Detailed temperature measurements in growing Antarctic sea ice allow to fit directly the thermal conductivity at a variety of temperatures, depths and gradients.
Some evidence has been presented for nonlinear thermal effects in sea ice, raising the question of the importance of brine movement for heat transport. More analysis of the temperature data is planned.
Local salinity and ice density changes with depth are unknown, and do have a significant effect on fitted thermal conductivity. Future experiments would benefit from such information, although since these properties change with time, and involve destructive methods, it is difficult to measure them in conjunction with temperature measurements.
The work of Woods and Linz (1992) raises interesting questions about whether there is a critical Rayleigh number for the onset of convection in tilted brine tubes. It is planned to extend their model to the case of brine tubes in sea ice, and to undertake a numerical and asymptotic analysis of this model.