Geology Reference
In-Depth Information
04
05
06
07
08
Greenland
Greenland
Greenland
Greenland
Greenland
4.0
0.0
5.0 m
Overall
Thickness (m)
My ice
3.5
Fy ice
3.0
0.83 m
Tr end = -0.17 m/yr
2.5
2.0
1. 5
Figure 10.36
Trends of MY, FY, and total ice change in the Arctic Ocean during winter, obtained from ICESat‐1
sea ice thickness estimates from five ICESat‐1 campaigns between 2004 and 2008. Note the decline in MY ice
thickness, which is reflected in decrease of the overall thickness, but the FY ice thickness does not demonstrate
the same decline [
Kwok and Sulsky
, 2010]. (For color detail, please see color plate section).
footprint of SIRAL is 250 m in the along‐track and
about 15 km in the cross‐track directions. Similar to the
altimeters onboard ERS‐1, ERS‐2, and ENVISAT, the
SIRAL operates in the Ku‐band (13.4 GHz). The repeat
cycle of Cryosat‐2 is 369 days with a track spacing of
over 4 km. This means that mapping ice thickness over
the entire polar region requires numerous passes that
have to be acquired over several weeks. The first map of
sea ice thickness obtained from the Cryosat‐2 mission
over the entire Arctic basin was produced by the Centre
for Polar Observation and Modelling at the University
College in London. It shows the average ice thickness at
each observation footprint during the months of January
and February 2011. Ice thickness in narrow passages
such as those existing in the Canadian Arctic Archipelago
region cannot be produced because of land contamina-
tion. Annual ice thickness maps of the Arctic region are
a valuable source for monitoring interannual variability
of Arctic ice.
Sources of uncertainty and errors in estimating ice
thickness from radar altimetry are presented in
Tonboe
et al
. [2010]. They adopted a few approaches: (1) a simu-
lation of radar scattering from bubbly MY ice to assess
the effect of snow and ice properties, (2) field campaign
measurements to determine the effect of snow depth on
the ice floe buoyancy, and (3) the high‐resolution
Radarsat to study the errors due to the heterogeneity of
the surface cover within the altimeter's footprint. Their
field data revealed that the spatial variability of snow
depth could produce spatial and temporal uncertainty of
about 0.3 m in the estimated ice thickness. Another 0.3 m
uncertainty is caused by unknown variability of snow
density. One of the highlights of the study is the emphasis
on the spatial variability of the scattering surface, i.e., the
horizon where the freeboard is measured even when the
ice and snow surfaces are geometrically flat. In addition
to the well‐established factors that affect the ice floe
buoyancy (namely, ice density, snow density, and snow
depth), the simulations assert that the radar penetration
variability and preferential sampling errors are no less
factors affecting the freeboard measurements than those
affecting the ice floe buoyancy.
10.4.4. SAR Observations
Ice thickness can be inferred from a single‐channel
SAR image using ice type identification as a proxy indica-
tor. Since ice types cannot be uniquely identified based
