Geology Reference
In-Depth Information
fractures and the migration of fracture interactions cause orthogonal intersections to
evolve towards interangular interactions. Analogy is made with basalt rock that forms
hexagonal columns upon cooling. The complex-systems approach recognizes that spatial
variability in snow cover and vegetation also infl uences fracture dynamics; in particular,
the snow-stress feedback is complicated, site-specifi c, and not universal.
In the ice-free areas of Southern Victoria Land, Antarctica, evolution of the polygonal
net has been examined on land surfaces that range in age from 10
−3
to 10
−6
years by R. L.
Sletten, B. Hallet and R. C. Fletcher (2003). Several stages of polygon evolution are identi-
fi ed. In the mature or end phase, as typifi ed by Beacon Valley (age 10
−6
years), sand-wedge
polygons are 10-20 m in dimensions. If an average growth rate of 0.6 mm (see below) is
then applied, the ratio of polygon size to wedge-growth rate suggests that the entire land
surface has been reworked or recycled by sand wedges on time scales of 10
−4
to 10
−6
years.
Although such a dramatic interpretation of landscape appears unrealistic, it must be
remembered that the ice-free terrain of Beacon Valley is probably the oldest land surface
on Earth that has experienced continuous cold-climate conditions for several millions of
years. It appears that, over time, there is a progressive change from curvilinear and quasi-
orthogonal intersections towards 120° intersection angles and, eventually, to regular fi ve-
and six-sided polygons. Some support is given, therefore, to the numerical modeling
approach described earlier. It may be that hexagonal patterns develop best in homogene-
ous material subject to long periods of uninterrupted and uniform cold-climate conditions,
as in the Antarctic Dry Valleys, while orthogonal patterns are immature and develop in
heterogeneous materials that experience changing environmental conditions.
6.2.4. Polygon Morphology
Typical of many polygonal systems is a raised rim on either side of the fi ssure. This is true
for both ice- and sand-wedge polygons. Sometimes the rim may be as much as 0.5 to 1.0 m
high. Even in newly-forming permafrost, as on the recently-drained lake bottom at Illisar-
vik, a shallow ridge a few centimeters high formed adjacent to the initial frost cracks
during the fi rst winter (Mackay, 1980a).
A commonly held view is that the double-raised rims are caused by the accumulation
of either ice or mineral soil within the crack, thereby forcing adjacent frozen material
upwards. This must be questioned because it is now thought that lateral thermally-induced
movement of active-layer material occurs from the polygon center to the periphery.
This is inferred from measurement of the distance and tilt of steel rods inserted into
permafrost on either side of the fi ssure (Figure 6.5A) (Mackay, 1980a, 2000). The move-
ments refl ect summer warming and expansion of the active layer outwards from the
middle of the polygon. In the Mackenzie Delta region of Canada, a movement rate of
0.25 cm/yr was estimated for one polygon, implying a coeffi cient of thermal expansion of
about 1.7
10
5
°C. The tilt of spruce trees adjacent to polygon rims (Kokelj and Burn,
2004) also suggests this sort of movement. The implication is that shearing occurs at the
active layer-permafrost interface. At the same time, the progressive increase in width of
the wedge itself must be accompanied by some deformation not only of the wedge but also
of the enclosing ground.
This must also be the case for the shallow ridges that typically border many of the
sand-wedge polygons in Antarctica (Sletten et al., 2003). Here, the development of the
polygon net involves long-term convection-like cycling of material through the polygon
(Figure 6.5B). Measurement of the spacing between steel rods hammered into the per-
mafrost on either side of contraction cracks indicates average rates of surface widening of
wedges of between 0.1 and 2 mm/year. A systematic tilt of the rods is interpreted to refl ect
×
