Time-dependence of the spatial pattern of accumulation rate in East Antarctica deduced from isochronic radar layers using a 3D numerical ice-flow model

In East Antarctica surface mass balance data can only be obtained from the sparsely distributed ice cores when considering time periods greater than a few decades. Observations of internal layers measured by airborne ice penetrating radar, in principle, permit extrapolation of mass balance information from these ice cores. We use radar survey lines gathered in the 1970s, and a three‐dimensional numerical model, to investigate the feasibility of such extrapolation, seeking to match the calculations of englacial layer geometry with observations. First, we justify the use of a three‐dimensional model by showing that simple vertical flow models cannot explain all the observations and that horizontal advection is a significant glacial process. Then we examine processes that affect calculations of layer geometry, finding that spatial accumulation‐rate patterns are extremely important while geothermal heat flux and flow mode (sliding or internal deformation) are of substantially less importance. Where the layer is from the Last Glacial Maximum (17.5 ka), we find a very good match between the spatial pattern of accumulation rates inferred from this layer and the modern spatial pattern of accumulation rates. When considering deeper layers from beyond the current interglacial, we find that a different spatial accumulation‐rate pattern must have existed, in addition to the known change in accumulation rate from ice cores. The glacial spatial accumulation‐rate pattern would have had proportionally greater accumulation at the South Pole than now, compared with the Vostok and Dome C ice cores