2018-05-28 Event: CO/KS/NE/WY

[John Farley]

Am I wrong on having my doubts about SPC's 10% tornado area in CO today? I can see some of the logic, as some of the factors that Brett Nikeson mentions in his excellent discussion in yesterday's thread on the WY setup will be in play in this area today. But there area also a lot of negative factors such as cloud cover and capping (see NWS Denver/Boulder's latest discussion update), and the HRRR has neither much instability nor much precipitation in the area SPC has highlighted. Rather, if you believe the HRRR, the places to be would be either in far east-central CO near the KS line, or way up in the northeast corner. I made the decision yesterday to sit this out because where it looked best then (and still does to me, at least off the HRR) was around an 8 hour drive for me and did not seem worth it. Second guessing that now based largely on the SPC outlook (they know a lot more than I do!), but I am seeing a lot that could go wrong with that outlook from my admittedly less knowledgeable perspective.

 

[Quincy Vagell]

This seems to be to be a classic 10% tor day around this time of year in northeastern Colorado. Shear looks very favorable and low-level moisture levels are seasonably high. With that said, the area is socked into clouds and boundary layer heating is minimized. When I say "classic 10%," I mean that it's almost anyone's guess how it verifies. Almost any solution is on the table, although it seems unlikely that it ends up being a big day, meaning a localized outbreak.

Still, given upslope trajectories and locally enhanced low-level vorticity, there very well could be a few tornadoes. Will they be high contrast and/or last more than a couple of minutes? Who knows. High resolution radar simulations can't always pick up on small scale details, such as what may transpire in northeastern Colorado, so I wouldn't let that alone call this event off. If storms can develop, mature and remain relatively isolated, then it will probably be a "good" day for chasers in the area. Even if there are only a few sloppy, brief tornadoes in one general area, that could verify the 10%.

Northwestern Kansas looks a bit more interesting with boundaries in play, even better moisture and large CAPE profiles. The issue there is that there may be too much convection too fast, leading to messy storm modes. Farther south down the dryline, the story looks similar, fast upscale growth.

The Oklahoma/Texas panhandles looked interesting and still do, but the disconnect between stronger forcing to the northwest and better moisture to the east may be too significant. There may very well be a few isolated "tail end charlie" storms in that vicinity, but for limited boundary layer moisture and somewhat weak forcing aloft, any storms may struggle to mature/sustain themselves.

One thing that the KS/OK/TX area does have in play that hasn't been there in many events this year is favorable low-level flow. The low-level jet should really start cranking after 00z, so who knows... Sometimes the HRRR convects too much and if western Kansas doesn't go into MCS-mode too quickly, then that might be the area that performs the best.

 

[Dan Robinson]

The "triple point" of intersecting boundaries north of Burlington is intriguiing to me - it's been sinking southward with time, which isn't good - better for it to stall. The outflow boundary component crossing I-70 just east of the CO/KS border has been slowly orienting itself more favorable for deviant motion. Like Quincy said, CAMs are having a hard time with this and I'd just as soon throw them out at this point for realtime obs -way too many fine-scale details for the models to pick up on. A discrete storm on the boundary northwest of Burlington would be my play for the day. We'll have to see if more of the low-level clouds can clear out to the NW, time is running out for that to happen.

 

[Quincy Vagell]

The "triple point" of intersecting boundaries north of Burlington is intriguiing to me - it's been sinking southward with time, which isn't good - better for it to stall. The outflow boundary component crossing I-70 just east of the CO/KS border has been slowly orienting itself more favorable for deviant motion. Like Quincy said, CAMs are having a hard time with this and I'd just as soon throw them out at this point for realtime obs -way too many fine-scale details for the models to pick up on. A discrete storm on the boundary northwest of Burlington would be my play for the day. We'll have to see if more of the low-level clouds can clear out to the NW, time is running out for that to happen.

Agreed with the southward progression. One has to wonder if the boundary is just a little bit too late, meaning that there is some relatively cool/stable air north of the boundary. It is late May, so recovery seems likely, nonetheless, and the environment under/near the boundary is strongly unstable with limited capping.

The CAMs do well overall, especially with more regional/synoptic events. They struggle with boundaries (they really struggled with the southeastern Wyoming environment yesterday) and don't always have the right idea with convective evolution.

A recent example was Friday in the southern Plains, where the HRRR showed a mass of convection developing, but there were only a couple of isolated cells that went up. That was a more marginal environment and today is different. Still, who knows if the models are simply over-convecting.

The best thing to do is rely on obs, satellite and radar from here on out.

If there's one silver lining to a day like this, it's that it really makes you think and challenges you as a chaser. The 10% tornado area isn't the slam dunk target and we all know that SPC does not issue convective outlooks for chasing prospects.

 

[JamesCaruso]

Our initial target this morning was in an area bounded by Limon, Denver and Fort Morgan. Among other factors, models showed temps too cool in southwestern Nebraska, and too hot in southwestern KS. HRRR seemed to show persistent UH - at least until the last run I looked at before heading out, the 15Z run - coming right off I-25 and moving northeast, to the northwest of Limon. There was also a persistent UH signal around Gove/Dighton KS but this seemed to be in an area of more widespread convection and a north/south oriented line of storms. I was happy to see the SPC 16:30Z outlook upgrade the tornado probabilities in the exact area we were targeting. They did mention the cloud cover but expected it to burn off. En route, seeing the stratus was still as thick as ever, we stopped in Arriba, about 25 miles east of Limon and 55 miles west of Burlington. Limon has NNW surface winds right now, whereas Burlington looks good with backed winds and 80 over 60. As I write this, skies are brightening in Arriba. We will hang tight for a bit. Thinking the area that looks best at the moment is my crudely drawn ellipse in this image, which dovetails with Dan’s post above, but assuming the cloud deck clears I would prefer heading west and getting the additional orographic influence of the Palmer Divide and the ability to do more with less moisture. So we will continue to wait and see how things evolve.

Not at all regretting the decision to blow off southwest KS as not only are temps in the 90s but winds are already veering in the warm sector.

 

[John Farley]

Looks like that last outflow boundary blasting in from NE and KS got the job done when it hit the Cope storm. Probably on some of the storms NW of there, too. Obviously, I really screwed up blowing this one off. This reinforces what was said in the posts above about mesoscale processes.

 

[JamesCaruso]

John, don’t worry about it, I was out there and still screwed it up. Ended up going too far east to the storm north of Goodland, and when that died it was too late (and too few road options) to get to the landspouts in the Seibert/Cope storm. Two terrible days, first missed the tornado near Cheyenne yesterday, and now this...

 

[Quincy Vagell]

We all have our challenges from time to time. The best thing to do is brush it off and get back out there, ready and in position for the next chase. I targeted Colorado on Sunday and went for the southern play yesterday. I bailed north toward the Kansas part of the outflow boundary in the afternoon, but when it was clear that new outflow was surging southwest, I went back south toward the panhandles.

 

[JamesCaruso]

Thanks Quincy. Learning from mistakes is the important thing. And it’s not always easy to quantify what’s a “mistake.” In other words, even in hindsight, it wasn’t necessarily wrong to target an area that didn’t produce. There’s an element of chaos theory, where two targets can be equally valid/logical even if only one of the two actually produces.

On 5/28/18, I think my chase partner and I probably made the wrong decision when we left Burlington and went east. We did that because we had tunnel vision with regard to the KS portion of the OFB. Burlington was starting to dry out, so we went east toward higher dewpoint air, further influenced by a cell popping north of Goodland and, upon that occurring, mass chaser departures from the Loves truck stop in Burlington.

As we went after the cell north of Goodland, I knew it was heading into bad outflow air, but figured if it died we could turn west and head toward the northeast CO storm. (That didn’t work out because of the motion of the Cope/Seibert storm and limited road options.) But I didn’t think the environment in that area was *that* much better than north of the KS OFB. However, in a meso discussion, SPC did say the outflow was “less prevalent/hostile” in northeast CO. I guess that’s something I’d like to understand better, because based on the surface obs (below) northeast CO was clearly being influenced by the same cold pool, correct? Is it simply that northeast CO was further from the source of the outflow??

.

 

[Quincy Vagell]

@JamesCaruso when it comes to outflow, the general rule that I follow is that the newer the outflow is, the more disruptive its cold pool will be to the surrounding environment. The older that it is, the less influence it has on the thermodynamic environment.

I went back and looked at satellite imagery from mid-afternoon, around 21z. Notice how the "new" outflow from northern Kansas convection is impinging on the original outflow boundary and has reached the general vicinity of Goodland, KS. The outflow was still about two hours away from impacting the storms in northeastern Colorado. Even though there still was cold pool interaction with time, this portion of the new outflow was delayed 2-3 hours from when it impacted the northwestern Kansas environment. Right there, you have a gradually "weakening" cold pool moving into an area that had the extra time to recover from previous outflow, so it would take more influence to substantially disrupt the environment. Combine that with the fact that storms in northeastern Colorado were already maturing and there was not much influence.

On the surface chart you posted, notice how Colby, KS was 64F at 22z. Northeastern Colorado was in the 68-73F range. In addition to the better thermodynamic environment, mesoanalysis indicated that there was substantially more low-level vorticity in place in Colorado vs. Kansas.

In hindsight, if there wasn't the secondary outflow, who knows if the northeastern Colorado storms could have become even more intense. Sure, there were several landspouts and at least one, true mesocyclonic tornado, but maybe there would have been more of a supercell tornado threat if the environment was not disrupted. Even with relatively modest thermodynamics in northeastern Colorado, the degree of low-level shear was supportive of tornadoes, even if most of them were of the landspout variety.

 

[John Farley]

@Quincy Vagell Good discussion, thanks! What I wonder on the other hand is, if that new outflow boundary had not hit the storms on the original one, would there have been tornadoes at all? They seemed to occur about the time the new boundary hit the storms that went up on the original one (and also storms farther northwest). And of course boundaries are important in creating landspouts, the dominant tornado mode yesterday. So it may be that this boundary temporarily provided the extra boost of vorticity needed to make those storms tornadic.

 

[Dan Robinson]

The Siebert storm formed north of the original outflow, then slowly backbuilt southward until it was able to get updrafts right on it. The west-moving outflow from the Kansas MCS was threatening to blow through the entire area, but it thankfully slowed enough to allow the Seibert storm to proceed unimpeded. I targeted Burlington, and once I saw the west-moving outflow I knew it was game over there. At that point the only option was the Seibert storm. When in doubt, I always choose the storm on an (old) outflow boundary rather than ones in an open warm sector..

 

[Jeff Duda]

The Siebert storm formed north of the original outflow, then slowly backbuilt southward until it was able to get updrafts right on it. The west-moving outflow from the Kansas MCS was threatening to blow through the entire area, but it thankfully slowed enough to allow the Seibert storm to proceed unimpeded.

I would disagree on semantics here. The updraft pulses on the Siebert storm were almost certainly directly forced by the OFB, so I think it's misleading to say the storm formed north of the outflow. Upper level winds pushed all updrafts north of the OFB, but since the only updraft forcing was at the OFB itself, that basically forced the storm to stay anchored to the boundary (because as soon as any given updraft pulse moved north of the OFB, lack of CAPE killed it off).

This storm was very interesting indeed. I don't know if it was ever truly a sustained supercell. It seemed to generally show transient supercell characteristics, and most of the tornadoes were non-mesocyclonic, likely spinning up due to an updraft pulse sucking up the vorticity along the OFB. It did seem that one mesocyclonic cycle occurred, but that tornado seemed to be pretty rain-wrapped. And the show definitely seemed to end once the new pulse of outflow from the east intruded on the environment.

 

[JamesCaruso]

Thanks all for the excellent discussion. Quick follow up, question, the fine line depicted in this radar image is the “first” OFB, not the “second,” correct?

I think our first error was heading east from Burlington to begin with; we knew the air behind the OFB was too cool/stable but with Burlington drying out we wanted to get more into the warm sector south of the OFB, which as Dan noted was probably a bad idea. To be honest I can’t even remember if the northeast CO storm had already formed when we first bailed east. I do know it had formed once we went north out of Goodland, and that’s when we figured we could always cut west if the storm north of Goodland died. But we were way too late by then.