Possible snow?

11:00PM update:
The 00z NAM came out wetter and farther west with the rain snow line over us, but the freezing line to the east, meaning a major ice storm. It is useless to worry about the rain snow line at the moment though. The line will shift around in the coming days. The simulated radar is quite impressive though.

The GFS gives us nothing again. Why is it so insistent on forcing it out to sea?

It would be interesting if the GFS turned out to be right...

ORIGINAL TEXT:
We got a little snow tease today. About 1/2 an inch of snow fell across the region. The is what is left of what had the potential to be a big snowstorm a week out but alas, the timing was off and it never got started. Next week, another storm is forecasted by models to have the potential to impact the Mid Atlantics somewhere around the wed/thurs timeframe.  As usual, the models have had issues with resolving the different pieces of energy that may come together to form the storm. There are a host of issues this storm has to overcome. The energy that  will hopefully come together to form the storm are still over the Pacific ocean. since this is a place where there is very little data collection, leading to initialization errors, which will add up.  Second, timing and temperatures will be an issue, as they have been all winter. The arctic high that is pumping cold air will be retreating east as the storm approaches. If the storm comes late, the high might warm us above freezing and we will get rain. The final issue is storm track. If the storm tracks too inland as the 0z Euro advertised, then we will get rain as warm maritime air wraps around the storm. If the storm tracks too far east like the GFS is saying, then we will get nothing. any model runs should be taken as guidance. Nothing will happen exactly as the models show.

Snow maps are calculated with 10:1 ratios. Actual ratios may be lower due to temperatures and melting. Any snow that falls will be wet and heavy.

ECMWF(Euro)
The Euro has been amazingly consistent with giving us a snowstorm even from a week out.

The 00z Euro tracked the storm further inland than previous runs, which would give us a couple of inches of snow before turning over to rain. This inland track is quite unlikely since initially there will be 2 lows, one in the gulf and one off the florida atlantic coast, that are fighting to gain dominance. Only if the low in the gulf low assumes dominance quickly will the storm track this far west.

The more recent run, the 12Z puts us in the bulls eye, with an unusually strong 988 Mb storm off the coast. The 12z Euro depicts an all snow storm with liquid equivalent precipitation in excess of 1 inch. This is the reasonable "perfect" scenario. 

The 12z ensemble mean lends strong support for the operational run that gave us the "perfect" snowstorm, with a swath of heavy snow up the east coast. The ensemble runs are low resolution runs using slightly different data. This attempts to account for possible data errors. The mean takes the average conditions produced by all the ensembles. If it matches or is similar to the the operational run, then the event is more likely to happen than if the operational run varies widely from the ensemble. The wider the variation, the less confidence the the run there is.

NAM

The NAM ended while the storm was still impacting the region but it draws in warm air during the storm, possibly inducing a changeover to sleet or rain, which cuts down on snow totals in the DC metro and points east while dumping on the areas west. The low in this case hugs the coastline. This warmer scenario is the most likely seeing that temps have been a problem all winter, when major precipitation events are in the vicinity. One thing I have against the NAM is that it is a short term, high resolution model and it tends to get very inaccurate beyond about 48 hours. This storm is at the end of its run, at 84 hours.

CMC
If the Euro is perfection, the CMC is the worst case scenario. ever. It bombs the low off of NC and brings it up the coast to Ocean city as a 981 Mb low then brings it father up to New Jersey as a 969 Mb low and finally across the Canadian Maritimes as an 961 Mb low. This is an absurd strength for a cyclone at this latitude. This is only Due to the strength of the low and the track off the coast, temps will not be an issue although the a shift east will mean we are left out of the heaviest snow. To give it credit, it forecasted the big rainout last week while the other models were screaming snow plus most of the other snow events. The 12z CMC run can be summarized as "high off crack" but it shows that the ingredients are there for an amazingly strong storm if the timing is right.




GFS
The GFS is the outlier here, bringing the out to sea. The 18z run was an improvement, giving us 3 inches using the 10:1 ratio (too high). The 18z and 06z runs are done with less data than the 12z and 00z, meaning that they will be less accurate.  However, there were no major differences between the 12z and 18z runs. The gulf low is not strong enough and is drawn out to sea by the FL low and give us close to nothing. The GFS has been very inconsistent this winter and quite inaccurate with mid to long term forecasts. At least it has backed off the notion of a big rainy noreaster.


Summary:
There is a potential for a major snowstorm but at 4 days out, it is impossible to know what will actually happen. The storm's track, strength and precip type will be affected where the arctic high is, the speed of the system and whether it can get organized fast enough. To get the snowstorm, we want the storm to organize quickly, move in while the cold air is still in place, and stay long enough to drop copious amounts of snow. I recognize that this is a tall order here in the Mid-Atlantic. Hopefully the GFS will join the EURO/NAM/CMC camp and forecast some more snow.

Thanks for reading!

-Alex