Modelling flow and accreted ice in subglacial Lake Concordia, Antarctica

More than 150 subglacial lakes have been discovered in Antarctica so far. Due to obvious challenges with exploration, numerical modelling remains one of the major tools to acquire information about those hard-to-access objects. Until now only the huge Lake Vostok has been investigated in detail. Thi...

Full description

Bibliographic Details
Published in:Earth and Planetary Science Letters
Main Authors: Thoma, Malte, Grosfeld, Klaus, Filina, I., Mayer, Christoph
Format: Article in Journal/Newspaper
Language:unknown
Published: 2009
Subjects:
Online Access:https://epic.awi.de/id/eprint/21178/
https://epic.awi.de/id/eprint/21178/1/Tho2009c.pdf
https://doi.org/10.1016/j.epsl.2009.06.037
https://hdl.handle.net/10013/epic.33476
https://hdl.handle.net/10013/epic.33476.d001
Description
Summary:More than 150 subglacial lakes have been discovered in Antarctica so far. Due to obvious challenges with exploration, numerical modelling remains one of the major tools to acquire information about those hard-to-access objects. Until now only the huge Lake Vostok has been investigated in detail. This paper focuses on Lake Concordia - the second largest subglacial lake in Antarctica over which substantialgeophysical data has been collected. This lake is covered by about 4000m ice and is located near Dome C. In order to apply numerical models to the hard-to-access Antarctic subglacial lakes, decent geometries and boundary conditions are required. In this study we present the results of airborne gravity inversion, suggesting that this lake has an area of 617 km^2, a volume of 31 km^3, and a maximum water columnthickness of 126 m. This bathymetry is used as geometry input for an established 3D-numerical lake-flow model to simulate the circulation and basal mass balance. Compared to our model studies of subglacial Lake Vostok, we obtain a general circulation pattern that is significantly weaker (due to the smaller size of the lake) and of reversed orientation (due to the reversed ice surface tilt). The modelled mean horizontal and vertical velocities are in the order of 0.2 mm/s and 0.5 μm/s, respectively. The larger molecular convective velocity estimations (1.35 ± 0.13 mm/s and 0.81 ± 0.08 mm/s) are similar to Lake Vostoks. The modelled average melting and freezing rates are 4.3 ± 1.1 mm/a and 1.1 ± 0.3 mm/a, respectively, and the corresponding fresh water gain is 58 ± 27 dm^3/s. Integration of the modelled freezing and melting along prescribed ice flow lines allows us to calculate the distribution and thickness of accreted ice at the ice sheet bottom. We estimate a volume of 2.6 ± 2.0 km^3 (8.3 ± 8.2 % of the total lake volume) occupying the north-eastern corner of the lake covering an area of 159 ± 48 km^2 (26 ± 9 % of the total lake area). With about 16.800 ± 7.600 years, the residence time of the lakes water is significantly shorter than Lake Vostoks.