The National Weather Service has a constantly updated snow total map (see below) that shows how variable the snow was. Around 2 inches on Bainbridge and near Tacoma, but 2-5 inches in Seattle (I had 4 inches in northeast Seattle). But amounts increased to the east and north, with around 6 inches in Bellevue and Everett, and even more near the Cascade foothills.
The UW forecast model for the critical 24-h period (4 PM Sunday to 4 PM Monday) made Sunday morning only showed 1-2 inches over Seattle and Everett.
The forecast starting 12-h later (4 PM Sunday) was much better, but that was only available a few hours before most of the snow fell.
But this forecast was not an easy one (dependent on the existence of hard to forecast snow bands rather than larger-scale fronts), so I would argue it was fairly successful...particularly since the temperature and winds were fairly well predicted. Much better than we could have done a decade before.
The roads are a mess though--even major ones like 405 and its on-off ramps. Did the SDOT and WSDOT pre-treat the roads last night with deicer solutions? I didn't see such activity last night driving around, but I could have missed them. But pre-treatment is REALLY important in this kind of situation, starting with warm roads followed by a major chill-down.
We have some dry days ahead, but a another, possibly larger, snow event is predicted for Thursday/Friday. For example, bellow is the 24hr snowfall forecast ending 4 AM Friday. Huge amounts on the east side of Puget Sound, with 4 inches extending to Seattle. Too soon to be sure about this one, but keep it in mind.
Uh Cliff? Its still snowing on Beacon Hill.
ReplyDeleteYeah it's not showing any signs of stopping on First Hill either.
DeleteGlad to hear it's over but here in NE Bellevue it's still snowing hard with 12 inches accumulated so far.
ReplyDeleteSDOT is going to have to answer some tough questions after this one. I live on 15th Ave NE in north Seattle (a 4-lane arterial that carries 4 bus routes), and I see no evidence of any pre-treatment here. It's now just shy of 9:00 AM, and we have 3 inches or so of accumulation on elevated surfaces and at least an inch accumulated on the road, and it hasn't been plowed yet. One Bus Away has times that are all over the place, with the express buses running anywhere from 8 to 52 minutes late. I have decided that discretion is the better part of valor, and elected to telecommute today, but it's a mess out there on the roads for the folks that don't have that option.
ReplyDeleteWSDOT doesn't look like they did much better - the Seattle area incidents list is long on the website - currently 15 blocking incidents listed.
I know it's hard to maintain a proper snow response capability in a city where decent snow accumulation only happens a few times a decade, but the preliminary indication is that our DOTs could have done much better on this one.
The fun part, of course, is that Seattle closed the schools for a couple inches of snow.
ReplyDeleteFun for kids. Tough for parents.
DeleteIt is still snowing (not heavy, but kind of like a drizzle in Sammamish, and the clouds look like there is more to come. My patio snow height is about 6-7 inches, I have a feeling this isn't over :).
ReplyDeleteHmm, what is the Snowish word for “drizzle”? After 20 years in Seattle, I only speak Rain, as well.
DeleteSDOT laid down salt last night about 9pm. By 6am the road was snow covered. No follow up treatment of any kind yet. Still snowing in south Seattle.
ReplyDeleteWind is ferocious in Lynden, snow all but blown away, having coffee induced memories or spring robins chirping with nary a breeze. Also the two tone whistle of the black capped chickadee.
ReplyDeleteThanks for the bird report.
DeleteSnowing lightly and blowing moderately in tacoma, best snow day I have experienced in more than 5 years.
ReplyDeleteCliff, did the spinning low pressure not move to the south as originally anticipated? In Marysville we have about 7" and still snowing quite well.
ReplyDelete"Was fairly successful"
ReplyDeleteIt's time to start owning up to the GFS's failures. It was not successful. As you noted, the models predicted very little in puget sound, and had the smallest totals in the north sound. And that was less than 24 hours before the event!!
A big part of this is your messaging, Cliff. You're not properly conveying the uncertainty when you say things like, and I quote:
"Now, there is on outlier forecast that I should mention. The European Center model, which is lower resolution that the UW model, but often produces an excellent large-scale forecast, is going for much more snow over the interior (see below)---like 3-4 inches over Seattle for the period ending 10 AM Monday. This represents the extreme high end possibility."
The extreme high end possibility? No... forecast models can miss all in the same direction. There is substantial uncertainty in forecasting snow in the northwest, and you know this. To say "3-4 inches" was the extreme high end possibility is just poor messaging on your part. Seattle already has 4 inches and the amounts are increasing. Areas north and east of Seattle have substantially more, and you didn't properly convey that this was a possibility to consider.
It's not solely your fault of course, but these defensive post-mortems when you quickly try to say "hey actually the forecast was pretty good, honest!" while the snow isn't even done yet... they're incredible weak and ill-timed, to say the least.
I'm with you on this one
DeleteSame
DeleteHmm..after living in the PNW for 65 yrs. my rule of thumb has always been if they report that there is just a slim chance of a snow event...we get dumped on. Conversely, predictions of a whopper often dont pan out...😊
DeleteI just measured 10" at my Clearview home and it is still snowing. It is a beautiful light powder that I never (but would love to) encounter on the west side of the Cascades, as it is only 23F here also. Trying to forecast accumulation amounts must be near impossible if one has to consider water content.
ReplyDeleteNot ready for a postmortem here. Snow event is still going strong in Tacoma. It's been coming down steadily for the past couple of hours in central Tacoma!
ReplyDeleteI so appreciate your blog and especially enjoy hearing about what the different models are forecasting!
ReplyDeleteI have 6 inches on my parking spot(cement) and 7 inches on a hard dirt spot, 8 inches in the grass at Magnuson Park, on the lake. I walked through parts of the park is more snow. Absolutely insane amount of snow for sea-level in Seattle. I've lived in the city for 12 years now.
ReplyDeleteTo say the model was fairly successful I think is a bit off. The city and suburbs were hammered. 7-10 inches north of Seattle. This was a major event.
Yesterday it was written on this blog that .5 inches for Seattle/Puget sound and European was 3-4 inches. Now you are writing that Seattle was forecast for 2 inches. But we got hammered.
Forecasting is extremely difficult, but this forecast was wrong. That's okay, but gotta own up to it more and learn from it. Its almost as if nobody looked outside yesterday or called a friend. You were predicting .5 inches yesterday but people had already gotten that much.
Basically no snow here in Victoria (there's a dusting that could be mistaken for frost).
ReplyDeleteCliff, I don't see how this was fairly successful. I'm looking out over 8-10 inches in the upper half of the Issaquah Highlands. Correct me if I'm wrong, but we aren't supposed to get above freezing which means most of these inches will stick around for another day or so. With this in mind I have doubts about schools opening up tomorrow and perhaps even Wednesday (although it's too early to tell). Another system is coming (perhaps a stronger one you say?) which could possibly compound what we are dealing with already. I made it a point to stock up for a couple days over the weekend, but not for a week...
ReplyDeleteMost of the models missed the mark and that has real time consequences on the accuracy of preparedness. What is the point of accurate forecasting if not to be accurately prepared?
Still snowing pretty hard in west Seattle! Four inches in my yard so far .
ReplyDeleteThe temperatures Thursday and Friday are uncertain according to NWS. I wonder what's going to happen there, but it looks warmer, with temperatures rising instead of falling after the precipitation. I'm quite interested in that, since I have a flight Friday morning!
ReplyDeleteOnce again, the European model seems to have performed the best for a PNW snow event.
ReplyDeletePerhaps the other models can learn from this?
At my Bow place it is not unusual for the forecasts - when the Frazier outflow is involved - to understate the amount of snow and be a bit high on the temperature .. which is lower and stays lower longer. I think some of the uncertainty is due to the hilly terrain around Bayview ridge and the fairly extreme low tides in Padilla Bay (there was a minus tide at 10 pm last night.)
ReplyDeleteEric, the instability of the low and where it will go is very hard to predict. I feel Cliff does a great job at his craft and his uncertainty was clearly defined. He was implying the cold air mass and wind which was spot on.
ReplyDeleteGreat job Cliff
Re lack of pretreatment on roads, my guess is combination of lack of early warning plus superbowl. I am actually grateful for absence of "snowmageddon" media coverage. It is what it is. I would have hit the grocery store, but now I can take a walk in the snow today.
ReplyDeleteCould you comment on why the Euro model did so much better at this forecast? I'm reading a book on Hurricane/Superstorm Sandy and was struck by how the Euro model was so much more accurate than the U.S. models. I know they have more powerful computers but there has to be something in the software that handles uncertainty like snow and hurricane tracks better. Can you explain their "secret sauce" in a way that we might be able to understand?
ReplyDeleteWind blowing hard in Shoreline and power is out. This ain't over up here.
ReplyDeleteWhere can I find the snow coverage map? It looks fairly accurate if it indeed says there is 9" in the Woodinville area. I'd like to see more detail if I could.
ReplyDeleteAside from snow, this is a fairly significant wind event in the Bellingham area. Gusts over 40 mph at the airport started last evening at 7 pm and are continuing at that level. 50+ mph gusts started at 1 am and continued until 8 am. So, this is pretty long duration.
ReplyDeleteMy Davis weather station, which is somewhat sheltered compared to the exposed airport, peaked at gusts of 43 mph around 6 am, so we are about 10 mph less than the airport.
As far as winds are concerned, the NWS wind advisory was pretty accurate.
The most PSE power outages have been here in Whatcom County.
And it is still very cold (19F at 11 am).
The forecast temperatures were quite off for Redmond close in near Microsoft. The forecast called for just under freezing as the low point overnight (30-31) which would leave the surface too warm to accumulate much but in fact I woke up to 24 degrees. At 11am it's still only 25.
ReplyDeleteHow could you possibly argue that this was forecast fairly successful?
ReplyDeleteSeriously,I prefer to over-prepare. But this time, I decided to believe the less serious snow level predictions. Living on well water and on a steep hill, we normally pay attention if we get warnings. Stuck at home now, we can only start bottling water for when the now flickering power goes out.
ReplyDeleteThe big picture: If we wait for prediction "certainty", it's too late to really prepare. The same goes for "global climate change", getting vaccinated, and other potential disasters.
still snowing at 11:45am in Seattle
ReplyDeleteI got 6.5 inches by 8:10 AM today on the Bothell-Mill Creek line.
ReplyDeleteIs this is case where the European model was more accurate? Do you think it does tend to be more accurate for snow forecasts?
ReplyDeleteWoof! I understand the difficulty in forecasting PNW snow events, but this was a particularly brutal miss. For several days we were hearing about potential for an uncertain snow event, but forecasters insisted it was pretty sketchy and trace-3" at best, followed by some cold.
ReplyDeleteI also understand it was Sunday (and Super Bowl Sunday as well) but it was snowing at 5 PM Sunday and really didn't stop for 18 hours, blanketing the entire Puget Sound area and creating treacherous driving conditions, and much lower temps than advertised. Even in the middle of the event, forecast updates were slow to update (or failed to update at all), and the winter weather advisory wasn't really posted/extended until after the window had practically closed.
I got a gut feeling when it started snowing during the SB that the naysaying forecasts were wrong, and that gut feeling was more accurate than the sum of all forecasts that failed to recognize a snow event as it was already happening. Hard to call that a win, in my mind.
It also seems like the forecasts caused WSDOT to prepare inadequately. Even at 9 am in Pierce County most major thoroughfares had not been plowed or deiced, which is completely unacceptable for this type of event.
I absolutely agree, sure it was a tough storm to predict but why the slow updates and lack of advisory when it became clear things could be serious.
DeleteCliff, I'm a fan of your work and your efforts. But similar to Eric above...what separates pretty imprecise forecasts from just guessing? Or i guess better said, when the forecasts point to some number, say 1" along with some words about "with a lot of uncertainty", nobody listens to the latter part. So if you said, "we really don't know, but it might be 1 inch", isn't that a more realistic view of these snow events?
ReplyDeleteAnd/or, "when it comes to snow events" we really don't know..
I’m surprised there was no discussion as to the “why” of the miss. The original forecast right up to the afternoon update called for mixed snow and rain. As opposed to the European model, which had mostly snow. Instead, it seemed like in many places it never rained, going straight from a thin icy sleet to wet snow. And the nighttime snow band that both models predicted was significantly stronger and longer than predictions.
ReplyDeleteThe was a Capital Hill Weather Gang blog (Washington Post) entry after a similar event in DC that talked about the difficulties in predicting mixed precipitation. Basically the snow/rain boundary is very sensitive to small changes in temp and pressure and therefore the maximum errors are magnified.
That said, this was a bad miss. There should have been some warning 24 hrs out that this was in the realm of possibility. Particularly since daytime highs were correctly predicted to be solidly in the freezing range, meaning little to no help from Mother Nature melting snow. The lack of preparation has made what would be a very handleable event into a disaster.
If the NWS had used the EMS forecast (low res and all), the roads would be salted and the plows would be out. And that’s what proves it’s a bad miss: the consequences.
I think you could fill Husky Stadium with big misses on lowland Puget Sound snow forecasts. Anyone who has lived here for a few years comes to learn that. It is usually either cold and dry or warm and wet. Because snow is a marginal event, local effects and minor changes in altitude can make a big difference in accumulations. So, when there is some snow in the forecast and it actually happens, I think they are doing pretty well by historical standards. We just don't get enough of it so that people and drivers get enough to experience to handle it well, so any snow makes for a lot of disruption. Expecting local governments that hardly ever see this to be perfect is not very realistic.
ReplyDeleteI work in Green Lake and it was sticking there by 2pm yesterday. Here at home in Maple Valley, we've got 5" of light powdery snow that hasn't stopped since yesterday at 4pm, and a steady strong breeze. This is much more than I'd expected and I've lived here (Seattle and Maple Valley) my whole life.
ReplyDeleteThey had salted bridges over the highway by 1pm yesterday. There was salt all over Yesler into the ID. We know snow is hard to predict here. Kudos to you for tracking so many models and let us know how you came up with your prediction scientifically. Your audience is welcome to use the same evidence and come to a different conclusion, but they should not blame you for their lack of effort. Thank you for this sevice.
ReplyDeleteBoth with wind gusts and now snow depth, there's a large area around Kent, mostly east and SE, that is completely unrepresented when looking at totals.
ReplyDeleteThe entire hwy 18 corridor has only 1 spot with a number. I'm in the lea hill area, and this reminds me of the recent wind storms where we had 60mph+ gusts, but it was not mentioned anywhere in the news or these wind maps.
Now for snow as well.
I'm curious why this area is lacking representation about these weather events. Is it a lack of weather related equipment for the area?
Forecast was for wind, cold and snow. That’s what we got. Seems a little odd to be whining about minor misses when predicting the future. Have any of you seen a BBC UK national forecast? ‘Wintry showers on higher elevations across N England and Scotland’ is about as pinpoint as it gets. Somehow they manage and go about there day.
ReplyDeleteWell said PBBallard.
DeleteWhen the forcast predicts more damaging weather events that turn out to be mild people complain too. The world is full of experts.
DeleteI agree with all of the above. I hope we're beyond looking for a scapegoat!
DeleteLike DonB above, I'd really like a pointer to that snow coverage map. And yes, 9 inches in the Woodinville area is what we got.
ReplyDeletemaybe now that we're in the period of anthropogenic climate disruption, it's time to take more seriously those models (like the european one) which predict (accurately, it seems) more 'extreme' events than we might be used to here in the PNW....
ReplyDeletebecause the ability to forecast from the more 'conservative' models seems a tad inadequate now... that might be because we're no longer living in the world/climate system those models were built for... maybe the european model is more 'sensitive' to the new reality in which we find ourselves...
just a thought...
I grew up in Seattle living there from the 1930s to 1960 and nearby for most of the following years. Being interested in weather from early on, I followed the weather forecasts closely. Snow, being a relatively rare event, was seldom forecast accurately. I remember the big blizzard of Jan. 13, 1950. The forecast that morning was for snow changing to rain by afternoon with warming temperatures. After school that afternoon I walked home in about 12" of powder snow, blown into 2 foot drifts by 30 mph north winds and the temperature was 13 degrees when I got home. Now that was a major forecast bust. It is still difficult to accurately pin down snow events in the Seattle area, despite great improvements in computer models and the existence of satellite photos. It may always be thus, but thanks to Cliff Mass and other meteorologists at the University of Washington in their work in developing the current set of small-scale computerized models, the forecasts of snow and all other weather parameters in this tough-to-forecast area are much better than in the past.
ReplyDeleteJim Holcomb
Thanks Cliff. I was able to predict from your blog posts who would be out and who would be in to work at my company on N. Pacific Street in Wallingford on Monday and who would be out today due to ice. My staff live north and south of Wallingford, from Lynnwood to Renton.I've followed along especially since I caught wind of upcoming possible snow so I could have some prediction work load wise for this week.
ReplyDelete