V31C-3045
Modeling mechanical and thermo-mechanical erosion by flowing lava at Raglan, Cape Smith Belt, New Québec, Canada

Wednesday, 16 December 2015
Poster Hall (Moscone South)
Vincenzo Cataldo1, David A Williams1 and C Michael Lesher2, (1)Arizona State University, Tempe, AZ, United States, (2)Laurentian University, Mineral Exploration Research Centre, Department of Earth Sciences, Goodman School of Mines, Sudbury, ON, Canada
Abstract:
The 1.5-D Williams et al. model of thermal erosion by turbulent lava was recently applied to the Athabasca Valles lava channel on Mars, in an attempt to establish the importance of thermal erosion in excavating this ~80-100 m deep outflow channel. The modeled erosion depths (0.4-7.5 m) are far less than the depth of the channel which, combined with the short duration of the eruption, suggests that mechanical erosion may have had a greater role. Several studies suggest that mechanical erosion by lava is more important in channel-tube formation than previously thought, under certain circumstances. How would we be able to distinguish between mechanical and thermal erosion? By investigating model results when substrate properties change, as we move from a consolidated, mechanically strong substrate to a partially consolidated or unconsolidated, mechanically weaker substrate. The Proterozoic Raglan komatiitic basalt lava channel of the Cape Smith Belt, New Québec, Canada is a complex erosional environment involving invasive erosion of both sediment and gabbro substrates - which makes it a critical test case. The lava eroded an upper layer of soft sediment, with erosion at the tops, bottoms, and sides of the conduit, through underlying gabbro, and then burrowed laterally into underlying sediment, a scenario requiring a two-dimensional modeling approach. Using the available field data, we will simulate two-dimensional thermomechanical and mechanical erosion interfaces on all sides of a turbulent lava flow by creating a finite-element mesh. The mesh will be defined by the geometry of the lava flow at those lava conduits for which data on lava and substrate composition, lava thickness, slope of the ground, conduit area and volume, and lava flow length are available. Ultimately, this model will be applied to lunar sinuous rilles and martian lava channels for which the use of a two-dimensional approach is needed.