Measuring Snowfall…er…Depth…er…

How Accurate Are Snow Measurements?

Southampton PA, USA — A recent blog article on the Philadelphia Inquirer website entitled Snow: A measure of history” raised the question about how snowfall depth is measured accurately.

Of course, anyone measuring something has to always(?) ask:

Is it accurate? Then How Accurate?

Since accuracy is a qualitative term, the answer a metrologist would expect would be quantitative. After all, if a measurement is expressed in units of depth, then the accuracy should be also in the same units with an uncertainty figure, such as: “The depth is 0.3 meters, ??0.1 meters or ??33%”.

That’s not the way meteorologists apparently report their numbers, and as the blog article implies, for some very understandable reasons.

Basically it’s hard to be precise in measuring snow depth.

To quote the article: “Snow measurement…is far more art than science”.

So what do they do and how do they do it?

The National Weather Service in the USA has a webpage entitled “Snow Measurement Guidelines” ( that describes their practice guidelines and a step by step recommendation to best record and report not one, but three parameter related to snowfall. They are:

1. Snowfall: Measure and record the snowfall (snow, ice pellets ) since the previous snowfall observation (24 hours). (See Snowfall section below for details)

2. Snow Depth: Determine the depth of the new and old snow remaining on the ground at observation time. (See Snow Depth section below for details)

3. Water Equivalent of Snow: Water equivalent of melted snow collected in the gauge since the last observation. (See Water Equivalent section below for details)

In an earlier blog article “Snow: A measure of doubt” by the same author, Tony Wood, the Weather Columnist for the Inquirer, this topic of measurement variables was rained, especially the revised practice guidelines of the NWS.

The latter are free and in a downloadable PDF format at:

BOTTOM LINE: Snow measurement is more of an art than science and takes a lot of educated estimates. The NWS has been refining the process and is striving for better, more consistent results, but nothing close, as far as we can see, to the stage where they can publish the types of data a metrologists would prefer to see.