High Density Snow

4th February 2023

Another slightly damp day in Lochaber today.

The Cairn Slope down from Nid Ridge.

A fair bit of grass on the surface of the snowpack.

The experimental method. I measured snow density using a small commercially available kit. A cutter (not shown) cuts a block of snow of known volume (not sure the actual volume, but would say around 10cm3). This block of snow is then placed in a bag which is weighted using a spring balance which is scaled in terms of density rather than weight.

As the snowpack was, to put it diplomatically, not the most technically interesting I decided to do a bit of an investigation into the density of the snow. Although snow density is not something I measure day to day as a forecaster, a general knowledge of it can be useful for context. I found the slight softer surface layer to have a density of around 380 kg/m3, and the lower icier layers to be around 450 to 500 kg/m3 (there was much more scatter in these results). The density of pure ice (discussed in more detail below) is around  917 kg/m3, this means that even the denser lower layer, the snowpack is roughly half as dense as the material that makes it up, or to put another way, that half this snow is actually free space.

The table below shows the literature values for various types of snow and ice.

Typical densities of snow and ice (kg/m³)
New snow (immediately after falling in calm) 50-70
Damp new snow 100-200
Settled snow 200-300
Depth hoar 100-300
Wind packed snow 350-400
Firn 400-830
Very wet snow and firn 700-800
Glacier ice 830-917

Source: Paterson, W.S.B. 1994. The Physics of Glaciers.

Unsurprisingly the Lochaber snowpack is currently quite dense (5 to 10 times that of cold new snow in calm condition). Looking at the transition from dense snow to ice there are various stages (assuming that the snow does not melt first, which in in this case it is likely to do).

Névé is a granular type of snow which has been partially melted, refrozen and compacted, but  is generally not yet multiyear.  I am not sure why Névé was missed of the Paterson table above but tends to have a density greater than about 500 kg/m3.

Firn (from Swiss German firn “last year’s”, cognate with before) can be loosely defined as dense, multiyear snow ice.  It tends to be denser than névé and is an intermediate stage between snow and glacial ice. The upper density of firn is well defined as 830 kg/m3 as and is associated with closure of the pore space.

Glacial Ice Pure ice has a density of about 917kg/m at 0°C and at atmospheric pressure. Ice behaves like most solids, with increasing density as temperature drops, with the density of pure ice being 920 kg/m3 at -23°C and values reach 922 kg/m3 for the coldest ice to be found, in the Antarctic ice sheet. Ice density also increases slightly with pressure having a density of 921 kg/m3 (at 0°C)  under the load of 4 km of ice, which is typical of the East Antarctic plateau. However, glacial ice will tend to contain air pockets which will reduce the density to within the range quoted in the table above.

Comments on this post

  • Sam
    4th February 2023 5:41 pm

    Great blog BF. As you said – not the most interesting of snowpacks at the moment – so I really enjoyed your technical input. Thanks

    • lochaberadmin
      4th February 2023 6:30 pm

      Glad you enjoyed it. Lets hope the snowpack get more interesting soon!

  • Matt Dalby
    4th February 2023 7:37 pm

    I’d always thought that neve was a French word that meant the same thing as firn, and firn was just old snow that had undergone one or more melt freeze cycles and didn’t have to be multi year snow. The density of the snow pack is well worth noting as it’ll affect who quickly the snow pack will decrease in depth during a thaw. Denser snow will decrease in volume more slowly than less dense snow, at any given temperature, as it contains more ice and less air and therefore requires more energy to convert it to water.

  • James
    4th February 2023 8:54 pm

    +1 for the great blog. Snow density and survival time in burial would be an interesting area to drill down into. Interesting comparison between Canadian (Maritime snowpack) vs Swiss (Continental snowpack) here by Haegeli et al…

    Haegeli P, Falk M, Brugger H, Etter HJ, Boyd J. Comparison of avalanche survival patterns in Canada and Switzerland. CMAJ. 2011 Apr 19;183(7):789-95. doi: 10.1503/cmaj.101435. Epub 2011 Mar 21. PMID: 21422139; PMCID: PMC3080528. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3080528/

    • lochaberadmin
      5th February 2023 8:25 am

      Hi James, thanks for the link. An interesting idea, but I suspect there might be too many other factors (shape of landscape, victim activity, forest cover, rescue ability, range of snowpack characteristics across Canada, etc) for a clear correlation to be apparent. However, an interesting paper looking paper (not read it yet) and I might well be wrong in everything I said above.

Got something to say? Leave a comment

    Latest Lochaber Avalanche Report
    RSS Feed
    Keep up to date by subscribing to our RSS feed
Service funded by sportscotland
Forecast data supplied by the Met Office
SAIS Sponsors