As of the date of posting, meteorological summer is officially over and fall, which is essentially a transition period between summer and winter, has begun. As the traditional warm-season climatic fare, such as severe thunderstorms and 25+ C temperatures become scarce, I’m starting to pine for a white substance that hasn’t attempted to brighten the ground here in Possum Lake in nearly five months: snow.

That last occurrence was nothing to get worked up about, considering that it failed to accumulate, but the March 14-15 system was completely different.

The evening forecast on March 13 by Environment Canada echoed the predictions by the mainstream media and many of the recent global model runs (excusing a couple of seemingly fluke runs that depicted 60-75 cm in Montreal. this had happened several times that winter without results so the prudent move was to ignore them). Those forecasts were maintained through EC’s 11 am update the next day.

Winter Storm Bust

This forecast depicts an “ordinary” winter storm with 15-25 cm and wind. In most people’s minds, that would be a nuisance that would probably extend their commutes from 30 to 40 minutes or something like that.

Behind their backs, a short-range model trend emerged, reviving the extreme totals. It was not just a couple of rogue SREF members either:


An entire suite of ensembles was on board with over 60 cm in parts of the Montreal area (and most importantly, my backyard) as of the 00Z (8 pm) run on March 14 (13).

By 7 am the next day, I was rather elated to see that a consensus of short-range models from the NAM to the SREF were showing 60-70 cm totals. The RGEM and HRDPS were, unsurprisingly, somewhat restrained but still had me shovelling 30-40 cm if correct. I still had a dash of skepticism though because I still remember when the NAM spit out 70+ cm for storms in December 2013 and March 2014 which ended bringing me about 25-30 cm.

At 7:20, light, granular sub-atomic flakes started falling onto the barren landscape that had been freezer-burnt by a week of sub-freezing temperatures, including highs below -10.


After about 6 hours of this very light snow and growing anxiety, the snow intensified, the wind picked up, and it finally started to accumulate.

By 4:30 pm, blizzard conditions had started and road conditions became poor. 5 cm had fallen.


Snowfall rates were around 3 cm/hr at this point. The forecast from EC had been changed to a storm total of 27-32 cm as well, which will also prove to be almost as inadequate as the first call.

At 6 pm, the forecast changed yet again, to predict an additional 35 cm on top of the 11 that fell by then. That was a better call for most of Greater Montreal, but not all parts…


After 7 pm, the deformation band had become quite formidable over the south shore of Montreal, producing rates between 10-18 cm/hr for the next four hours whilst meeting the blizzard criteria at 8. The storm started to pull away in the overnight hours but continued to produce decent snows through 4 pm on the 15th.

On the morning of the 15th, I woke up to this:


In the end, what the general public thought would be a 15-25 cm storm ended up being a 77.8 cm storm. During the day on the 14th, over 65 cm fell, unofficially breaking the 51.1 cm 24-hour record set on March 4, 1971, and coming in second to the all-time record total snowfall of 86.3 cm, set between December 26-28, 1969, both set at YHU, a weather station that no longer reports snowfall (instead, they obscenely underestimate liquid equivalents).

The impacts were quite significant, inlcuding the closures of all public schools (including post-secondary institutions) the night before, something that hasn’t happened since the 1998 ice storm and a major traffic jam which saw drivers getting stuck in the snow for over 12 hours ( All in all, the area was completely shut down from the evening of the 14th until the next afternoon when things started to return to normal despite wraparound snows.

Here is the final summary for the storm in Quebec:

Weather summary
for Quebec
updated by Environment Canada
at 5:13 p.m. EDT Wednesday 15 March 2017.


A snow storm is affecting the province of Quebec since yesterday
March 14th. Exceptional amounts of snow have fallen and winds caused
widespread blowing snow giving zero visibilities. The system is
still active this morning, therefore this summary is only partial.

The following is a summary of weather event information received by
Environment Canada as of 5 PM on March 15th 2017.

1. Snowfall summary in centimetres

* Laurentians: 15 to 25
* Lachute-St-Jérôme: 45 (Oka)
* Mont-Tremblant: 12 to 15
* Lanaudière: 10 to 20 in the east, 40 in Repentigy
* Montréal/Laval 38 (PET Airport), 47 (downtown), 73 (Saint-hubert),
68 (Beloeil)
* Vaudreuil/Richelieu Valley: 50 to 75
* Drummondville-Bois-Francs: 60 to 70
* Eastern Townships: 35 to 60
* Quebec City: 30-50
* Charlevoix: 40 (massif)
* Mauricie: 15 to 20
* Laurentians Wildlife Reserve: 34
* Saguenay-East (Mont-Édouard): 30
* Témiscouata: over 30
* Kamouraska-Rimouski: 16 to 20
* Gaspé peninsula: 20
* Gaspésie provincial park: 18 to 20

Roads on the north and south shores of Québec and Natashquan are
either closed or have fair to zero visibilities. Roads on the south
shore between Villeroy in Bois-Francs and Gaspé are closed or have
fair to zero visibilities.

2. Maximum winds (km/h):

* Lachute-St-Jérôme: 50
* Lanaudière: 52
* Montreal/Laval: 70
* Vaudreuil/Richelieu Valley: 60
* Drummondville-Bois-Francs: 70
* Eastern Townships: 50 to 60
* Beauce: 52
* Quebec City: 80 over north shore, up to 120 over south shore
* Mauricie: 60
* Montmagny-L’Islet: 80 to 110
* Riviere-du-loup/Kamouraska-Rimouski: 85 to 95
* Matane-Cap-Chat: 90
* Saint-Anne-des-Monts: 70 to 80
* Gaspé/Percé/Chandler/Baie-des-Chaleurs: 70 to 89
* Baie-Comeau: 80
* Tadoussac: 90 to 100
* Pointe-des-Monts/Baie-Trinité: 107
* Sept-Îles: 60

Please note that this summary may contain preliminary or unofficial
information and does not constitute a complete or final report.


The YUL and downtown (8 km from Possum Lake) numbers were debunked by other observers who recorded 50-60 cm in those areas.

The final EC forecast did alright in areas such as Laval and the eastern, western and northern fringes of the island of Montreal where amounts were likely around 45-50 cm.

However, it is important to note that this was an incredibly difficult storm to forecast, which did highlight several important, overlooked factors that impact snowfall rates and storm conditions both locally and universally:

  • what can make up for poor ratios
  • the ideal track for a major blizzard in the St Lawrence Seaway
  • Why does the valley get the strongest winds during coastal storms

Forecast ratios were, in fact, rather high, as shown in this sounding from the height of the storm (8 pm March 14 in Maniwaki; in Montreal the temperature profile was similar)


As you can see, the dendritic growth/-12 to -17 C zone was 3.5 km (10 000 ft) high. The depth of the DGZ is the biggest factor in determining snow ratios; ratios of 40:1 have been associated with DGZ heights near 13 000 ft, so a conservative estimate of the ratios with this storm would have been about 20-25:1. Accounting for winds I predicted a ratio of 15:1. The low-level wind profile also depicts plenty of moisture advection typical of the SW section of a classic comma-shaped cyclone. The strong upper-level jet also favours deformation.

However, strong winds aloft were present in the St Lawrence Valley, oriented in a NE to SW direction, same as the river itself, resulting in the winds being channelled in a “wind tunnel” effect such as those experienced between high-rise buildings and under overpasses. Sustained winds peaked in the 30-40 km/h range in Sherbrooke and Ottawa, whilst they exceeded 50 km/h on the south shore of Montreal, gusting to over 80 km/h.

Consequently, flake fracture cut down on snow ratios, resulting in a recording of 11.53:1 at my location. However, the lower ratios did not prevent nearly 20 cm/hr rates from occurring because of the impressive dynamics within the DGZ that resulted in a very high precipitation efficiency. Deformation resulted in the stretching of snowflakes which meant that they were of a considerable size and fell at a substantial quantity.


A strong feed of moisture fed directly into the St Lawrence Valley as shown here and the 700 mb low was located in western Massachusetts which tracked into the Eastern Townships by the next morning. The heaviest snows typically occur about 200-300 km NW of the H7 low as shown by this snowfall total map:


Whilst orographic lifting is likely responsible for the 100+ cm amounts in NY, a distinct corridor of snowfall totals in the 68-78 cm range occurred on the south shore of Montreal east into the Richelieu Valley as the surface low was in the final stages of bombogenesis as it tracked from eastern Long Island to Rhode Island to Central Maine.


The strongest forcing is well east of Montreal in NS, but the 850 mb layer is above freezing there. Here it’s well entrenched in the DGZ, which means that there is strong forcing in that zone that will support the growth of many dendrites. Meanwhile, there is a transition between weak and strong mid-level winds in the Richelieu Valley, creating a deformation zone (reds, pinks and greens on the map; which cover everyone that got 60+ cm). As shown in the sounding, the low-level wind profile that evening was almost unidirectional, which is a factor crucial to the organization of long-lived, persistent lake-effect snow bands.

300 mb

The 300 mb pattern shows a great deal of divergence in a left-exit region, which helps structure the upper portions of the snow bands, much like the same strong jet-streak in the summer further tilting a storm’s updraft to help produce a very long-lived supercell. The strong winds stretch the plentiful snowflakes which, simply put, is part of the process of deformation.

Whilst this post is not intended to explicitly explain the science behind these snow bands, it should help other snow lovers determine where the heaviest snow bands are to occur and what is usually associated with them.