uality of
iceborne
age data
red and
ons and
spatial
'oncept).
use of
-weather
etection,
anges as
tation in
eoscopic
cameras
plex and
r (SAR)
Besides,
inoramic
elligence
88 aerial
amera in
> sensing
e of the
verage of
> sensing
ographic
2, 9)
yographic
] direct
yographic
discussed
a, Sharov
ammetric
) analyze
ents. The
void any
] for joint
size. For
| scale of
n 1993 at
500 DPI,
'd to the
y f camera
tic stereo
sampling
ed at the
of stereo
| borders,
coastlines, hydrographic network, morains, etc., were manually
vectorized in spaceborne images, and the resultant graphic
features were registered to the corresponding aerial images by
uniting homologous tie points with a root mean square error of
48 pixels or € 12 m on the ground. Finally, the relative
differences in position of “old” and "new" natural boundaries
were measured. The measurements, have shown essential
changes in the glacial environment and shorelines of FIL.
Several specific values of glacial changes revealed in this study
are summarized in Table 4.
Table 4. Planimetric values of glacial changes detected in FJL
(1953-1993)
N | Island Glacier name | Glacier Mean value
(No), type* size. km 2 of change,
n km
exposition
l Becker Suvorov,ice cap 25.5/8 -1.0,..1.2
2 | Brady N2, outlet SO0/NE 1-025...037
3 | Champ N12, outlet 31.7/ NW - 0.42
N13, outlet 18.8/N - 0.8
4 | Greely N12, outlet 7.41] W - 0.5
5 | Hall Sonklar, outlet, | 46.7/SE | -2.0...4.0
ice shelf
Hoffmann | Adamov, isl. Ice 52.8 - 0.3...0.45
7 | Hohenloe | N5/N6, ice cap 14.0 x
Hooker Yurij, outlet 12.5/ NW - 1.4...1.6
Obruchev, outl 47.375 - 1.8
Sedov, ice cap 10.2 - 0.1...0.12
Churlyonis,i.c. 11.3 = XX
9 | Kane N2, ice cap 1.6/8 30.1...0.13
10 | Karl N17, outlet, | 26.2/ NW -1.0...1.1
Alexander | ice shelf
11 | Kuhn Snow field 0.85 / E
N4,slope.glacier 0.4 / SW - 0.05
12 | Leigh N8, ice cap 13.178 - 0.25
Smith snow field 0.178
13 | Lutke N4, ice cap 8.2 - 0.05
14 | McClintock | NI8, outlet, 45.6 / N - 0.75...0.9
ice shelf
15 | Nansen N9, ice cap 5.8 -0.75...1.2
16 | Prince N24, outlet, | 42.6/E -0.65...0.7
George L. | ice shelf
17 | Salisbury Eastern, outlet, 49.0/ E + 0.5
ice shelf
18 | Wilczek Famous, outlet | 382.4 /E - 1.0
Land Impetuous, 137.9/N * 0.6
outlet, ice shelf
19 | Ziegler Natalie, ice cap 2.7 - 0.15...0.3
*) data are given in accordance with the Catalogue of Glaciers in
FJL (1965); **) planimetric changes could not be detected.
Table 5. Areal changes in glacier cover in FJL in the course of
1953-1993
N | Parameter \ Year 1953* | 1993 Difference
km? (96)
1 Ice-free land area, 2,399 2,432 +33
km? (1.38%)
2 | Glacierised area, 13,735 13.525 -210
km? (1.53%)
3 | Total area, km” 16,134 15,957 2177
(1.1096)
4 |Index ofglaciation, | 85.13 84.76 - 0.37 96
%
*) data adopted from the Atlas of the Arctic (1985).
Table 5 gives the changes in total land area and glacier cover
having occured in FJL in the period from 1953 to 1993. The
present areal extent of glaciation in FJL was automatically
measured on the basis of a controlled mosaic composed of five
rectified KATE-200 images. The image distortion caused by the
influence of the Earth's curvature and camera tilts was
compensated. Moreover, the image data were processed in two
different ways to enhance ice-free areas or glacial bodies. The
precise delineation of all glacier areas was performed and ice-
free areas, glaciers and aquatories were represented by different
opaque colors.
In total, seven different image histograms were calculated, each
representing the number of classified pixels depending on the
delineation approach. 75,818 pixels, or 0.8% of the total number
of pixels, which could not be definitely classified, were
proportionally distributed over all three classes. The average
value of the pixel size was defined as 98.3x98.3 square meters
on the ground, and the areal extent of glaciation in FJL as well as
the total area of the archipelago was determined by averaging
between seven estimations. The accuracy of the final estimation
is evaluated as + 0.8%, which is not worse than that of manual
planimetric measurements in aerial photographs and available
maps performed by O. Vinogradov and T. Psaryova in 1965.
The same image-to-image registration technique was
successfully applied for uniting the KATE-200 image mosaic
with the available topographic map of FJL published in 1989/90
at 1:500,000 scale. The planimetric accuracy of this combination
covering the entire archipelago is characterized by a rmse value
of € 125 m. Figure 4 shows a small fragment from the resultant
composite, in which planimetric changes of shoreline at Hall
Island are clearly visible.
Fig. 4. Fragment from synthetic map representing coastal
changes at Hall Island
The data obtained lead to the conclusion that the most essential
changes in glacial cover and in the total area of the FJL
archipelago have resulted from marine abrasion and active
calving of ice shores. The amount of inland glacial retreat is
evaluated as being, at least, six times smaller than the amount of
withdrawal attributable to the calving of ice shores. A
comparison of old aerial images and available maps with
CORONA imagery taken in the first half of the 1960-s did not
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 205
