Far-field Gravity and Tilt Signals by Large Earthquakes: Real or Instrumental Effects?

Abstract

A wide set of dynamics phenomena (i.e., geodynamics, Post Glacial Rebound, seismicity and volcanic activity) can produce time gravity changes, which spectrum varies from short (1… 10 s) to long (more than 1 year) periods. The amplitude of the gravity variations is generally in the order of 10 8…10 9 g, consequently their detection requires instruments with high sensitivity and stability: then, high quality experimental data. Spring and superconducting gravimeters are intensively used with this target and they are frequently jointed with tiltmeters recording stations in order to measure the elastogravitational perturbation of the Earth. The far-field effects produced by large earthquakes on records collected by spring gravimeters and tiltmeters are investigated here. Gravity and tilt records were analyzed on time windows spanning the occurrence of large worldwide earthquakes; the gravity records have been collected on two stations approximately 600 km distant. The background noise level at the stations was characterized, in each season, in order to detect a possible seasonal dependence and the presence of spectral components which could hide or mask other geophysical signals, such as, for instance, the highest mode of the Seismic Free Oscillation (SFO) of the Earth. Some spectral components (6.5’; 8’; 9’; 14’, 20’, 51’) have been detected in gravity and tilt records on the occasion of large earthquakes and the effect of the SFO has been hypothesized. A quite different spectral content of the EW and NS tiltmeter components has been detected and interpreted as a consequence of the radiation pattern of the disturbances due to the earthquakes. Through the analysis of the instrumental sensitivity, instrumental effects have been detected for gravity meters at very low frequency.