Tharsis Montes Ice Sheet Models at High Obliquity Driven by Gcm Results

Introduction: Fan-shaped deposits on the flanks of the Tharsis Montes volcanoes [1-3] have been interpreted to be glacial in origin, and are likely Amazonian in age [4]. The deposits contain three distinct facies [5], each of which can be associated with different glacial processes [1]. Earth analogs to these three facies have been identified in the Dry Valleys of Antarctica and their climatic implications described [6-7]. In previous work we have shown that fundamental differences between the atmospheric snow accumulation environments on Earth [8] and Mars [9-10], combined with the University of Maine Ice Sheet Model (UMISM) [8,11-12] constrained by geological observations [1-5], allowed us to characterize the mass balance of the Martian ice sheet by two equilibrium lines, and that glacial accumulation is favored on the flanks of large volcanoes, not their summits as seen on Earth. In addition, we have shown that coupling this mass balance parameterization to sample spin-axis obliquity histories [13-14] leads to chronologically reasonable glacial episodes with a maximum configuration that is in accord with the geological observations [15]. However, we found this to be true only for repeated advance and retreat during multiple 100 Ka obliquity cycles where the mean value is in excess of 45. In our past reconstructions, we used UMISM constrained by the geological record to define the parameterization of mass balance. While this approach gave a good fit to the geologic record, it yielded little information about how the atmospheric circulation of Mars was actually changing during the periods of high obliquity. In this contribution we describe the use of results from a focused run of an atmospheric general circulation model (GCM) for Mars at high obliquity [16-17]. This GCM, run for a high-obliquity climate, favors deposition of snow on the northwest flanks of the Tharsis Montes due to upwelling and adiabatic cooling of moist polar air as it rises up the slopes of the volcanoes. Predicted, rather than parameterized, accumulation rates are used to drive UMISM, and the resulting ice sheets are compared to the geological evidence. This allows us to assess the validity of the GCM results, and also to assess both the spatial geometry in terms of areal extent and ice volume, and the temporal response in terms of how long such a high obliquity climate must exist to create ice sheet imprints that are in agreement with observed landforms. Modeling: UMISM uses the well-tested shallow-ice approximation [8,11-12], which is suitable for modeling ice sheets on Mars because the cold temperatures and low accumulation rates make wet-bed sliding unlikely. In the shallow-ice approximation, a combination of mass and vertically-integrated momentum conservation yields a time-dependent partial differential equation for the ice thickness that requires as source the net mass balance (the difference between a positive deposition of ice and any negative removal over an annual cycle) at each point in the domain. In this simulation, the mass balance distribution is obtained from the results of a GCM run for a high-obliquity climate. Mass Balance Distribution: The GCM used was the Martian Global Climate Model of the Laboratoire de Meterologie Dynamique [18-19], a well-tested model able to adequately simulate present Martian climate. A high-obliquity climate was simulated for 45 degrees obliquity (near the most probable value of 41.8 [13]) with a spatial resolution of approximately 2 degrees. The simulation showed net accumulation of 30-70 mm/yr on the western flanks of Olympus, Ascreus, Arsia, and Pavonis Montes, largely due to adiabatically cooled westerlies blowing upslope. The model was relatively insensitive to obliquity, as long as it was greater than 40 degrees, but mildly sensitive to atmospheric dust load, an unknown at high obliquity.