There's no viz-ness like snow viz-ness
Look, I'm not happy about that joke either. But sometimes the creative process floats one over the plate and you just have to take a swing at it, earning a bloop single. Anyway, this week's theme is snow, which doesn't have anything to do with baseball, but the creative process can also pitch some nasty, unhittable stuff.
Chapter 1
When it has snowed in the past
I grew up in Rochester, in western New York. It's about an hour east of Buffalo, if you don't know, meaning that it is just slightly further away from the barrel of the gigantic snow cannon that is Lake Erie. While I don't remember a lot of massive blizzards as a kid, I do remember that snow was pretty much a constant during the winter. Enough to allow me to put off cleaning up after the dog for months at a time, for example.
It was enough, too, to inculcate a sense that snow was proper. That winter, and the Christmas season in particular, needed it. When I lived in California, I certainly didn't complain about the climate, but every October or November I would start to feel snow pangs, ones that would probably have dissipated rapidly if I'd had to, say, shovel a driveway every day, but I didn't.
So every year, I also look forward to the first big snow, an occasion that is frustratingly variable in its arrival. But I am me, as you might have gathered, and that means that I determined to figure out when I could expect that first snow to occur.
The National Oceanic and Atmospheric Administration (NOAA) has collected data on precipitation (and temperature and all the other things you'd assume) at weather stations around the world for decades. This allows us to track weather events over time in a useful way — like when it first snowed each year.
There are a few important caveats here. The first is that each station is or was in operation in a specific place for a specific period, meaning that there are often gaps. Another is that "first snow" is itself relative. Are we talking a dusting? A foot? As you can see, I set a baseline of a tenth of an inch, enough to cover the ground but not requiring an actual blizzard.
You can also see that there's a lot of variability. One year, the first snow wasn't until April. In another, it came in early October. You can also see that the earliest snow seems to have shifted later in the year, perhaps hinting at a pattern that we'll come back to.
But enough about me! I also pulled data from the 70,000-plus historic weather stations in the continental U.S., some 40,000-plus of which have records of snowfall. So here are the earliest and latest snowfalls on record near you, assuming you can find where you live on a static map of the lower 48.
As an aside (and in keeping with the putative intent of this newsletter), it was a pain to figure out the color gradient for these maps. I needed to make it easy to differentiate between early, middle and late snowfalls but also effectively capture the regional patterns. The final result is imperfect, but I think it works?
Anyway. What's interesting about the first map above is how you can see both the snow-heavy Rockies and the snow-less parts of Arizona and Florida. You can also see that there are a lot of outliers, little red dots (meaning later first snows) in a sea of blue (earlier ones). This is generally a function of stations with fewer historic records.
Now back to the change over time. To illustrate that, I made maps showing the average first snowfall by decade across the country. The number of stations with data, you will see, varied by decade, making comparisons imperfect.