I’ve moved the forecast portion of my blog banter to the off-topic (political) blog, but my forecast is, unfortunately, going to remain very uncertain for the time being. A complex weather-maker is forecast to take shape over the Coastal Atlantic waters Saturday Night and Sunday which may potentially impact the heavily traveled I-95 corridor and points east with significant snows in the run-up to Christmas (making this a very important storm to forecast well since many travel plans hang in the balance).
As a “weather weenie,” I pride myself on getting the word out about an impending weather threat as early as possible with some minimum amount of confidence and then sticking by the logic and hard work that went into that forecast, rather than waffling with every model run. I believe, in fact, that the goal of every meteorologist should be to warn the public as early as he or she can while minimizing the costly false alarms. The National Weather Service does an unquestionably fantastic job under difficult circumstance, but I do feel that they fear being wrong a little much on occasion (because they face constant pressure to provide accurate forecasts and to avoid setting costly preparedness plans into motion when confidence in their forecast is not extremely high). As a scientist, however, certain storms force me to concede that there are still inherent limits on my ability to make early forecasts. This is one such situation.
I had originally planned to produce a snowfall forecast graphic for this event tonight after I’d ingested some of the global model output from the 7 PM runs, however, there remains a tremendous level of variability from run to run within any given weather model and a huge spread between the various models at any given time. For one forecast cycle (this morning’s model runs) we had good agreement that lined up well with my basic intuition about the kind of weather that should happen given the overall pattern in which we currently reside. You have all doubtlessly noticed how unusually cold it’s been in Central Asia, Europe and Eastern North America this month, for example. This extreme cold is related to a rare occurrence in the atmosphere. Three separate times now in two winter seasons, the temperature of the middle atmosphere has warmed far beyond climatological average for this time of year over the North Pole, forcing all of the cold air that would normally sit over the pole down over the continents to such extremes that we’ve had to define new scales to describe these oscillations of air masses over the pole. A rule of thumb in this business that usually works well – arctic air likes to leave a carpet (of snow) in front of it as it arrives. I’ve been expecting a major snowstorm to hit the Eastern US around this time for many weeks.
But pattern recognition can only take you to the point where you can make a pretty good guess about the position of troughs and ridges in the flow – it can’t give you a perfect forecast for any given event or any given time. Unfortunately, this high variability in the results of numerical model simulations is a sign that the success of this storm in producing heavy snow depends heavily on factors that are highly sensitive to the initial state of the atmosphere…and we don’t have good enough instruments, data recovery methods and observations to perfectly depict the initial state of the atmosphere. In other words, where this storm goes and how much snow we get can change dramatically based on tiny changes in the strength of very small rotations in the atmosphere…this storm is a strong argument for the butterfly effect. Tiny errors in our read on what’s happening in the atmosphere can grow to huge errors in even a 60 hour weather forecast in situations like these.
This is a long winded way of saying…I just don’t know yet whether this is a major event or a near miss. I got excited (as a lover of snow) when I saw the models come around to my way of thinking from two weeks in advance, but I have to admit, after looking at the most recent set of models, that I am not ready to make a specific snowfall forecast. Hopefully, I’ll have something for tomorrow night.
In the meantime, I urge any readers to pay careful attention to national weather service or media outlets regarding the evolution of this forecast. I will say that in my gut, I believe this will be a major weather maker for at least the New York City and Boston areas and a measurable snowfall further south into DC and Philly, but it remains very possible I could be wrong. Stay tuned.