I have been measuring daily precipitation for eight years for CoCoRaHS, a “citizen science” weather data-gathering program. The biggest thing I have learned is to not take snow measurements very seriously.

The days are long gone when I’d brag that my town got 12 inches in the last storm and your town only got 11. Now I know that both those numbers have, shall we say, significant error bars.

The problem, of course, is that snow blows around. It piles up here and is dispersed from there, a pattern that can differ from storm to storm depending on the direction of the wind. Plus, over time it settles under its own weight – it can make a difference whether you measure the minute the snow stops falling or whether you wait until the following morning.

The result is that most of the time there is no real answer to the question “how much snow did we get?” – the only answer is “I measured XYZ inches at this specific location at this time.” And I do mean specific location, because even when there isn’t much wind snowfall depth can vary by two or three inches in locations that are just a couple yards apart.

There are definitely days when the measurement I take at my usual location seems to under- or over-count the inches of snow that fell on my property. I use the figure anyway because the more important thing is to be consistent over time, a relative value rather than an absolute one. That’s why I measure at roughly the same time every morning, as well, even if we’re in the middle of a storm.

I bring this up now because of an interesting (if rather wordy) report from NOAA about what was allegedly the biggest-ever snowfall in Pennsylvania: A staggering 50.8 inches that was said to fall in 24 hours in Erie.

You can read the whole thing here. In summary, they rejected the claim because the snow observer hadn’t been trained in NOAA’s systematic methods (e.g., use of a “snow board” that was cleared four times a day for total accumulation) and had never done snow measurement before – in partiuclar, the observer held down the board with a traffic cone to keep it from blowing away, which could alter how much snow accumulated there. Also, other measurements nearby were suspiciously different, and the snow wasn’t melted down to give backup data on moisture content.

Next time you hear all these suspiciously precise numbers from TV weather folks about snowfall, take it with a grain of salt. Road salt, I suppose.