HR: 0830h
AN: V51E-0328    [PDF]
TI: Laboratory experiments on columnar jointing
AU: * Goehring, L
EM: goehring@physics.utoronto.ca
AF: University of Toronto, Department of Physics 60 St. George Street, Toronto, ON M5S 1A7 Canada
AU: Morris, S W
EM: smorris@physics.utoronto.ca
AF: University of Toronto, Department of Physics 60 St. George Street, Toronto, ON M5S 1A7 Canada
AB: The mechanism causing columnar jointing has remained an enticing mystery since the basalt columns of the Giant's Causeway in N. Ireland were first reported to science in the 17th century. This phenomenon, in which shrinkage cracks form a quasi-hexagonal arrangement, has been shown to produce columns in starch, glass, coal, sandstone, and ice, as well as in a variety of lava flows. This suggests that this pattern-forming process is very general in nature. However, most studies of columnar jointing have been confined to field studies of basalt flows. Following Muller, we have experimented with desiccating corn starch in an effort to understand this pattern from a more general point of view. The diffusion and evaporation of water in starch is thought to be analogous to the diffusion and extraction of heat from a basalt flow. By combining direct sampling and x-ray tomography, fully 3D descriptions of columnar jointing were obtained with starch samples. We have characterized the pattern with several statistical indices, which describe its structure and relative disorder. These methods can resolve the ordering of the colonnade near the free surface. We identified two distinct mechanisms by which the mean column area increases during pattern evolution. We found both a slow, almost power-law increase in column area, as well as episodes of sudden catastrophic jumps in scale. The latter suggests that the column scale is not a simple single-valued function of drying rate, but rather a metastable state subject to hysteresis. Such metastable behaviour might explain a fundamental question about columnar jointing -- why the columns are so regular in the direction of their growth. Moreover, these experiments may help discriminate between the various theoretical models of this pattern forming process. Finally, our results lead to predictions that could be tested by field measurements on basaltic colonnades.
UR: http://www.physics.utoronto.ca/nonlinear
DE: 5104 Fracture and flow
DE: 8429 Lava rheology and morphology
DE: 8499 General or miscellaneous
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting