14-15 December 2013 Snowstorm

This storm was a double barrel (Miller Type 'B') storm with the main upper low tracking from the southern plains through the Ohio River valley and southern Great Lakes. Meanwhile, the upper jet dynamics were sliding across the central and southeast United States out onto the middle Atlantic region with a secondary coastal low developing off the NJ coast and tracking out to Cape Cod. This storm took place over a weekend which minimized impacts from schools and normal business traffic across the region. However, it was the 2nd to last weekend before Christmas in terms of shopping potential for the region.

The main impact from this storm occurred during Saturday night. Initially, the snow was driven by isentropic lift with widespread light snow developing throughout the day Saturday. By Saturday night, all of the models were indicating several frontogenesis bands to impact the region. The complicated factor was the extent of potential mid-level warm air entrainment across our southern zones as we and the Weather Prediction Center were expecting a period of ice across our southern zones. While little to no freezing rain occurred, sleet was the primary winter type as this shifted northward to the south and east of the immediate capital region. As the secondary low developed, the mid-level wind field also increased with the Berkshires and southern Green Mountains reported wind gusts to between 40 and 50 kt. The Bennington, VT ASOS (KDDH) had a wind gust to 58 mph early Sunday morning, and North Adams (KAQW) also had a gust to 46 mph. As was mentioned in the Area Forecast Discussion a couple of times, near blizzard conditions were possible and these wind gusts and accompanying visibilities we might have approached this criteria. By definition, blizzard conditions are 3 consecutive hours of heavy snow or blowing snow with visibilities of 1/4 statute miles or less with sustained or frequent gusts of 35 mph or greater.

Much of the model guidance (outside a few of the GEFS ensemble members) suggested that the coastal low would track south of Long Island and then northeastward towards Cape Cod and eventually the Gulf of Maine. Throughout the forecast process, it was difficult to discern to what degree of potential phasing there would be between the northern stream system across the southern Great Lakes and the southern-stream dominated coastal low. A continued great way to discern phasing potential between two separate low pressure features is by looking at the PV trace (typically PV at 300mb works fine) through time utilizing the Volume Browser. As the event neared, both the GFS and ECMWF depicted a slight "treble clef" appearance to the northern stream Potential Vorticity anomaly, which favors a "capturing" of southern stream energy. This semi-capturing of the southern stream energy allowed the coastal low to actually track on the westward (closer to Long Island) spread of the guidance, allowing slightly warmer midlevel air to advect northwards into the Capital District and even as far northwards as the Lake George/Saratoga region. This resulted in riming of snowflakes, especially towards the end of the event. Although this did lower snow to liquid ratios down from 15:1/even 20:1 in some locations early in the event, the capturing of the coastal low allowed the best FGEN in the 850-700mb layer to pivot across our forecast area, leading to the intense mesoscale band that moved across the Capital District during Saturday night.

The frontogenesis band that developed Saturday night was quite impressive with over 40 dBZ returns tracking from south to north. At times, our snow rates were 2 inches per hour and with temperatures remaining in the single digits and mid teens, the fluff factor was high. Outside of lake effect, the combination of these radar returns along with very cold temperatures were likely the result in the higher snowfall totals that were observed. However, as the night progressed, and the midlevel warm air crept further northwards, more riming occurred with snowflake growth resulting in a decrease in snow to liquid ratios. When the 12z snowfall and rainfall measurements were taken Sunday morning, the snow to liquid ratio was actually found to end up near climatological normals (exactly 10:1).

This was a multi-band event, with a final intense single band forming over the region early Sunday morning. The placement of this intense single band was difficult to pinpoint 24-36 hrs before the event. Most of the guidance initially kept this potential band/banding south and east of the Capital Region, which ended up being further north and west. The other odd thing about this event is that the 850 hPa/700 hPa/500 hPa low never closed off over the region. The secondary wave was very intense with strong bands of snowfall Saturday night into Sunday morning.

2. What was learned from this event:

Dual-pol radar was utilized during the event to determine that the very high reflectivity returns where in fact all snow (and not melting flakes or sleet). This helped give confidence for snow rates when issuing Special Weather Statements.

Towards the end of the steady precipitation, sleet was reported for several hours at Albany airport. While there was sleet just south and east of here across the mid-Hudson valley and Taconics, model soundings, observations at our office and dual-pol radar didn't indicate sleet here or at the airport. At the time when the sleet was reported, there were heavily rimed snowflakes occurring at our office, which was a result of snowflakes falling through a heavily saturated layer. These flakes could be mistaken for sleet, as they were coated in a layer of ice. The flakes were white in color and easily accumulated with the rest of the snow - proving it was in fact snow (sleet would bounce, would be translucent in appearance and not freeze into a solid mass). In addition to providing inaccurate information to the public and aviation interests, this misidentification could have an impact on winter storm research and the ongoing QPE Field study. This is something for people to keep in mind during future events, as the observation may not meteorologically be exactly what is occurring.

The primary low ended up being stronger and had more of an impact than models were indicating in the days leading up to the event. In fact, the frontogenesis associated with the primary low was clearly separated from the frontogenesis caused by the coastal storm during the evening hours of Dec. 14. The 00-06 hour NAM40 and GFS40 analysis/forecasts on Dec. 14 depicted this nicely. Moisture and additional ascent from the coastal low impacted the area more during the overnight hours. Even though heavy snow with rates of 1-2 inches per hour occurred for several hours, these "bands" were fairly large and were oriented more west-east and tracked south to north. Banding from this event would fall under the "Transient Band" classification. This seems to fit with the conceptual model from the research which would make the primary low the main player for much of the event. Further evidence for the transient band motion classification is seen from snowfall totals, which were fairly uniform in a fairly large area where banding occurred and coincided with the max snowfall (see GIS snowfall map).

 

Forecast Model Guidance 1 to 3 days in advance - Note the significant differences in storm track,

precipitation placement and low-level jet characteristics from the various sets of guidance.

 

Above:  Four-panels of MSLP from the GFS (upper left), ECMWF (upper right), GFSEnsemble (lower left) and 850 hPa temperatures from the GFS and ECMWF (lower right) showing the 144 hour forecast (left), 84 hour forecast (middle) and 48 hour Forecast (right) valid at 12Z 15 December.

Above:  Four-panels of probabilities for 1 inch liquid equivalent precipitation (top GFSensemble on left and top SREF on right) and 850 hPa winds (bottom GFS on left and bottom NAM on right), forecasted at 84 hours (left) and 48 hours (right) valid at 12Z 15 December.

 

Above:  Four-panel of quantitative forecast precpitation (QPF) from the 1200 UTC 13 December GFS (upper left), NAM80 (upper right), ECMWF (lower left) and NAM12 (lower right).

Above:  Time sections for Albany, NY from the 1200 UTC 13 December GFS (left) and NAM (right).

Above:  Time sections for Glens Falls, NY from the 1200 UTC 13 December GFS (left) and NAM (right).

Above:  SREF probability for > 4" of snow in 12 hours from the 1500 UTC 12 December SREF (left) and 1500 UTC 13 December SREF (right).

Above:  Loop of GFSEnsemble MSLP and anomaly ending 1200 UTC 15 December.

Above:  Loop of GFSEnsemble Precipitable Water (PWAT) and anomaly ending 1200 UTC 15 December.

Above:  Loops of GFSEnsemble 850 hPa winds and anomaly (left) and 925 hPa winds and anomaly (right) ending 1200 UTC 15 December.

Above:  Loops of GFSEnsemble 850 temperatures and anomaly (left) and moisture convergence and anomaly (right) ending 1200 UTC 15 December.

Above:  GFSEnsemble probability for 0.50" liquid equivalent precipitation in 24 hours (left) and 1.00" liquid equivalent precipitation in 24 hours (right) ending at 18Z 15 December.

Above:  Loops of plume diagrams for Albany, NY from the SREF ending at 0300 UTC 15 December (left) and GFSEnsemble ending at 0000 UTC 15 December (right).

Above:  Weather Prediction Center 24 hour snowfall accumulation 50th percentile, 48 hour forecast valid 1200 UTC 15 December.

 

Forecast Model Guidance at maximum impact to the NWS Albany forecast area - Even within 12-24 hours

from impact, the sets of guidance were still showing noticeable differences in forecast output.

 

Above:  Forecasted heights and anomalies at 500 hPa from the 1500 UTC 14 December SREF (left) and 1200 UTC 14 December GFSEnsemble (right).

 

Above:  Forecasted MSLP and anomalies from the 1500 UTC 14 December SREF (left) and 1200 UTC 14 December GFSEnsemble (right).

Above:  Initialized temperatures and anomalies at 850 hPa from the 0900 UTC 15 December SREF (left) and 1200 UTC 15 December GFSEnsemble (right).

Above:  Initialized winds and anomalies at 850 hPa during the warm advection phase of the storm from the 1500 UTC 14 December SREF (left) and 0000 UTC 15 December GFSEnsemble (right), dominated by V wind anomalies.

Above:  Initialized winds and anomalies at 850 hPa during the upper deformation phase of the storm from the 1500 UTC 14 December SREF (left) and 0000 UTC 15 December GFSEnsemble (right), dominated by U wind anomalies.

 

Above:  Forecasted probability of 1.00 inch liquid equivalent precipitation in 24 hours from the 1500 UTC 14 December SREF (left) and 1200 UTC 14 December GFSEnsemble (right).

 

Above:  Plume diagrams for Albany, NY from the 1500 UTC 14 December SREF (left) and 1200 UTC 14 December GFS Ensemble (right).

 

Above:  Plume diagrams for Binghamton, NY from the 1500 UTC 14 December SREF (left) and 1200 UTC 14 December GFS Ensemble (right).

Above:  Plume diagrams for Monticello, NY from the 1500 UTC 14 December SREF (left) and 1200 UTC 14 December GFS Ensemble (right).

 

Observational data

 

Above:  Soundings for Albany, NY from 1200 UTC 14 December (left), 0000 UTC 15 December (center) and 1200 UTC 15 December (right).  Much of the saturated layer was within the dendritic growth zone early, then warmed later, but still remained below freezing.

 

Above:  Soundings for Upton, NY from 1200 UTC 14 December (left), 0000 UTC 15 December (center) and 1200 UTC 15 December (right).  There was more steady warming of the column at Upton, NY.

Above:  Satellite loops of Water Vapor Imagery (left) and Infrared Imagery (right).

Above:  Regional radar loops from the National Centers for Atmospheric Research (left) and College of DuPage (right).

 

Above:  Surface map loop.

Above:  GIS snow map of the NWS Albany, NY forecast area.

Above:  NESIS category 2.