Geology Reference
In-Depth Information
O/C - CLMR
O/F - CLMR
R/C - CLMR
Snow - I
O - FRZL
R - FRZL
1. 3
4.4
4.3
Avg. permittivity (F/m)
Avg. loss (F/m)
1. 2
4.2
1. 1
4.1
1. 0
4.0
3.9
0.9
3.8
0.8
3.7
0.7
3.6
5
7
9
3
4
11
12
13
1
2
6
8
10
0.6
3.5
Figure 3.34 Variation of average permittivity and loss within the top 0.15 m of FYI cores with different crystallo-
graphic structures. Data were obtained from Lancaster Sound, central Arctic, in May 1993. The numbers attached
to the core at the bottom are the average salinity in the top 6 cm. Nomenclature of ice crystallographic types
(shown at the top) is explained in the text [ Shokr and Sinha, 1995].
appear to be slightly better than using the random-
needle assumption. This applies to both permittivity and
loss (for permittivity, the root mean square of the differ-
ence between measured and calculated values is 0.18
and 0.29, respectively). It should be noted that accurate
evaluation of using the oriented needle assumption
requires measurements of the tensor components of the
dielectric constant. The component parallel to the elec-
tric field direction can be twice as high as the perpen-
dicular component, especially at temperatures above
−6 °C. Taking a simple arithmetic average of the two
components is not the most accurate way of combining
them into a single observation. The two apparent clus-
ters in the columnar ice data in Figure  3.35 represent
data points obtained from two different years (May
1992 and 1993). Obviously, the ice was more saline for
the cluster with higher permittivity and loss.
Permittivity of MY hummock ice (Figure  3.36)
shows that all assumptions of air bubble inclusion
shape in the PVD models are equally successful in
approximating the measured values. This is mainly due
to the fact that permittivity of air (1.0) does not
significantly contrast with that of pure ice (3.15).
Moreover, the composite permittivity is isotropic since
the two orthogonal components from the oriented nee-
dle assumption are found to be nearly equal. This elimi-
nates the need for an appropriate assumption to model
the highly asymmetric and convoluted air bubbles in
hummock ice. Permittivity of hummock ice is relatively
easy to model because this ice is strictly a two‐phase
material (pure ice and air) and the permittivity of air is
independent of temperature. Melt pond ice data pre-
sented in Figure 3.36 show that using the brine pocket
as dominant inclusion for this ice type produces better
results than using the spherical air bubble assumption.
It can be seen also that the linear model is fairly suc-
cessful in reproducing both permittivity and loss for
both hummock and melt pond ice. This model seems
to be successful when brine volume fraction is suffi-
ciently small.
 
Search WWH ::




Custom Search