to Sea-Ice Nomenclature Overview page
Sea ice displays rather complicated
mechanical properties. At high strain rates, ice behaves
elastically, ie all deformation below a critical yield stress is
reversible. Under sustained loadings, ice behaves
inelastically, resulting in large irreversible strains. Under
constant stress, ice creeps viscoplastically, resulting in a strongly
non-linear relationship between strain rate and stress.
There are numerous possible stress
states of sea ice, among these include constant strain, constant
stress, tensile, compressive and flexural deformations. Here, I
have chosen to focus on one stress state - the uniaxial vertical
compressive strength - and investigate its variation with
porosity. The uniaxial vertical compressive strength is the
maximum stress that can be developed at a specified strain
rate. I have chosen to compare the compressive strengths at a
strain rate of 10-3 s-1. This is
primarily because data is available for this strain rate for a wide
spectrum of ice types with different porosities and crystal
texture. At strain rates greater than 10-3
s-1, brittle fracture takes over, while
at strain rates less than 10-5 s-1, creep
rupture is the failure mechanism. For practical reasons, it is
difficult to test accurately on cylindrical speciments at strain
rates greater than 10-3
s-1. A strain
rate of 10-3 s-1 is where ice response is
linear-viscous and is a reasonable approximation to the rate at which
ice floes interact and deform under wind and wave action.
Porosity has an important influence on sea ice strength. Brine and air inclusions have negligible strength compared to the ice matrix and they act as points of weakness. Inclusions are ubiquitous within the ice matrix, both microscopic (eg, brine lamellae) and macroscopic (eg, brine drainage channels) scales. The influence of porosity on the strength of sea ice can easily mask the influence of other microstructural properties, such as grain structure and texture and particulate inclusions. Hence, I have chosen to characterize the uniaxial compressive strength of different ice types according to their porosity.
Porosity values are calculated from
density, temperature and salinity. In relatively new young ice,
it is often assumed that gas volume is small and that brine volume
can account for the majority of the porosity. The theoretical
density-volume fraction equations of Cox and Weeks (1983) have been
used to calculate the porosity. Relative brine volume
Vb/V can be calculated from the density
r in
Mg m-3 and salinity of ice
in ppt Si
:
Vb / V = rSi / F1(T)
where F1(T) is a coefficient dependent on temperature. (Cox and Weeks, Equation 5)
At the temperature of -5 deg C, F1(T) has a value of 91.3 Mg m-3 .
For the present study, density,
temperature and salinity data of new young ice are obtained from
Tucker et al. (1991). Ice was sampled in the winter marginal
ice zone in the Fram Strait. For multi-year ice, Richter-Menge
et al. (1987) has been cited for ice sampled from the southern
Beaufort sea.
|
|
|
|
(at -5 deg C) |
COMPRESSIVE STRENGTH (as deduced from porosity) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Strength estimates for Nilas, Pancake, 1st Year Ice are deduced from Vaudrey (1977) as quoted in Mellor (1986) (Figure 39).
Porosity and Strength
measurements for
Multi-Year ice are obtained from Richter-Menge et al. (1987).
(Figure Ice samples were collected in 1981 in the southern Beaufort
Sea.
To: Section A. Sea Ice
Microstructure
To: Section B. Microstructure vs. Formation
mechanisms and Albedo
To: Section D. References