What would you have done??

[JeremyS]

I'm starting this thread to get a few opinions of some of the chasers on the forum. I had a pretty frustrating couple of days chasing 6/28-6/29, where I struck out on the tornadoes. This is a very humbling "hobby" of course an no one is ever 100% for seeing tornadoes on chase days, but obviously all you can do is keep learning. I thought I would post this question/situation on what happened yesterday to see what other people "think" they would have done.
After missing 3 Iowa tornadoes the day before by a matter of minutes/miles, my chase partner and I went out again yesterday(6/29). We targeted northeast Nebraska along the boundary that ran east to west into northwest Iowa. I wasn't real optimistic heading into the chase for tornado chances, but you never know.
If any of you were watching yesterday or were even out chasing this should sound familiar. There was an initial cell that formed about 20 miles west of Sioux City. We were right there as it formed. I was on the SPC mesoanalysis site and looked up shear and helicity parameters. Bulk 0-6km shear was initially lacking but I could tell was going to get better with the approaching wave. Effective helicity was rather low in the immediate area though. One thing that captured our attention though was an increased area of helicity values(200 m/s2) on the mesoanalysis that ran along the Missouri River in southeast SD.
The storm continued to strengthen as it headed towards Sioux City. It was in an area of good surface based CAPE that stretched well into northwest Iowa. So initially we stayed with it a little bit as it moved along at 15-20 mph to the east.
At some point though, we decided to let it go as it really looked like it would head further into an area where the storm SRH was very low. There were more storms to our west and to our northwest up in S. Dakota. The storm however did start some attempts at brief, either very slowly or even non rotating wall clouds that usually fell apart pretty quickly. There was even some rotation showing on the velocity scans on radar that were slowly getting better.
Then a tornado warning came out for it just as we had started to head west away from it and head towards the Yankton area in the higher SRH values. Since we were so close to the tornado warned storm we turned back around to follow it through S.City. There were reports of up to 4" hail as it moved through town. Still though, the wall cloud attempts were pretty feeble. But we couldn't leave the storm. It's hard to leave a tornado warned storm when you're right there! Eventually we moved into Iowa where about 20 miles southeast of S.City the tornado warning was finally dropped. At that point we knew to leave the storm and looked back to the northwest. Almost comically, 10 minutes after we left the storm it went tornado warned again and then a radar confirmed tornado was reported. DOHHHH!!
There was a rather discrete storm to the west of Yankton that didn't look that impressive on radar but was severe warned for wind. So we finally headed towards the Yankton area about 50 miles away. Mesoanalysis still showed SRH values of 2-300 right along the river. As we blasted northwest and made it onto I-29 in S.Dakota we could see the storm to our west and the base of the storm. As we approached the exit to head west off of I-29, we could now see a wall cloud under the base and sure enough we could see a tornado under the storm from about 18-20 miles away! There was a confirmed tornado with the storm at this time. The tornado only lasted about 2 minutes longer once we could see it before it lifted but had been on the ground for longer prior. As we approached Vermillion, we were about 8-10 miles from the storm and it made another attempt at a tornado with a bowl funnel and you could see dust swirling underneath(I love how flat S. Dakota is). The structure was amazing and I wished we would have stopped to take pics, but we wanted to get closer.
Of course as could almost be expected, once we made it through Vermillion and crossed the river back into Nebraska, the storm rapidly fell apart and we didn't see anything interesting further after that.

So if I haven't totally lost your interest yet, what do you think you guys would have done?
Would you have stayed with a tornado warned storm that you were right on despite it not showing anything real promising visually(still looked somewhatgood on velocity scans though), and in an environment that on mesoanalysis that didn't look that good for tornadogenesis?
OR
Would you have bailed on it and gone to the Yankton area where there were some weaker storms initially, but were in a better environment(SRH on SPC mesoanalysis)?

We left the storm feeling defeated and headed back towards Omaha. Near the town of Ponca, I then got a speeding ticket for going 75 in a 60. However, the cop was very nice and lowered it all the way to a 65 in a 60 for a $59 ticket. So I guess it didn't end as badly as it could have. Then after getting home to Omaha, we had golf ball to baseball sized hail move right through the center of town! Pretty significant damage everywhere!

 

[Ben Holcomb]

Never leave a storm with a rotating wall cloud.

With that said, it's hard to say. I would have checked echo tops, vil and ZDR on oax and fsd. Those three things can give you an idea of updraft strength.

The other thing to remember is that mesoanalysis is model output and not necessarily gospel.

Sent from my XT1650 using Stormtrack mobile app

 

[John Farley]

FWIW I probably would have stayed with the initial storm longer than you did. It is not unusual for storms where the tornado warning has expired to cycle back up again and get new TOR warnings and sometimes produce. I tend toward a "don't leave fish for fish" (fishing expression but applies to storm chasing, too) kind of attitude. It is always a judgement call and I was not there, but I probably would have stayed with the initial storm longer. That said, I do not think you should feel "defeated" - you did, after all, see at least one tornado and maybe 2. A bowl funnel with dust swirling underneath is a tornado in my book.

 

[Quincy Vagell]

I have two responses, but first, let me comment that I had very similar experiences to yours over the past few days. The only difference is that I didn't catch a tornado yesterday, so kudos there.

This is going to be a very long reply, but the summary is that chasing is challenging, especially if one chases mainly for tornadoes. Everything has to go right for tornadoes to form and that doesn't even include other things that could go wrong, like messy storm modes, longevity of tornadoes, terrain, road networks, traffic, busted model forecasts, poor/missing observational data, etc. Each setup is unique and below I'll eventually explain why the past two days included different environments with approaches that may have worked one day, but not the other.

Specifically about yesterday:
There are two ways to look at it.

The first being strictly meteorological, in the moment. When the discrete cell went up in far northeastern Nebraska, it wasn't long before I jumped on it. It was along a boundary and was isolated. Even as it moved east and other storms fired, the lead cell looked the most organized and had little to no convective contamination out in front, so a case could be made that this was the cell that would have the most longevity and potential to become/stay severe. It did, but as you mentioned, it never really had the "look," even though some of the low level rotation SE of Sioux City looked like it might finally produce.

The second viewpoint deals more with mesoanalysis and convection allowing models. The HRRR is king with convective evolution, especially between 3-12 hours out. It seems to be about 70-80% accurate (subjectively) with convection, it's placement, mode and intensity. On the high risk tornado day in OK this May, it kept on insisting messy storm modes, despite an extreme parameter space. For several runs in a row yesterday, it kept showing the most intense cell forming NW of Yankton and dropping SE toward Sioux City. More or less, that was the cell that produced the tornado you mentioned. SRH and low level shear were better up there than points south, but I hesitated due to limited "room to breathe" and concerns about cell longevity. While the lead cell by Sioux City was more isolated, the low level shear was not as impressive as farther NW.

Chasing is tough. I've been regularly chasing since 2014 and this year I've finally been more daring (based largely on experience) about abandoning storms, even if they are tornado warned and/or have the most chasers on them.

Conventional wisdom says that you favor the storms that are most discrete and isolated, but the past two days have proven that this isn't always the mandatory approach, especially if there is substantial low-level shear.

I've come to accept that if one chases regularly (both low and high end events), they'll probably not see a tornado more often than they will. Maybe I'm wrong, but even some of the veteran or otherwise successful chasers have had bad "luck" not only this week, but this year as well. (There seems to be a lot of bad luck since 2014, mainly since tornado counts in the Plains have more or less been below average in favorable chasing terrain - still with a few historic events, like Dodge City and Pilger) Aside from 6/12/17 (even some chasers bungled that event), I haven't seen any conclusive tornadoes since March.

It's been a tough season for tornadoes and chasing, but I'm still learning. All chasers can learn from each chase and get more strategic in the future.

It's hard to specifically give advice on the past two days. Chasing is an art and for me, it comes down to the best balance of assessing mesoanalysis, short range CAM guidance, radar/satellite, experience and patience.

I find that choosing a target after carefully assessing the environment and sticking with it is often the best move. In 2014-2016, I found myself going back and forth too much, considering too many targets and often staying with a storm too long, only to have to adjust back 50-100+ miles in a short span of time.

While the HRRR is a huge tool that I heavily weigh my chase targets on, it's not the only tool. Just like mesoanalysis alone isn't always the most helpful. Sometimes bad data gets ingested and I like to supplement surface maps with it as well. Tuesday was a great example of why severe indices are often misused and over the past two chase years, I've learned to use them with more of a grain of salt than I did in the past. https://twitter.com/stormchaserq/status/879895015446663168 I've also learned the importance of assessing low level shear, 0-1km. Today for example, SW Oklahoma looks primed for severe, but 0-1km shear looks sub-marginal, despite decent 0-3km shear and favorable deep layer shear with large CAPE.

In addition to 0-1km shear, I also try to make a point of assessing a "tornado checklist" that I'm working on as a gut check with storm chasing. I went into most chases this year with tempered tornado expectations, accepting that overall, any particular setup really needs to thread the needle to produce tornadoes, let alone photogenic ones. (On the checklist, a work in progress, I assess instability, shear in different layers, shear vectors, SRH in different layers, LCLs, dew-points, mean mixing ratios, lapse rates, height tendencies, wind speeds at 850mb and 500mb, etc.) I honestly don't complete the checklist every event because it's time consuming, but it's helped me narrow the focus to several specific parameters and how they come together, and not just using the plug and chug CAPE/shear combo with a side of SRH. (Not saying you do that) As part of the tornado checklist, I also make note of red flags and things that could go wrong, as another way to assess various targets and have a realistic expectation of what the day may have in store, regardless of SPC outlooks, recent/prior days chasing, etc.

Even those events that do produce, often one has to be very precise about their target, AND hope that both terrain and road networks cooperate. Wednesday in Iowa was tough since the tornadic supercells just SW of Des Moines were closely clustered, had somewhat limited windows to produce tornadoes and rolling hills along with dirt roads were also limiting factors. One would have had to commit to the initial cell in far SW Iowa fairly early in order to catch it. I actually saw that the low level environment was extremely favorable out there, but I overlooked the fact that very low LCLs combined with substantial low-level shear meant that any discrete storms would have a high likelihood of producing a tornado early on in the life cycle.

No two setups are the same. The more one chases, the more they can learn from experience to make the best decisions on a given chase day. Tornadoes formed very quickly on Wednesday, while Thursday involved a bit more patience.

I'm okay with the fact that most of my chases will not include tornadoes. That makes it more challenging and tests my knowledge and patience. If all else fails, I'd rather avoid a messy target/storm and focus on where I may have a better shot at seeing something photogenic, other than a tornado, although this last piece of advice wouldn't have been very helpful for the past two chase days.

 

[JeremyS]

Thanks for the replies so far! Great thoughts and point of views to consider. I've caught myself thinking, am I becoming a storm snob where I only want tornadoes and can't enjoy the chase or just a beautiful storm? For the most part I don't feel I am that way as I've been chasing for 10 years now and every year going into the season it's all about just getting out and seeing storms and chasing again. I chased last week on a day where I knew there wouldn't be tornadoes because I just wanted to go out by myself and see a storm since it had been about 4 weeks since I had chased. I think this year has been one of those years where "desperation" to see a nice tornado affected my mood after yesterday.
Wednesday we didn't see a tornado but there was some amazing structure. Yesterday as we approached the storm by Vermillion, that was honestly one of the coolest looking storms I can remember ever seeing in person(again wish I had stopped right there to take pics), before it rapidly fell apart. Have to remember those things about chasing and not just the tornado misses!
I think the tough part to this week was being so close to having a couple of really great "tornado days", and instead missing out on Wednesday and seeing the tornado yesterday from a long distance with no chance of getting any pics or videos.
On to the next chase!

 

[Jeff Duda]

Storm chasing is a lot like poker. It's not that hard to learn how to do and you can be successful at it right off the bat. Sure, people who are more knowledgeable and experienced will have a slight leg up on noobs, but when it really comes down to it, luck contributes a sizeable fraction towards total success.

Tornadoes are a feature on scales smaller than we observe other than with our eyes. Consider this: no model forecast, satellite images, radar observation (save for something within 10 miles of the radar site, and even still only large tornadoes), or METAR observation can truly resolve a tornado. A tornado depends on so many small scale factors that evolve faster than model forecasts, surface obs, and "survellance mode" radar products update. So not only are we not observing a lot of the features that are critically important for tornadoes, any time we DO measure them, they tend to change before we can do much about it. A lot of stuff happens that is undetectable by your eyes and your senses that impacts whether a supercell produces a low-level meso and whether a low-level meso becomes tornadic.

When you're in an environment that is not overwhemlingly obviously supportive of a widespread tornado outbreak, the luck factor becomes so much more important to determining success. Don't get too down on yourself about what happened. I have done the exact same thing multiple times (I think even this year I did it still). It probably has little-to-nothing to do with your meteorological knowledge or analysis skills and way more to do with happenstance that led you to the point where you were. Life is chaotic - even a small change can cause a dramatically different result. You can't control that.

 

[JamesCaruso]

Chasing is a frustrating avocation for the reasons illustrated above, probably bordering on psychologically unhealthy for a Type A guy like me ;-) Among many, many failures over the past 20 years, one particularly troubling back-to-back pair of days illustrating the "can't catch a break" (and can't always apply lessons-learned) scenario was in 2013. I went after the first storm to go up in my target area, followed it north too long and too far, and missed Rozel (which was right in my same general target area). So the next day, hanging out somewhere near Enid IIRC, in a target area validated by an SPC MSD, I decide to stay put and NOT to go after the first storm that formed further south along the dryline, and end up missing Shawnee.

So some days you choose A but the answer was B instead. So the next day you choose B but should have chosen A. It's absolutely maddening!

But chasing is like baseball, where a 300 average is pretty darn good (not saying that's the actual percentage, but you get the idea).

What's frustrating is that it always seems like other chasers are doing better. But that's only because you take any storm and *someone* is going to be on it. But although it may sometimes seem like it, it's not the same person every time. And you don't always hear about every chaser's unsuccessful chases. In the social media age, we all show our highlight reels, not our bloopers. We tend to forget that even the chasers that seem the best are only batting 300. Like Jeff said, a lot of it is luck, and the chasers that seem the most successful are improving their odds just by being out there a lot, whereas others that may just take a two-week chase vacation and get 7-10 chase days a year at best (or, this year, two) are just never going to be in the game often enough to capitalize on luck. I'm not saying skill isn't required, or that there's no way to be "good" at chasing, or that some chasers aren't more talented than others, but you know what I'm saying. It's great to try to learn from mistakes and improve, but sometimes we'll just never know exactly what happened. I've often wished their could be some published narrative of exactly what happened on a given day - sort of like an SPC outlook or MSD, or a local AFD, written after the fact. Still waiting for that product ;-)


Sent from my iPad using Tapatalk

 

[Shane Adams]

From reading your post, it sounds like you had constant data in the car, were constantly checking it, and made most (if not all of) your decisions off it. I've been in situations many times where you're caught between two storms and both are within striking distance, so I understand the frustration. You asked what I would've done (differently)? As soon as I got on the first storm, I would've put the detailed data away and just focused on the sky, with only occasional glimpses at basic radar to check for other storms in the area (possibly out of good visual range due to haze/precip/etc). I try to get a generalized understanding of the environment region-wide in my target area, so I can just focus on whatever storm pops closest to me, instead of trying to pick and choose minute to minute what storm is best based on trying to pinpoint exact parameters on a microscale level.

The worst thing I can personally do out there to screw up a chase is to constantly check data and react to parameters on a screen. The best parameters in the data don't always correspond to the best storm of the day, and especially when you're already there and on it, I'd take a close visual over looking for data. Seems to me the biggest mistake you made was changing your mind the second time; sounds like if you had commited to the west storm after you initially left the first one, you probably would've had a good shot at the tornadoes.

Another thing I would've done differently is, I would've stopped as soon as I had a visual on the tornadoes. Distant tornadoes pose a risk of not lasting until you get as close as you want, which leaves you with nothing. I prefer to stop, zoom, and document the tornado as opposed to wasting its lifecycle trying to force a closeup.

 

[Dan Robinson]

Many excellent points in this thread. My thoughts:

- Storms often cycle up/down in intensity and/or visual appearance a couple of times before they produce. Don't leave a storm in a good environment unless it's showing signs of completely drying up (vanishing VIL core or shriveling updraft diameter) or if another storm to the south its about to choke yours off.

- Stick with a storm on an outflow boundary or warm frontal zone rather than one in an open warm sector.

- A warm front/stationary frontal zone might be on the "fringes" of better parameters (helicity, STP, etc), but storms have a much better chance at producing in these areas than in the "bullseye" of the parameter. If this is the case, don't count a storm out just because it looks like it is exiting a parameter bullseye. That might just mean it's encountering the frontal zone, where the chances of producing a tornado are the best. If you know where the frontal zone/outflow boundary is, you can "put the data away" as Shane says and focus on the visual.

At some point you have to just commit to a storm based on the best information you have, and you aren't always going to pick the right one. Choosing the wrong one happens to all of us.

 

[Jeff House]

Strongly agree to use visual structure cues. We thought about bailing on Rozel when the cell to the northwest was producing. However we visually inspected improving mid-levels. Base was still high, but the mid-levels were getting close to classic. Then the low-levels improved several minutes before radar did so. Remember the scans are delayed. Thankfully we stuck with it and saw the complete Rozel show.

Radar VIS and surface usually is the only data we dissect by mid-afternoon. Target decisions should be made at lunch. If the HRRR shows a good cell that you believe is on a boundary intersection, it probably offers the best odds. Go by the boundary intersection, not the exact HRRR placement. Just track that boundary intersection..

Finally the whipsaw question. What strategy today? I mostly favor a systems based chase strategy. Boundary intersections are my method. Two days before DDC we missed the TX Panhandle tornado by not following our system. 14Z HRRR had a great cell on the boundary intersection. Later runs lost it, but the boundary intersection remained (barely) discernible on VIS and surface.

Next day we did see Woodward because we stuck to our system, while baking in the sun until 23Z. Then the system paid off bigly with DDC. Of course just like any systems based method, there are exceptions. Still, the system makes a good reality check in the heat of battle.

 

[JeremyS]

Chasing is a frustrating avocation for the reasons illustrated above, probably bordering on psychologically unhealthy for a Type A guy like me ;-) Among many, many failures over the past 20 years, one particularly troubling back-to-back pair of days illustrating the "can't catch a break" (and can't always apply lessons-learned) scenario was in 2013. I went after the first storm to go up in my target area, followed it north too long and too far, and missed Rozel (which was right in my same general target area). So the next day, hanging out somewhere near Enid IIRC, in a target area validated by an SPC MSD, I decide to stay put and NOT to go after the first storm that formed further south along the dryline, and end up missing Shawnee.

So some days you choose A but the answer was B instead. So the next day you choose B but should have chosen A. It's absolutely maddening!

But chasing is like baseball, where a 300 average is pretty darn good (not saying that's the actual percentage, but you get the idea).

What's frustrating is that it always seems like other chasers are doing better. But that's only because you take any storm and *someone* is going to be on it. But although it may sometimes seem like it, it's not the same person every time. And you don't always hear about every chaser's unsuccessful chases. In the social media age, we all show our highlight reels, not our bloopers. We tend to forget that even the chasers that seem the best are only batting 300. Like Jeff said, a lot of it is luck, and the chasers that seem the most successful are improving their odds just by being out there a lot, whereas others that may just take a two-week chase vacation and get 7-10 chase days a year at best (or, this year, two) are just never going to be in the game often enough to capitalize on luck. I'm not saying skill isn't required, or that there's no way to be "good" at chasing, or that some chasers aren't more talented than others, but you know what I'm saying. It's great to try to learn from mistakes and improve, but sometimes we'll just never know exactly what happened. I've often wished their could be some published narrative of exactly what happened on a given day - sort of like an SPC outlook or MSD, or a local AFD, written after the fact. Still waiting for that product ;-)


Sent from my iPad using Tapatalk

I know what you are saying James. It does always seem like everyone else is out there bagging tornadoes, while you're missing out. I hate to say this, but I do take solace in seeing people's misses on chase days. Hope that doesn't sound too horrible, but I guess it reminds me even the "good" chasers that I follow will miss tornadoes!
I make it out somewhere between 5-15x per year and at the end of the season feel like if I've batted .300, it's pretty hard to really complain.
So far this year I'm 3/8, so again nothing to whine about(too much)! I went out yesterday and missed the big tornado in NoDak, but was on an absolutely amazingly structured mothership tornado warned cell that formed just before sunset near Brookings, SD. Amazing lightning show too!

 

[JeremyS]

Strongly agree to use visual structure cues. We thought about bailing on Rozel when the cell to the northwest was producing. However we visually inspected improving mid-levels. Base was still high, but the mid-levels were getting close to classic. Then the low-levels improved several minutes before radar did so. Remember the scans are delayed. Thankfully we stuck with it and saw the complete Rozel show.

Radar VIS and surface usually is the only data we dissect by mid-afternoon. Target decisions should be made at lunch. If the HRRR shows a good cell that you believe is on a boundary intersection, it probably offers the best odds. Go by the boundary intersection, not the exact HRRR placement. Just track that boundary intersection..

Finally the whipsaw question. What strategy today? I mostly favor a systems based chase strategy. Boundary intersections are my method. Two days before DDC we missed the TX Panhandle tornado by not following our system. 14Z HRRR had a great cell on the boundary intersection. Later runs lost it, but the boundary intersection remained (barely) discernible on VIS and surface.

Next day we did see Woodward because we stuck to our system, while baking in the sun until 23Z. Then the system paid off bigly with DDC. Of course just like any systems based method, there are exceptions. Still, the system makes a good reality check in the heat of battle.

The thing I really feel like I messed up on was the visual clues on this storm. Even though it was tornado warned, it just never had that look. No obvious areas of rotation and the wall clouds were brief and usually not rotating. Of course that can change rapidly and I guess that's why we stuck with it as long as we did. One thing I forgot to mention, that I think Dan did, was new convection to the south choking off the storm. That actually happened to the Sioux City storm and that was about the time the warning was initially dropped and when we did leave. However, the storm did reorganize as the new convection was absorbed by the parent cell and then that is when it was warned again and had a radar confirmed tornado with it.
The HRRR did show the Yankton storm the northwest on several runs, which in retrospect should have been another reason to maybe bail on the Sioux City storm.
I love reading all the responses and individual thought processes though. Thanks again for everyone's input!

 

[Ethan Lang]

My chase partner knows very well that I always have trouble deciding on which cell to commit to. This past year I had bad luck with this. I agree that it is very hard to leave a tornado warned storm but once I look at all of the meteorological data and get a better understanding of what is happening I then just go with my gut and most of the time it works out.

 

[James Wilson]

Am I the only one that just stays on the cell with more lightning? LOL!

 

[Taylor Wright]

In my opinion, always go with what looks best visually, not on radar. As others have a mentioned if you have a beefy rotating wall cloud, you should never leave it. June 22, 2016 comes to mind, where you had a sloppy, HP looking cell on radar (the storm we were on) then a classic, stationary storm with a beautiful hook echo to our south. We were flying to get to the southern storm, but stopped when we noticed our storm, which looked like a lima bean on radar, had a beautiful wall cloud and RFD cut. Because of this, we stayed with the northern storm. Guess which one produced the first tornadoes? We saw two tornadoes before the storm became entirely rain wrapped and almost killed us with CGs. Oftentimes radar can be very deceiving. Unless your storm is completely shriveling up or becoming completely linear, or there's a prominent tornado producer within 20 miles and yours is struggling, you shouldn't leave it.