Six-Degree-of-Boradband Seismogeodesy by Combining Collocated High-Rate GNSS, Accelerometers, and Gyroscopes

Jianghui Geng,Qiang Wen,Qijin Chen,Hua Chang
DOI: https://doi.org/10.1029/2018GL081398
IF: 5.2
2019-01-01
Geophysical Research Letters
Abstract:Combining collocated high-rate Global Navigation Satellite Systems (GNSS) and accelerometers produces broadband seismogeodetic displacements. However, accelerometer data must be heavily downweighted due to their baseline errors originating primarily in instrument rotations, and therefore their contribution to seismogeodetic displacements is seriously underestimated. We further introduced a gyroscope into this classic seismogeodesy to mitigate baseline errors and formulated advanced six-degree-of-freedom (6-DOF) seismogeodesy without undervaluing accelerometer data. A shake table holding one GNSS antenna, four accelerometers, and one gyroscope was used to simulate waveforms from the 2010 Mw 7.2 El Mayor-Cucapah earthquake. We found that the displacements derived from the 6-DOF seismogeodesy were up to 68% more accurate than those from the classic seismogeodesy over 0.04-0.4 Hz. Moreover, broadband rotations containing the permanent components were also generated, which were unachievable by integrating gyroscope data. We believe that the 6-DOF seismogeodesy is capable of improving both source rupture studies of large earthquakes and high-rise monitoring under strong seismic waves. Plain Lauguage Summary The faulting mechanism of large earthquakes has always been an enigma to the seismology community. One important reason is that we cannot measure accurately the low-frequency (e.g., <0.1 Hz) displacements at near-source stations using existing techniques. In particular, seismometer recovered low-frequency displacements are normally biased and incomplete due to intrinsic instrument limitations; Global Navigation Satellite Systems is able to obtain unbiased broadband (e.g., 0-5 Hz) displacements, but they are over 2 orders of magnitude noisier than seismometer measurements. We therefore develop an advanced technique that combines collocated Global Navigation Satellite System receivers, accelerometers, and gyroscopes to capture highly accurate earthquake-induced displacements and also rotations, spanning the frequencies from 0 to tens of hertz. We demonstrate that the accuracy of low-frequency displacements can be improved by up to 68% in contrast to traditional measuring techniques, and the permanent rotations can also be recovered, which can never be achieved by only inertial sensors (i.e., gyroscopes and accelerometers). We believe that this new technique can assist in deepening our understanding of the rupture process of large earthquakes, and the broadband rotations have the potential to unveil more unknowns about earthquake sources and Earth media beyond our present knowledge. We can get more prepared for future destructive earthquakes.
What problem does this paper attempt to address?