19

I have not been able to find an estimate of how many known volcanoes host perennial ice masses, and it would be very useful to me. Does anyone know of any attempts to do this?

f.thorpe
  • 13,583
  • 2
  • 46
  • 89
foobarbecue
  • 2,583
  • 1
  • 20
  • 30
  • 3
    I'd be willing to bet that GIS layers for both of these things already exist... just needs somebody to overlay them :) – Semidiurnal Simon Sep 17 '14 at 12:01
  • Good point. I'll try later today with data from http://nsidc.org/data/docs/noaa/g01130_glacier_inventory and http://www.volcano.si.edu/ – foobarbecue Sep 17 '14 at 13:47
  • Well, I'm having a hard time finding a GIS layer with reasonably complete glaciers. I think I may need to make something myself from http://neo.sci.gsfc.nasa.gov/view.php?datasetId=MOD10C1_M_SNOW . – foobarbecue Sep 18 '14 at 05:30

1 Answers1

19

21.7%, by my calculations (338 / 1556 holocene-active volcanoes). I calculated perennial snow by combining 6 weeks of MOD10A2 data from winter and summer weeks in 2014 to figure out which pixels had snow both in the summer and the winter. My source code for that is available as a gist (an older version accidentally classified cloud as snow; that's been fixed). Then I checked if these pixels fell within 10km of the volcanoes in the Smithsonian Institution's Global Volcanism Database. I also plotted the woefully incomplete Global Land Ice Measurements from Space glacier polygons.

There are clearly a few false-positives on ocean islands... I think this is due to MOD10A2 marking choppy surf as snow. Looking into that. I intend to publish this and will put a reference here if it is published.

Here's what the map looks like for part of Alaska:

enter image description here

And the whole world:

enter image description here

foobarbecue
  • 2,583
  • 1
  • 20
  • 30
  • 1
    Snowy in Northern Ireland in Aug? Maybe you should double check if the map is correct. – BHF Sep 25 '14 at 08:24
  • 1
    Ah, thanks for catching this. Standby for new map. I think I accidentally lumped cloudy pixels in with snowy ones. – foobarbecue Sep 25 '14 at 17:55
  • 2
    I worked on this some more and published it as a paper: https://doi.org/10.1016/j.jvolgeores.2017.01.017 – foobarbecue Oct 29 '17 at 23:00