Geology Reference
In-Depth Information
Table 8.1
Backscatter coefficient lookup tables from scatterometer measurements of different ice
types in different seasons developed for use in the GPS of the Alaska SAR facility.
σ
0
(dB)
Thickness
(cm)
Season
Ice Type
X (HH)
C (VV)
L (HH)
Winter &
early spring
MY
>220
−3.6
−8.6
−7.0
TKFY
70-220
−14.2
−11.5
−23.4
TNFY
20-70
—
−13.3
−23.4
OW
0
<−29.7
≤−19.7
≤30.7
Late spring
MY
>220
—
−10.7
−15.5
TKFY
70-220
—
−13.2
−23.4
OW
0
<−29.7
≤19.7
≤30.7
Early summer
MY‐TKFY
>70
−15.9
−16.3
−15.1
TNFY
20-70
−15.1
−13.1
—
OW
0
<−29.7
≤−19.7
≤30.7
Mid‐summer
MY
>220
−15.7
−16.3
−22.7
TKFY
70-220
−15.7
−14.7
−19.8
TNFY
20-70
−14.7
−13.1
−19.8
OW
0
<−29.7
≤−19.7
≤30.7
Late summer
MY
>220
—
−16.8
−21.9
TNFY
20-70
—
−18.4
−28.1
OW
0
<−29.7
−19.7
≤30.7
Fall
MY
>220
−6.9
−10.5
−17.3
TKFY
70-220
−13.5
−12.5
−18.4
TNFY
20-70
−14.4
−19.7
−18.4
OW
0
—
≤−19.7
≤30.7
Note:
Data are compiled from tables presented in
Onstott
(1992).
support the SAR data validation and ice‐type algorithm
development at the SAR Geophysical Processor System
(GPS) of the Alaska SAR facility (ASF). The tables are
also presented in
Onstott
[1992] and recompiled here in a
brief form in Table 8.1. The X‐band data are valid for
incidence angle 25° while the C‐ and L‐band data are
valid for 23° and 35° to coincide with the angles of ERS‐1
VV and JERS‐1 HH SAR systems, respectively. The sea-
sons presented in the table are defined as follows (applied
to the Arctic region only): Winter and early spring denotes
cold conditions in November-April. Late spring indicates
warming condition in May and June. Early summer, mid-
summer, and late summer are for moist snow observed in
early June, peak melt in late June and July, and postpeak
melt in July and August, respectively. Fall indicates freez-
ing conditions in September and October. The backscat-
ter from the L‐band in winter and summer is notably
lower than the X‐ and C‐bands.
Another dataset of radar backscatter was acquired dur-
ing the Winter Weddell Gyre Study (WWGS'92) in 1992 in
the Antarctic using a ship‐borne C‐band microwave scat-
terometer [
Drinkwater et al
., 1995]. The scatterometer had
dual polarizations: VV and VH. Data were obtained at a
variety of incidence angles between 15° and 70°. Graphs
showing results from this experiment are compiled in
Drinkwater
[1998] and reproduced in Figure 8.3. The fig-
ure shows the angular variation of backscatter coefficient
for four types of ice: white ice, smooth FY ice, rough FY
ice, and MY ice. White ice has high concentration of min-
ute air bubbles (micron-scale). In this figure it refers to
thin FY ice of thickness between 30 and 70 cm. Rough FY
ice refers to heavily deformed, ridged, or rubble surface.
The data in the figure were obtained from ice that exceeded
60 cm in thickness. The cross‐polarization return is signi-
ficantly lower than the co‐polarization return, but the
latter drops considerably at higher incidence angles for
white ice and smooth ice so that it becomes similar to the
co‐polarization return. A remarkable observation from
this figure is the indistinguishable backscatter from rough
ice and MY ice in the ERS‐1 incidence angle range.
The Seasonal Ice Zone Experiment (SIZEX 92) was the
first ERS‐1 SAR ice winter validation experiment in the
Barents Sea [
Sandven et al
., 1999]. Backscatter signature
from several ice types and open water were obtained after
image calibration and normalization to an incidence angle
of 23° (center of the swath). Results of backscatter (C‐
band) from different ice types in winter as well as smooth
MY ice and FY ice floes in summer are shown in Figure 8.4.
Note the wide range of backscatter coefficients from thin
ice types including new, Nilas, grey white, and frost-flower-
covered surfaces (−25 to −4.5 dB). In the summer, back-
scatter from FY and MY ice completely overlap and
occupy a relatively narrow range. A striking observation
from this figure is the wide range of backscatter from OW
