OceanStWx

Yeah, there were some late closures there that were congrats Nova Scotia, but if we can get that going between MTP and CHH it becomes a rosier picture.

Oh we know there's a bias keeping the QPF too close to the low center for one. But even the 12z GFS has a decent banding signal from central LI through BOS despite what it's QPF/snow output shows.

Sorry, should've included a trigger warning for the snow starved.

As an example, here is February 1 of last winter. Northern NJ is showing an EFI around 1, meaning every EPS member was more extreme than the model climate. And with a SOT value of 2, the top 10% of EPS members were also quite extreme. And well...

There's some good news I guess. Based on the 00z EPS, the extreme forecast index is okay for the AEMATT crowd. This index essentially shows how unusual the EPS forecast is based on the last 20 years of the model climate in the 5 weeks centered on the date (the climate built by many model runs, not state of the climate according to the model). Values between 0.5 and 0.8 are unusual, above 0.8 very unusual or extreme. A value of 1 means that every EPS member is more extreme than the model climate. The thing to keep in mind is that significant snowfall, even in the winter, is unusual. Most days it doesn't snow. But there is a signal for some significant impacts possible according to the EPS. Also note that this index doesn't say how extreme. It could be a little more extreme than the model climate, or a lot more extreme. That's where the black lines come into play, that's the shift of tails (SOT). A value of 1 indicates that the 90th percentile of the forecast is one times the distance between the 90th and 99th percentile of the model climate. A value of 10 would be ten times. A SOT appears when at least 10% of EPS members are forecasting an event greater than the model climate extreme. Say the model climate extreme is 6 inches and the difference between the model climate 90th and 99th percentile is 1 inch, with a SOT value somewhere between 0 and 1 (call it 0.5) this 90th percentile forecast would be 6.5 inches. Notice how the SOT values are higher and the EFI higher down around DC. That's because snow is more extreme in that area than New England, and more likely to be an extreme event based on forecasts.

Okay, now the 00z Euro is the biggest run of our lives.

Sorrey aboot all that snow I dumped on you last run, eh.

It's also a problem that they've been taking 72 hours to get here from PDX. With such flat flow small errors are magnified when things are moving quickly.

The way our grids our broken up this is Thursday night. The nights run 6 pm to 6 am. So 4 am technically sneaks into Thursday night.

It's a numbers game. If BOS pulls 3" a week in the heart of winter it's virtually impossible for them to have a below average snowfall season.

00z clusters sort of flipped from 24 hours ago. It was a 55/45 splits OTS vs tucked, and now that's 46/54. Members that liked the tucked clusters included 65% of CMC (), 55% of GEFS, and 48% of EPS. But that still just means that on the whole the ensemble suites remain pretty split on which way to go. The good thing is that the agreement between members that produce more QPF than the mean and those that don't. The flatter the flow the more like it is to scrape and scoot. Pump the downstream ridging a little and deepen the trof and the QPF is higher than the mean.

Heights across the Puget Sound area look pretty close to reality with the 12z GFS.

That aspect of the pattern is a limiting factor for sure. But if you can get a real dynamic deepening in our backyard it will want to cut across thickness lines and every hour counts.

Absolutely. GFS bias is more progressive anyway. But I can already see the sparkle in the weenies' eyes of 2 ft in 9 hours or something like that.

Great look to the mid levels, but boy that GFS is flying. In and out of BOS in less than 12 hours.

Not especially hard to rime snowflakes enough to produce a graupel-like hydrometeor, but there was just so much supercooled water thanks to the convection that more true ice than "Styrofoam" started falling.

He had an 80% chance of no pants on below that NBCCT graphic.

You know we'll get at least one run like this.

That's where all the variance is coming from too. That handoff from the North Pacific into the western "ridge". Still looking at tomorrow/tomorrow night when some balloons may be able to sample some of that. All the spread is in the area of a tucked solution. WPC cluster analysis is nearly split down the middle. 55% of the members farther east, 45% tucked in closer.

That's probably going to be the biggest story in post-event reviews. Not buying the snow threat wasn't great, but then no real mention of a power outage threat. I think you're being a little harsh there. Kuchera and 10:1 were pretty similar maps. But that depth map isn't that far off reported totals as it winds down. DCA reporting a snowfall and depth of 6.7/7. The over 10" amounts are just south of the DC area like shown on the 02/12z ECMWF. Maybe the depth is going to miss the highest amounts, but overall that map will be vastly better than the 10:1 or Kuchie maps. Of course that's why we don't like to forecast with clown maps to begin with.

I’ll never not think of tornado controversy with that town.

Nice MAUL down in AL. No wonder they are pulling TSSN.

Maybe Scooter can bag an OES band and still end up at 2 inches.