Snow for Alabama: Can the Models Do the Talking?

December 24, 2010 · 0 comments

Alabama hasn’t had a lot of luck with White Christmases. In Birmingham, for instance, it’s only dusted snow once, and that wasn’t an actual measurable accumulation.

It’s no wonder then that the prospect of snow tomorrow has sent chills of excitement down the spines of Alabamans who have had their eyes glued to the computer models this week.

On his AlabamaWx blog, ABC 33/40 broadcast meteorologist James Spann and Tim Coleman have been trying to temper excessive expectations for days now, even while trying to patiently explain the promising but multifarious model output. On Wednesday, for instance:

There is very little skill in forecasting winter storm events in Alabama until you get with about 48 hours of the event. Nobody knows the exact snow placement and amount this early. Even the know-it-alls don’t know, even though they will never let you bebutlieve it (those of us that have been doing this a long time professionally have had enough doses of humility over the years to be firmly out of the know-it-all camp). We can begin talking accumulation placement tomorrow when that 48 hour window opens up.


The NAM and the GFS, the two primary American models, show very limited moisture, and not much more than a dusting of snow for the I-20 corridor. The deepest moisture will be over the southern half of the state, where initially the precipitation will fall in the form of rain.

The ECMWF and the GEM, the European and Canadian models, are a little more bullish on moisture for North Alabama, but it is still limited. Both of these models suggest enough snow to get 1/2 to 1 inch on the ground. Which, if happens, would be historic for Birmingham. Up north, everybody would completely laugh at the fuss this is creating.

All this talk about what the computers say apparently gets a violently different reaction from folks depending on the stakes. Therein lies a lesson in communicating with science as we approach a meeting devoted to the topic. Wrote one commenter yesterday on AlabamaWx:

Yeah I’m.pretty upset that one.minute the models are right on for a winter storm then the next it flakes out. It literally crushes a lot of peoples wants and all but atleast we did have a chance a day ago! Now its all a good memory.

Just the day before, Spann wrote:

I am amazed at the angry tone of e-mails this evening… some are simply livid that I am not predicting a big Christmas day snow storm that would be historic for Alabama. I will probably never understand why winter weather brings out such passion and emotion. Seems to be more intense every year. Never was like this in the “old days”… one guy called me an “idiot of historic proportion” because “his forecast” was for 6 inches of snow for Birmingham. Wow.

Apparently, despite the cool dispassion of mathematics and computers, it is actually easier for people to rant at computers churning out uncertainty than at humans making reckless guesses. As Coleman cautioned on Tuesday:

Honestly, I don’t know exactly what will happen, and I teach Atmospheric Dynamics.  This is a complicated system, with multiple distrbances and wind maxima aloft.  These computer models are calculating Newton’s laws of motion, thermodynamics, and radiative transfer laws, at thousands of spots across the northern Hemisphere, every 1 to 5 minutes.  A lot can go wrong.

Now that it looks like snow will indeed fall as far south as central Alabama (still probably just a dusting), feelings are smoothing over. No need to bash the models and go out on a limb.

In fact, the models are actually doing what some humans at one time might have missed entirely: showing the potential for snow even though this is not a typical Alabama snow situation. Spann explained this morning:

This is not the classic “Gulf low” snow setup for Alabama… instead the snow will be squeezed out by the upper trough. There isn’t much moisture to work with, but this kind of setup can coat the ground with snow pretty quickly, especially if there are snow showers involved. And, we have seen forecast soundings from time to time that actually show 100 j/kg of CAPE, meaning the air will be unstable enough for showers of snow. Watch the Weather Xtreme video and you will see the RPM model shows a nice batch of light snow/snow showers over much of North and Central Alabama late tomorrow afternoon.

Snow from Gulf lows just happens to be the topic of a student poster presentation at the upcoming AMS Student Conference (Sunday 23 January). Ismari Ramos of the University of Puerto Rico and J.M. Coyne studied the paths of the usual Gulf lows as they approached Alabama with wintertime moisture.

A southwest to northeast track pattern was seen for these systems. All the west Gulf lows have different routes as they move inland. Roughly 37.5 % of the storms enter through Alabama, 29.2% through Florida, 25% through Louisiana and Mississippi with 8.3%. Upper level features like troughs play a major role in the intensification of associated surface lows. Many of the cases showed similar characteristics of strong jet stream winds approaching the base of the upper level trough which helped to strengthen both the trough and associated surface low.

So hooray for the models for boldly doing what they’re supposed to do, climatology aside. Maybe mentiong models doesn’t help becalm the public, but without models such exquisite anticipation and gnashing of teeth would not be possible between meteorologists and their fans. As Cliff Mass points out in a summation of the role of models in forecasting on his blog last week and how they’ve advanced meteorology lately,

So next time you are thinking about making fun of meteorologists or laugh at the classic jokes (you guys get paid for forecasting 50%!, meteorologists have a pair of dice in the backroom, I wish I could get paid for being wrong all the time, etc), bite your lip!

Or at least hope Birmingham gets some snow this time around.