# A better way to measure snow

I’m sure you’ve either over heard or taken part in a conversation that goes something like this:

A: Can you believe it was 110 F in Las Vegas today?

B: That’s high, but it’s also a dry heat, so it doesn’t feel as hot .

A: True, but it’s still very hot.

To help resolve these comparisons, we use the heat index, which accounts for the relative humidity and temperature. At the other end, we have the wind chill, which predicts how cold you will feel by using wind speed and temperature. The key thing that comes out of each of these estimations is an adjusted temperature that accounts for the wind speed or humidity to give you a real-feel temperature.

I think these approximations are a great way to easily convey many pieces of information into a single number.

So what does this have to do with snow? Well, perhaps it’s because I’m a relative novice to the whole concept of shoveling sidewalks, but what I want to know when it’s going to snow is not how many inches of snow I will receive, but the amount of time and effort it will take me to clean my driveway and sidewalk before I need to head to work.

Unfortunately, inches (or feet if you are in Buffalo) is not an effective measure because snow can be light and fluffy or heavy, wet snow. I haven’t survived too many winters, yet, but it’s clear to me that the heavier snows are much more difficult to clear.

The relevant metric should not be inches of snow, but mass of snow. (It’s at this point that we could consider a more complicated formula with wind speed, to account for snow drifts in this driveway-clearing snow metric). Getting everyone to refer to snowfall by mass is a loser’s battle, just look at how difficult it is to get Americans to use metric units.

So, I propose a snowfall index similar in nature to the heat index or wind chill in that the snow total is adjusted by the density of the snow to give a real-feel snowfall total.

We already have tabulated the snow density based on the temperature, so it’s really a mater of combining the temperature at which the snow falls with the total inches in a meaningful way. I will start by picking a reference temperature of 10 F. This temperature will be used to provide a mass equivalence for the snow index.

Using 10F as a baseline, I used the liquid equivalent to determine a snow index factor to multiply any snowfall total.

SFI = (Actual Snowfall)*(3e-5*T^3+1e-3*T^2+1.9e-2*T+6.968e-1)

Actual snowfall is in inches and T is in degrees F. It may be aggressive to use the equivalent density to adjust the snowfall totals. After all, 3 inches of snowfall at 32F would give a snowfall index value of 9 inches, which may be a little extreme, but snowfall totals are in desperate need of a wind chill equivalent, and this is my first stab at it.

So, will it actually take me 3x longer to clear snow at 32F than 10F? I doubt it, but this winter, I will put it to the test.

Advertisements