156
M.R. DRINKWATER
a.
White Ice
b.
Smooth First- Year Ice
0
0
. -
~
Freq, .4.3 CH.
~
Freq . • 4.3 CH.
·10
.
· 10
. Mean vy
Mcan vv
C
. Mean ~y ii
. Mean hy
..
'0
-20
'0
-20
IE
3
tS . . ·30
1
-30
j
~
-40
I
-40
..
~
=
•
-sss ":;1\1 ~ · 16.4 dB
·23.H W'kIS-27.SdB
-50
-50
10
20
30
40
50
60
10
20
30
40
50
60
Incidence AIIgle (O J
Incidence Angle (OJ
c.
Rough First-year ice
d,
Multiyear Ice
0
..
Freq . • 4.3 CHz
-,
Freq,. 4,) GHz:
=-10
. M ...... =. Meon ..
:s
. Man lw :s
.
"5
"5
Meanhv
"ij
-20
'u
IE
l
~
u
u
~
-30
. . -30
"
' 5
!
I
"5
~
I I
-B
... -40
-40
..
..
= · 11 ,7 S ~~SRI S · IS ,I dB
=
·50
-50
10
20
30
40
SO
60
10
20
30
40
50
60
Incidence Angle (')
Incidence Angle (0)
Fig. 5. Mean WWGS'92 shipborne scatterometer ice signatures, from a white ice, b smooth first-year
ice, c rough first-year ice, and d multiyear ice. Curves are fitted exponentially and error bars indicate
the standard error of the data. Vertical lines delineate the incidence angle range of ERS SAR (20 0 :0; e
:0; 26 0 ) and overlapping shaded pdf's show the probability distribution from SAR pixel samples in the
locations of shipborne scatterometer measurements. A range of observed SAR values is shown for all
sampled images for each ice type
expected to fall in the range -B.B ~ crOShip ~ -16 dB, based on the surface scatterometer
measurements, but this is heavily dependent on the physical surface characteristics.
A typical medium thick (67 cm) first-year ice sample acquired on July 13, 1992 at
65.99'S, 33.53'W had interleaved bands of frazil and columnar sea-ice crystal growth
and 12 cm of layered snow. The snow-ice interface temperature was -13' C, despite air
temperatures lower than -30' C, and an accompanying salinity at the ice surface of 14
psu. Angular depth-hoar crystals developing under the strong thermal gradient in the
basal snow were of significantly higher salinity than the ice surface itself, having typical values of around 27 psu. This highly saline, rough scattering interface at the base of
the snow probably accounts for extremely high backscatter values in the range 20-26'
in Fig. 5b. Likewise, the remainder of the snow had significant salt content, and airborne
M.R. DRINKWATER
a.
White Ice
b.
Smooth First- Year Ice
0
0
. -
~
Freq, .4.3 CH.
~
Freq . • 4.3 CH.
·10
.
· 10
. Mean vy
Mcan vv
C
. Mean ~y ii
. Mean hy
..
'0
-20
'0
-20
IE
3
tS . . ·30
1
-30
j
~
-40
I
-40
..
~
=
•
-sss ":;1\1 ~ · 16.4 dB
·23.H W'kIS-27.SdB
-50
-50
10
20
30
40
50
60
10
20
30
40
50
60
Incidence AIIgle (O J
Incidence Angle (OJ
c.
Rough First-year ice
d,
Multiyear Ice
0
..
Freq . • 4.3 CHz
-,
Freq,. 4,) GHz:
=-10
. M ...... =. Meon ..
:s
. Man lw :s
.
"5
"5
Meanhv
"ij
-20
'u
IE
l
~
u
u
~
-30
. . -30
"
' 5
!
I
"5
~
I I
-B
... -40
-40
..
..
= · 11 ,7 S ~~SRI S · IS ,I dB
=
·50
-50
10
20
30
40
SO
60
10
20
30
40
50
60
Incidence Angle (')
Incidence Angle (0)
Fig. 5. Mean WWGS'92 shipborne scatterometer ice signatures, from a white ice, b smooth first-year
ice, c rough first-year ice, and d multiyear ice. Curves are fitted exponentially and error bars indicate
the standard error of the data. Vertical lines delineate the incidence angle range of ERS SAR (20 0 :0; e
:0; 26 0 ) and overlapping shaded pdf's show the probability distribution from SAR pixel samples in the
locations of shipborne scatterometer measurements. A range of observed SAR values is shown for all
sampled images for each ice type
expected to fall in the range -B.B ~ crOShip ~ -16 dB, based on the surface scatterometer
measurements, but this is heavily dependent on the physical surface characteristics.
A typical medium thick (67 cm) first-year ice sample acquired on July 13, 1992 at
65.99'S, 33.53'W had interleaved bands of frazil and columnar sea-ice crystal growth
and 12 cm of layered snow. The snow-ice interface temperature was -13' C, despite air
temperatures lower than -30' C, and an accompanying salinity at the ice surface of 14
psu. Angular depth-hoar crystals developing under the strong thermal gradient in the
basal snow were of significantly higher salinity than the ice surface itself, having typical values of around 27 psu. This highly saline, rough scattering interface at the base of
the snow probably accounts for extremely high backscatter values in the range 20-26'
in Fig. 5b. Likewise, the remainder of the snow had significant salt content, and airborne