Towards Estimating the Salinity of Subglacial Lakes from Aerogeophysical Data

Subglacial lakes are now being recognized as an integral part of the global cryosphere and the global climate system. Knowledge of the physical, chemical and biological processes operating within these features is crucial for addressing questions about the presence and functioning of life in subglacial lakes. However, little is known about the prevailing in situ environmental conditions. Existing airborne ice-penetrating radar and laser altimeter data over large subglacial lakes can be used to estimate the salinity without penetrating the lakes. The underlying assumption for estimating the salinity from aerogeophysical data is that the ice sheet above large subglacial lakes is in hydrostatic equilibrium and the hydrological potential of the lake surface is constant. Because the flexural support of the ice near the shoreline reduces the load and thus the ice overburden pressure the data analysis has to be restricted to regions several ice thicknesses away from the shoreline allowing this analysis only over the largest know subglacial lakes. In order to estimate the salinity from aerogeophysical data several effects have to be included in a refined calculation of the hydrological potential: 1) A firn layer has to be included in the calculation of ice thickness from radar wave travel times. Because the velocity v of electromagnetic waves in firn is faster than in ice (vfirn > vice) a firn correction has to be added to the preliminary ice thickness estimates. Including a firn layer will impact the ice overburden pressure estimate by changing the total ice thickness estimate and reducing the load by a near surface layer with lower density. 2) Lateral changes in the radar wave velocity for the deeper section of the ice sheet have to be accounted for. 3) The vertical density structure of the ice ρice = ρice(z) has to be included in the calculation of the ice overburden pressure pice. The water density is a function of the salinity, temperature, and pressure, i.e. ρwater = ρwater(S, T, pice). It is important to note that the horizontal and vertical variations in all parameters that impact salinity estimates cannot be neglected. Many of the unknown parameters involved in the salinity estimate can be predicted at least within certain boundaries. I will present a strategy towards estimating the salinity of subglacial lakes from aerogeophysical data, that carefully balances the unknown parameters involved in the calculation.