Saturday, December 13, 2014

December 19-21 Potential Winter Storm

Model guidance is supporting the risk of a winter storm in the December 19-21 timeframe.
I'm trying to decide how to start off this post, since simply going through model guidance won't work. I'll begin by explaining each graphic, doing a compare/contrast as we do so.

Tropical Tidbits
The latest GFS model has a trough pushing into the Midwest on December 20th, as the blue colors and depression of contour lines shows. This trough is beginning to close off after becoming negatively tilted, and is pushing northward as a result. The surface low takes a track through the Ohio Valley on this run, dropping appreciable snows from east Kansas into southern Michigan. North Missouri sees snow over 6" from this system.

In this overview of the pattern, we can diagnose a few items arguing for this more inland track, as future guidance I'll show you will depict an East US track. First of all, we have another strong trough dropping into the West US, trying to work its way southward. This will try and force a ridge to develop in the central and eastern Rocky Mountains, as we can already see above. However, as we see a deep upper level low over Greenland, this ridge won't be able to exert too much influence (that ULL over Greenland defines the positive phase of the North Atlantic Oscillation/NAO, notorious for keeping the jet stream very 'non-wavy')
What does make a ridge form, however, is the troughing in the Bering Sea into the Gulf of Alaska, a textbook positive East Pacific Oscillation (EPO) pattern, which will act to limit cold air reserves in Canada, but also try to direct storm systems northward. The latter influence is seen well in that ridge centered over New York, just east of the storm system in question. That same northward influence could happen along the East Coast, but for now, the GFS favors this solution.

Tropical Tidbits
Here's the ECMWF 500mb geopotential height anomaly forecast, valid at the same time as the GFS graphic. We see more than a couple significant differences here.

First and foremost, the storm system in question seems to be developing a second piece of energy, signified by that second dip in the contour lines along the Gulf Coast. Looking at the forecast following this timeframe. it looks as if that second piece of energy will act to pull the main trough east, and develop it into an East Coast system. This is far different from the GFS, which maintains a single trough.
Additionally, we see the entrance time of the second trough into Western North America has been slowed by about a day. This allows the ridge in the central and eastern Rockies to flourish well into Canada, where the ridge up there is also quite a bit stronger than its GFS counterpart. Consequentially, the storm system is suppressed to the south in the ECMWF forecast. Also, note the lack of a ridge to the east of the storm system in the ECMWF image, compared to the GFS. This, combined with the extra piece of energy along the Gulf Coast, appears to favor an East Coast solution. The solution results in this pattern, valid 24 hours after the image above:

Tropical Tidbits
Purely for comparison, here's the GFS snowfall forecast I mentioned earlier, expressing the solution in opposition to the ECMWF:

Instant Weather Maps
I've been looking back and forth between the ECMWF and GFS images I've shown above, trying to figure out which one I think is the most valid, and I can't decide.

On one hand, the ECMWF projection appears to be obeying the positive EPO signal, as exhibited by troughing along the west coast and a strong ridge in central Canada.

On the other hand, the GFS is doing well with the emergence of a ridge just east of the trough, possibly as a result of the negative PNA orientation out west, as the aforementioned second storm system drops into the West.

It will ultimately depend on the timing of when the trough drops into the West, just how strongly the atmosphere responds to the +EPO signal, and (of course) if that secondary piece of energy forms along the Gulf Coast. It's worth noting none of the model guidance is having any consistency with the ridge in Canada, purely by looking at run-by-run comparisons. Additionally, today's 12z ECMWF run (examined above) is the first one to have that secondary piece of energy develop to the south of the storm, something that does not bode well for any consistency that either the GFS or ECMWF may have built up. I'll pass on giving my opinion right now, because this is a truly grotesque set-up.

To summarize:

- A winter storm is possible for the Central or East US in the December 19-21 timeframe.
- Model guidance is expressing little to no consistency on a defined track for this storm.
- Cold air availability will eventually become a concern.
- Anomalously low confidence exists.

Andrew