Geology Reference
In-Depth Information
March 18 2003
9+
145
RADARSAT:
21:50 GMT
4. 1.
7
554
SSM/I:
21:45 GMT
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
64
68
72
76
80
84
88
92
96
100
To tal ice
concentration
NI concentration
YI concentration
FYI concentration
Figure 10.22 Radarsat‐1 image with a single ice analysis polygon from CIS Radarsat image analysis (top left)
panel. Results of total ice concentrations and concentrations of three ice types from ECICE using the input satellite
observations combination mentioned in the text are shown in other panels. Note the nonuniform distributions of
YI and FY ice concentrations within the polygon. The dark line in the top left panel marks a possible split of the
shown polygon into two types. (For color detail, please see color plate section).
FY ice medium (code 1), and five‐tenths of FY ice thin
(code 7). The calculated total ice concentration agrees
with the CIS estimates. The algorithm outputs zero NI,
which again agrees with the CIS chart. However, while
CIS estimate calls for three types of FYI in the polygon,
EICE shows appreciable value of YI concentration. The
remarkable observation in the figure is the split between
the calculated YI and FY ice concentrations at the two
sides of the dark line shown in the top left panel. It sug-
gests that this polygon could have been divided into two
parts separated by that line in order for each part to have
a more‐or‐less uniform spatial distribution of ice types
and concentrations. Figure 10.23 shows clearly the value
of the subpolygon information revealed by incorporating
information from the SSM/I footprints. This should sup-
port the operational analysis of SAR imagery data when
estimated parameters from different algorithms are over-
laid on the SAR image.
10.2.2.4. Comparison of Passive Microwave Algorithms
The wide array of passive microwave algorithms for
sea ice concentration has stimulated studies on intercom-
parison and identification of the best algorithm for use
under a given ice regime. A number of studies were
launched to compare two or more algorithms [ Andersen ,
2000; Belchansky and Douglas , 2002; Meier , 2005; Shokr
and Markus , 2006; Andersen et  al., 2006, 2007]. In the
absence of validation data, intercomparison of results
Search WWH ::




Custom Search