Modelling the effect of atmospheric pressure variations on gravity

Abstract

A model is developed to describe the effects of local and global atmospheric pressure fluctuations on local gravity. The calculated admittance of gravity to the pressure tides S1–S4 at Pinon Flat, California is compared to those measured using the superconducting gravimeter. The theoretical and measured admittances are in good agreement for the S3 and S4 tides. The difference between theory and observation at S2 can probably be explained by the ocean tide at that frequency while the discrepancy at S1 remains unexplained. Techniques for the removal of random pressure variations from gravity records are described. Using statistics derived from spherical harmonic decompositions of the Earth's pressure field, it is shown that more than 80 per cent of the pressure induced gravity variations should be removed from a gravity record by employing a single barometer and least-squares procedure. It is further shown that adding a ring of barometers should increase the effectiveness to about 90 per cent. The amount of barometric pressure 'noise' which remains after subtracting the simultaneous record of two gravimeters is calculated as a function of their separation and indicates that significantly less noise is present in the difference than in the individual records for separations of up to 17° of arc.