Antarctic Geothermal Heat Flow, Crustal Conductivity and Heat Production Inferred From Seismological Constraints on Crustal Composition and Lithospheric Thermal Structure

Geothermal heat flow (GHF) is a crucial parameter governing ice flow rates at the base of the Antarctic Ice Sheet. Tectonic history biases the crustal distribution of heat-producing elements, and the pattern of mantle convection influences regional thermal structure, leading to significant intracontinental variations in Antarctic GHF of order 100 mW/m2. However, extensive ice cover across Antarctica severely limits the ability to directly measure GHF or crustal composition. Geophysical proxies are therefore required to access information pertaining to the lateral structure of GHF and its potential impact on ice sheet dynamics, important for assessing present and future thresholds in Antarctica. Previous studies have used geomagnetic data to infer the depth above which ferromagnetic structure is locked in, corresponding to the ~850 K isotherm. Others have relied on the sensitivity of seismic velocity to thermal structure to model local variations in surface temperature gradient. Both approaches require assumptions on crustal properties, which are typically chosen ad-hoc, and may affect GHF estimates in a significant and non-systematic manner. Other studies have used the observed covariation between lithospheric seismic velocity and GHF in regions with high measurement densities (e.g., continental USA) to map Antarctic seismic structure into GHF. This introduces a dependency of inferred Antarctic GHF on the range of tectonic environments sampled by the continental region used to derive the empirical relationship. Here, we adopt a distinct approach, in which Monte Carlo sampling is used to include crustal conductivity and heat production as free parameters in a numerical modelling procedure that fits theoretical geotherms to new probabilistic seismic inferences of upper mantle temperature structure beneath Antarctica. By integrating empirical constraints on crustal conductivity derived from P-wave velocity data, we are able to build distributions of covarying crustal conductivity, heat production, and GHF. This allows us to generate a model of Antarctic GHF which is complementary to that of other studies, and includes an estimate of lateral uncertainty structure based on the sensitivity of thermal gradients to crustal composition and anelastic deformation at seismic frequencies.