MOESM3 of Effect of newly refined hypocenter locations on the seismic activity recorded during the 2016 Kumamoto Earthquake sequence Tomoko Yano Makoto Matsubara 10.6084/m9.figshare.c.3791047_D3.v1 https://springernature.figshare.com/articles/journal_contribution/MOESM3_of_Effect_of_newly_refined_hypocenter_locations_on_the_seismic_activity_recorded_during_the_2016_Kumamoto_Earthquake_sequence/5052526 Additional file 3: Figure S2. Histogram showing distribution of standard deviation of the difference from the hypocenter depth of our final result to the depth relocated due to 100 different velocity models in region (SR2). Top panel is the background seismicity (JUICE catalog) and bottom panel is the seismicity of aftershocks. This test implies that the hypocenter can be located its depth about 1 km in error due to an incorrect velocity structure. In other words, it is less likely that the hypocenter depth is more than 5 km apart from the final result because of the velocity structure. The 100 models of velocity structure are generated by changing the moho depth from 10 km to 40 km from the original velocity model (Ukawa et al. 1984). 2017-05-30 05:00:00 The 2016 Kumamoto Earthquake Double-difference method Active fault zone The Beppu–Shimabara graben