Over the last few days we’ve been mentioning that a storm modeled to pass our latitude on Saturday has steadily been trending west, and over the last couple cycles of models, we’ve started to see several of our forecast models showing direct hits on the region, especially on today’s 12z runs. In this update, we’ll provide an overview of the setup and lay out some of the scenarios depicted on guidance, as well as assign some weights to each to give you an idea of what we’re thinking. Pending this not falling apart on modeling over the next 24 hours or so, a first call map will be issued tomorrow (Thursday) night.
This storm is a fairly classic Miller A, that is, a traditional “coastal storm” that forms in the gulf and moves east to the Atlantic before turning the corner and coming upstream towards us. The low is formed by the initial phasing of two streams out west(which will happen tomorrow), and is then eventually pulled up the coast by the phasing of a piece of the northern stream into the low, tugging it north up the coastline instead of east and out to sea. The other piece of the puzzle is if the low can close off at the 500mb level of the atmosphere, because if it does, it would result in very strong dynamics, dynamic cooling, and a very heavy CCB(band of heavy precipitation), which our area would be right in the bullseye of. If the phase of the northern stream is delayed or is sloppy, the system will be able to escape out to sea as it will be weaker, and furthermore will not be as strong as the energy contained in the system will be less. In general, the models have the storm passing within the vicinity of the 40/70 benchmark on Saturday afternoon.
Also complicating matters is the fact that we will not have an antecedent airmass in place that is sufficiently cold to support snow to the coast, and that there is no high pressure to the north that would be able to supply this cold for us. As a result, we are looking at the storm itself to dynamically cool the column, which most of the forecast guidance shows. However, if the storm is too far east or is weak, we will not see enough in the way of dynamics to cool the column, and the end result will be rain for a good portion of the state. As of now we are not favoring that scenario, but it is a possibility and as of now the easiest way for this storm to bust.
The Models and Scenarios
As usual, the models can’t agree on anything, and this storm is no exception. Three distinct camps of guidance exist with this storm, each of which would bring us a different outcome. Let’s walk through them one by one.
Camp 1: Miss out to sea
The GFS and the UKMET have both been in this camp for their past several model cycles, with each showing a low placed much further to the east than the other models. This is mainly because the storm is much less organized aloft than on the rest of the guidance, and as a result the phase is sloppy, the low cannot strengthen and intensify, and as a result it is able to escape east and drop little to no precipitation on our area. Furthermore, because the dynamics are not there due to the low being east, any precipitation would likely fall as either rain or non-accumulating snow, perhaps a slushy inch or two in some spots. Saturday would be a mostly cloudy day with a chance of showers towards the afternoon.
However, we don’t believe this solution is likely for several reasons. First off, both the GFS(Especially since it was “upgraded”), and the UKMET are known to be erratic models, and both have been flipping and flopping all over the place with this storm. Furthermore, both are known to have progressive biases in their forecasts, and in recent events in this pattern both have verified too far southeast(For our rains last weekend, the GFS was dry at this lead time.). Additionally, they are outliers against the trend on all other guidance towards a further NW and amped storm, and generally one discards outliers when forecasting. For these reasons, although we can’t ignore these solutions entirely, we give them a 20% chance of verifying.
Here’s an image of the GFS low position and precipitation type/amounts. The UKMET is almost identical.
This camp has the highest concentration of the models, including the EURO and it’s ensembles, the NAM, the GGEM and it’s ensembles and the JMA. These models all have an organized storm aloft and a well-timed phase, which allows for a mature cyclone to come up the coast and pass ontop of or just west of 40 lat 70 lon. Most of these models either close off H5 or come close to it, and when combined with the dynamics, they are able to cool the column to a point where most of the precipitation would be heavy snow. There is a risk of coastal areas, especially in CT, mixing with sleet or rain for a portion of the event, but they would also be closer to the higher snowfall totals which would make up for the mix. In this situation, we’d see the lightest totals in the NW corner of the state where a couple of inches would fall, with a general swath of warning level (6”+) snows for the majority of the state, with amounts of 10”+ possible but confined mainly to NE areas. This is a blend of the NAM and the EURO given that I think the GGEM is too amplified, if the more amplified GGEM verified (15% chance), double those totals at least.
Snow would move in early Saturday morning and persist for most of the day Saturday before moving out late Saturday night. We would likely see a mix with or change to rain/sleet on the coastline before a shift back to heavy snow later in the day as the low bombs and temps crash east. Travel would be significantly impacted due to heavy snow at rates of possibly an inch or more per hour. Snow would likely be gone by daybreak Sunday.
As of now, we think this is the most likely solution for several reasons. First off is consistency. The vast majority of the models support this solution, including our best model verification wise, that is, the EURO. These models have been more or less locked into their track since yesterday afternoon, while the GFS has been all over the map in terms of its solutions. Additionally, the pattern that we are in has shown to support amplification as has the seasonal trend, with most storms tracking slightly further NW than modeled in the mid range and also verifying wetter than modeled(As is common in an El Nino, especially when the subtropical jet is active as it has been this winter). As a result, we are going with a 65% chance of this scenario verifying.
Here’s images of the NAM and GGEM models for the storm. Click each individual image to enlarge it.
While this is a lonely camp with only one member, it bears mentioning due to the fact that other models have been trending west over the last few cycles and that trend could continue. The NAVGEM model is well to the west of the benchmark, and as a result introduces warning level snows to NW areas and has a much more prolonged period of mix/rain for southern areas. Basically, the reason this would verify is that there is little stopping the west factors we talked about above from continuing other than a shortwave which forces the storm east, and if that wave weakens, the storm has open season to come west. As of now that wave remains strong on most models and as a result am not favoring this solution, but the end result would be warning snows for all areas, split in half in southern areas by six or more hours of mix/rain. Will weight it at 15% for now.
Here’s a shot of the storm on the NAVGEM. Precipitation type is not shown(So the storm isn't all rain!)
When combining these outcomes and tweaking these numbers slightly to match the preference of the forecaster, you get the following map.
As always, we love questions, so feel free to ask us either by leaving a comment on this post or the Facebook post. Until tomorrow, thank you for reading Southern Connecticut Weather!