Friday Storm Surprise


Well isn’t this storm a real treat to forecast. My forecast from yesterday is going to bust in a big way. No denying that – this forecast has been a train wreck from the beginning. While we’re much better than we used to be – meteorology continues to be a humbling science!

All that said, I for one am excited about a Friday morning snowstorm! This storm has trended stronger and farther west just about every run since Tuesday. The reason why is that a powerful upper level disturbance will swing through and produce a rather expansive – and impressive – swath of snow.

There will be a few issues to contend with here. One issue is going to be timing the transition from rain to snow – and even a bit of sleet. This forecast sounding for the area around New London shows that potential around 4 a.m. tomorrow with the atmosphere hovering around 0c.

Screen Shot 2016-02-04 at 8.12.13 PM

Expect a real mess tomorrow morning across the state. Heavy, wet snow will cause issues and I expect roads to be quite messy in most areas. There are several things that make me quite bullish on this event – including our short range ensembles all showing a very significant storm.

Screen Shot 2016-02-04 at 8.45.24 PM

This plume diagram from BDL shows mean precipitation around 0.8″ with many higher members! That would be approximately 8″ of snow though some will be lost to mixing at the onset.

I expect this storm will be a fun one for us weather geeks and a real pain for commuters tomorrow.

Tough Friday Morning Forecast

Some of our computer models have quite an intriguing solution popping up Thursday night and Friday morning. For example, the ECMWF (Euro) shows at least 6″ of snow in southeastern Connecticut by Friday morning! Other models, such as the GFS and NAM and our in-house WRF, show little if any snow in Connecticut.

What we’re dealing with is something called an anafront. This type of cold front is where precipitation falls on the cold side of the boundary. I’ve included the NAM computer model showing the location of the front at 4 a.m. Friday (note the wind shift way offshore and the Theta-E gradient nearby).


Now, here is the accumulated precipitation between 1 a.m. and 4 a.m. Friday. Notice how far the precipitation falls northwest of the front’s location! Virtually all of the precipitation is way northwest of the cold front.


So why does this happen? One of the reasons is that there is a strong upper level disturbance scooting through Friday morning which provides a source of lift you can see that as we reside in the right entrance region of a pretty impressive jet streak (~180 knots).


These anafronts are finicky and difficult to forecast. The northwest extent of the precipitation is always a bit of a question mark and some of the models tend to overdo how far west the precipitation gets. We’ve seen many of these events in the past and they frequently tick east or offshore at the last minute – but not always!


This is really a tough one. There’s a large spread in guidance and I’m honestly not sure what to expect. The most likely scenario is the one I’ve highlighted above. A reasonable worst case scenario would be 1″-3″ in Hartford/New Haven and 3″-6″ in eastern Connecticut. An equally plausible scenario would be little/no accumulation anywhere in Connecticut with the steady snow staying east.

We shall see!

Friday Fun?

Could we be dealing with another storm later this week? There’s the potential for one with the European model ensembles strongly hinting at that potential. 

You can see a pretty classic east coast storm. This is an ensemble mean which is showing a number of members close to the coast and an equal number offshore. One thing to note is there’s not a tremendous amount of cold to the north (no strong/cold) high pressure system so if the storm were to hug the coast you could envision precipitation type issues at the coast.

While the Euro ensembles are bullish the GFS ensembles are not. The Euro ensembles show a 40% probability of seeing 0.5″ of liquid in Hartford (that would equal about 5″ or 6″ of snow) the odds on the GEFS are much, much lower.

I wouldn’t be surprised to see this tick further northwest but there’s a whole lot of uncertainty still here. Stay tuned!

What a Storm!

Whiskey and Hippo enjoy the snowstorm in Southington.
Whiskey and Hippo enjoy the snowstorm in Southington.

For some areas – this storm was an overperformer. In New York City – it was a historic overperformer! In northwestern Connecticut – nary a flake to be found.

So, what was the deal? As expected the storm tracked way south (farther south than most blockbuster storms in Connecticut do) but the large circulation was able to produce a band of very heavy snow across the south coast. Along the Massachusetts border, dry air in the boundary layer effectively killed the storm. Check out this observation from across the border in Westfield, MA during the height of the storm showing a huge temperature/dew point spread.

KBAF 232353Z 02014G20KT 10SM OVC034 M03/M16 A2979 RMK AO2 SLP096 T10331156 11022 21033 50010

Very dry air was able to sink south across the border with gusty northerly winds. You can see where there is snow and no snow on the ground from this morning’s visible satellite image.


So – which model “won”? No model ever wins. I think meteorologists in general do a poor job explaining how we use computer modeling. At any given location we can show nearly a dozen of site specific model snowfall forecasts – most of which are useless when used alone. While we do look at raw model precipitation output there’s a whole lot of “value added” by meteorologists including knowing which models tend to work best in a given scenario, watching out for pitfalls like dry air in the boundary layer, looking for signals for mesoscale banding, forecasting snow crystal growth and efficiency, etc. If we just look at the raw model QPF forecasts you can see that no model did particularly well – but one model did quite poorly. The NAM!

Screen Shot 2016-01-24 at 8.22.19 AM Screen Shot 2016-01-24 at 8.19.04 AM

Our forecast for the event was decent given the tight gradient. As I’ve mentioned our forecast was pretty much a Euro ensemble/GFS ensemble blend which was relatively consistent. Here’s a list of official and CoCoRaHS observations.

  • Milford – 11.4″
  • Shelton – 10.7″
  • Portland – 10.0″
  • Bridgeport – 9.5″
  • Prospect – 8.7″
  • Southington – 8.5″
  • Moosup – 7.5″
  • Dayville – 7.3″
  • North Grosvenordale – 7.0″
  • Berlin – 6.8″
  • East Killingly – 6.8″
  • Oakdale – 6.1″
  • West Hartford/Elmwood – 5.5″
  • Brookfield – 5.1″
  • Westbrook 5.0″
  • East Hartford – 5.0″
  • Staffordville – 4.2″
  • Newington – 4.2″
  • Norwich – 4.0″
  • West Hartford (2.7 NNW) – 3.8″
  • Stafford Springs – 3.8″
  • Enfield – 3.1″
  • Enfield – 2.2″
  • Collinsville – 1.7″
  • Windsor Locks (BDL) – 1.5″
  • New Hartford – 0.9″
  • Litchfield – 0.4″
  • North Granby – Trace
  • Norfolk – 0.0″

Below is our initial forecast from Wednesday afternoon (almost 72 hours out). We gradually increased totals from Hartford on south and never changed northern areas – in fact decreased totals. I’ll grade the forecast as a B or B-… not the best but not bad all things considered.

Initial snow forecast released Wednesday afternoon