Ambient noise and data quality analysis of strong motion stations in Inner Mongolia region

Along with the construction of the project of National Earthquake Intensity Rapid Reporting and Early Warning, Inner Mongolia has established 193 strong motion stations and a data center. This paper calculated the acceleration power spectral density （PSD） of countinous waveform records of the 193 stations in 2022 by using the power spectral probability density function （PDF）. By drawing the distribution maps of PSD, power spectral PDF and RMS value of ambient noise in different frequency bands in different time-space with different-type accelerometers, the variation characteristics of ambient noise at the strong motion stations in our studied region is evaluated, and then the PSD anomalies are analyzed. Accordingly, the constructive suggestions are put forward for the anomaly detection and strong ground motion data processing. The results show that the diurnal difference in PSD of ambient noise at different stations with different seismometers mainly appears in the frequency band above 1 Hz, and the diurnal PSD difference in different regions differs largely. The ambient noise values in the data recorded by seismometers QA-2g, JS-A2 and TDA-33M at frequency 2 s show little change with space and direction, so we can get reliable anomalies from the observation data by monitoring the ambient noise at frequency 2 s. The change of ambient noise with high frequency band at strong ground stations mainly comes from the change in environmental noise, but the influence of instrument self-noise cannot be completely ignored. Anomalous strong ground data is mainly resulted from large zero point shift of accelerometers, accelerometer failure, pulse interference and incorrect system parameters. Therefore it is recommended to optimize the observation environment of the stations, further standardize the instrument installation and debugging, and to configure and review the JOPENS system parameter.