Tornadoes and Outflow Boundaries

[Justin Wiley]

Ok guys, I’ve been following severe weather for a while in my short life. I’m definitely not a pro so I apologize in advance if this seems “stupidâ€.

Is there any direct correlation with tornado intensity or tornado frequency with outflow boundaries? I noticed discussion in the historic chase event thread of 3/13/90 and Hesston that the storm was aided immensely by an outflow boundary. I think the Andover tornado spun up on one too maybe? How often do outflow boundaries lead into the formation of tornadoes? Is there a better a chance a violent tornado can form on one of those as opposed to some other kind of a boundary? Is there any difference with structure or formation when a tornadic supercell forms on an outflow like Andover as opposed to say like the Red Rock storm (which I’m assuming wasn’t formed on one?) Can they really make or break a big tornado event?? And finally, do supercells intentionally form on the outflow boundaries in response to them or can they just by sheer coincidence bump into one of them and then enhance tornado output/intensity?

Sorry for that barrage of rambling questions. I’m just really curious about the play on tornadoes and outflows. Probably simple and common stuff to you guys but it’s something I just never really looked into too much.

 

[Chris Allington]

I dont know that you could find much correlation between tornado intensity and outflow boundaries...

That being said, the increased low level shear along a boundary such as thunderstorm outflow or a warm front can definately help focus low level rotation and allow a tornado to form. Outflow boundaries can also help increase surface convergence to allow convection to break the cap and provide a focus for thunderstorm development.

 

[William Reid]

Chris gave a quick and concise answer while I was typing up the extended version:

*****************

I guess I'll take a stab at this one. I may stray off into the obvious, so bear with me...

Ok guys, I’ve been following severe weather >for a while in my short life. I’m definitely not a pro so I apologize in >advance if this seems “stupidâ€.

>Is there any direct correlation with tornado intensity or tornado >frequency with outflow boundaries? I noticed discussion in the historic >chase event thread of 3/13/90 and Hesston that the storm was aided >immensely by an outflow boundary. I think the Andover tornado spun up >on one too maybe? How often do outflow boundaries lead into the >formation of tornadoes?

-----------------

I can't really give you a number or a percentage, and I'm not familiar with any outflow boundary help on the Hesston and Andover days. Perhaps the best way to approach this is to think of (most) tornadoes as rather small and fragile by-products of its "mother" cell or storm system. Conditions in the lowest levels of the atmosphere (say, from the surface to a couple of thousand feet up) need to be VERY favorable for tornado formation ---- even for tornado formation beneath supercells. A balancing act of sorts among the temperature, moisture, and wind fields in and around a storm is necessary in order for a tornado to form and to persist.

Some outflow boundaries --- some new ones and some old ones --- tend to be locales where the low-level "meteorology" can greatly enhance tornado development.

So the best answer to your question so far is --- "it depends". It depends on the outflow boundary and whether it is a good boundary or a poor boundary for tornado-making.

-------------------

> Is there a better a chance a violent tornado can form on one of those as >opposed to some other kind of a boundary?

------------------

The answer to this one is a shaky "yes." From my experience, on the big days and outbreak days with numerous strong-to-violent tornadoes, outflow boundaries are of little or no consequence. If the atmosphere wants to make violent tornadoes, then it doesn't need a boundary. Outflow boundaries can be well-suited for tornado-making if they are oriented somewhat parallel with the mean flow aloft, allowing storms to "latch on" and ride the boundary. The same is true on occasion for warm fronts. Cold fronts and dry lines are "boundaries" which are often good for storm development, but tornadic storms tend to move off of these types of boundaries, so that it is not the boundary itself which aids in tornado formation. So, chasers playing the dry line or cold front don't sit right on the boundary and hope for tornadoes, whereas a chaser could position himself or herself on the warm front or outflow boundary to improve tornado-intercepting odds.

Outflow boundaries tend to be more important on the "subtle" slight-risk days when the overall synoptics are favorable for supercells (if storms can form), but the overall low-level regime is weak or generally unfavorable for tornadoes. However, an outflow boudary in this situation could well be the difference-maker in locally providing a very favorable environment for tornadoes. In addition, the boundary may also be where convective development is most likely, given the local low-level convergence. The May 24, 2008 event in northern Oklahoma (pig farm tornado) comes to mind. So, outflow boundaries can aid in storm development AND tornado formation.

---------------

>Is there any difference with structure or formation when a tornadic >supercell forms on an outflow like Andover as opposed to say like the Red >Rock storm (which I’m assuming wasn’t formed on one?)

-----------------

Again, I don't know the role that outflow boundaries played on this day (my guess is that they were of little consequence, but I could be wrong). Assuming that the storm that formed ON the outflow boundary is able to REMAIN on the outflow boundary for a while, then it would be favored for tornadic production as opposed to a cell that is not associated with an outflow boundary.

An outflow boundary can locally enhance inflow with its stronger backed winds, and can result in locally lower cloud bases if one side of the boundary is relatively cool and moist. Outflow boundaries tend to be places where local low-level vorticity is enhanced, and tornados like that. So, you could have two similar non-tornadic supercells plugging away and working on similar air...and if one of them comes upon an old outflow boundary, with vorticity-rich air and higher relative humidities, then that one should be favored for tornado production.

--------------------------

>Can they really make or break a big tornado event??

--------------------------

I guess it depends on your definition of "big tornado event". As a chaser, you can only be on one storm at a time, so it doesn't matter if there are 100 tornadoes that day or just one ----- all that matters is what your storm does. But, yes, there likely have been many days with numerous tornadoes which would have been tornado-less had outflow boundaries been absent.

----------------------------

> And finally, do supercells intentionally form on the outflow boundaries in >response to them or can they just by sheer coincidence bump into one of >them and then enhance tornado output/intensity?

-----------------------------

Both occur. Outflow boundaries can be excellent foci for convective development. On days with strong caps, it may only be along the old outflow boundaries where development is possible. And, once a supercell gets cranking after forming on the outflow boundary, it immediately becomes a prime tornado candidate as it is sitting on the boundary.

There are a bunch of ways to get supercells and outflow boundary interaction. Some "new" outflow boundaries might not be favorable at all if they are cool and rushing away from their parent storms and undercutting others. The Last Chance, CO, tornado that I saw in July 1993 was from a storm that formed on outflow from nearby storm that had been drifting along for hours. The initial storm formed on the dry line on the Colorado Plains and moved slowly east for several hours, and had problems making tornadoes. Its cool and moist outflow bumped into the nice moist southeasterlies to its southwest, and a new storm went up as sunset approached. The new storm rode the outflow boundary from the old storm and produced a couple of large tornadoes. The two storms were perhaps 20-30 miles apart.

The more typical way to get a "good" outflow boundary is from morning convection that spills out cool and moist air. Eventually this air stalls out as it comes up against the moist shoutherlies or southeasterlies to its south. Ideally, the boundary will remain in place through the afternoon, the low clouds associated with the cool air burn off, and the boundary bakes for a fews hours under the high sun. The convergence along the boundary might aid in storm development, and the backed winds (and the higher RH) on the cool side of the boundary would enhance low-level helicity values beneath the updraft.

---------------------------

>Sorry for that barrage of rambling questions. I’m just really curious about >the play on tornadoes and outflows. Probably simple and common stuff to >you guys but it’s something I just never really looked into too >much.

---------------------------

I hope that my rambling helped some. It is always a good idea, especially on the "marginal" days, to examine the satellite, radar, and surface data to see if any outflow boundaries are in play. Some are obvious, and some are barely detectable.


Bill Reid

 

[Mikey Gribble]

Chris basically answered it. Doswell has a lot of this stuff in his stormscale book, which is awesome btw. It's like 30 years old, but it's still a great read.
Why do storms produce tornaodes so frequently along outflow boundaries? Think of it this way. There is already preexisting vorticity along that boundary. Normally as storm has to get going for a couple hours and get organized before it tornadoes. Part of the reason is because it needs to begin to generate its own vorticity along the forward flank downdraft. When a storm fires along or hits a boundary they can often go tornadic very fast because they have a great source of vorticity available to feed off of (ingest through the updraft). It's been a year or two since I read that book so I may have butchered what Doswell said, but that is the basic idea.

Like Chris said there is often enhanced convergence along outflow boundaries so storms are likely to fire there. Storms almost always develop along boundaries (whether you can recognize the boundary or not). Think about it. You have a wide area like say half a state. Say this entire area has the exact same temperature profile throughout the entire depth of the atmosphere. Well in the boundary layer, which is the air right above the surface, there may be a front, a dryline, and OFB, a differential heating boundary from earlier in the day or even an OFB from the day before that you can't even detect on the surface network any more. The storms are going to form in those places first. There are reasons why storms develop in certain places and not at others.

I know Chris got it right, but hopefully that helps a little more. Get Tim Vasquez's forecasting handbook if you haven't already. Before you read that though go get an intro to meteorology book like meteorology today. You don't even have to read all the chapters in the intro book. It takes a lot of time and hard work to learn this stuff and get decent at forecasting. I've been at it for 8 years and I still struggle constantly. Learning how to forecast is a never ending process. As soon as you get cocky or think you know something mother nature force feeds you a big piece of humble pie (or at least she does to me).

 

[Jon Miller]

What do you look for on sfc data that would indicate a possible OFB ?

Thanks
Jon Miller

 

[Andrew Ryan]

http://www.stormtrack.org/forum/archive/index.php/t-3511.html

I think they are usually the easier to identify them on radar rather than sfc data because the can occur in between stations. But to sum up what was said in the link, you are looking for strong temp and dewpoint gradients and changes in wind directions that aren't associated with cold fronts, drylines, etc.

 

[Greg Blumberg]

Andrew is correct. It is better to find them out on radar, just makes sure to check other observations (satellite and surface data) to reinforce your knowledge. These other obs can make up for the limitations of radar. If there's an outflow boundary far from the radar, the lowest scan might not see it. Also take into consideration that all sorts of boundaries (cold fronts) can show up on radar.

But you really want to know about seeing them through surface data, just look for a contrast of cool dry air and warm moist, sorta like a mini-cold front.

 

[Mike Johnston]

W/ outflow boundaries, as w/ other subtle atmospheric features (gravity waves, bores, meso-lows, etc.), the challenge is detection in real time:

- As noted above, radar is often your first clue. Look for a very thin, curvilinear feature in the hours preceding peak heating.

- Be sure to put the radar in motion (loop it) to verify some consistency of the feature and its movement. It's direction of movement will most likely have a "life of it's own", independent of obvious synoptic scale boundaries.

- The actual radar returns may very well be reflecting subtle features such as flying insects congregating along the boundary, rather than actual precip.

- By using a little rough extrapolation, you may be able to estimate the location of the boundary at peak heating, take a look at what the RUC model shows w/ regard to other important ingredients (CAPE, shear, etc.) at that time and location, to help in your targeting.

- Although not to be ignored at other times of the year, these features (along w/ differential heating boundaries, etc.) seem to take on a more prominent role in late Spring (ie. June) when there is usually plenty of boundary layer moisture around, but locating the spark (source of lift) is the key.

 

[Greg McLaughlin]

As Justin Teague puts it, "Outflow boundaries are tornado racetracks."

Here are some dates with tornadoes associated with outlfow boundaries. These are just off the top of my head.

3/13/90: Hesston/Gosell, KS F5

4/26/91: Andover, KS F5 Red Rock, OK F4 (I believe both supercells developed near remnant outflow boundary from morning convection and became tornadic after moving across the boundary.

5/27/97: Jerrell, TX F5 (storm cycled mesos to the southwest as it backbuilt along the boundary)

6/23/98: Columbus, NE F4 (infamous "Crazy Farmer" video)

6/01/99: Checotah, OK F3* (storm backbuilt along boundary to southwest producing at least 7 tornadoes)

5/01/03: Muskogee, OK*

6/13/07: Orienta, OK* (storm produced several tornadoes as it drifted along boundary. 500mb winds were only 20kts!)

5/24/08: Lacy/Hennessey, OK* (see Storms of 2008!)

*events witnessed by Justin Teague and/or myself



These are just a few quick dates that I can think of off the top of my head. Almost all of these dates featured cyclic tornadic supercells anchored on a boundary.

From my experience you want to chase a storm that roots along a boundary and deviates from the mid-level flow. Obviously this can be dependant on how the boundary is oriented to the flow. I will never pass up an outflow boundary if I believe storms will initiate along the boundary and anchor to it.

A supercell along an outflow boundary can get some of that helicity ;0

 

[Justin Wiley]

Thanks guys. I think you pretty much answered all of my questions. I've always focused on the 'main' boundary setups (i.e. cold fronts, warm fronts, dry lines) and never really thought too much about the outflow boundary setups.

A couple quick follow-ups:
If a supercell forms along the dryline and then moves off of it, and there's not a warm front or outflow boundary present ahead or along it, does that mean the supercell is likely to weaken or be less likely to produce a tornado?

and..
Is it safe to say that an isolated area of early morning thunderstorm activity is somewhat of a good thing in an area that is 'primed' for potential severe weather/tornado development later in the day because of the possibility of developing outflow boundaries? I've always thought that you wanted a severe weather area primed up and not filled up with any type of convection for fear of atmosphere disruption or stabilization.

 

[Justin Wiley]

As Justin Teague puts it, "Outflow boundaries are tornado racetracks."

Here are some dates with tornadoes associated with outlfow boundaries. These are just off the top of my head.

3/13/90: Hesston/Gosell, KS F5

4/26/91: Andover, KS F5 Red Rock, OK F4 (I believe both supercells developed near remnant outflow boundary from morning convection and became tornadic after moving across the boundary.

5/27/97: Jerrell, TX F5 (storm cycled mesos to the southwest as it backbuilt along the boundary)

6/23/98: Columbus, NE F4 (infamous "Crazy Farmer" video)

6/01/99: Checotah, OK F3* (storm backbuilt along boundary to southwest producing at least 7 tornadoes)

5/01/03: Muskogee, OK*

6/13/07: Orienta, OK* (storm produced several tornadoes as it drifted along boundary. 500mb winds were only 20kts!)

5/24/08: Lacy/Hennessey, OK* (see Storms of 2008!)

*events witnessed by Justin Teague and/or myself



These are just a few quick dates that I can think of off the top of my head. Almost all of these dates featured cyclic tornadic supercells anchored on a boundary.

From my experience you want to chase a storm that roots along a boundary and deviates from the mid-level flow. Obviously this can be dependant on how the boundary is oriented to the flow. I will never pass up an outflow boundary if I believe storms will initiate along the boundary and anchor to it.

A supercell along an outflow boundary can get some of that helicity ;0

Thanks for that interesting piece of info. Lots of strong and violent tornadoes on that list. Obviously the outflow boundaries are virtually impossible to predict, but once there, are a very key ingredient/aid into the formation of supercells/tornadoes. I can't believe I've overlooked this type of stuff for so long.

 

[Dan Robinson]

As Mikey mentioned, a dryline storm that remains isolated will start creating its own boundary/vorticity after some time. The Mulvane day was a good example. OFBs often mean great chase days, as Greg said. Another one was the Hill City day 6/9/05. That storm initiated and later moved into an OFB, then went crazy.

Some of the best chase days I can remember were ones where we woke up under an MCS moving out from the previous day, then drove west or south a little to pop out into sunny skies. Followed a few hours later by showtime. Seems like an MCS that lasts all night is a sign that a lot of parameters that you want are in place (good low level jet, deep moisture, upper support etc). As long as gets out of there by noon A good squall line will also leave subsidence in its wake early on, which helps clear things out for the next event.

 

[Jeff Snyder]

Justin,

Good questions! The previous posters in this thread have made some good comments and provided good answers, but I'll provide a few comments as well. OFBs are something that you learn rather quickly (after a few "surprise" events, a la 5/24/08) never to ignore. Greg's partial list of noteable chase days should be evidence enough that any chaser should pay careful attn to OFBs, particularly those OFBs that are aligned along (or at least near) storm motion and that are associated with some destabilization on the cool side of the OFB.

Oftentimes, moisture improves east of a dryline, and it's common for stronger ~850 mb flow to reside 50-100 miles east of a dryline. Therefore, we often look to drylines as a source for convergence and storm initiation, and these storms may well move into an environment more favorable for tornadoes some distance east of the dryline. Away from the Plains, drylines are often less common, so we look to cold fronts to provide a source for initiation for discrete supercells. Of course, you can look for cold fronts for initiation in the Plains too, but, anecdotally, drylines tend to provide better probs for discrete convection compared to cold fronts.

Now, warm fronts can also serve as a source for initiation. However, from my experience, forcing along warm fronts tends to be more broad, making it more difficult to pin-point likely initiation locations. Depending upon the storm motion relative to the warm front (e.g. the orientation of the warm front to the storm motion), developing shows and storms may move quickly across the warm front and into the deeper cold air N of the warm front. Many times, however, chasers look for warm fronts as a location of enhanced tornadic supercell potential, as low-level shear tends to be enhanced along and to the cool side of the warm front. A storm may develop near a surface low on the dryline but move towards a warm front... As with OFBs, some storms can "lock onto" a warm front, which often provides increased tornado probability.

Re: early-day convection. For many decent and better setups, there will be a strong low-level jet through the mid-morning hours, and elevated showers and thunderstorms are often observed closer to the nose of the LLJ in a region of strong(er) moisture and warm air advection. This overnight and early-morning convection, as you noted, can "prime the pump" by providing a nice boundary for later convection can interact with. Of course, outflow from morning convection can also completely scour moisture from the (prior) target area and all but kill a potential chase day. It often depends upon the strength of the cold pool associated with the convection, and it is affected by how quickly convection clear during the morning (and early afternoon). I've seen setups ruined not only by a scouring OFB, but also by overnight and morning convection that never really clears the target area, which often inhibits heating and destabilization. Note that forecast models may not pick up on OFBs since OFBs are caused by convection that is often very difficult to forecast accurately 24-36+ hours ahead of time (as we all know!).

 

[Greg McLaughlin]

There have been some excellent posts on this thread. Another outflow boundary event I just discovered was the Picther supercell on 5/10/08. After looking over visible satellite loops I discovered an outflow boundary was produced by elevated convection in southwest MO that morning. This boundary drifted south and west and was oriented nw to se. The Pitcher storm had developed along the Oklahoma/Kansas border and was tracking in an easterly direction when it came under the influence of the boundary. At that point the storm anchored on the boundary and "right turned" to the southeast. NWS Tulsa did a presentation on this storm and showed velocity scans which indicated the storm had an intense low-level meso, but lacked a deep meso for quite some time.

It is my belief that the vorticity with the boundary was ingested into the updraft and quickly developed into an intense low-level meso which produced the long track violent tornado which hit Pitcher and moved into southwest MO.

 

[Brad Nelson]

Ouflow boundary and Hugo Tornado

The Hugo, MN tornado from last year is another good example. The storm rooted on an outflow boundary in the northern Twin Cities metro from convection that fired to the north of the area during the morning. The boundary never moved all day and, I believe, was the driving force behind the development of this tornado due to the enhanced low level shear the storm encountered on the boundary. Here is the surface plot from 21Z from that day showing the easterly surface flow that was in place due to the outflow boundary. I placed a red dot on the map showing the relative location of Hugo.