A slush pool on the National Mall reflects the Capitol, February 14, 2014. The Capitol has a red glow from the light of the setting sun.  The snow melted quickly in Washington after the temperature reached 52 degrees during the afternoon of February 14.  (Kevin Ambrose)

It was a classic battle of the weather models.  Several days before the storm, the European weather model simulated a storm in the southeast U.S. that tracked it up the East Coast.  At the same time, however, the GFS weather model showed no storm.

Run-after-run, the models held firm with their widely different solutions to the future weather in Washington.  Storm or no storm?  Everyone knew that at least one of the models would have to give.

About two days before the event, the GFS “caved” and the model finally found the storm.  Unfortunately, after the GFS showed the storm, it simulated about 50 percent of the precipitation in D.C. that was output by the European and other models.  There was model consensus for a storm, but there was not consensus for the amount of precipitation.

A snowman with exceptionally skinny legs sits on a park bench at the National Mall, February 14, 2014.  This snowman has partially melted into the puddle that is located under the bench.  (Kevin Ambrose)

As the date neared for the storm, some of the weather models bounced around with the location of the rain/snow line.  That created additional doubt in forecasting the snow amounts, particularly for areas close to Washington, D.C.

Most weather models pushed the rain/snow line north and west of Washington, D.C. during the middle of the storm.   The rain/snow line was expected to move to about 25 miles west of Washington and then move back east of Washington as the storm moved away.

To additionally complicate the forecast, a dry slot was expected to occur during the middle of the storm followed by a band of potentially heavy snow to end the storm.  It was a tricky forecast that was covered in detail by the Capital Weather Gang.

Where to begin? The owner of this car in Oakton, Virginia evaluates the tremendous effort necessary to dig out after the snowstorm, February 14, 2014. (Kevin Ambrose)

Temperatures were very cold before the storm.  At least there was confidence that heavy snow would break out everywhere at the start of the storm but a changeover to rain was expected for D.C. and points to the east by 7 a.m. on Thursday, February 13.

As the storm approached the area, a few models showed truly impressive snow totals north and west of D.C. that exceeded 15″.  The GFS, however, held firm with a precipitation output which was significantly less.  Most meteorologists increased their forecast for snow totals, even exceeding a foot in the northern and western suburbs, but few forecasters went with the really big snow totals that were output by a few of the models.

As snow moved into the area, the GFS bumped up its precipitation amount.   There was finally model consensus that the area would receive a very juicy storm, but who would see all snow and who would turn to rain?  It became a nowcasting exercise to track the rain/snow line.

Two days before the storm, the European model had a good estimate for snowfall.

Heavy snow overspread the entire area during the night of February 12.  The temperatures were in the 20s.  But as dawn approached, National Airport flipped to rain at 35 degrees while Dulles Airport reported heavy snow at 22 degrees.  The rain/snow line moved into eastern Fairfax County but didn’t move west for hours.

By mid-morning on the 13th, the dry slot moved across much of the D.C. area and the rain/snow line quickly jumped as far west as Leesburg. But, by that time, most of the heavy precipitation had already fallen as snow across the northern and western counties.  The changeover to rain, however, reduced the snow amounts for the southern and eastern counties.

The storm ended with a burst of wet snow that ranged from a coating to two inches.  The areas to the north and west of Washington that received the most snow at the start of the storm fared the best with the additional snowfall at the end of the storm.  Southwestern Virginia did particularly well with the snow burst.  Roanoke reported six inches of snow falling within several hours.

A snow depth comparison timeline from Snochi and Snowmageddon. As a point of reference, the snowman is 19″ tall. If you look closely, the snowman has a National Weather Service ruler strapped to his side. These photos were taken near Oakton, Virginia. (Kevin Ambrose)

Ultimately, the models that showed the big snow numbers before the storm were closest to correct.  Areas north and west of the beltway received what can be called a historic snowstorm while areas near the tidal Potomac River and to the east received a major snowstorm followed by rain.  Here is a link to area snowfall totals.

It was a very complicated snowstorm that was forecast well by CWG.  After a vote, the storm was named Snochi.

Warmer temperatures followed the storm and high temperatures in the 50s were widespread across the area on Friday, February 14.  A significant amount of the snow melted, but many areas still maintained a substantial snow pack into the weekend.

The simulated radar for 3 am on February 13, 2014 by the NAM weather model. The NAM performed well.

The photos in this article show scenes from the National Mall, the Tidal Basin, Mount Vernon, and western Fairfax County.  Check out the different snow amounts at the different locations.

Hopefully, our next major snowstorm be a little easier to forecast!  Model wars and tricky rain/snow lines are tough on both meteorologists and snow enthusiasts.  But, as many of us know, winter storms in Washington are often very hard to forecast and rain/snow lines love to set up over the I-95 corridor.

Clearing the driveway in Oakton, Virginia, February 13, 2014.  The snow depth on the driveway was 16″.  The investment in the snow blower finally paid off.  (Michael Ambrose)

The HRRR model did a good job with a short range snowfall simulation.

A Snowmageddon and Snochi comparison: The left photo is from February 7, 2010 and the right photo is from February 14, 2014. These photos were taken at the same location near Oakton, Virginia. (Kevin Ambrose)

Slush and snow on the Mall. Both of these photos were taken February 14, 2014. (Kevin Ambrose)

The Jefferson Memorial at sunset, February 14, 2014.  Slush and ice covered the Tidal Basin.  (Kevin Ambrose)

Sunset at the Tidal Basin, February 14, 2014. (Kevin Ambrose)

Presidents’ Day at Mount Vernon, February 17, 2014. (Kevin Ambrose)

Opposite views of Mount Vernon on Presidents’ Day, February 17, 2014. (Kevin Ambrose)

Waiting for the snow to melt, February 17, 2014. Four days after the storm, this truck in western Fairfax County is still locked in deep snow. (Kevin Ambrose)

A snow pile occupies parking spaces at the Fair Oaks Mall, February 16, 2014.  Despite the warm-up in the forecast, large snow piles will remain in our area for many days.  (Kevin Ambrose)