2011-05-11 FCST: KS/OK/TX

[Mike Johnston]

From a quick look at the recent 09 UTC SREF, my biggest worry for Wednesday is the availability of good helicity. Everything else looks good from the perspective of a nicely negative tilted trough, SBCAPE around 2000 J/kg, and a good position of the surface low. I'm seeing effective shear of 40 knots and issues with lower than wanted helicity values. Good helicity isn't depicted from the SREF's 0 - 3 km helicity probabilities. It's depicting low probabilities of the 0 - 3 helicity even reaching over 200 J/kg. What's even more interesting is that the 12 UTC NAM is depicting some subtropical jet interactions with our shortwave.

I'm tying the low helicity value to an uncertainty of the wind field forecast. What's odd is that the SREF (and 12 UTC NAM for that matter) has a mean 1000 mb Colorado low, but we're dealing with a negatively tilted trough! One would expect a much deeper low given the negative tilt and jet streak. I think that the helicity and shear values will be something that will be figured out the day of, rather than forecasted. Maybe I'm misdiagnosing a feature as I'm doing this analysis really quickly, but something doesn't smell right with the recent model solutions. I'm looking for different kinematics on Wednesday.

Yeah, the shear is a bit elusive, but if you take a look at Earl Barker's "Situational Severe" graphics (based on the NAM now 60 hours out), there is a definitive but small area of 55-60kt deep layer shear indicated over far NC Oklahoma and SC Kansas. This is also co-located w/ 0-1km helicity values of around 300 m2/s2 and the best overall wind crossovers (surface-850-500.) Also underneath what looks to be a good upper-level divergence. I know these bullseyes can easily go poof from one run to the next, but a good sign even so. I'm talking about an area roughly from Wakita, OK to Anthony, KS.

Link:
http://www.wxcaster.com/svr_awareness_model.php3?model=NAM212&fcsthr=60
(time-sensitive)

Interested to see if Mikey Gribble has anything to chime in on this, as he seems to have pretty good forecast instincts especially for areas in his own neck of the woods.

 

[Greg McLaughlin]

The 00z NAM and 12z NAM appear to have come up with some erroneous solutions regarding Wednesday's setup. The soundings from last night's run didn't look right. Echoing what Greg Blumberg said, because of the dynamic upper trough, it seems reasonable to assume wind fields and shear profiles will look better than what the NAM is currently forecasting. The NAM also kicks things off in Oklahoma really late (after 00z) which doesn't make much sense. I expect initiation to occur in the typical 20-23z time frame. The 12z NAM 500mb RH charts show a nice area of convection firing in central Oklahoma with a nice "mesoscale dryslot" so-to-speak on the southwest flank of this convective shield. This looks quite promising, however it behooves me not to look too much into such details at this point. Wednesday holds quite a bit of potential, and I suspect the environmental parameters may very well be "better" than what the NAM is showing. I do really like central Oklahoma at this point.

I do know there is a website which includes discussions on how the models initialize after each run. Does anyone have a link to this?

 

[Kelly Sugden]

I don't like how 09.12Z GFS/ECMWF are initiating convection early around 21Z while the 09.12Z NAM shows it as early as 18Z across Kansas. The vortmax is still over SE Colorado however significant CAA at 500 hPa is probably the result for the QPF. Could be an early show. Hopefully is just an odd model perturbation. Otherwise, central Kansas near the triple point is looking good attm.

 

[Trey Thee]

This setup looks decent overall. It has some issues, somewhat limited CAPE for example,

Helicities are good but nothing spectacular. There's an interesting area in N Central OK that some others have pointed out. It will be interesting to see if this little piece of data sticks around with future model runs. Central KS into S NE look pretty solid as well. Although there are also some issues there that still need to come together. I don't see anything that just screams at me right now. But it looks like a decent setup capable of a few tors in OK.

 

[Logan Karsten]

With the 00Z NAM coming in, I am still seeing big shear/helicity lingering issues in KS near the triple point. Although there will be some decent flow at 500 mb at 00Z, the overall flow with the trough actually decreases from 00Z to 06Z, which worries me. In addition, the flow at 700 mb never exceeds 30-35 kts, same with 850 mb. While I see supercell plays with shear perpendicular to the dryline, I am beginning to question exactly how many tornadoes a setup like this is going to produce. I feel like the window of opportunity will be fairly narrow near the triple point, as opposed to a large swath along the dryline. This was my biggest fear going into this week regarding the overall pattern. We have a ton of moisture pooling in the plains, but by the time the main energy finally ejects out into the plains, the upper level flow associated with the trough is a shadow of what it was a few days ago. It also doesn't help that by 00Z Thursday, the 500 mb trough is beginning to cutoff.

 

[Greg Blumberg]

I will say again what I said before. Something does NOT smell right with the model solutions we're seeing.

I've just taken a look at the 00 UTC NAM and the 21 UTC SREF. I'm very close to just throwing them all out the window. I'm gonna go on a limb here and say that the ensembles are not doing a good job at assessing the uncertainty of this event. It's amazing to see how many signals you can look at that say that the surface low pressure system should deepen (or at least redevelop and then deepen) in the models and yet it doesn't! I'm seeing a combination of warm air advection, lee troughing, and positive differential vorticity advection in the 00 UTC NAM. There's also a jet streak and a negative tilt! The cyclones are also stacked westward with height! I have no idea why the model isn't responding with a deepening low pressure.

I'm looking at a comparison between the 00 UTC OUN sounding and that depicted by the 00 UTC NAM. Some of the initialization errors I'm seeing are appalling. The moisture depth of the boundary layer is off by at least 50 mb. And the 700 mb dewpoint is several degrees off!

Also, I do want to post the quote from the OUN Forecast Discussion:

"WEDNESDAY LOOKS LIKE TROUBLE. `NUFF SAID."

I agree, given the synoptic setup. The thermodynamics (CAPE values) and low level wind field depicted by the models does need to shape up if this is to come true.

 

[Royce Sheibal]

I agree on the models looking strange. My best guess is that the models are interpreting a large mass of rain or possibly leftover MCS, and that's throwing it for a loop. I'm going to be targeting SC NE/NC KS, and I'm going to get there early. That area tends to be a honeypot for triple-point storms, however I've got a bad feeling I'm going to be following HP's all day.

 

[Tim Supinie]

I'm looking at a comparison between the 00 UTC OUN sounding and that depicted by the 00 UTC NAM. Some of the initialization errors I'm seeing are appalling. The moisture depth of the boundary layer is off by at least 50 mb. And the 700 mb dewpoint is several degrees off!

I'm not seeing that at all, from the Hoot website. Both the 00 UTC NAM and OUN sounding have the best moisture to just below 800 mb, and the 700 mb dewpoints are in the -6 to -8 C range on both of them. And anyway, if we're talking about moisture depth, we'd want to be worried about initialization errors south of OUN, rather than at OUN, as those errors will propagate northward with the flow. In the LCH area, switching to TwisterData, the NAM did initialize it a bit too moist from 850 mb to 800 mb and then too dry from 800 mb on up. In the CRP area, the NAM did initialize a bit too much moisture right near the surface, but not extraordinarily so (maybe a couple degrees C), and there's more going on near the surface than advection, anyway.

 

[Doug Drace]

This is a strange looking day to me right now, particularly in KS. Skew-T's/Hodo's arent all that impressive to me, especially in the KS portion of the Moderate Risk. Maybe its my less technical meteorlogical forecasting skills as compared to some, but this vertical wind profile from 500mb-100mb has no directional turning with height/westerlies...something I have never seen before, especially on a day in which potentially numerous tornadoes are forecasted (SPC Moderate Risk 45%, along with numerous news stations). This first Skew-T is from the Pratt-Kingman, KS area. Correct me if im wrong, which I very well could be, but this looks more like a Line Segment/Linear Profile (through the 500-100mb portion of the Skew-T, nothing but southerlies from 500-200mb). With that said, from the surface to 500mb, there does appear to be nice backed surface winds that begin to veer nicely up to that 500mb range, along with some decent increasing speed with height. Honestly though, I would rather, plus, have too stay in OK on this one, due to my obligation to the news station, and what appears to me, to be a more favorable, "Classic" Vertical Wind Profile further south of I-40 in OK near the Lawton area. These profiles are consistant with both NAM/GFS. I just dont like what im seeing north, could be wrong, but MUCH better Directional Shear, Discrete Tornadic Vertical Profiles in OK vs KS, particulatly WC OK-SW OK. Speed Shear also looks better in OK vs KS as well...Opinions???

Pratt-Kingman, KS NAM 0z

Lawton, OK NAM 0z

 

[Andrew Ryan]

This is a strange looking day to me right now, particularly in KS. Skew-T's/Hodo's arent all that impressive to me, especially in the KS portion of the Moderate Risk. Maybe its my less technical meteorlogical forecasting skills as compared to some, but this vertical wind profile from 500mb-100mb has no directional turning with height/westerlies...something I have never seen before, especially on a day in which potentially numerous tornadoes are forecasted (SPC Moderate Risk 45%, along with numerous news stations).

You don't want directional shear in your upper levels. Winds backing with height is bad news. You want speed shear in the upper part of the atmosphere.

This first Skew-T is from the Pratt-Kingman, KS area. Correct me if im wrong, which I very well could be, but this looks more like a Line Segment/Linear Profile (through the 500-100mb portion of the Skew-T, nothing but southerlies from 500-200mb). With that said, from the surface to 500mb, there does appear to be nice backed surface winds that begin to veer nicely up to that 500mb range, along with some decent increasing speed with height. Honestly though, I would rather, plus, have too stay in OK on this one, due to my obligation to the news station, and what appears to me, to be a more favorable, "Classic" Vertical Wind Profile further south of I-40 in OK near the Lawton area. These profiles are consistant with both NAM/GFS. I just dont like what im seeing north, could be wrong, but MUCH better Directional Shear, Discrete Tornadic Vertical Profiles in OK vs KS, particulatly WC OK-SW OK. Speed Shear also looks better in OK vs KS as well...Opinions???

I think you are confusing the wind profile with the shear vector. Your shear vector will usually determine your storm mode. If it's parallel to the front, you're more likely to a linear mess, while perpendicular is favorable for supercells. With the current setup, it's kinda of in the middle, especially close to the dryline. That could be problematic with lots of storms going up and having one storm becoming dominant. But I wouldn't use the model solution to verbatim. Although it has a stronger sfc low for the 12z run, I still think it's underestimating the deeping the low will undergo, IMO. We will see.

 

[Andrew Stoller]

I 100% agree Doug. The wind profile further north above 500 mb I think is do to the very negatively tilted/meridional flow of this tough. The SPC mentioned that, and that's exactly why I'll be targeting central to south central OK. In fact, my initial target is going to be Lawton! Given the strong dynamics of this system and the veering in the low levels, I think there's good potential for initial tornadic storms in central KS near the triple point, but they'll be messy and will likely go more into line segments and lower ends severe/heavy rain event after initiation. It seems with systems like this, there's always more longer lived tail end discrete storms which form to the south where better directional shear exists. The Tushka event on April 14th this year is a prime example.

 

[Greg McLaughlin]

I would think backing winds from the 500-300mb layer would tend to favor splitting cells with deviant right splits becoming the more dominant supercells. A deviant motion tomorrow would likely enhance storm relative shear. I do think convection will fire along the dryline down to near the Red River. I second again what Greg Blumberg is saying, the synoptic pattern looks impressive, indicative of a significant event in the southern Plains.

 

[Tim Supinie]

For what it's worth, I hope the 12Z NAM solution verifies over the 12Z GFS's for a comparison of this morning's model runs. The NAM's crossovers are *much* better, and the shear vectors are mostly perpendicular to the dryline, where as the GFS's are about 45 degrees to the dryline. I'm not sure why the NAM's surface winds are backed so much, though. This would imply strong isallobaric forcing, which ... doesn't appear to exist much on the NAM (it keeps the synoptic surface low pressure center between 1000 and 996 mb between 18Z Wed and 00Z Thu).

One thing I don't like, though, is that the NAM breaks out precip by 21Z in C OK, and then it dies. Though given that it's the NAM, and it's convective parameterization scheme isn't that great, I'll chalk this up to that. Plus, the NAM has some very odd small-scale shortwaves moving across OK around then or shortly before that could be partially responsible. I'm not convinced that's actually going to happen because they kind of appear out of nowhere, and they're on the scale that they're not well-resolved by the model, anyway. But I guess we do need to be aware of the possibility of early initiation if the cap isn't strong enough.

I will probably do a better analysis this evening, once the 00Z model packages come out. And as always, final call on go or no go will be made in the morning.

 

[Doug Drace]

Andrew R...I totally get what you are saying. Maybe I mis-stated the point I was trying to make, which was: From 500-200mb in the KS portion of the risk, upper winds are due south, not westerly as would be the case in a Big Tornado Outbreak scenario. I never said Winds should be backed in the upper levels, I said 'veering' winds, or 'westerly' winds are completely lacking from 500-200mb in KS on those particular Skew-T runs. Which is why I like the southern end of the risk in SW OK a LOT more for longer lived Discrete Supercells. In SW OK there appears to be a more Classic Vertical Wind Profile with SE Backed inflow winds at the Surface, which VEER to SWerlys with height throughout the vertical profile. As well as Stronger Speed Shear in the Southern/Central end of the risk. Hope that clears things up. Again im No Expert, just appeared to be a glaring difference in the 2 locals overall vertical profiles.

 

[Adam Simone]

Andrew R...I totally get what you are saying. Maybe I mis-stated the point I was trying to make, which was: From 500-200mb in the KS portion of the risk, upper winds are due south, not westerly as would be the case in a Big Tornado Outbreak scenario. I never said Winds should be backed in the upper levels, I said 'veering' winds, or 'westerly' winds are completely lacking from 500-200mb in KS on those particular Skew-T runs. Which is why I like the southern end of the risk in SW OK a LOT more for longer lived Discrete Supercells. In SW OK there appears to be a more Classic Vertical Wind Profile with SE Backed inflow winds at the Surface, which VEER to SWerlys with height throughout the vertical profile. As well as Stronger Speed Shear in the Southern/Central end of the risk. Hope that clears things up. Again im No Expert, just appeared to be a glaring difference in the 2 locals overall vertical profiles.

I also noticed this and am kind of concerned about it since I'll only be able to chase in Kansas. I'm a rookie when it comes to forecasting though so I hope I'm missing something. Winds below 850 look to be SE which is good from my limited understanding of severe weather forecasting but they look south slightly SW above 500. If they were more westerly it would be better right or is it just more important that the wind increases in speed with height. I get off work early tomorrow at 11 and plan on targeting somewhere to the west of Salina or Wichita. Ill look at the 0z runs and decide.