GLIMS Update: GLIMS+ STARs; attempt to reduce snow saturation

[Jeffrey Kargel]

Dear
GLIMS consortium members,

 

This
message is mostly informational, but those of you wanting

to adjust some characteristics of GLIMS+ images should read

carefully and respond within the next
few days with any changes. GLIMS+ requesters also may wish to review the attached
file for accuracy before we submit this imaging plan late next week for review, approval, and implementation by ASTER Mission Operations.

 

The
attached file of "GLIMS+ STARs (ASTER Science Team 

Acquisition
Requests)" was produced through the efforts of quite 

a few
members of the GLIMS consortium.  RC Chiefs and 

stewards
contributed the requests based on gaps in existing 

ASTER
coverage or special needs for enhanced coverage of high-

priority
areas; I compiled the requests and revised some of them 

in
consultation with the requesters; Greg Leonard spot-checked 

the
requests for accuracy and produced a text file, with unform 

format
(in a few cases combining small-area requests into larger 

polygons),
which Bruce Raup then used to calculate the gain 

setting
requests.   Please be reminded that in addition to this 

supplemental
request, the regular GLIMS STARs (which was 

revamped
completely and functioned in 2008 and was updated to 

correct
some problems in 2009) are in effect.

 

We are
trying something new with this set of supplementary 

"GLIMS+"
STARs.  The change should result in nonpolar glacier 

images
that have much-reduced saturation on snow-covered 

equator-facing
slopes.  The gain settings have
been computed 

with a
modified algorithm that includes a “slopeshift” term.  As 

many
people have noted, even GLIMS-mode images (which are 

designed
not to saturate for horizontal snow-covered surfaces) 

commonly
include significant saturation if slopes angle toward 

the sun
by more than a few degrees.  So the added "slopeshift" 

will
reduce image saturation at the slight expense of giving lower 

signal:noise
in the darkest areas (rock, water, and shadow 

areas).
 Please note that for many non-GLIMS images (e.g., 

Global
Map images), gains are set for rock targets, and we have 

no
control over that (such that snow and ice areas are usually 

saturated
no matter what the slope).    

 

We are
incorporating default values of the slopeshift  of -15 

degrees,
which means that equatorward slopes of at least 15 

degrees
will remain unsaturated.  In
practice, for the majority of 

images
only equator-facing slopes steeper than 20 or 25 degrees 

(and
even greater at low latitudes) will be saturated; this 

compares
to standard GLIMS-mode images which saturate with 

fresh
powdered snow surfaces if they slope sunward more than a 

typical
6 degrees or so, which can be 30% of the accumulation 

zone in
rough terrain.   For DEM
production and studies of the 

accumulation
zone, the -15-degree slopeshift value will greatly 

improve
image quality. 

 

The
exceptions are in Greenland, the Canadian Arctic islands, and 

Antarctic
Peninsula, where most of the glacier area is low-sloping 

large
outlet glaciers, ice shelves, or otherwise have small areas 

that
slope more than a few degrees.  In these areas the default 

slopeshift
value is 0 degrees, meaning that the gains are 

calculated
just as they have been for the normal GLIMS STARs 

(based
on the central latitude of a polygon and the middle date 

of the
imaging season).

 

The
requesters of the individual GLIMS+ STARs may email Greg 

Leonard
(gleonard at email.arizona.edu) with any adjustments you want
to 

the
slopeshift values for your GLIMS+ requests.  For example, if you have 

been happy with GLIMS gain settings and
don't care about some 

saturation
in the snowy areas (for instance, if you're studying mainly 

debris-covered areas and glacier lakes), then maybe you
would prefer a 0 slopeshift value.  In exceptional case involving the most
extreme rugged and snowy topography, you may wish larger slopeshift values than
15 degrees.  In Greenland, if
occasional slight image saturation has been a problem, we can incorporate a
small slopeshift of -5 or -10 degrees.

After one season of Northern Hemisphere imaging (by
northern fall), we will review our results and see whether regular GLIMS
imaging should be modified to include finite slope-shifts such as these, or
whether we want to retain standard GLIMS-mode gains (with no slopeshift).  This can be done on a region-by-region
basis.

 

Sincerely,

 

Jeff Kargel





_________________________________________________________________
Windows Live™ SkyDrive™: Get 25 GB of free online storage.
http://windowslive.com/online/skydrive?ocid=TXT_TAGLM_WL_SD_25GB_062009
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://nsidc.org/pipermail/glims/attachments/20090605/d9cd438c/attachment.html>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: GLIMS+STAR-withGains&Slopeshift.txt
URL: <https://nsidc.org/pipermail/glims/attachments/20090605/d9cd438c/attachment.txt>