NUMERICAL MODELING OF THE LAST GLACIAL MAXIMUM YELLOWSTONE ICE CAP

We use two 2-D numerical glacier models with complementary strengths to address several pressing questions related to the LGM Yellowstone Ice Cap. First, we assess the potential range of paleoclimate states that promote the formation of the ice cap using an implicit 2-D ice flow and energy balance model (Plummer and Phillips, 2003). The detailed surface mass balance scheme also allows us to determine the likely locations of initial glacier growth and subsequent coalescence (e.g. the Beartooth plateau, Absaroka mountains). Second, we explore explanations for spatial asynchronies in the Yellowstone Ice Cap LGM moraine ages. The maximum terminal moraine ages from the western portion of the ice cap are one to four thousand years younger than ages on the east side of the ice cap (Licciardi and Pierce, 2008). As a potential explanation for this discrepency, we assess the importance of orographic precipitation in promoting the westward growth of the ice cap through time (e.g., Pierce, 1979) and the potential for a thickening ice cap to reduce leeside accumulation leading to the contraction of eastward-flowing outlet glaciers. For this exploration, we use an efficient semi-implicit 2-D ice-flow model with a surface mass balance scheme that allows inclusion of a simple algorithm for orographically-forced precipitation (e.g., Roe, 2005). Finally, we couple the semi-implicit 2-D ice-flow model to a fully implicit solution for flexural isostatic response to the evolving ice load, and test the sensitivity of the ice cap to a generous range of plausible solid Earth rheologies. This provides initial estimates of the effect of flexural depression on ice dynamics, potential locations of proglacial lakes, and the location and magnitude of the forebulge. This modeling effort was inspired by the instruction of Bill Locke at Montana State University and the work of Ken Pierce and Joe Licciardi, whose modeling and cosmogenic radionuclide dating efforts inform and constrain our research.