Most of the models are in fairly tight agreement, with the exception being the NAM. Since the NAM has a very wet bias, it is OK to weight the forecast 90% away from the NAM at this time, and then if need be, adjust upward later on. Most modeling also agrees that the northern half of the state may see an inch or so more than mapped at this time, but not much more than that. Again, the exception is the NAM. If the NAM were to verify, blizzard conditions would probably be felt throughout the state. Again, the forecast is weighted heavily against that- for now.
So what should we look for? How do we determine which guidance is right and wrong? Most of the mesoscale guidance, such as the NAM, that bring very heavy precipitation into the southern half of the state and significant precipitation into the northern half of the state, have the snow starting much quicker than the other guidance. The faster the snow starts, therefore, the more accumulations we'll see.
The reasoning here is fairly simple... Obviously if snow falls for a longer time, there will be more accumulation, but it is deeper than that, too. If the snow makes rapid progress northward, it means two things: One, the low pressure system is moving further north, and Two, the moisture is encountering less resistance as it moves into drier air. So that is definitely going to be something to watch for and we'll know by midnight tonight.
So for now, the basic reasoning remains the same. Snow amounts will remain unchanged, but we will need to closely follow radar and observations to see whether the very snowy models are verifying or if the drier models are verifying, or if it's somewhere in between. No changes will be made for now, but for now, I would think small upward adjustments in accumulations would be more likely than downward, especially on the south coast.
The NAM will not be used much in this package, but because of its consistency and other models getting wetter, we should at least consider it in the backs of our minds as a possibility and watch to see what real-time trends to. But of course, it's a lot easier to adjust upward as the event unfolds if necessary, rather than make a huge adjustment upward and then scale back again.
There is going to be a huge cutoff to the snowfall amounts. It is entirely possible a place on the south coast gets 10", and just on the other side of the Merritt Parkway, only a couple inches fall. Neither of these amounts are in the forecast or expected at this time, but I simply am using this as a reference for Connecticut residents to point out how close the line may be between big accumulations and not much at all. Of course, there will be a very small area that receives something in between, but that area is going to be so small it is impossible to fit on the maps we produce.
It will become very windy along the south coast, regardless of how much snow falls. Winds could gust to 50 to 60 MPH right along the coast. This could cause power outages, especially if significant snow falls as well. Winds like that can cause power outages without any additional help, and if heavy snow falls in addition to those winds, it will make any power problems worse.
Coastal flooding will certainly be an issue along the south coast, as well, as tides are already astronomically high.
For now, the vast, vast majority of any precipitation that falls will fall as snow. There is a small chance of a few sleet pellets mixing in at times along the immediate south coast, as with almost any nor'easter, but the percentage of sleet anyone can expect at any given location and the chancy nature of it happening would yield me to just go with a straight snow forecast.
This product will be updated again in a few hours with longer term information, as well as some maps. I may very well update this again later on as newer model data comes in. If we need to update the snow map at all, that will also be done later.