psuhoffman

I went to Oakton. ‘97.

It's a very marginal setup...the HRRR and GFS aren't in actuality THAT far from the other guidance...they are just 1-2 degrees colder and have the precip come in a little harder initially...that combo is the difference between a thump snow along our NW fringe regions...and not. It's not like the guidance is showing some vastly different outcome...its just the minor differences have rather significant impacts on the ground truth in this case.

There is a reason I let my slightly above average snowfall forecast ride even while many were jumping ship over the last 2 weeks. My gut says something is different. We will see, maybe I just ate something...

Storm cancel December Cancel Winter Cancel

This is perception bias. It does happen in reverse. The vast majority of our actual snowstorms were not on the guidance 174 hours out. And very often not showing that much snow...even at 72 hours. They trended into something in the last 48-72 hours... just like most of our "fantasy digital threats" trend away form something...because the truth is the odds of the guidance being exactly right from range is low. Very rarely do we get a snowstorm where the guidance nailed it from really far out. The first Feb 2010 storm...January 2016 were rare exceptions not the norm. And the reason it feels like we get way too many "false threats" is also perception bias. We register every potential threat as if "the models say it's going to snow" WHich models? Did they all? And when...did they have a threat at day 8 then lost it by day 6...which day is the one that matters...if we count every day through the whole winter where a couple models spit out a permutation with snow...as a legit threat and expect that snow to happen...that is on us...that is a failure of our perception not guidance. All that said our guidance is not perfect...its flawed...and we need to continue to work to improve our ability to forecast both with models and through other means. But you're compounding this by falling victim to typical perception biases.

I think the only hope for a snowier outcome (and even then it’s mostly for the NW fringes of the forum, is for it to come in hot and heavy and the dynamic cooling and mixing of the column keeps it isothermal near 32 long enough to get the classic thump surprise. We’ve seen it happen. But it’s not something we can nail down ahead of time usually.

What part of Hanover are you?

There is a breakfast place in Manchester called Dutch Corner. Christmas Tree Park a few blocks up the road might be a nice place for a Jebwalk also. If you end up in this area.

Yes it’s better if it snows before mid December in our northern and colder regions but it’s even better if the storm hits everyone. I know it was in jest.

When I isolated all our December snowfalls and looked at what factors seemed to be most common by far the PNA and EPO had to be favorable. With the caveat that the AO is still the most common indicator. But that is universal across every possible pattern and time of year. If the AO is positive it's just hard for the jet stream to be suppressed enough to get a system to stay under us. But in terms of the pacific (EPO/PNA) v atlantic (NAO) the pacific was definitely more important. Not necessarily both but it was almost unheard of to get a snowfall early in the season if both the EPO and PNA were unfavorable. Having the NAO also certainly helps...but the NAO alone without any help from the PNA/EPO doesn't do us much good in December.

Another way we scored some snow (and let me emphasize "SOME", we're talking 1-3 or 2-4 type events here usually) during that pattern was with a follow up wave timed up right behind a cold front. The cold to our NW can press east behind a lakes cutter but it's not going to last...but if you get a strung out boundary with multiple waves we can get clipped with one.

I predicted slightly above normal snowfall this year

The monthly snowfall composites back that up

The weighted analog mean produces the follow snowfall predictions BWI: 18.1" IAD: 18.3" DCA: 14" December and January average below normal temps. February much above normal. My current snowfall forecast is slightly above this but I simply added a small bump for the "we're due" index. We've been pretty unlucky lately, maybe we get one good hit and this is that rare nina that ends up slightly above normal snowfall. They do happen once in a while and we are due for one. But for what it's worth the composite mean beat my "gut" adjustments 3 of the last 4 years. So... My call BWI: 20" IAD: 23" DCA: 15"

I ran the composites using my weighted full analog method. It didn't really change the outcome much. December through mid January still looks pretty good. February looks brutal. Is what it is. Definitely not a non winter look though. Winter Mean Dec Jan Feb

Last year 2009 was my top analog, and I actually think that worked out pretty well. This year it's closer to the bottom of the list and barely made the cut. I thought I would have more time once soccer ended, I coach both my kids teams, but I took on a lot more at work and honestly I'm not sure if I'll have time to do a full in depth seasonal forecast but I did take an hour to identify the analogs using my formula. I think I am going to run a new anomaly mean plot where I weight them.

I am starting to lean towards something like this as the mean winter pattern. December-January coldest, February warmest, March is a wildcard.

Did you cancel winter yet?

That’s more important to snow around here in December than the NAO

2019 following the weak nino

I’m not talking about the longer scale -PDO. Within that are shorter term extreme phases. That’s what’s been killing is. We can snow in a somewhat negative PDO. Look at the 1960s! But past mini extreme -PDO periods (typically 4-7 years where the PDO is below -1 most of the time) are always god awful for snowfall. We have to be coming to the end of this mini super negative PDO phase. None going back 125 years last longer than about 6-7 years.

For the whole mid Atlantic 1996 For Manchester MD Feb 9-10 2010 We started with 20” otg, then this crazy convective band set up along the thermal boundary the evening of the 9th. We got 12” in 4 hours and had about 14” new snow that evening before the lull then the CCB associated snowfall developed right over us and snowed all day on the 10th. True blizzard conditions, wind, freezing cold, another 16” or so fell. We ended up with about 30” which was way over any expectation going in. Had close to 50” otg when it was over.

Yes. But that can’t continue forever. Look at all the past similar extremely negative PDO periods. There longest such recorded periods and closest to this one were in the 1950s and 1970s. But those didn’t extend past 6 or 7 years. The extreme -PDO. So we have to be coming to the end of this current cycle soon. There is absolutely no precedent for this continuing much longer.

Historically I don’t disagree. But we’ve said this quite a few times over the last 10 years and it seems our “hit rate” has become pretty low in a more zonal suppressed “big bowl” pattern also, to the point maybe it’s better to just take our chances on the big hit anymore. But since we don’t control what we get it doesn’t matter. Yea I’d take a high chance if getting some moderate 3-5” type events over a very low chance at a 10”+ but it doesn’t seem to be that way anymore. We seem to waste the “high probability” looks just as often as the big hit ones lately.