SPC bust on 4/26/17

[Chinook]

The SPC busted their 15% hatched tornado outlook hard on Wednesday. What went wrong? It seems like every interesting storm avoided the moderate risk area. I have rarely seen the SPC bust this bad with the all-important tornado risk. They sometimes have a cap bust in the Plains, but in recent years, the SPC has been mainly conservative when they see some possibility of a cap bust in the Great Plains. In my opinion, the high risk days of April 2nd and 5th were kind of off. But even April 5th high risk day wasn't so terrible (see below), just the 30% was pretty overdone.

 

 

 

 

 

[Quincy]

The issue seemed to be convective mode and jumping fast on a conditional threat. Given the early morning HRRR runs, it seemed possible that pre-frontal robust convection (supercells) would initiate in a favorable environment for significant tornadoes. Once the 06z outlook jumped on 15% tor/MDT, it would be hard for SPC to backpedal without appearing to flip-flop, as it wasn't until midday/early afternoon that CAMs and observational trends shut down the already conditional threat for robust, discrete convection. I don't like to micro-analyze decisions made by the experienced SPC team, as I have often gotten too excited about a setup that ended up busting, but I would have supported a 10% hatched tornado outlook at 06z. The 15% honestly shocked me, as it was highly conditional. This is especially true since they didn't go with a wind-driven moderate risk via squall line, but instead on the supercell potential. By reading the SPC text, they did seem to hit on the uncertainities, especially within their 13z outlook. The other issue may be that some forecasters will go with 2% or 5% tornado outlook, but mention a conditional threat for a significant tornado in the text. It may be the case that others would rather go with a hatched area, even if such a threat is conditional. 

On top of all this, wind profiles hoisted red flags as well, so combine less-than-ideal kinematics with a highly conditional discrete storm threat and you have the recipe for a bust.

Either way, it was one of SPC's worst "busts" in recent memory, but when looking back at the CAMs, the guidance did a fairly good job:

• HRRR only had a couple of outlier runs showing discrete warm sector supercells.
• NCAR ensembles favored a linear threat with nothing too impressive on the southwestern flank of the well-modeled squall line.
• 3km NAM never showed much pre-frontal supercell activity either.