How much snow did we get? You might be surprised how complicated the task of snowfall measurement, or estimation, can be. There’s more to it than simply sticking a ruler in the ground — and it’s an inexact science.

The weather station on the Montana State University campus is one of roughly 9,000 U.S. stations that comprise the National Weather Service cooperative observer network. Measurements occur only once a day at these stations, at 4 p.m., in the case of MSU.

Think of what can happen in the world of snowfall, during the 24 hours between measurements. Those snowflakes don’t just sit there after they fall. They’re constantly changing from the moment they hit the ground, due to a multitude of factors.

If the ground is above freezing the flakes may melt. Snow compresses and compacts under its own weight. Then there are the effects of wind, scouring snow from some areas, depositing it in drifts elsewhere. Finally, there’s sublimation — when ice crystals change directly to water vapor, bypassing the liquid phase altogether.

With the aim of making snowfall measurements comparable and consistent from station to station, the NWS says that observers should report the maximum snowfall accumulation during the 24-hour period. Sometimes that can be problematic.

In a simple case, say that it snows hard all day long, but the temperature is slightly above freezing so that most of it melts almost immediately. As far as the observer knows, no more than an inch of snow ever actually accumulated. The daily snowfall would be reported as one inch.

Now another scenario. The observer notes 0.3 inches of liquid precipitation in the gauge at 4:00 p.m., and is aware that the precipitation fell as snow overnight, when the temperature hovered around freezing. But that snow has melted, and the observer has no idea about the actual maximum accumulation. What now?

Observers can estimate snowfall using a snow ratio, which converts a known amount of liquid precipitation to inches of snow. For example, at temperatures near freezing, a ratio of 10:1 is used – that is, 10 inches of snow to 1 inch of liquid. As the temperature drops, the ratio increases. At 10 degrees, the ratio jumps to 30:1.

In our example, the observer could multiply the liquid precipitation by the applicable snow ratio (10:1) and come up with an estimated snowfall of three inches. However, what if the ground was warm and some of the snow melted immediately, which would have reduced the maximum accumulation? The observer should take that into account. I told you it was an inexact science.

A quick note: preliminary November data from MSU indicate that the streak of consecutive months of cooler than average high temperatures has now reached 18.

Greg Ainsworth keeps an eye on local weather and climate. Contact him at