SPC Day 3 Moderate Risks

There was also a MDT on SWODY3 valid for 1-2-06 (as Jim Tang, and Jonathan Racy, noted in the 4-13 FCST thread) --> http://www.spc.noaa.gov/products/outlook/archive/2005/day3otlk_20051231_1100.html

Note that both of the MDT risk Day 3 outlooks occurred before the SPC reorganized the probabilities and the relationship between probabilities and categorical risks. It appears that 35% severe prob was sufficient for a MDT risk before last spring; as Jon Racy noted in the FCST thread, a MDT risk on a SWODY3 now requires 45% hatched probability. See HERE for the categorical risk - probability breakdown.

On a similar note, Day 2 outlooks can contain HIGH risks as well, if they are accompanied by 60% hatched severe probability. I really don't think we'll ever see many HIGH risks on Day 2 outlooks, given that many environments that would justify a HIGH risk also tend to be quite difficult to forecast with enough confidence to issue the highest of categorical outlooks. Heck, we have a tough time with high risks on Day 1 outlooks, and there's a reason why high-end severe weather or tornado events are rare! I think the best candidate for a High risk on a SWODY2 would be for a derecho event, which may be a little easier to forecast (with sufficient confidence) than a violent tornado supercell event.
 
Day 3 MDT and day 2 HIGH risks should be relatively rare for one important reason - it's *much* more difficult to back out of an outlook than it is to upgrade later on day 1. Another thing that you must keep in mind - the same set of forecasters is not working every event. Everyone has some sort of bias(es) in their forecasting approach, and who you work with can also have a substantial impact on the outcome. Some of you that like statistics and have followed the SPC outlooks for many years have probably noticed that a higher-end outlook is more probable when particular forecasters are on shift, and vice versa. Considering the (in)frequency of big outbreaks and constantly changing combinations of 20 SPC forecasters, and some inconsistency is virtually guaranteed.

Rich T.
 
Back
Top