12z Mid-Day Model Update

Just wanted to show what some of the mid-day model guidance is showing.  I'll have my second call map out tonight after 18z and some of the 00z models come in around 1030.

12z NAM
Here's a map showing 60 hour totals through 84 hours of the nam.  Notice the extremely heavy totals over much of central and south-east Virginia.  This run of the nam has an extremely tight gradient on the western side.  For example, Lynchburg gets well over 1.00" qpf which with 10:1 ratios translates to 10+ inches of snow, while an area like Blacksburg gets around 0.5" qpf (5" of snow with 10:1) and Bristol only 0.25" qpf (2.5" of snow with 10:1).

I do think the nam is suffering from some errors here.  Looking at the surface map, you can see that the low takes a pretty good track to give areas along and west of I81 more than just 0.5" of qpf.  What I think is happening is called convective feedback.  You can see a very high qpf area in purple just to the east of the low.  This can cause the model to get confused, and fail to determine where the low pressure placement should be.  You can see how this causes the low to become pretty elongated on the eastern side, putting more precipitation emphasis on that side of the low while taking away from the western half.

With said errors pointed out and out of the way, here's the 12z nam total snowfall.  Keep in mind that a lot of the numbers in central and southeast Virginia are skewed a bit because I don't think the model is doing a great job distinguishing between what falls as snow and what falls as a mix.  This still produces a good snowstorm from I81 and points east to Richmond.  West of I81, like I mentioned before, I think there's some feedback errors that keep the amount of precipitation too low.

12z GFS
So onto the GFS..  here's the total qpf it's outputting from this storm.  You can see the gradient isn't quite as sharp as the nam, but still generally 0.5" of qpf or less for areas west of I-81.

Looking at the surface, once again I believe the gfs has the same problem as the nam and is suffering from convective feedback issues.  You can see the higher qpf area in purple east of the low that just messes everything up.  I believe this is wrong.

12z gfs snowmap (although I think western fringe is still wrong and eastern areas are too high)

Now that the American models are out of the way, time to get onto the foreign models.

12z Canadian
Here is the 12z Canadian.  The track of the low is very similar to both the gfs and nam, maybe slightly west, tracking the low right up over VA Beach and just off Delmarva.  Unlike the American models, the Canadian doesn't have crazy qpf amounts to the east of the low, and instead has the more typical comma head like structure that we usually see with east coast snowstorms.  This allows precipitation to stretch as far west as eastern Kentucky and all of West Virginia.  This greatly increases amounts for far western VA west of I81.  I also think this does a better job showing the mixing scenario that will occur along I95 after a good front end thump.

5 hours later...

Total snowfall map from 12cmc...This is much closer to what I see happening, although I'm still not sure the I95 corridor in VA sees quite that much.

It's worth noting that the 12z ukmet is very similar to the 12z Canadian.  Still waiting on the 12z Euro to come in...I'll be updating this as it does.  Right now it appears to be american models vs foreign in regards to the western extent of precipitation.

1:15pm update: Euro 12z
The Euro is in and it is even more amp'd than last nights runs.  The Low tracks just west of the outer banks as a 996 mb low...before bombing out to sub 992 just off the Delmarva.  Like I said before, I think the American models are wrong here with how east they track the system.

Snow Totals from 12z Euro: I do think these are likely too high ....but shows more agreement in the western track.