Table 3. Ground and SAR Data For Target
Area G3.
Ground SWE Digital
Sheet # (cm) Number
Raw Filtered
65
reflectors | 103
43
22
22
1 21 24 30.0
9 19 36 312
3 23 31 28.4
4 22 23 253
S 24 22 28.8
6 25 41 33.0
7 25 35 34.1
8 24 25 30.7
9 17 25 27:7
10 25 25 26.7
11 24 29 26.2
12 21 24 25 4
13 20 24 284
14 2 24 25.8
15 22 30 25.0
16 17 17 271
17
25
39
reflectors | 126
108
ze eio >
je
20 +
T
10
T
15
T T
20 25
DN (nadir, HH)
T
30
T
35
T
40
45
Figure 2. Linear regression of SWE vs.DN-s
Similar equations were developed for each test
area, both with HH and VV polarizations.
RESULTS
The ground survey gave detailed information on
snow distribution (depth, density, SWE) and on
the underlying ground cover (soil rock, ice,
vegetation type). The SAR data must be
correlated with all of the snow parameters, the
underlying surface conditions and with the
incidence angle and polarization of the RADAR.
Since the preparation of the necessary computer
programs and the development of the absolute
calibration procedure took a very long time, the
data has only been partially analysed to date.
Ground Data.
We had a very good distribution of SWE within
and between test lines (Table 2). The snow
depth should be over 20 cm to have an effect on
the SAR return. In our case this varied between
3] and 138 cm. The SWE vaiues had a range of
9 to 38 cm with averages of 12.9 to 24.5 cm of
individual test lines.
The snow pit data provided detailed information
on snow layering characteristics. All 22 detailed
snow pits had at least 3 distinct snow layers.
Thirty two per cent had four distinct layers and
only one pit had five distinct snow layers. The
mean thickness of the three main snow layers
from the bottom of the pit to the surface was 22,
25, and 24 cm respectively. In the snow layer
closest to the substrate, the snow temperature
showed the least variation and averaged -1.5 °C.
The temperature of the snow in layer three near
the surface was very close to the ambient
temperature. The SWE variation was minimal in
the layer close to the substrate and highest in the
second snow layer with variation further
decreasing in the newer snow on top of the
profile.
SAR Data Correlation With Ground Data.
The C- band SAR return from snow covered
areas is made up of volume scatter from the
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B7. Vienna 1996
C E >
d:
SC
Cc
ac