Determination of the frequency response of seismic and infrasonic IMS station sensors using a traceable on-site calibration approach

Abstract

As part of the joint research project "Metrology for low-frequency sound and vibration - 19ENV03 Infra-AUV" laboratory calibration methods for seismometers and microbarometers in the low frequency range down to 0.01 Hz have been developed. These procedures provide the possibility of traceable on-site calibration during operation for field sensors of the Comprehensive Nuclear-Test-Ban Treaty Organization’s (CTBTO) International Monitoring System (IMS). The traceable calibration allows for accurate amplitude and phase information as well as for an assignment of uncertainties in amplitude and phase. Thereby, data quality and the identification of treaty-relevant events is improved. The on-site calibration procedure requires a reference sensor with a precise and traceable response function which is provided by the newly developed laboratory calibration methods, as well as the record of sufficient coherent excitation signals within the relevant frequency range. The reference sensors can be installed as transfer standards co-located to the operational IMS station sensors without disturbing their regular measurements for treaty validation purposes.At IMS stations PS19 and IS26 in Germany we performed on-site calibration tests with both seismometers and microbarometers calibrated in the laboratories at PTB and CEA, respectively, using signals from different natural and anthropogenic excitation sources. Following the approach of Gabrielson (2011) with modifications from Charbit et al. (2015) and Green et al. (2021), the gain ratio between the station sensor under test and the reference sensor is calculated. By multiplying the gain ratio with the precise frequency response of the reference, the frequency response function for both magnitude and phase of the station sensors including site-specific factors such as the wind noise reduction system or possible effects of pre-amplifiers and data loggers are determined.We present calibration results derived from the comparison of IMS station sensors with the laboratory-calibrated instruments along with the nominal responses. The results show agreement with deviations of less than 5% from the nominal response function for frequencies below 10 Hz for all components. The traceable determination of the response for the individual components in detail improves the sensor quality; subsequently waveform amplitudes can be estimated correctly.