Stressed Scarps at the North Pole of Mars

In this study, we simulated the thermal behavior of steep scarps at the north pole of Mars with a 1D semi-implicit thermal diffusion model with radiative boundary conditions at the top surface and negligible heat flow from beneath. The thermophysical properties of the scarp face are those of water ice at 200K overlain by a neglible dust cover. The steepness of these slopes means that they exchange reflected and emitted radiation with surrounding flat terrain as well as open sky. We separately simulated the temperatures of the surrounding terrain (assumed to be dark sand when defrosted) to calculate the upwelling fluxes onto the scarp face. Near-vertical surfaces at the pole have similar illumination geometries to flat ter-rain at the equator, but with much larger atmospheric path lengths making their heating very sensitive to interannually-variable aerosols. We calculate scarp and surrounding flat surface heating with a 32-stream pseudospherical radiative transfer model using the pyRT.DISORT code (https://github.com/kconnour/pyRTDISORT) and atmospheric states extracted from the Mars Climate Database (https://www-mars.lmd.jussieu.fr/mcdpython/).

The thickness of the dust cover is a crucial controlling factor on the thermal behavior of the ice and it strongly affects the season at which the scarp loses its CO2 frost cover. HiRISE images show the scarp has already defrosted by Ls 350, so the dust cover must have negligible thickness, although it is still sufficient to warrant using a dusty albedo. Shallower polar slopes (e.g. the spiral troughs) appear unfractured. Our modeling suggests this is due to thicker dust covers, possibly because they do not experience the scouring avalanches of steep slopes.

We follow the approach of Mellon [1997] to solve for the time varying stress in an initially unfractured viscoelastic solid. No lateral strain can occur, so surface-parallel thermal expansion and contraction is opposed by elastic stresses over short timescales that decay over longer timescales due to grain-size-dependent viscous effects. The thermal history at each depth can be used to calculate these stresses, as well as surface-normal displacement. We use a Zenner pinning approach with NPLD dust abundances derived from RADAR sounding data to constrain ice grain sizes to be 10–1000 um.

During much of the northern summer, diurnal temperature oscillations in the ice are large and associated surface stress can vary by several MPa and alternate between extensional and compressive. Compressional stresses occur during warmer periods and are thus more effectively viscously relaxed than extensional stresses. Colder ice allows for the accumulation of greater extensional stress during polar night.