Geoscience Reference
In-Depth Information
In this case, data handling was made with ArcGIS software, and particularly
consisted on two processes: In the first stage, digital topographic maps' were repro-
jected (from PSAD 56 to WGS 84); in a second stage, the vector files were gathered
by each ASTER image, so all the vector data over the remotely sensed data were
stored in the same folder. This processing and handling are relevant for the RASTER
data orthocorrection and also for the vector data adjustment.
Aerial photographs: Aerial photographs used in this study were from a FONDEF
flight in the late 1990s. The aerial photographs were used because they have a
better resolution than satellite imagery, but also because photographs from years
before the ASTER data are available can be used to “validate” what we can rec-
ognize as a glacier. This is necessary because for the inventory to be included in
WGI/WGMS/GLIMS, the inventory must consider “every perennial snow and ice
patch”, so the photographs enable a multitemporal reference in order to confirm
perennial characteristics.
Satellite imagery: ASTER data were chosen for this analysis, because of
their spatial resolution (15 m), and also the possibilities for combining other
studies carried by other research teams in the fields of biology, peat lands mon-
itoring, highlands vegetation, primary productivity in mid altitude valleys, and
hydrology. This remotely sensed data were also chosen because they enable the
construction of a Digital Elevation Model (DEM) from the same datasets by the
provider.
Raster data were obtained from ASTER imagery; such imagery has been well
proven in cryosphere studies (Raup at al.
2000
; Racoviteanu et al.
2008
; Paul et al.
2002
) and actually has been proposed as a base for GLIMS project (
www.glims.org
).
Imagery data have been processed for all ASTER datasets, but in this specific case,
only the very near infrared dataset (VNIR) has been considered. This is because
the inventory at this stage has been carried out only using manual delineation. In
spite of this, all images bands were processed. Also aerial photographs have been
used in the process, however, the aerial photographs were not rectified, and only
used to compare its area to the corresponding area from the already rectified and
ortocorrected ASTER imagery. Also, DTM data, which have been used for the
orthocorrection key process (Paul
2001
), this process has been carried out with
“ASTER DTM” application, developed by Brazilian company Sulsoft for ENVI
software in an IDL created graphic user interphase (GUI). At a later stage, both
raster data types (Imagery and DTM) have been combined for undertaking the ortho-
correction process with the formerly mentioned software. In this process, raster and
vector files are used. In the first stage, a vector file with a homogeneous presence in
the selected imagery must be used. In this specific case, the drainage network from
IGM (1:50,000) was used, as its coverage on the Chilean side is the most homo-
geneous, from the Chilean-Argentinean border to the mid-altitude that is located
in the images. Also, DTM data, which has been used for the orthocorrection key
process (Paul
2001
), this process has been carried out with “ASTER DTM” applica-
tion, developed by Brazilian company Sulsoft for ENVI software in an IDL created
interphase.
