Some Thoughts about Backing Surface Winds

[Bob Hartig]

I just finished writing a post in my blog about why veered surface winds don't necessarily put the kibosh on tornado potential for a synoptic setup. It's a matter not so much of wind direction as of what the wind is doing. This is a subject I've pondered for a while, and while I'm no whiz kid at forecasting, I've observed enough that at this point I have some thoughts to share, along with a few maps to serve as examples. I welcome addenda as well as critique.

 

[Jeff Duda]

Bob, this is a very interesting topic and one worth discussing. In your blog post, you certainly nailed certain ideas about southeasterly surface winds favoring higher storm-relative helicity and providing moisture advection. A good question to help your understanding would be to ask why southeasterly winds do these things. The answers to these questions partially involves location/region.

In the Great Plains and Midwest, southeasterly surface winds promote moisture advection because southeasterly is the direction of the most accessible source of moisture for the Great Plains and Midwest - the Gulf of Mexico. Although the Atlantic Ocean is almost as close, you don't get moisture advection from easterly winds usually since the air coming from there must top the Appalachians, which usually results in drying due to topographical lift over the mountains.

Since synoptic scale storm systems in the Great Plains and Midwest tend to come from upper level disturbances moving eastward across the CONUS, and since QG theory indicates that DPVA results in UVV, this indicates that mid-upper level flow in the region is mostly out of the SSW, SW, or WSW. Thus, having southeasterly surface winds lengthens that hodograph even more, thus providing more shear.

I can think of two more good reasons I like southeasterly surface winds for chasing. 1) The major surface boundaries that force thunderstorms to form usually have an orientation such that southeasterly surface winds result in stronger surface convergence, which results in a more likely chance of convection forming (i.e., NOT cap busting) despite the presence of resistance, or CIN, to the forming of convection. 2) Since mid-upper level winds usually have a strong zonal component (i.e., west-east) that drives storm motion, having southeast winds at the low levels (not just the surface) will help reduce the forward move speed of storms when you average the wind vectors in the cloud-bearing layer (which is one way of computing storm motion vectors). Thus southeasterly winds are more likely to result in slower storm speeds, which is always nice for chasing.

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

The above discussion highlights why southeasterly surface winds are so desirable...for the Midwest and Great Plains. Southwest winds can still result in tornado days if there is enough shear above. In cases like that you would hope that mid-upper level flow is more westerly or northwesterly (e.g., northwest flow events). Unfortunately, having southwest surface winds in this region frequently results in dry air advection, but on short time scales this won't be enough to totally kill of instability or increase capping.

Southwest surface winds can be much more favorable for severe weather than southeasterly winds in other regions, such as the Ohio Valley region or the southeast US. While mid-upper level flow is generally also southwesterly, it can be more veered due to the evolution of many shortwave troughs that cause severe weather one day over the Plains, then again the next day to the east. Southwesterly surface winds will promote moisture advection just fine far enough east, because the Gulf of Mexico is still at least partially at the source of southwest winds. Also, southeast winds in the Ohio Valley region would indicate winds coming over the Appalachians which would indicate drying, so those aren't as good for moisture advection. In general, southwest winds are generally still good enough to get sufficient shear for tornadoes in the Ohio valley and in the southeast US.

 

[Bob Hartig]

To be clear, Jeff, I'm an ardent admirer of southeasterly surface winds. My point was that the absence of them isn't always a deal breaker, particularly east of the Mississippi where southwest winds don't automatically equate to dry desert air, and where GOM moisture can fetch from the south and even southwest rather than the southeast. You've hit upon this point in your last paragraph. Things differ from region to region.

I'd much rather have nicely backed surface winds any day of the week. But I've concluded that in the Great Lakes and much of Dixie Alley, it pays not to write off a system too quickly just because surface winds are veered. The examples I've provided in the post demonstrate this pretty well. Note that the storms were in Illinois and Michigan, not Oklahoma or Texas.

 

[Chris Vagasky]

I've much rather have nicely backed surface winds any day of the week. But I've concluded that in the Great Lakes and much of Dixie Alley, it pays not to write off a system too quickly just because surface winds are veered. The examples I've provided in the post demonstrate this pretty well. Note that the storms were in Illinois and Michigan, not Oklahoma or Texas.

I've posted some "historic" Michigan tornadoes in my blog lately, and many of them have occurred with S or SW surface winds:

West Bloomfield F4

for example.

 

[Scott Weberpal]

Veered surface flow was a major factor in the lack of tornado production yesterday. Yes, southwest surface winds can get it done, but typically northwest flow aloft is needed to have the appreciable deep-layer shear for deep/strong mesocyclones. I even made a comment on facebook yesterday that I didn't think there would be much in the way of tornado activity in Alabama or Mississippi because the surface flow was A) veered (in regard to flow aloft) creating straight hodographs, and B) the surface winds were quite weak.

I can think of a handful of strong tornado/oubreak events in the upper-midwest and great lakes region that with northwest flow aloft, were able to produce with southwest surface flow.

June 23, 2004 (20+ tornadoes in WI) http://www.spc.noaa.gov/exper/archive/events/040623/index.html
July 13, 2004 (Roanoke, IL F4) http://www.spc.noaa.gov/exper/archive/events/040713/index.html
July 18, 1998 (Oakfield, WI F5) http://cimss.ssec.wisc.edu/oakfield/cs1.htm

 

[Brian Stertz]

The 4/2/06 set up was like that as well...sw surface winds and 260 degree 500mb winds...plus a nice theta poke. This yielded a very photogenic supercell and long tracked tornado in NE Arkansas into the MO Bootheel. This one comes to my mind when breaking conventional though on sfc winds.

 

[Brandon Sullivan]

Looks like some of the major points were covered.. (Convergence along a boundary was my main concern, except for the obvious low level SRH factor)

Also, a sustained and powerful supercell can also have local impacts on its environment.. I've seen a supercell back it's own winds..I'm not saying go chase every SW flow event, but if you get a powerful/sustained supercell, especially one that can root along a boundary, then sw winds do not totally hamper things.. Additionally, mesoscale influences that would cause a storm to slow down has dramatic influences SRH values, since its obviously a Storm Relative motion calculation.

 

[Skip Talbot]

Great post Bob and others. I agree that the cardinal direction of the surface winds matters little when compared to what the upper level winds are doing, especially east of the plains. One thing I always look for when picking out chase targets is not so much, how backed the surface winds are, but what the surface winds are doing over time. Many setups that look like they are going to be great start out with strong ESE surface winds, but as time goes on they veer and go slack. You wind up fighting a losing battle trying to get a storm to mature and produce before your directional shear is cut down to nothing. Another common problem I've seen is that the winds back nicely deep in the warm sector. That's not where storms wind up firing, however. They'll often go up right on the dryline/cold front, or west of the moisture axis where surface heating and instability are greater, but the winds are much more veered. So even if you've got 250 1km SRH and 20 knot SE surface winds 50-100 miles east of the dryline (which looks great on model plots), if your storm winds up firing on the western edge of the warm sector, right on the boundary, they might actually be experiencing SW winds and a fraction of that helicity. I always try to pick a target where the surface winds are becoming more backed with time: deepening surface low, storms approaching a warm front, storms moving off the cold front/dryline, etc. It seems like these setups wind up being the big tornado producers, while ones trending in the opposite direction bust or are mostly non-events. Assuming you've got your deep, rich moisture in place, its not so much the specific direction of the winds for a given setup/environment, for where storms initiate, or even for where the storm is at the moment. Its more about how those winds are changing over time and whether the wind barbs on your surface chart are turning counter clockwise in your favor or against clockwise against it.

Some recent examples I can think of that probably played a major a role in whether or not there were tornadoes would be:
Last Tuesday. Storms went up along the instability axis in south central Nebraska and north central Kansas. Winds were terribly veered out that way (not to mention there was a big dewpoint depression). Once the storms moved into an environment of backed winds and richer moisture northwest of Omaha wrapping around to southwest Iowa, they really took off, organized, and became tornado producers. A great example of surface winds backing with time, relative to the storm, as the storms moved into a better environment. Had that dryline surged east with the storms, I think we would have been chasing disorganized, undercut, high based junk all day.

June 17, 2010. I know most people were expecting storms to go up right on that cold front/dryline. I know I was, and so was SPC judging by their probabilistic graphics that hugged the boundary. Winds were much more veered out west by that boundary and as a result many more expecting more of a linear event. When storms fired on the prefrontal trough, well ahead of the cold front/dryline, they found themselves in a much more favorable environment right off the bat with richer moisture and strongly backed surface winds. They were also much closer to the warm front draped northwest to southeast across Minnesota and a significant tornado outbreak resulted as the storms moved northeast into increasingly backed winds and greater directional shear.

 

[mikegeukes]

A Synoptic Climatology of Northwest-Flow Severe Weather Outbreaks. Part II: Meteorological Parameters and Synoptic Patterns
Robert H. Johns
http://journals.ametsoc.org/doi/pdf/10.1175/1520-0493%281982%29110%3C1653%3AASCONF%3E2.0.CO%3B2

A Synoptic Climatology of Northwest-Flow Severe Weather Outbreaks. Part II: Meteorological Parameters and Synoptic Patterns
Robert H. Johns
http://journals.ametsoc.org/doi/pdf/10.1175/1520-0493(1984)112<0449:ASCONF>2.0.CO;2

Mike

 

[Bob Hartig]

Skip, you've made some really great observations! Thank you. And thanks to all who have contributed to this thread, or who may yet choose to contribute. I love this kind of interaction, and I invariably wind up learning from it!

 

[Rob H]

NW flow events are for sure something where you want to see veered winds, but just like with supercell type, there are shades of grey in between. One of my favorite chases ever was a MDT risk on 6/25/10 in MN. I haven't seen many zonal outbreaks like this:

Hardly any chasers out, insane lightning, and some of my best wall cloud pictures ever. Everything was completely rain wrapped with these HP beasts that were just scraping the ground, but it was a great day.

 

[Lanny Dean]

...Also, a sustained and powerful supercell can also have local impacts on its environment.. I've seen a supercell back it's own winds..I'm not saying go chase every SW flow event, but if you get a powerful/sustained supercell, especially one that can root along a boundary, then sw winds do not totally hamper things.. Additionally, mesoscale influences that would cause a storm to slow down has dramatic influences SRH values, since its obviously a Storm Relative motion calculation.

Edit.....looks like Skip already mentioned this sorry.

Of course we are talking about a typical noted patern but I totally agree with Brandon. There have been many, many tornadic events that I can remember and actually witnessed first hand with SW flow, yet the supercell still produced because it helped to create it's own enviorment per say. It is my thought that with increased strength, storm speed and other mesoscale details can certainly "cause" the suprecell to have backed it's winds thus increasing inflow parcels.
As we all know strength in inflow has a direct relationship with strength of the updraft/total storm.

I have never been one to downplay an event or possible event simply due to SW winds. Most noteably with NW flow alfot. Of course I am not saying that I would rather have SW winds at H85 as opposed to due south, I think we all would rather see south or even a southeasterly component but I won't count the SW flow out either. I think my one rule for questioning wheather I think a SW event may produce or not is the placement of existing boundries (someone already covered it) but I will tend to focus on the placement of that boundry in realtionship to any sort of backing winds.

However...this has also bit me a few times as well. Certainly not proud of the fact but this is why we missed the big Minnesota outbreak last year...I simply did not think that tornadoes would happen with the shape of the winds. If I remember correctly winds veered really bad over the suspected area. They did increase and even back well to the east of the boundry but I didn't think things would fire much less that far east to start.

 

[Paul Knightley]

It may be that, quite often, a veering in the flow typically means that surface convergence is tending to decrease. In addition, a veering flow in the absence of a frontal passage is usually an indicator that an upper-level disturbance has moved away, thus decreasing the chance of storms breaking out.
Another reason has already been mentioned: the fact that with many systems moving through, the 700-500 hPa flow is typically (but, of course, not always!) SW'erly - thus, a SW surface flow means that shear is lower than with a SE flow, and more importantly, hodographs are not as curved in low levels.