Seismic reflection soundings of ice thickness and seabed depth were acquired on the Larsen C Ice Shelf in order to test a sub-shelf bathymetry model derived from the inversion of IceBridge gravity data. A series of lines were collected, from the Churchill Peninsula in the north to the Joerg Peninsula in the south, and also towards the ice front. Sites were selected using the bathymetry model derived from the inversion of free-air gravity data to indicate key regions where sub-shelf oceanic circulation may be affected by ice draft and sub-shelf cavity thickness. The seismic velocity profile in the upper 100 m of firn and ice was derived from shallow refraction surveys at a number of locations. Seismic velocities in the water column were derived from previous in situ measurements.

Uncertainties in ice and water cavity thickness are in general < 10 m. Compared with the seismic measurements, the root-mean-square error in the gravimetrically derived bathymetry at the seismic sites is 162 m. The seismic profiles prove the non-existence of several bathymetric features that are indicated in the gravity inversion model, significantly modifying the expected oceanic circulation beneath the ice shelf. Similar features have previously been shown to be highly significant in affecting basal melt rates predicted by ocean models. The discrepancies between the gravity inversion results and the seismic bathymetry are attributed to the assumption of uniform geology inherent in the gravity inversion process and also the sparsity of IceBridge flight lines. Results indicate that care must be taken when using bathymetry models derived by the inversion of free-air gravity anomalies. The bathymetry results presented here will be used to improve existing sub-ice shelf ocean circulation models.