« Noises on! | Main | !yretsyM a s'tI »

Snow Stats

So this is going to turn into a different note than it was when I started it. You see, at some point yesterday I looked at the snow totals for the season here in the Hartford Area, wondering if my sense that it was quite snowy this year was an accurate sense or whether it was unduly influence by this year being more recent than many previous years. As seen on xkcd, of course, only not so much the climate change aspect (because (a) I am already convinced, and (2) I understand that a snowy winter don’t mean shit for the models, and (iii) I have been following sports in Australia this past summer and understand that the planet is round and that the axis is tilted from the plane of revolution) as the perceptual whatnot aspect.


There it was, the report up to and including the Fourth of February, and the season’s snowfall total was 25.4 inches. The normal snowfall up to that point is 23.5 inches, so I said something quite like We’re only a couple of inches over normal this year. Correct! Not wrong! My previous non-statistical impression that it was a snowy year was called into question by the facts! Because, undeniably, we were only a couple of inches over a normal year, and what’s a couple of inches? One light snowstorm that wouldn’t close the schools, that’s all.

So, there I was, vague impressions corrected by facts. And then I started thinking: If normal snowfall over that period is 23.5 inches, and this year we had in the same period 25.4 inches, that’s an increase of—what—eight percent? That’s a lot, isn’t it? We were eight percent above normal snowfall! It’s a very snowy year! I was right the whole time! Snow had been falling at a much greater rate than in most years, just like I thought. Quite snowy this year is an excellent description of eight percent above normal. Facts! Statistics! So much better than subjective opinion, eh?

Hm, now. Which is it? A couple of inches, that is, nothing to write home about, or 8%, that is, a goodly chunk? I have no idea. I mean, I could find out, presumably, with access to what are very likely public data tables, whether 8% above normal puts a year into the top quintile (super snowy), the second quintile (quite snowy) or the middle quintile (meh). Or I could just look at the standard deviation and say ooh, outlier or enh, normalish. But without that? Averages have betrayed me once again.

The thing is, though, that there’s a different lesson to learn about measuring the year-to-date versus normal, which is that as I was looking it up, snow was falling, and not the 0.4 inches that is normal for the Fifth of February but 10.2 inches. So as of the end of that day, we are at 35.6 inches total as against the benchmark of 23.9 inches normal, or (rounding up to) an extra foot of snow. Or, if you prefer, almost half again the normal allotment; fifty percent above average! Wow! A very snowy winter indeed!

Or… if I was counting pre-hatch chickens on Wednesday morning, are the chickens coming home to roost now? If a normal year is around 40 inches of snow—is it? I’m having trouble finding the official data, but it looks like that’s about right. So we’re only at 35.6 inches. If we only get 4 more inches of snow before Spring, we’re under normal for the year. Is that likely? Well, it wouldn’t be a crazy outlier, having one more small snowstorm in February and two more in March, and only a dusting in April. It’s not unheard of. It wouldn’t even make the news, I think.

Tolerabimus quod tolerare debemus,


Data on how long the snow stays on the ground and how deep it has tended to be would also be relevant to an examination of how snowy a winter is. If you got 10 inches of snow that melted steadily away over the next three days, and then went a week without snow, and then got 2 inches of snow that stayed on the ground for two days before melting, and then no snow for a week, you would have gotten a foot of snow but only had snow cover for 7 days, with an average snowpack of around 1.3 inches/day. On the other hand, if you got 6 inches of snow, which stayed on the ground, and then 10 days later you got another six inches of snow on top of that, and all of that snow stayed in place to the end of the a three-week period, then you would have gotten the same amount of snow as the first case, 12 inches, but the average snowpack would be around 7.7 inches/day, and the snow piles would be much more impressive, too.

The NWS does keep snow height stats (currently seventeen inches, although of course they aren't counting the part where we've piled the snow to get it off the driveway) but the historical snow height doesn't show on the report. I agree that the really relevant stat is days-since-snow-height-was-zero, also known as where's the lawn? and that at least here in southern New England the real mark of a snowy winter is that the lawn is gone for the whole season. That would be terrific information on that history sheet as fuel for the whinging. This year, we had visible lawn all the way until the big storm a couple of weeks ago, interrupted only for a few days at a time, which may have accounted for it not feeling like a particularly snowy season.


Is it a nice powdery snow that you can easily shovel, sweep, or move with an electric snowblower? Or a dense painful snow (perhaps with a layer of ice mixed in) that hurts to even think about tackling?

Or as the kids think about it: Is it a useless powdery snow that just blows up in your nose? Or a lovely packable snow that you can use to make snowballs and snowmen and igloos?

Or as the folks with bad knees think about it: Is it a high traction snow that has fallen over clear pavement? Or a low traction snow that is obscuring large patches of ice?

My sense of a snowy winter is how many days the snow has truly interfered with what I arbitrarily felt like doing that day, and how good a day the backup plan turned out to create. NWS is doing a terrible job of tracking that.

Comments are closed for this entry. Usually if I close comments for an entry it's because that entry gets a disproportionate amount of spam. If you want to contact me about this entry, feel free to send me email.