Long-period lunar Earth tides at the geographic South Pole and recent models of ocean tides

For many years the ocean tide models of Schwiderski (1980) were the standards used by the Earth tide community to interpret deviations of observed Earth tides from predictions on the basis of earth models constructed by seismologists. Recently, the TOPEX/POSEIDON altimeter mission provided new and improved information on pelagic ocean tides, which led several research groups to generate new models of the major oceanic tides. This in turn renewed our own interest in the observations of long-period lunar tides at the geographic South Pole that were reported many years ago with an attempt to interpret the deviation from predictions using the Schwiderski models. We used four different models of the fortnightly (M-f) and monthly (M-m) ocean tides to calculate their attraction and loading effects at the South Pole and compared the results with the observed gravity tides. In our earlier interpretation we did not realize that for long-period ocean tides the so-called 'Greenwich' phase does not refer to the phase of these tides at the latitude of Greenwich, but to the phase at the equator. This resulted in a mistake in the relative phases of Earth tide and ocean effect at the South Pole. For M-f we now find that all models predict the phase lead of the observed versus theoretical tides within the formal uncertainties; however, the amplitude is still underpredicted by 1.5-2 per cent. This could be due to several reasons: instrument calibration, errors in the body and/or ocean tide models, relaxation of the Earth's elastic properties, and the huge ice sheet of Antarctica. These possibilities are discussed. For the M-m tide the observed amplitude is well predicted within error; however, the uncertainties in the measurements are rather large.