SPC outlook policies changed

[Andy Wehrle]

Several people have commented on today's issuance of a moderate risk on the SWODY3, noting that they have never seen such a thing before. The reason is because it has not been done before. One of the forecasters there who belongs to this forum posted awhile back regarding the changes in outlook guidelines that went into effect this year:

As I understand it,

-a moderate risk/35% zone/significant severe zone can now be issued on the Day 3 outlook

-a high risk can now be issued on the Day 2 outlook, presumably in conjunction with a 45%/sig probability (they have used that probability on the Day 2 before, but only as a moderate risk).

-a Public Severe Weather Outlook will be issued for all moderate risks (provided the upgrade to MDT is issued at or before 1630Z) in addition to all high risks.

Personally, I agree with the introduction of higher probabilities/risk categories for the day 2/ day 3, but I think they should be used only for those situations where there is unusually high confidence of a major severe weather outbreak/long-track strong to violent tornadoes for the period in question. For example, if one is seeing the kind of signals in the models that forecasters were seeing on April 24, 1991 or May 2, 2003, a Day 3 MDT would be in order.

Perhaps one of our SPC insiders could clarify?

 

[Jeff Snyder]

One of my big wonderings is if there will ever be a time when one is confident enough in a model solution to trust a heightened risk (HIGH on SWODY2 or MDT on SWODY3) for 36-84 hours out. While many high-end severe events are "synoptically-evident", more often than not, there are still mesoscale influences that can make or break a big event. Since numerical models don't really handle mesoscale environments well beyond 24 hours, I wonder how much confidence one can place on a 2-3 day model solution regarding storms which are very much local processes (for tornadic supercells anyways, though derecho/MCS are not as much). It seems that there are troubles verify high risks, NOT because the forecasts are "bad', but because the mesoscale environment that often 'makes or breaks' an event are poorly handled by models beyond 12hrs. The categorical risks have been in the past based partly on the confidence that a given event will unfold. We've all read discussions which mentioned that a further upgrade may be needed if models appear that they will verify. This idea makes sense, since the response by EMs/spotters/media/public for MDT and HIGH risks is sufficient enough to want to ensure that the event will actually play out as expected.

I'm not a huge fa nof lowering the MDT criteria, and I'm not a huge fan of the changing of PSWOs for MDTs now. LOL Then again, I haven't really heard the reasons behind the changes, and I certainly assume that there are very valid reasons. We've already have numerous PSWOs this year, and I don't even read them anymore. I can't imagine that the response from chasers/spotters/EMs/media/etc to PSWOs is the same this year compared to years past... I think the SPC has, in the past, issued 0-3 PSWOs a year, and we've seen how many already this year? They are no longer "you better do absolutely everything you can to pay attn to the weather today" type of products as they have been in years past owing to their rarity. If it's a matter of the public wanting an idea that a more signficant severe threat is forecast, and they don't understand the convective outlooks, I'd think that they'd be better off reading the Hazardous Weather Outlooks issued by their respective NWSFO, as they are already geared for the public, than lowering the MDT/PWSO threasholds.

I remember back in the late 90s, the SPC was experimenting with two different products -- one product focused on how widespread the severe activity was expected to be (isolated, scattered, widespread, etc), and the other produce focused on the magnitude of severe activity expected (marginal, significant, extreme, etc). It seems that the probability graphics have replaced these, and they seem to be effective at conveying not only whether a high-end (hatched) event is expected and how widespread the activity is expected (2, 5, 15, 25, 35, 45%), but also what "type" of severe activity is expected (hail, wind, tornado).

 

[Warren Eckels]

Not specific 2-3 days out

The Day 2 Convective Outlook map shows probabilities that some sort of severe weather will take place, but it does not show exactly what is expected. Severe thunderstorms without large hail or tornadoes are severe enough to warrant elevated risk.

Of course, the forecast text may well expect tornadoes:

NUMEROUS SEVERE TSTMS CAPABLE OF PRODUCING TORNADOES...AS WELL AS VERY LARGE HAIL AND DAMAGING WINDS...CAN BE EXPECTED ACROSS PARTS OF THE SRN PLAINS ON FRIDAY AND FRIDAY NIGHT.

 

[Rich Thompson]

Andy,

I'm coming off a mid shift and I'm not all here...but I believe you've got it all correct!

Rich T.

 

[Mike Johnson]

Yes, and look what happened.... a big "dud" of an event - to use the wording from an AMA discussion after Friday's bust.

The SPC usually does a good job.... but three days out is a reach.

Mike