To the problem of noise reduction in sea-level records used in vertical crustal movement detection

The aim of this investigation is to develop a simple technique that would allow us to use the sea-level records for detecting contemporary vertical crustal movements of duration from several months to several years. The choice of auxiliary data needed for any such analysis is restricted to the regularly available meteorological data to make this approach possible in routine search for precursory movements in earthquake-prone areas. A linear mathematical model is designed to evaluate the effect of atmospheric temperature and pressure variations, river discharge, long periodic tides and Chandlerian motion. Spectra of the residual sea-level variations are also shown. It is concluded that local episodic crustal movements of a magnitude larger than some 10 cm may be detectable by this approach. If finer resolution is needed then it it necessary to also account for steric level, wind, and sea-current variations, for which data are largely non-existent.