Surprise Snowstorm for Western Washington

WOW! What was expected to be nothing more than a few scattered snow showers for Western Washington ended up being a bona-fide blizzard for the Northern Interior and a major lowland snowstorm for most locations north of Olympia. I was actually up in Seattle for the weekend doing some recording with my funk band Circular Reasoning (stay tuned for our second album!) so I was lucky enough to see some flakes fall, and even luckier to make it back to Portland before the real heavy snow started falling late Sunday night.

Even though most models were showing a paltry trace-2″ for most of Western Washington (with the notable exception of the northern Olympic Peninsula, where a Winter Storm Warning was in effect for up to 8 inches of snow), there was a significant amount of uncertainty in ensembles. Additionally, the ECMWF (European) model was going for several inches of snow throughout Puget Sound – much closer to reality than the paltry amounts the American GFS model and even the vaunted University of Washington WRF (which is initialized from the GFS) model were showing, providing yet another example of the superior accuracy of the ECMWF compared to all other global models.

So, what exactly went wrong with the forecast, and why did Washington see so much more snow than modeled? Simply-stated, it was because a persistent, poorly-predicted band of snow stalled over the I-5 corridor. More specifically, a feature known as a deformation zone strengthened over Puget Sound and sat over the area for hours on end. Moreover, exceptionally strong northeasterlies poured out of the Fraser River Valley into Western Washington, cooling temperatures well-below freezing and ensuring that snow stuck to roads at sea level, not grassy surfaces above 300-500 feet as was forecast.

Let’s start by just taking a quick frame-by-frame look of the general evolution of the large scale pattern as predicted by last night’s 00Z GFS run. This run actually did a decent job with snow accumulations, but seeing as the model finished running around 9 pm PST, it wasn’t particularly useful from a forecasting and preparedness perspective for the Monday morning commute!

The slideshow below shows the surface low pressure system tracking along the coast and weakening in the process. At first glance, this looks like a poor track for lowland snow, as the low is too far offshore to bring much moisture into the region. However, the southward track of this low created a northerly pressure gradient, giving the absurdly cold arctic air in British Columbia an opportunity to flood across our northern border.

Of course, the weather over a region isn’t just determined by the conditions at the surface. Let’s dig into something a little less intuitive but far more enlightening – some 500mb absolute vorticity charts. 500mb (millibars) represents a pressure level in the atmosphere that corresponds to roughly 18,000 feet (meteorologists like to use pressure instead of height as a vertical coordinate), and because density is a function of temperature (as described by the ideal gas law, or, as it’s known in the atmospheric science sphere, the “equation of state”), colder airmasses have lower 500mb “heights” than warmer airmasses.

Vorticity is a more complicated calculation – I still have nightmares with vorticity proofs from my atmospheric dynamics courses at the University of Washington – but it basically just describes the tendency of an air parcel to curl.

The following charts show “frontogenesis” over the I-5 corridor, which is a fancy word for the strengthening of a front. This is shown by the gradually strengthening vorticity gradient extending from Bellingham south through the Willamette Valley. The mechanisms by which this frontogenesis occurred are well above my pay grade, but from reading the National Weather Service forecast discussions it sounds like it was affiliated with a “deformation zone” aloft, which is a complicated concept that describes how air parcels are distorted in different directions as a result of convergence and divergence in the upper atmosphere. Shout out to Logan Johnson of the Seattle NWS forecast office for pointing me in the right direction here!

When we take a look at 6-hour precipitation totals ending at 10 pm Sunday and 4 am/10 am Monday, you can see the precipitation associated with the surface low itself as well as the precipitation that lingers over Western Washington Monday morning due to this frontogenesis/deformation zone.

Remember that intense northerly pressure gradient I was talking about? Well, it resulted in some insane winds for the Northern Interior. The Anacortes/San Juan Ferry clocked a gust of 90 mph earlier today, and Bellingham had windchills of -4 earlier this morning! And though pressure gradients and winds have relaxed compared to earlier today, temperatures are still well below freezing across Western Washington. With a fresh coat of snow and mostly clear skies expected tonight, temperatures should drop into the low 20s for Seattle tomorrow morning, with even colder weather in outlying areas.

More snow later this week?

Cold and dry weather is expected to persist through Thursday, but more snow looks likely for Western Washington Friday and Saturday. Portland will be on the fringe, but areas above 500-1000 feet may see sticking snow during that time frame. Much colder than-average weather should persist next week. After taking a long leave of absence, winter has finally arrived here in the Pacific Northwest.

And last but certainly not least, here are some pictures of the snow around the region!

And finally, some eye candy from NASA’s polar-orbiting TERRA satellite. This shot was taken yesterday and shows “cloud streets,” which are formed when cold air moves over relatively warm water and picks up moisture in the process. If these look familiar, it’s because they are the same phenomena I talked about in my last blog on the polar vortex that brought record cold to the Midwest last week.

Thanks for reading! Hopefully my next blog will feature Portland instead of Seattle. 🙂

Charlie