How do they do that?

[lemon boy]

So (Hacksaw) how do the SNOTEL sites measure the water content?

[pechelman]

not Hacksaw, but there are a few ways it could be done.
Pickup the Snow Booklet if you're more interested

I think its got something to do with the fact that permittivity, dielectric constants, and surface absorption coefficients/albedo all varying with water content and density of the snow.

but ill just stop there because i dont really know.
if hacksaw doesnt hop on here before this evening, ill look at my copy of the Snow Booklet when I get home. I remember there being like 3 or so different ways to analyze water % via surface techniques.

[FrankZappa]

Weight I would assume. [/shrugg]
or - Chipmunk's gotta have somethin to do in winter... And CM labor is cheap.

[Hugh Conway]

Here's a company that sells instrumentation describing some SnoTel sensors
http://www.rickly.com/MI/SnowCover.htm

Pickup the Snow Booklet if you're more interested

Most worthless snow related publication ever

[pechelman]

if its so worthless, why was I able to provide some insight on this question?

its a book about snow measurement, which honestly, isnt a very interesting subject, so I dont expect it to get widely acclaimed. Its an easy read, concise, and helps to explain somethings with a reasonable amount of detail with regard to meteorology, snow morphology, and snow measurement.

i particularly liked the graphs in the early section of the book relating snow grain structure, snow density, and snow water content to temperature. I thought those were pretty USEFUL.



edit:
more info on deh pillows. prize goes to FZ!
http://www.wcc.nrcs.usda.gov/factpub/ads/adsintro.html

[lemon boy]

per that site then it's simply a measurement by weight (displacement). thanks.

edit pech- the pillows at the bottom talk about measuring water content.

[pechelman]

yea i see that now
didnt scroll down all the way

did some googlizing and found this
so I wasnt too far off on my other methods
http://ieeexplore.ieee.org/iel2/1064/7705/00322643.pdf

[pechelman]

and to take your question a bit further LB

How accurate are these pillows when the snowpack gets really deep?
Since a lot of the weight is then supported by the snowpack itself, it seems like the pillows would become less accurate as the pack gets deeper and has stronger\harder layers to distribute the load away from the sensor.

edit: partial answer to my question
http://books.google.com/books?id=izs...kn-k4#PPA37,M1

[lemon boy]

I'm guessing that, as they're positioned in flat spots usually, the bridging doesn't have too much of an effect given that the entire area is under the same new load.

One of the reasons that I ask is generally it seems that the WC measurement is a more accurate predictor of how much new snow you're likely to find than the snowfall measurements (which IMO tend to underestimate new snow by a good ways)

[mmcpheet]

and to take your question a bit further LB

How accurate are these pillows when the snowpack gets really deep?
Since a lot of the weight is then supported by the snowpack itself, it seems like the pillows would become less accurate as the pack gets deeper and has stronger\harder layers to distribute the load away from the sensor.

This is true as bridging can occur over the snow pillows. Many of the sites (at least in California) also share portions of a snow course, where a snow surveryor takes a monthly manual measurement using a snow tube set. The results of the snow course are then used to provide a "ground truth" of the snow pillow readings.

Some more info here: http://cdec.water.ca.gov/snow/info/HistSnowSurvey.html

[pechelman]

sounds like bridging effects are mitigated by making a larger array of pillows by butting them up against one another. it does sound like pillows typically overestimate snow from a few various effects such as snow creep, moisture\runoff consolidation & pooling, and issues with the pillow\tubing\hydraulic lines\air undergoing thermal expansion & contraction with ambient changes.

so you just use WC?
seems like thered be a lot of error in that when assuming a set percentage water in the snowfall. (ie 10% or something)

or do you do something else?

[lemon boy]

Tis a dark and unwholesome art, guessing snowfall by water content. But I usually use around 5ish% as the average and adjust up or down from there based on temps. Yes, it is very crude. But, I'm also not often surprised to find a foot + when the snotel says 3-4" but is indicating .8" or more of WC. :shrug:

For instance, look at the berthoud site:
http://www.wcc.nrcs.usda.gov/nwcc/sn...ays=7&state=CO

There is simply no way that they got .7" of snow between the hours of noon and 11pm on Nov. 30 given that the pillows picked up .5" of WC. 5-6" is way more likely (warm storm).

And that's not to say that I don't look at the snowfall amounts, if nothing else they're good indicators of base and b/c I think they systematically underestimate snowfall when they're showing a good pileup you know that it's going off.

[lemon boy]

Oh and it is definitionally a much better indicator of new snow load.

[ronco]

So (Hacksaw) how do the SNOTEL sites measure the water content?

The pillow measures snow pack weight and some sites make an ultrasonic measurement of total snow depth. If you then look at the deltas in weight and depth of some time period, say 12 hours, you can get a pretty good estimate of the density/water content of that snow fall.

ronco