I noted a discussion last night regarding the SPC's forecasted outlook for May 13th, 2009. I believe that the discussion was going on in this thread, but I cannot seem to find it anymore, perhaps it was deleted.
Regardless, it seems as if there can be alot of debating whether or not a risk that the SPC puts out "verifies" or not, with many claims like "a moderate risk used to mean ___" made. It seems clear that last night, and many of these instances, people can let their emotions get in the way on either side of that argument. Today, I decided to objectively verify whether or not the moderate risk "verified," for just for the sake of fulfilling everyone's curiosity.
The SPC's definition of the convective outlooks, per their own FAQ page, is based on the individual probabilities of hail/wind/tornado within 25 miles of a point. This diagram comes to mind:
http://www.spc.noaa.gov/misc/prob_to_cat_day1_seetext.jpg
Did May 13th, 2009 MOD verify?
Based on this, a moderate risk means that there is a 15% chance of tornadoes, a 60% chance of severe hail, and/or a 45% chance of severe winds, all within 25 miles of a point. Using GIS, the 1630z MOD risk outline was drawn and it's area calculated. The MOD risk was ~156,763 sq miles in area. The area created by a 25 mile buffer around a point is 1,963 sq miles. This would create ~80 "probability circles" within the bounds of yesterday's risk area. Averaged over the risk area, the tornado probability was about 7%, the hail probability was about 45%, the sig. hail pobability was about 9%, and the wind probability was about 35%. Multiplying these percentages by the number of "probability circles" yields the number of evenly spaced reports averaged over the risk area that is required to verify the risk. Based on the SPC's definitions and the risks shown above, yesterday's MOD risk was low-end, and the area southwest of central Missouri only became a moderate risk due to the hatched (sig severe) area for hail.
For the moderate risk to "verify" according to the SPC definitions there would have needed to be 5.6 tornadoes, 35.9 severe hail instances, 7.2 significant severe hail instances, and 27.9 severe wind instances.
*The obvious caveat to this is that severe weather does not occur as points, it occurs as lines or polygons.
Based on this, the risk likely verified for tornadoes (23 reports), definitely verified for hail (127 reports), maybe verified for sig. hail (7 reports), definitely verified for wind (144 reports).
