Re: Glacier Color

[Maurice Barnhill <mvb@UDEL.EDU>]

Wes Davis wrote:
> >    This discussion on sky color has been interesting and informative.
> > I have a similar question whose answer I thought I knew, but now
> > have some doubt about.
> >
> >    This summer I traveled to Alaska where I was able to land by
> > helicopter on the surface of a glacier.  I had seen the faces of glaciers
> > by ship and observed the blue color of the face, but on the top the color
> > of the ice in shallow crevasses was spectacular.  It almost seemed as
> > though a can of blue paint had been poured into the crevasse.  These
> > crevasses were not deep, in most cases being no more than five or six
> > meters.
> >
> >    Now, the standard explanation I have read for the blue color holds that
> > the intense pressure due to the weight of the glacier somehow changes the
> > crytalline structure of the ice.  Another explanation claims that glacier
> > ice is no different from ordinary ice - there's just more of it.  Yet even on
> > the surface (where, presumably, the pressure is small) the blue color is
> intense
> > and even small pieces of ice are blue.  I saw an iceberg that had melted
> > away until a very thin sheet jutted up out of the water.  Even though it was
> > almost transparent, it was also blue.
> >    Thanks,
> >    Wes Davis
The determination of the color of glacial ice -- white vs. blue -- turns
out to be very interesting and not in fact due to crystal structure per
se, despite the implications of the frequently-quoted explanation.
Several years ago I went on a tourist trip to Antarctica, heard the
crystal-structure explanation from one of the group leaders, and was
skeptical that there was any ice structure that would do the job.  The
leader, a geologist, gave me a paper to read on the subject.
Unfortunately I don't remember the reference, but I remember the
explanation well.

The blue color of ice that has been under pressure comes about the same
way as the blue color of the ocean and large lakes.  Water absorbs light
towards the red end of the visible spectrum, although rather weakly.
Given enough water, the only remaining light is blue.  In the ocean,
light incident on the top is eventually scattered back out.  In the
meantime, at least if the water is clean and not too sandy, most of the
red has been absorbed, and you see only the remaining blue.  You can
tell that the color is not due to having blue colors scattered backwards
because at depth the color of the remaining light is still blue.  If the
blue had been backscattered the unscattered red would be what you would
see at depth.

Natural ice mostly comes from snow and sleet, which fall as small
particles.  Much air is trapped as the ice consolidates, and the trapped
air scatters light pretty much independent of color.  As long as there
is a lot of air, the light you see scattered back to your eyes comes
from a relatively small distance into the ice, and it hasn't passed
through enough ice for the red colors to be absorbed.  Thus it looks
white.  In a glacier, much of the ice is subjected to  high pressure.
The air bubbles are all squeezed out.  Then when the glacier breaks up
some of the air-free ice winds up where you can see it.  In this ice
incident light penetrates deeply enough before being scattered back to
the eye that significant amounts of red light are absorbed going in or
coming back out.  Hence what you see is blue.  There is a clear clue
that this mechanism is correct rather than the process of having
backscattering of blue light from near the surface.  When you see light
coming out the opposite side of a fairly thick piece of glacial ice, you
see a blue color, and not the red that you would see if blue had been
scattered out of the original beam.  In particular, the blue color you
see in crevasses is clear evidence of absorption of red colors in light
that entered at the exposed surface.

I took pictures of ice bergs that show blue colors in spots that should
have been illuminated mostly from inside the ice.  None of them are
entirely satisfactory, since I was far away from the bergs and could not
shade a spot from the sun so as to protect against having large amounts
of back-scattered light messing up the observations. I have, however,
seen photos of ice caves, and in caves the illumination is quite blue.
Thus I am satisfied that absorption of red rather than scattering of
blue is the correct explanation for the color of glacial ice.
--
Maurice Barnhill, mvb@udel.edu
http://www.physics.udel.edu/~barnhill/
Physics Dept., University of Delaware, Newark, DE 19716