at my target area. On February 28, the NAM forecast the 0-3km CAPE to be pooling on a cold front/warm from draped from northern Illinois southwest across northern Missouri, and another pool further south into the warm sector across southern Illinois and southern Missouri. There was a noticeable hole across east central Illinois into Indiana, however, and that’s an area that I suspect would not support a tornado threat, despite being in the center of SPC’s 10% hatched outlook region.

The lid strength index is essentially showing the width of the capping inversion as depicted on a skew-t forecast sounding. It’s a measure of how strong the lid is that suppresses surface based storm initiation and surface updrafts. I want to see this value below 2 for a decent shot at robust, surface based storms, and areas where it’s 0 or less are often areas where the NAM is trying to initiate storms. This plot gives me a good idea of where and when storms might form, and whether or not that they can support a tornado threat. Areas along the warm front/cold front were again highlighted, and down south into the warm sector. In the center, however, it appeared that a blob of cooler, stable, and capped air was passing through, no doubt the chewed over air mass from the morning MCS. That’s an area where I wouldn’t expect to find a significant tornado threat, despite the significant tornado parameter spiking off the charts there and this being centered in the Storm Prediction Center’s outlook.

I scrutinized the visible satellite loops and surface station plots the morning of February 28. North or south? That’s often the age old question when it comes to picking a storm chasing target. I held my ground in Springfield for as long as I could, not wanting to commit to a target until I was sure. Areas of clearing were evident at both the north and south ends of the outlook area. Agitated cumulus fields percolated in areas of clearing where solar heating was destabilizing the atmosphere. To the east, the surface stations showed the winds backing from southwest to southeast indicated favorable directional shear for supercells and tornadoes. It was the warm front on the northern end of the target that most had my attention. A finger of low level lapse rates was plotted on SPC’s Mesoscale Analysis extending across northern Missouri and poking areas of higher moisture to the east along the Mississippi. The pattern indicated an area to me that probably fire storms off first, right along the IA/MO/IL corner. Strong directional shear measured by the effective storm relative helicity was plotted along the warm front now draped across the Illinois River Valley. Storms firing on that corner along the river but be headed right toward to the warm front and this directional shear.

The High Resolution Rapid Refresh had a solid grasp on the pattern and was indicated a strong signal along the warm front target. Discrete supercells were forecast to track from north of Peoria toward Will County in the later afternoon and early evening. This forecast model can be really hit or miss, but sometimes it’s spooky good in nailing the locations of specific storm cells. February 28 was one of those spooky good days. Overlayed on top of this updraft helicity plot are the actual locations of the EF3 tornadoes that formed on this day.

Everything was coming together now for a tornado event, and indeed the Storm Prediction Center had the area highlighted for a tornado watch. The HRRR’s solution and the forecast made sense: storms would fire in the agitated cumulus field over northeast Missouri, track across the warm sector and toward the warm front draped from northwest to southeast across Illinois. Storms would start to rotate and peak right on the front in the region along the I-80 and I-39 corridors from Peoria to Bloomington. I had my target now and was heading out the door, having held my ground at the house until midafternoon.

The Forester wasn’t equipped for chasing yet, missing the large touchscreen I normally use, laptop, power inverter, and camera mounts. I would be doing this chase using only my phone and Radarscope app. This is what it showed as I left house and headed north up 55 toward my target. A broken line of storms, some already supercells, was in progress tracking northeast along and north of the warm front.

While approaching Lincoln, I saw a discrete storm on the radar that was going up near Jacksonville. It would pass close to Springfield. I had flashbacks of March 15, 2016 when I left the house to chase storms to the north, only to have a tornado pass close to my backyard. I could easily turn around and intercept it. I took an exit and pulled over to watch the radar and mull over the target. The storm was by itself and would probably stay discrete, while storms to the north were already stepping on each other’s toes. I thought to myself: “No, your target is good. Stick with it.” I got back on the interstate and continued north. Within an hour or two the cell near Springfield withered and died in the blob of capped air I had identified on the lid strength index and 0-3km CAPE plots.

Running north out of Bloomington on 39, a storm near Ottawa was tornado warned. I’d be playing catch up trying to intercept it, it was over the difficult terrain of the Illinois River, and was semi embedded. I decided to let it go and pick up a newer cell with more favorable chase conditions. A severe warned cell was taking on supercell characterizes, was more discrete, and positioned ahead of the line as it tracked north of Peoria. I knew better than to race across the river and get on it early, however. There are few river crossings and if the storm wasn’t ready yet, I’d be left behind as the storm produced on the warm front east of the river. I exited at Minonk and stopped to refuel as the storm continued to organize before I headed west to pick it up on the east side of the river. The cell held a brief tornado warning, and a report of a brief dust whirl tornado came in, but that was it and the warning was allowed to expire with only a severe thunderstorm warning remaining. The storm needed time to cycle and reorganize.

I was going to try to catch the storm on the river, but the terrain started to get really squirrelly within a couple miles of it, so I backed off and headed north a bit trying to find a nice open vista. North of the small town of Washburn I had my view. A classic supercell came into view, miles away to the southwest still and I waited for it there. Illinois based storm chaser and meteorology professor Walker Ashley passed by and said hi. We didn’t know it of course, but we were minutes away from an amazing tornado encounter.

Setting up downstream, I actually expected the storm to pass by to my north on its original east-northeast trajectory. But intensifying supercells often curve to the right, and this one did as well. I wound squarely in the inflow notch. Bands of inflow fingers fed into the base. An RFD clear slot and horseshoe shaped updraft base wrapping around it was faintly visible, but from my position a few miles downstream to the northeast and within the notch, I did not have a good visual on this structure. I anticipated the mesocyclone and area of interest for tornadogenesis to be further to the right along the base relative to my view. I couldn’t make it out, but low level rotation was rapidly increasing, yet the storm still sported only a severe thunderstorm warning.

And so I wasn’t expecting it when a debris cloud started to kick up my southwest, further left of the precipitation and out in the clear air than I would have thought. There was no mistaking the intense motion, however. This was a tornado. The debris cloud spin-up was followed by a snaky rope funnel that danced and slid up and down within the column of rotating winds. Local spotters were on the ball. The sirens in town sounded within seconds of spin-up. It took the weather service a couple minutes to issue a tornado warning for the already in progress tornado, however.

I positioned the tripodded camcorder on the tornado and zoomed in. The spinning cloud of dirt rapidly expanded yet the funnel retreated. The air filled with pieces of debris as the vortex impacted a structure on the ground.

The initial location of tornadogenesis surprised me, so I was immediately focused on which way the tornado was moving. Initial motion seemed to be to the left. I got to watch the tornado for about a minute. Using the tree as a stationary reference point, I compared the motion of the debris cloud against that fixed point. It didn’t appear to be moving left or right at all, only rapidly growing in size. I knew what that meant. The tornado was getting closer not just bigger. I could even hear it now. The waterfall rush sent chills down my spine as I knew I was in the path and needed to act immediately on my escape route.

I threw the tripod in the car, camera still attached, and raced east down the road almost a mile. Then I stopped on the side of the road, yanked the camera off and got out to shoot the tornado handheld. I was expecting the tornado to be immediately to my west and moving off to my north. But it was in the same position as before, off to my southwest, and even closer this time. How could this be? It was like the tornado was *chasing me*. The waterfall roar was distinct. The vortex crossed behind a farmhouse, dwarfing it, the air filled again with debris. Alarmed, I was about ready to jump in my car and start racing east again. It was a really good thing I didn’t.

I held my ground and watched the motion. It was definitely moving left now, and would pass me to the south, although at dramatically close range. Had I raced east, I would have been on a collision course with the fast moving tornado.

The tornado sped between me and the town of Washburn, most fortunately missing its center to the north. The roar of the wind mixed with the wail of the sirens in town. Now clear of the path, I watched transfixed.

The debris cloud crossed highway 89 off to my southeast, striking the power lines and causing them to arc with momentary bright green flashes. Now on the front-lit, backend side, the tornado churned as billowing brown clouds of dirt. White pieces of structure flittered and tumbled against the dark backdrop.

The funnel cloud reformed as a huge bowl, the debris cloud reaching up to kiss it. An intense inflow jet, the “ghost train”, screamed in feeding the tornado from the southwest, the dry dirt acting like a tracer for the invisible ground hugging winds. It looked like it was about to impact a farmstead, but the tornado missed to the south, passing behind the farmstead from my vantage and absolutely dwarfing it in scale.

I snagged a wider shot with my phone, showing the parent supercell structure above the tornado as the column of wind now filled in with brown dirt.

I set back up to get a tripodded shot, but the fast moving tornado was racing way from me to the east. It appeared that it was still growing in size and intensity, looming in the distance against the foreground scene.

After shooting for a couple minutes, I yanked the camcorder off the tripod, jumped back in the car, and raced east down the back roads trying to keep up with it. The tornado still appeared to be growing in size and was approaching wedge status. I was intently watching it, but then noticed debris in the road up ahead. That’s where the tornado crossed. The roof of an outbuilding, wagon, and some tree damage were blocking the road ahead and my attention was focused on that. I didn’t notice the downed power lines several hundred feet ahead of that damage. The thin lines were draped across and suspended several feet above the road, ready to clothesline an unsuspecting victim like an invisible tripwire. Fortunately I was already slowing down for the debris up ahead when I struck the lines. They hit the front end of the car and tangled on the right side mirror with a sickening metallic grinding and scraping. The impact was totally jarring, jolting my attention off the tornado and debris, and I brought the car to a quick stop as I recognized what had happened. The lines were draped across my hood and then went up at angle from the mirror. I put the car in reverse and tried to back out from underneath them, but they were snagged and not coming off by themselves. I got out to assess the situation, the large tornado receding away on the northeast horizon. The right side headlight cover had a crack and there were gouges in the paint down the right side, but otherwise the body appeared alright. It looked like you could just lift the lines off the mirror. Even if the power lines were completely dead, I knew better than to attempt something like that. I didn't want to become a storm chaser victim and take attention and emergency services away from other residents that would likely need it. I got back in and attempted a couple more times to maneuver out of the lines, and then going in reverse one last time, the tension on the line snapped my right side mirror off. The cover lay in the road, the rest of the mirror assembly still hanging from the body by the power cable.

Once free, I turned around and headed back to highway 89, which ran northeast out of Washburn. I wasn’t sure if I was still chasing or not. Many miles off to the east I spotted a tornado, a thin blue rope in the dusk light. I turned east toward Toluca to go after it, only managing to snag a shot of the dissipating debris cloud (visible on the right side of the image here next to the buildings). I thought it might have been the dissipating stage of the tornado I was just on, but it turned out to be a separate tornado.

By this point I had fallen well behind the storms. Light was fading fast as it was now after the winter’s early sunset. I lost visibility on anything up ahead. And my mirror was banging loudly against the side of the car in the wind. I decided to abort the chase, the tornado show was likely over anyway. I stopped at a gas station once I made it back to I-39. I either didn’t have a blade or didn’t want to mess with removing the dangling mirror assembly, so I just duct taped it down to the door so that I could drive home without it flailing around loose.

My first chase out with the new Forester and I managed to bang it up pretty well.

The Washburn tornado was rated EF3. Luckily it managed to not only miss town but also most of the farmsteads along its path. There were no fatalities or injuries reported with the tornado. Tragically deaths were reported with another EF3 that struck near Ottawa, IL on an earlier storm. The next few plots are courtesy Nick Nolte and his interactive event viewer web tool: http://tornado.nnwx.us/

The rope tornado I witnessed way off to the east was rated EF2, tracking near Long Point, IL.

SPC’s tornado outlook more than verified with an outbreak including a long track EF4 that struck downstate in the after dark hours.