"left" and "right" sides of glacier boundaries

[Bruce Raup]

Hmm, I think there is a bit of confusion about what we're talking about.
The issue is not what's on the left and right of the glacier, but rather
what's on the left and right of pieces of the outline.  I agree that it's
important to follow existing conventions, but I don't this the
hydrological one is the right one to use here.  For example, let's say one
segment of an outline is labelled "terminus".  Its elevation profile might
be U-shaped, so what would the hydrological convention say is "left" of
that boundary?

Since writing this, I've received notes from Hugh Kieffer (below) and
Graham's, sent to the list.  So excuse the overlap.

I agree with Graham that interior polygons (polygons around nunataks)
should be the opposite direction from the glacier outline, for the
area-calculation reasons he gave.  Enhancing GLIMSView to do this ordering
(if we agree on the conventions) would be great, but I've also already
implemented most of this functionality in the ingest software (which
I needed to do for other reasons).

More comments?

Bruce

>From hkieffer at charter.net Fri Feb 25 09:00:01 2005
Date: Fri, 25 Feb 2005 07:55:23 -0800
From: Hugh Kieffer <hkieffer at charter.net>
To: braup at nsidc.org
Subject: fuller discussion

Bruce:



There may be some confusion between "which" boundary and "which side
of a" boundary; it is the latter that is under consideration. Which is
the left side of a glacier is not under contention, everyone uses the
normal convention (left as you look down-flow).

An example of where the hydraulic convention breaks down is a playa (a
desert dry lake), where the boundary may be well defined, but there is
no "hydraulic left" because the flow is inward from all directions. A
GLIMS-relevant example is the terminus of a glacier; which (not where)
is the left side of the boundary? Under the convention proposed by
Bruce (polygons around glaciers proceed counter-clockwise when viewed
from above) the material on the left at a terminus is the glacier and
the material on the right might be an out-wash plain, seawater, or
another glacier.

I note however, that the proposed convention always maps
(topologically) the "left" side of a boundary to the "inside' of a
polygon, perhaps GLIMS should move to use of "inside" and "outside"
for polygons.

There will remain the odd case where a GLIMS line does not have an
"inside"; e.g., a firn line across a glacier. However, it seems that
in most (if not all) cases of unclosed lines relevant to glaciology,
either the hydralic left and right, or the topographically lower and
higher sides would be known. I propose that the "left" side of a
boundary be, in order of precedence:

"inside"   if the polygon has an obvious inside (as do all closed
                two-dimensional polygons)
   "left"   if not a polygon AND the line runs "downhill"
                (e.g. medial moraine)
  "lower"   if neither of the above

--------------------------------------------------
Hugh H. Kieffer
Celestial Reasonings .... Voice & Fax 775-882-3787
2256 Christmas Tree Lane ....... Cell 775-315-5135
Carson City, NV 89703 ....... hkieffer at charter.net
--------------------------------------------------

-- 
Bruce Raup                                               Phone:  303-492-8814
National Snow and Ice Data Center, U. of Colorado, 449 UCB, Boulder, CO 80309
http://cires.colorado.edu/~braup/