You know how the Heisenberg Uncertainty Principal says (roughly) that we can’t accurately know the velocity and the location of a sub-atomic particle at the same time because measuring one alters the other? Measurement of snow depth is like that, too: You can’t know both the depth and the continuous time of fall because the two measurements alter each other.

A case in point is the lovely snowstorm that the Northeast just had. My daily measurements for CoCoRaHS (community collaborative rain hail snow network) takes place between 7 and 8 a.m. Yesterday at 7:30, I measured 8.9 inches of snow that melted down to .57 inches of water. Today at 7:30 I measured another 4 inches of snow that melted down to 0.39 inches of water. Also today I measured the total amount of snow on the ground: 10.5 inches, which melted down to exactly 1.00 inches. Note that it has stayed well below freezing the whole time.

Doing a little addition finds a paradox: The numbers of melted water are sensible, within my margin of error: 0.57 plus 0.39 is 0.96, which is pretty close to 1.00. But the snow depth is wacky: how did 4.0 inches of snow on top of 8.9 inches produce just 10.5 inches? Where did the other 2 1/2 inches go?

Compression. (A reader points out that the actual term is “compaction.”)

Snow has weight. Every new inch that falls pushes down on the inch below and, especially when you have light fluffy stuff like this storm, flattens it a bit. This doesn’t affect the amount of moisture in the snow but it does affect the accumulated height. As a result, your measurement of the the amount of snow that falls depends on how often you measure it.

Snow totals are measured on a flat, white, horizontal board not too close to trees or buildings. It’s called a snowboard, which confuses snow sports fans.

Yesterday morning I measured the snow height on the board and then cleared it off. I went out again this morning and did the same thing. At the same time this morning I measured how much snow was on the ground near the board – but that snow included all of yesterday’s total which had been compressed by new snow, meaning it was no longer as deep as it had been when I measured yesterday. That’s how 8.9 + 4.0 = 10.5.

Furthermore, the 8.9 inches I measured yesterday had already been compressed during the 10 hours of storm up to that point. Imagine if I had gone out every hour overnight and measured how much snow had fallen onto the board, then cleared it off. Totaling up all those hourly measurements would have produced a number larger than 8.9. And if I’d gone out every 15 minutes, the accumulated total would have been even larger! That’s why the National Weather Service sets standards about how often snowboards should be cleared and measured during a snowfall, to provide consistency.

So which one of those numbers is the “real” snowfall total?

There is no answer – the physical measurement is entirely dependent on the temporal measurement. There’s an objective measure of moisture content but height is subjective.

Remember this when you hear people arguing about which town got the most snow.