A Comparison of Proton and Self-Calibrating Rubidium Magnetometers for Tectonomagnetic Studies

Abstract

A network of 27 proton magnetometers (PM's), designed to detect stress related magnetic events of crustal origin, has been operating near active faults in California for the past decade. We present here comparative magnetic difference field data from PM's used in this net with that obtained from new self-calibrating rubidium magnetometers (SCR's). The instruments were first compared over a 50-m baseline in an aseismic and magnetically quiet region in Colorado. For PM's having either a 0.25-nT or a 0.125-nT least count, the observed difference variations were 0.2-nT and 0.17-nT rms, respectively. For SCR's having a 0.001-nT least count and a 100-second averaging interval, the difference variation was 0.002-nT rms. Power spectra of these data indicate that the noise for the PM's is close to their least count limit. However, the SCR noise decreases at about 20dB per decade until it approaches its least count limit 40dB below the PM limit, for periods less than 30 minutes. A similar experiment was conducted using collocated SCR and PM pairs separated by 13km along the San Andreas fault. Power spectra indicate that both systems are equivalent and are dominated by external noise at periods greater than 4 minutes. Below 4 minutes the PM noise approaches its least count limit while the SCR noise continues to decrease at about 20dB per decade until it is 20dB below the PM limit at a period of 30 seconds. Improved discrimination of magnetic transients caused by fault activity with periods of several minutes to perhaps an hour appears to be possible with higher sensitivity magnetometers.