Microgravity surveying before, during and after distant large earthquakes

• Published in: Journal of applied geophysics Vol. 197; p. 104542
• Main Authors: Boddice, Daniel, Metje, Nicole, Tuckwell, George
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01-02-2022
• Subjects: Earthquake; Microgravity; Noise; Earthquake; Noise; Microgravity
• ISSN: 0926-9851; 1879-1859
• DOI: 10.1016/j.jappgeo.2022.104542

Microgravity surveying is a widely used geophysical method, especially for detection of deeper features which are difficult to detect with other techniques. However, the accuracy of the readings is strongly affected by the amount of microseism noise, which is typically rejected by using long integration times for each measurement cycle. Large seismic events such as earthquakes create large ground acceleration waves which have been known to affect gravity readings. While previous data which have been collected serendipitously, tend to only record the final readings, unique long-duration datasets during and after three earthquakes including the raw data (sampled at 6–10 Hz) from field gravimeters (Scintrex CG-5 and CG-6) in static locations recorded during surveys are presented. The aim of the current study is to characterise both long and short-term effects of the generated seismic accelerations on measurement accuracy and repeatability, and identify changes to commercial practice to mitigate these effects.
During earthquakes, the nature of the microseism noise was fundamentally altered by each of the different associated seismic waves. Minor effects were found for body P- and S-waves, but much larger effects were found for surface waves, especially the Rayleigh waves which gave errors many times those which would occur in normal conditions. These accelerations persisted in the data for several hours or even days after the earthquake affecting instrument performance. The main finding is that the optimum course of action is to identify the earthquake early by analysing the data in the frequency domain through FFT, switch to 60 s cycles and return to the base station until the strongest waves have passed. Surveying on subsequent days was affected by lower frequency free earth oscillations requiring removal of the unwanted signals using linear trends between at least hourly base stations. Using these techniques will both facilitate data collection as well as improve data confidence in these challenging conditions. Instruments operating in a gradiometer configuration were shown to be comparatively unaffected by the increased noise for typical commercial integration times paving the way for the next generation of instruments to operate successfully, even with this challenging environmental noise.
•Raw gravity data from field gravimeters were taken following distant earthquakes.•Strategies to mitigate noise from waves and free earth oscillations are proposed.•During the strongest waves, switch to 60 s cycles and return to the base station.•Linear trends between hourly base station readings are recommended to remove drift.•Instruments operating in a gradiometer configuration were comparatively unaffected.