[S-CG40] Science of slow-to-fast earthquakes

[SCG40-29] Micro Gravity Effect of Earthquake at Yemi Underground Lab in Korea

★Invited Papers

*Jeong Woo Kim¹、MJ Dehghan¹、A Dorjsuren¹、J Oh²、EJ Son²、W Kim¹、S You²、B Kim³、I Woo⁴ (1.Univ. Calgary, CANADA、2.NIMS, KOREA、3.GIST, KOREA、4.Kunsan National Univ., KOREA)

キーワード：Superconducting Gravimeter, Yemi Underground Lab, Korea, Earthquake Detection, Prompt ElastoGravity Signals

Since its installation at Korea’s Yemi underground lab (YemiLab, ~1,008 m below the Earth surface) in October 2022, the iGrav Superconducting Gravimeter (SG) (serial #001) has been collecting high quality micro gravity data at 1HZ sampling rate. Site noise characteristics were evaluated by an NM calculated from the raw data between the frequency band 0.1-100 mHz. The YemiLab NM, particularly within the seismic band, is significantly low and close to the New Low Noise Model (NLNM). YemiLab was built based on the active iron mine which produces ~0.7 million tons of iron ores per year, and the SG is located more than 700 m away from the current mining area. Effects of daily blasts for mining activity revealed that the effect was recorded in the dewar and barometer pressures as well as Tilt balance data, but no effect was recorded in the raw SG data, indicating the blasting vibration is absorbed by the SG tilt system. A calibration factor of -92.17 Gal V^-1 was estimated with a theoretical tidal analysis for the first 6 months of raw SG, which was later replaced with -94.38 Gal V^-1 estimated by a parallel measuremernt with FG-5. After removing all known environmental effects, residual gravity was calculated. The spectral analysis reveals several unknown frequencies in the power spectrum density of the residual gravity, which require additional analysis. Co-seismic gravity changes were investigated for several earthquakes. As a first step to develop a micro gravity-based earthquake detection method, we independently write a python code to extract Prompt ElastoGravity Signals (PEGS) based on the paper by J.-P. Montagner et al. [ 2016], which was confirmed by the original code provided by the authors of the paper with the 2011 Tohoku-Oki earthquake data. YemiLab’s unique layout and minimal noise will contribute to better to extract PEGS.