| Summary: | Subglacial heat flow is used as a boundary condition for ice sheet models and in understanding the tectonicdevelopment and properties of the lithosphere. Existing Antarctic heat flow estimates at continental scaleare based on univariate modelling of a geothermal gradient and do not agree. Disparities arise fromassumptions regarding lithospheric properties such as crustal heat production, upper mantle compositionand dynamic neotectonics. We employ a 'similarity approach' that compares Antarctic observables with observables linked to existinghigh-quality heat flow measurements from global compilations. Previous studies that use similarity tointerpolate heat flow values elsewhere, utilise datasets that do not extend to the Antarctic interior withsufficient reliability. Here, we optimise the similarity approach for existing Antarctic geophysical andgeological datasets by applying a careful sensitivity analysis and introduce weighting of observables.Observables used include topography, distance to volcanoes, geophysical data sets, and derived productssuch as depth to Curie temperature isotherm, seismic wave speed and curvature of gravity field. We alsoinclude geological observations. In total, 15 observables are used. The new heat flow map, Aq1, is presented together with uncertainty and measures of information entropyin widely used formats. We also provide the complete workflow as open source Python code relying on theagrid package. The complete computational framework allows for testing of alternative inputs and updatesas new data becomes available.
|
|---|