Minimizing the uncertainty in shallow borehole-temperature logging

The temperature distribution through the ice sheet is a record of past climate changes. This allows reconstructing the past surface temperature history from the vertical temperature profile measured in a borehole (‘borehole palaeothermometry’). Such a reconstruction from the inversion of the heat advection-diffusion equation is independent from the analysis of oxygen isotopes in the ice but requires high-precision measurements of the borehole-temperature on the milli-Kelvin level. 

The precision and accuracy of the measurements is influenced by the spatial and temporal (seasonal to multi-hour) variations of the borehole temperature, the influence of the measurement setup (disturbance of the temperature profile from the borehole or snow pit drilling), the heat transfer between the ice/snow and the sensor as well as the uncertainty of the measurements (depth uncertainty and sensor precision and accuracy).  

Here, we report on our efforts to quantify these uncertainties of near surface (10m) firn and shallow (~200m) borehole temperature measurements using a newly developed winch-based borehole measurement system as well as stationary chains of borehole/firn temperature sensors at replicate sites near the EDML drilling site, Antarctica.