I just wanted to comment on the SPC's forcast for today. They got it pretty much right, but I noticed that there were also quite a few tornadoes outside of the moderate risk area (most notably the long-track one in Arkansas). There were also quite a few high wind reports way east of the moderate area. I noticed that they moved the moderate risk area east quite a bit at the end of the day.
Having no in-depth meterological training, I won't criticize the SPC at all. They did a great job on March 12th, 2006. But I think they were a little off on the moderate risk area they were pushing today. Not only that, but with at least 11 dead and 60+ preliminary tornado reports I think today was absolutely a high risk day. I really have no idea why they didn't issue a high risk today.
So, up for discussion, should the moderate risk have been expanded to areas further south and east earlier in the day and why did the SPC not go with a high risk?
[/b]
Bryce,
I've worked at SPC long enough to not get too upset when folks criticize our forecasts. It seems everyone is an expert *after* the event
Anyway, I worked day shift yesterday, so I can comment on that part of the episode. A squall line swept across the majority of the MDT risk area during the early morning hours, leaving the 60+ dewpoints to the S of MO/IL. Still, steep lapse rates were present and the main shortwave trough was expected to pivot over the recovering warm sector in MO during the afternoon, when storm development was anticipated. Low-level shear profiles were modest in the open warm sector through the day (0-1 km SRH AOB 200 m2/s2), while the somewhat stronger shear was located along the warm front where the low levels had yet to destabilize. The MO/IA/IL part of this event was a race between destabilization along the warm front and the transition from a more discrete to linear convective mode.
Farther S, the question was convective initiation. The influence of the mid level trough appeared to extend as far S as nrn AR. There was not much of a low-level focus for initiation there, but afternoon temperatures got warm enough to reduce convective inhibition and allow things like convective rolls and terrain to intitiate isolated storms. The nrn AR environment consisted of strong deep layer shear and large CAPE (near 3000 J/kg for ML parcel), but similar low-level shear to MO and rather large T-Td spreads. Our guess was that the initial storms would mainly be a threat to produce very large hail, which they did. By early-mid afternoon, we noticed that the clouds and early precip across ern AR/wrn TN/wrn KY had held temperatures down, but winds were still backed to more southerly and dewpoints had remained in the mid 60s. We weren't sure how long the backed flow area would hang on after the precip moved away by mid afternoon and the clouds began to erode, but I thought enough of it to go TOR watch and mention the possibility of an increase in the tornado threat toward nern AR.
In retrospect, that favorable area did hold on and shear even improved during the evening, when the tornado threat appeared to peak. In the perfect world I would have issued PDS TOR before any cumulus clouds had formed, but there was just too much uncertainty. The environment was modified on the mesoscale by the early clouds/rain, and that is always going to present a problem for longer range outlooks. I also worked 3 May 99 back in the day, and I promise that *nobody* on this list would have believed a perfect forecast from that morning relying on subtle confluence boundaries, HCRs, a cirrus hole (none of which were resolved until late morning or afternoon), and a poorly forecast speed max!
By the way, the long track supercell cut across the MO bootheel, which was in the extreme srn part of the MDT risk. Why is this considered to be a bad forecast?
