Further analysis of the influence of site conditions and earthquake magnitude on ground-motion within-earthquake correlation: analysis of PGA and PGV data from the K-NET and the KiK-net (Japan) networks

We analyzed the within-earthquake correlation (peak ground acceleration PGA and velocity PGV) in Japan by taking into account local geological conditions. The database includes more than 23,700 records (K-NET and the KiK-net networks) from 84 events (M 4.2–7.4) occurring in 1999–2011. The correlation structure was investigated for residuals obtained using two recently developed ground-motion models for Japan. We have shown that the level of within-earthquake correlation may vary significantly depending on site classes, general geological conditions and earthquake magnitude. The results of the analysis are in general agreement with the findings obtained recently for Taiwan (Sokolov et al. in Bull Earthq Eng 10(5):1401–1429, 2012. doi:10.1007/s10518-012-9368-5). The high level of ground-motion correlation reflects the presence of a non-random component in the residuals that may be caused by the joint influence of surface soil and thick sediments, and this influence depends on the earthquake magnitude. Therefore, a single generalized model of within-earthquake correlation across geologically heterogeneous regions may not be adequate. The application of empirical correction factors, which may be a function of earthquake magnitude, source-to-site distance, site class, and average shear-wave velocity for a given station, allows for effective reduction in the level of within-earthquake correlation.