Three-channel correlation analysis: a new technique to measure instrumental noise of digitizers and seismic sensors
Sleeman, R., A. van Wettum and J. Trampert

Until 2006 the self-noise of instruments was usually determined by analyzing waveform data obtained from two collocated, 'identical' instruments. Correlation analysis could reveal the self-noise of both instruments, provided that the transfer function of both systems was known with high accuracy. This type of a-priori information however is not always known. In 2006 it was shown by Sleeman, van Wettum and Trampert that knowledge about the transfer function of the seismic instrument is not required anymore when dealing with 3 instruments. The correlation analysis of data from 3 collocated, 'identical' instruments direclty reveals the instrumental noise of the equipment based on the power spectral densities of the recordings only. Today, this new technique to measure self-noise has become a standard technique for manufactors of seismic sensors, and scientists.

Obviously, knowledge about self-noise is important in the interpretation of the USGS Low Noise Model (Peterson, 1993). In particular at lower frequencies (e.g. below 5 mHz) the Low Noise Model may need some minor revision as the noise level (at those frequencies) seems to be determined by instrumental self-noise. To estimate the self-noise of the STS-2, one of the most commonly used broadband seismic sensors, KNMI and ZAMG (www.zamg.ac.at) deployed a long-term experiment in the Conrad Observatory in Austria. In this experiment 3 STS-2 sensors were collocated during one year on one of the piers in the observatory, and protected from thermal radiation and convection by a new type of thermal insulation consisting of thin layers of neoprene. Results of this experiment are submitted to BSSA (Bulletin of the Seismological Society of America).

Reference:
Sleeman, R., A. van Wettum and J. Trampert, Three-channel correlation analysis: a new technique to measure instrumental noise of digitizers and seismic sensors. Bulletin of the Seismological Society of America, 2006, 96, 1, 258-271, doi:10.1785/0120050032.