I just don't get it....PDS

[Elinor McLennon]

Ok, what went wrong with the PDS in OK area? I know there were a few tornaoes but to hear the weather service, the weather channel and the local media, imminent Armageddon was on the way.
I don't know alot about forecasting and this may be the dumbest question/observation ever, but here it goes:
My thermometer at home read 74-75 all day. It was cloudy, and extremely windy all day. I tried to walk my small dog and she was getting blown off the walking path several times:) So, did the cloud cover and brutal "ground" level winds change the dynamics in the environment?
Last, I know we have beat this to death but I heard alot of people scoff this evening about "forecasters" over reacting and if they say we have a mod/high risk in Oklahoma that we won't get anything. I am afraid that people are so apathetic to the media "crying wolf" that there will be alot of injuries/deaths some day if we get bad storms.
I am renaming PDS to Pretty Dang Stupid after today. I'll cheer up by tomorrow.

[Joey Ketcham]

I don't think it's really crying wolf, after all there were tornadoes today.

Keep in mind that PDS watches are issued when atmospheric conditions can support long-lived, violent tornadoes. That doesn't necessarily mean it will happen, it's just wording enhancement to let people know that should tornadoes occur they can become long-lived and violent.

You have to remember that this is still a realitively new science and part of the on-going research is trying to understand why these atmostpheric conditions exist, but yet you don't get those deadly tornadoes that you would expect.

Just my .02 cents.

[Greg McLaughlin]

I believe this setup was an EPIC FAIL because we didn't have a good EML (capping inversion) in place. Too much convection fired early in the day and ruined the instability by the time the dryline mixed east. The only tornadoes formed along the outflow boundary where horizontal vorticity was maximized and could be ingested into an updraft despite the less than idea instability.

[Chris Allington]

I believe this setup was an EPIC FAIL because we didn't have a good EML (capping inversion) in place. Too much convection fired early in the day and ruined the instability by the time the dryline mixed east. The only tornadoes formed along the outflow boundary where horizontal vorticity was maximized and could be ingested into an updraft despite the less than idea instability.

Greg hit the nail on the head with this one... there was not capping and way to much linear forcing along the frontal boundary. As soon as storms started to go it all when into linear modes in a hurry.

EDIT: Weren't there even people talking about how the NAM forcasted junk convection to be ongoing in much of the warm sector all day? I think that was brought up in the forecast thread if I'm not mistaken.

[Elinor McLennon]

Yes, i believe it was mentioned about the NAM forecast.

Just seems to me that I can't think of a day with all the cloud cover and "hurricane" force winds from sun up to sun down turning into a major day. Seems like it's always hot, muggy and still.
The early morning junk confused me this morning....just didn't seem to add up for a major event this afternoon.
Sorry if I seem cranky.....just disappointed is all.....happy nobody was injured, disappointed from a chaser perspective.
Thanks for the replies and thanks for not hammering me about my forecasting inexperience.

[CZandbergen]

I am with Greg on this one. i was planning to chase till I saw the runs late last night. I figured there would be a couple good storms, but the aforementioned junk and a general model disagreement (and a probable 14 hour round trip drive) made me decide to sit this one out. It looked like a dynamic situation and it ended up being one from my seat back here in Tulsa. Good news is my forecast was good as a tornado was on the ground within about 20 miles of my target. I for one will catch the next one.

As far as the PDS goes, I have no comment. That is a pay grade far higher than mine....

[Joey Ketcham]

On the record, I do believe that today was a huge bust.

[Jason Foster]

To me....as a chaser....I don't pay much attention to the level of Risk beyond Moderate. The same holds true for PDS watches. While I do show excitement as soon as they are issued, in retrospect they do tend to be more disappointing for the kind of chasing I like. High Risks and PDS watches may have more to do with the SPC's public and government agency planning responsibilities than conditions for us chasers. I'm think it's been discussed in other threads, but I wanted to post here anyway.

[Jeff Snyder]

On the record, I do believe that today was a huge bust.

Earlier today, it certainly looked to me like there was an enhanced risk of strong tornadoes across the High Risk area. Very strong 0-1km SRH, moderate instability, 40-45 kt deep-layer shear, etc, all supported significant supercells.

That said, by 00-01 UTC, we knew that storm mode wasn't working out well, the line of semi-discrete storms in NW and NC OK was burfing cold outflow, and there was an expanding shield of weak convection SW of OKC. The supercells that had been moving northeastward out of TX and into the area SW of Lawton were dissipating (presumbly, there was a lot of seeding occcurring from the upstream storms). Surface obs in southwestern OK largely had Tds that dropped into the 60-61 with relatively cool temps. There was (and is) a small zone of undisturbed warm sector in the extreme SE TX PH east of the dryline (and west of the cold pool / outflow driven by the southwestern OK convection), and in far southcentral OK and northcentral TX was (and is) also undisturbed. However, Tds nearer I35 dropped a bit to the 59-60 F range. With these considerations, I found it very interesting that the SWODY1 maintained the high risk for 30% hatched tornadoes. The 30% prob implies a high coverage of tornadic supercells, which seemed quite doubtful given surface obs and radar. I do NOT criticize official forecasts often since tornadoes and severe convection are often very difficult to forecast. Heck, I'm wrong more often than not anyway, and these guys are pros. I fully supported the earlier outlooks, but I wonder how much continuity played in maintaining the high for the evening SWODY1. I do like how OUN NWSFO handled it in their evening HWO, mentioning only a "risk" (unspecified in terms of slight/moderate/high) and not mentioning any tornado risk overnight.

[Steve Miller TX]

http://en.wikipedia.org/wiki/List_of_SPC_High_Risk_days

[Derek Weston]

On the record, I do believe that today was a huge bust.

While I don't think it lived up to a high-risk billing in the way most people think of high risk... I don't know that it'd be considered a huge bust either.

There are, from what I understand, some still outstanding storm reports from the western OK area. (more tornadoes than the 1 report listed on the SPC) I say this because I know that many people judge such a risk on the tornado reports it generates.

Of course . . . the risk isn't to be judged by the actual result as much as its reasoning before hand. . . but if you don't participate in forecasting with a valuation of risk attached to your prediction... it's easier to sit back and criticize after the fact.

I have not been watching SPC forecasts or severe weather forecasting in general very long. However. . . I have noticed that there seems to be a fine line between major outbreak and "messy-junk" much of the time with these sort of dynamic setups that could potentially produce a "high" end severe outbreak. ("these sort" being a certain subclass, I suppose, of dynamic events -- I'll list a few examples)

Storm mode being the trickiest aspect to nail due to the nature of the forcing and directional shear we have to work with. (and that's probably simplifying things a bit)

Today ended up being messy and rather linear -- I believe it was another dynamic setup, June 5 of last year, that ended up somewhat similar. I think it was June 5 of 07 that also disappointed chasers -- a bust from the tornado perspective for these two.

Of course, we had another major outbreak a few weeks back that resulted in many tornadoes with fairly unidirectional shear and lots of forcing from a cold front in the Southeast, if I remember correctly. That day, to me, looked as though it could've been pretty messy. But it wasn't.

I think the super-outbreak in 74 featured a mixed convective mode due to the nature of the shear profiles and forcing mechanisms in place. (and there are more examples of this sort of dynamic situation resulting in a big outbreak)

Long story short: seems to me that it is hard to pick out some of the exact variables that will turn a possibly messy mostly linear 'outbreak' into a monster tornado producer.... beforehand.

Or, perhaps... the variables are properly identified and weighed... but the valuation (to the extent that it ends up in a high risk forecast) isn't exactly right on... or at least what you think it should be.

Some folks have pointed out in this thread some of the variables they thought might be problematic for this setup -- how confident were they that these variables would indeed prove to be problematic? And given your level of confidence... how do you then weigh the chances for a significant severe weather outbreak? Basically... how much confidence do you need to go high risk? What should result in high risk -- how much confidence ought a forecaster have in their forecast?

These are tricky questions to me, and I'm not exactly sure where I stand... honestly.

I want to say that certain types (some of them touched upon above) of high risk days are more inclined to bust than others. May 29 08, for example, looked a big tornado day to me -- I had high confidence. I didn't feel the same way about June 05 08... however... should it turn out... we've got an amazing environment for tornadic supercells and we could be looking at an enormous outbreak.

I guess the big question for me is... how much ought 'high risk' be about potential?

[shane turner]

I myself was wondering about PDS tornado watches and high risk days. It seems that quite a bit of moderate risk cases with 15% hatched areas for strong tornadoes has performed just as well as high risk days. There has also been a few slight risk days that have overperformed. I am no expert but do you think that a high risk day for severe weather is needed since they are only issued about three times a year. For instance the Greensburg and Parkersburg EF5 tornadoes came out of moderate risk days. Also the two EF4 tornadoes this year came out of moderate risk areas. I know the Murfreesboro tornado happened on a high risk day but came out of the moderate risk area. Even last year during the Super Tuesday Outbreak I think two of the five EF4 tornadoes came out of the moderate risk area. Other moderate risk days were March 1, 1997 and November 23-24, 2001 which had 3 F4 tornadoes from both events. Now I feel PDS watches are needed especially for an enviroment where there will be multiple strong/violent tornadoes but IMO there seems to be no difference between a high-end moderate risk or a high risk of severe weather. So it makes me wonder if there is even a need for high risks.

[Brian Kruse]

Hi Mike. Just wanted to let everyone know that we were 3 1/2 inches away from the "death zone" sunday. Saturday night we were within centimeters. The whole weekend seemed particularly dangerous.

[Steve Miller OK]

That's funny, I posed a similar question last night in the NOW thread regarding the 01z SPC update and it got erased. I thought maybe asking why this particular update was still a HIGH risk was against the rules of a WEATHER-related forum. I'm happy to see other people agree with the concern.
I have no opinion about the specific PDS watches issued yesterday other than to say 190 was somewhat justified but 192, I would have thought would be dropped, replaced or whatever the proper government wording for "downgrade" would be.

[Elinor McLennon]

Steve Miller's post showing the SPC high risk stats was interesting. It made me re-think my bad attitude towards SPC statements. The dang media hype is ridiculous...we're all gonna get hit by long-lived deadly tornadoes. But re-issuing watch til 3am puzzled me, especially when they were saying the threat was diminishing. Downgrade should have been done and if the media continues to stay on live basically non stop for 2 days and nothing happens then people will have a worse attitude that they already do about these watches.
I've had several calls today from people saying "i told you nothing was going to happen, it never does....they always get excited for no reason".

Hope everyone has a good day.

[shane turner]

Yeah Steve my post got erased too as well. It does not bother me it got erased by no means but I was just stating about the high risk as well. I thought my word usage was proper and did not use harsh words of any kind. Ellinor I also got from a nap yesterday around 1pm and saw the high risk and PDS watches and a squall line across the radar screen and also wondered if this was going to even produce strong tornadoes. The PDS tornado watch issued till 3am was also puzzling to me for it did not seem like much was happening. Even though I think SPC must have had good reasons to why they did issue them.

[Joey Ketcham]

I've had several calls today from people saying "i told you nothing was going to happen, it never does....they always get excited for no reason".


You will get that no matter what. I've had my own parents tell me that on big tornado outbreak days just because it didn't happen right over their head.

You simply have to remind them of 2 things:

1). We are still dealing with a relatively new science that we are still in the process of learning. That's why we have on-going field research going on to better understand why some storms produce tornadoes and others don't in environments favorable for tornadoes.

2). The whole point of a tornado watch is to alert people that the environment could support tornado development, there are no guarantees that tornadoes will happen. Tornado watches work just the way they are designed to.

[Brian Stertz]

I am thinking that there were several complicating factors on yesterday's set up. First of all, you had very little cooling aloft with the positively wave that seemingly grazed the edge of the warm sector. The supercells never seemed to want to get moving off the boundary and move east way from the cold air so this big caused problems IMO. This horrible cold pool that seemed to doom anything that was able to get sustained in NW Oklahoma and NE Texas Panhandle. The weak cap in warm sector was another problem as mentioned. It was far from the perfect set up but then again, it's not my game to judge as I was not sitting any forecaster hot-seat. Data was not too plentiful out in the field for me yesterday so that is my take after seeing what data I did access yesterday.

[Robert Dewey]

Around 150 reports of wind/hail in the MDT risk area... verified quite well.

Lack of tornadoes != bust

[Derek Weston]

why did I talk about high risks? ha

[Elinor McLennon]

When we made it to Yukon around 7:30ish and i heard the guys in Norman tell the EOC folks that threat was rapidly diminishing and then come home to hear PDS til 3am i just had to think "what the heck".
When i got home from church around 1ish and saw all the junk on radar i kept thinking maybe this is going to blow over then PDS would happen. Even though I knew the early stuff would probably wreeck the atmosphere.
I guess I'm a freak...being bummed about lack of tornadoes and the havoc they bring. Guess I need a much better attitude:) and possibly a new hobby.

I wouldn't want to be in the forecasters shoes and they must take alot of flack from people with my attitude:) I wouldn't want to have to make the final decision about watches, etc. I do respect them and realize it's not easy and it's not a perfect science.

[Jesse Risley]

Steve Miller's post showing the SPC high risk stats was interesting. It made me re-think my bad attitude towards SPC statements. The dang media hype is ridiculous...we're all gonna get hit by long-lived deadly tornadoes.



As has been stated before, I think you have to initially remember that the SPC doesn't forecast specifically for the chasing community. I look at it from this perspective - the parameters were certainly in place for a sustained severe weather outbreak, including some stronger tornadoes.

There are often other mitigating factors that are nearly impossible to forecast that will become an issue from time to time, as was the case yesterday.

Forecasters realized the potential for a sustained severe weather threat and alerted the public accordingly. I'm sure they hear about it quite often when it gets blown, but how often do they get thanked for being right on the money?

I really don't blame anyone for media hype, because dirty laundry sells.

[Mike Bennett]

I think it was the early convection due to lack of cap that made this a bust, respectively. However, with that being said I believe the SPC issued the High Risk after the early convection began and maintained it in the 2000 outlook and again in an outlook issued later in the evening.

[Mike Smith]

Let me offer a thought:

The objective guidance (per 14Z RUC) prior to the 1730Z 'high risk' convective outlook issuance was off-the-chart favorable for major tornadoes. CAPE 4000+j, helicities 400+, very dry air at 700mb across KS-OK (i.e., cap, little or no grunge convection). Surface temperatures were predicted to be in the mid- to upper 70's (which occurred in OK). Nice surface low and boundary.

Interestingly, the 12Z NAM yesterday was the opposite: Lots of grunge.

I have no doubt that if the RUC had been correct, yesterday would have been a very, very big day.

I, too, did not understand the PDS until 3am. But, I do understand the first PDS.

[Tim Vasquez]

Just as a note we also have a parallel "official" thread regarding the 4/26 activity in Target Area:
http://www.stormtrack.org/forum/showthread.php?goto=newpost&t=20421

Please use that for technical discussions of the 4/26 storms. Since this is the Beginner's Forum the discussions here should be limited to answering the poster's question about the PDS watch.

Tim

[Jay McCoy]

Looking at those stats the SPC is bating about .800 on high risk days so a bust now and then isnt bad overall. I dont exactly call 13 busts out of 50 high risks as "crying wolf". If they were 100% then that would mean our forcasting and understanding of tornadogenisis is perfect. We are far from that. Mother nature is still a difficult lady to understand so give the guys a break. Overall they do an excellent job. Maybe we didnt have 80 wedges chewing their way across Oklahoma but we did have a few very nice tornados right where they called for them. I for one never thought SW oklahoma was a worthy target. I had I-40 up to the kansas border picked the whole time.

I think the weak cap and earlier convection is what screwed us. BUt I do agree thney should NOT have continued the high risk in the evening update.. That made no sense except continuity of forecasts but that isnt always a good thing. I prefer accuracy over "continuity".

[Darrin Rasberry]

I dont exactly call 13 busts out of 50 high risks as "crying wolf".

I think the Wiki page should list June 5th of last year as a bust, since the broad 60/hatch Wind-High and the 30/hatch issued later that day didn't turn out much (although the mod risk part somewhat did for tornadoes). Schools were shut down that day across the area IIRC, and lots of other haywire things went on, but hey, it's the weather, and I agree that the batting average has done just fine.

Curiously, Craig and I realized driving home yesterday that the PDS on Sunday verified all the probabilities on the table. There were indeed two or more tornadoes, and one of them, should the reports in the thread for that day match the assessment, was EF2 or greater. The "long track" stuff ... well, not so much.

[Elinor McLennon]

Seems like the bigger tornadoes don't occur on days with so much cloud cover, temps around 74, 20 - 30 mph winds with gust higher than that......is that a stupid assessment for Sunday's setup in OKC?
People said it "felt" like a tornado day. Maybe out west but not here; it was a bit nippy with that wind and cloud cover.
Was it sunny and less windy when the tornado came around Roll, vici area?

[Jay McCoy]

Was it sunny and less windy when the tornado came around Roll, vici area?

The louds had broken and it had heated up nicely in Shamrock and the far westerrn part of oklahoma as the storms 1st fired. I cant say how warm it was in Roll before the storm got there as we got there at the same time were a bit busy shooting video and driving to notice the temp.

[Justin Wiley]

I think it all depends on the situation Elinor. I remember the the 4/26/91 event quite vividly. Some low-level cloudiness, mixed out by late morning, it was very warm and humid, and it was windy.

Obviously, more cloud cover usually means less instability and thus can turn potentially 'big' tornado days into nothing more than a few isolated hail/wind reports. Typically you want a nice low-level moist flow as long as the wind increases with height the higher you go in the atmosphere. So it being windy isn't necessarily a bad thing at all. It all depends on each specific setup. 74 degrees may be enough for one system that brings more colder air aloft and may not be enough for another.

[Chris Bethurum]

Was it sunny and less windy when the tornado came around Roll, vici area?

I was headed northwest on 270 within an hour of the tornado being reported on the ground. The most sun I saw all day was between Watonga and Reisling, but I still wouldn't call it sunny. There was that high thin broken layer and I'd say 30-50% coverage low-mid level broken stuff. I drove around in OKC around 1115-1300 and it was very windy, 25-35 from the SSE (my estimate) and it was completely overcast. The whole day I was thinking to myself, I can't see much happening today, it just doesn't look and feel the way I understand it should other than the strong moist low level winds.

Driving 270 I could see darkened sw-ne oriented linear bases within the whole overcast layer, but no real storm structure visible other than the lowered bases of the largely or completely hidden supercell structures of the storms that dropped stuff.

[Skip Talbot]

Seems like the bigger tornadoes don't occur on days with so much cloud cover, temps around 74, 20 - 30 mph winds with gust higher than that

With a southeast surface wind, the stronger the wind the better. The stronger the surface wind the more low level wind shear you have, and the better your chances for tornadogenesis.

[afischer]

In my experience, stronger sfc winds often (almost always?) reduce hodograph curvature and 0-1 km bulk shear, and thus SRH. I think the more that "off-hodograph" storm motion contributes to SRH (e.g. a strongly veered storm motion or relatively backed low-level winds)... the less of an impact a stronger sfc wind will have on reducing SRH. But every hodograph is different, so don't quote me on that.

Actually, several of the nastier tornadoes I can think of had light/decoupled sfc flow in the ambient environment (<12 kt)... including the 5-12-04 Harper F4, 5-22-04 Hallam F4, and the super-long-track Arkansas tornado from 2-05-08. This really serves to maximize the SRH, particularly in cases like 5-12-04 where the LLJ isn't all that strong.

[Jeff Snyder]

With a southeast surface wind, the stronger the wind the better. The stronger the surface wind the more low level wind shear you have, and the better your chances for tornadogenesis.

(Can a moderator perhaps split this off into a different / new thread?)

Not necessarily. Strong surface winds may create strong extremely-near-surface (e.g. 0-10 m) wind shear, but we typically look for wind shear in the 0-1, 0-3, and 0-6 km layers. If 6 km winds are 40 kts from the southwest, then southerly surface winds of 30 kts will create LESS wind shear than southerly surface winds of 5 kts. Think about this in terms of a hodograph -- the "longer" the vector between, for example, the 6 km wind and the surface wind, the stronger the wind shear between those levels. Strong low-level, but above-surface, winds (e.g. at 850 mb) are often desireable, but that's not really the surface wind. Let's look at any example hodograph.

Note that the hodograph below represents that of a VEERING wind profile (i.e. the winds veer from southerly to west-southwesterly at 6 km). This is a straight-line hodograph despite having winds that veer with height:

http://www.tornadocentral.com/now/hodograph_strength_of_sfc_wind.png

The red line represents a wind profile characterized by a surface wind of 30 kts from the south, a 3km wind of ~40 kts from the southwest, and a 6 km wind of 60 kts from the west-southwest. The red star represents an approximate storm motion for a right-moving supercell. Now, let's look at the blue lines... The only difference is that we now have a surface wind of 10 kts from the south, with the 3km and 6 km winds the same as for the red wind profile. Note that the change in the surface wind has had some change in the winds between the sfc and 3 km. Similar to before, the blue star is an approximate storm motion for a right-moving supercell. Owing to a WEAKER surface wind, the blue wind profile actually has more 0-3km SRH! Because the red wind profile yields a straight-line hodograph, there won't be ANY 0-3km SRH for initial convection; any 0-3km SRH will only arise when storm can split, whereafter the right-moving supercell will take a motion that takes it off the hodograph. In the blue case, even a storm that does NOT have a storm motion that deviates from the norm will ingest some 0-3km SRH. The blue wind profile does have slightly stronger 0-6 km shear, but they're pretty similar in that regard.

Similarly, let's look at an example of a couple of vertical wind profiles for which a VEERED surface wind will actually resulted in higher 0-3 km SRH than a backed surface wind:
http://www.tornadocentral.com/now/hodograph_dir_of_sfc_wind.png
Again, the red wind profile features southeasterly surface winds at 10 kts, south-southwesterly 3 km winds at 30 kts, and between southwesterly and south-southwesterly 6 km wind. So, the mid-level flow is a bit meridional, but this isn't uncommon sometimes, and often not "bad" given southeasterly surface winds. Yet again, despite both directional and speed shear in the red wind profile, this is a straight-line hodograph. So, unless you get a right-moving supercell (which probably won't happen until after a cell splits), there is actually 0 0-3 km SRH. Now, let's say the surface wind veers to the southwest and remains 10 kts -- this is the blue wind profile. Now, because there is some curvature of the low-level flow (e.g. no longer a straight-line hodograph), even initial convection (moving with the mean flow) will ingest some 0-3km SRH. If we take a supercell with minor right-moving characteristics, we note that the 0-3km SRH with this blue wind profile (i.e. veered sfc wind) actually has more 0-3 km SRH than the red wind profile. In other words, we would actually want a VEERED surface wind, since it yield better curvature to the low-level hodograph. In this case, the red wind profile does have marginally greater 0-6km shear.

From the two above idealized example hodographs, all of which feature winds that VEER and increase in speed with height (e.g. all hodographs have directional and speed shear). However, one can see that there are wind profiles for which weaker surface winds are better than strong surface winds. In addition, one can see that there are wind profiles for which a veered / SWrly surface wind yields a better hodograph (and greater 0-3km SRH) than a backed / SErly surface wind. This is not necessarily intuitive, but it brings about the importance of looking at a hodograph.

EDIT: The storm motions above are only eye-balled estimates, but they illustrate the point I think. The storm motion for the blue case in the 1st example may be too fast, but the point remains.

[Jeff Duda]

Jeff,
Very well explained examples of how to use hodographs to evaluate shear parameters. Perhaps Skip's comment was a little more general (I don't know for sure; he will let you know). I like seeing stronger surface winds with very strong low-mid level winds not only because of the enhanced shear, but because stronger surface winds will hold back the cold pool outflow from storms, keeping the storms upright longer and keeping them from becoming outflow dominant. They also serve to enhance surface convergence where the RFD hits the ground under/near the meso.

Others have mentioned this on the other thread discussing this date, but another problem seemed to be that the good shear was too low with the best shear below 3 km. For good sustained supercells, you want deep layer (0-6 km) shear to be high, and many profiles showed winds actually backing and weakening with height above about 700 mb or so. That definitely hurt the chances for supercellular activity.

[Darrin Rasberry]

Did that assume that the supercells would right turn (they didn't IIRC)?

[Jeff Duda]

Did that assume that the supercells would right turn (they didn't IIRC)?

When supercells turn right, the storm motion vector moves farther away from the hodograph so that the area increases and helicity increases (in general).