Estimating the diffusion in the deepest section of the Dome-C ice core using a new statistical method

The diffusion of stable water isotopes in firn and ice is a complex process which can smooth the measured isotope profile and thus remove the high frequency variations. While for the firn this process is relatively well understood and can be accounted for, this is not the case in the deepest parts of the ice core, where additional processes are introduced due to the increased temperatures near bedrock. Combined with the extreme thinning the deep ice has experienced over hundreds of thousands of years, variations up to centennial or even millennial timescales can be heavily attenuated. Understanding how to best recover this signal in the deep ice is crucial to get a reliable record in deep ice cores such as the Beyond EPICA Oldest Ice Core that is currently being drilled.

In order to reconstruct the climate signal of this old ice, an accurate estimation of the diffusion length is necessary. Current estimation methods are mostly suitable for firn and shallow ice as they are assuming a rather stationary underlying climate signal. In this contribution, we present a method which approaches the issue without assuming the low frequency climate variability is negligible. Using this method on the high-resolution Dome-C isotope data (doi.pangaea.de/10.1594/PANGAEA.939445), we provide an improved estimate of the diffusion length of the Dome-C ice-core. Both the diffusion length estimate of the deep ice in Dome-C as well as the new method are useful for the interpretation of future deep ice coring projects such as Beyond EPICA Oldest Ice Core.