On the minimum of vertical seismic noise near 3 mHz

Abstract

SUMMARY Acceleration power spectral densities of vertical seismic noise at the best seismic stations show a minimum near 3 mHz. We suggest that this minimum is caused by a cancellation near this frequency of Newtonian attraction vs. free air and inertial effects exerted by atmospheric phenomena on the sensor mass. Simplistic models of atmospheric phenomena are used to quantify this effect and examples are shown for special atmospheric events.

Clear evidence for the sign-reversal of the pressure admittance to gravity near 3mHz

“…Peterson (1993) and Berger et al (2004) show in their studies of broad-band seismic noise at many stations of the Global Seismic Network for many time windows that the lower envelope of vertical acceleration noise power spectral densities (PSD) has a minimum near 3 mHz not present in the horizontal components. This minimum in the vertical PSD and the difference between vertical and horizontal noise at these frequencies is nicely explained by the contributions from the atmosphere as described in detail by Zürn and Wielandt (2007) and . This special feature in the vertical noise PSD allows the clear detection of the Earth's background free oscillations ("Hum") in the vertical component records of many stations (e. g. Nawa et al, 1998;Suda et al 1998;Tanimoto et al 1998;Nishida et al 2002;Ekström, 2001;Fukao et al, 2002;Rhie and Romanowicz, 2004;Kurrle and Widmer-Schnidrig, 2006), while in contrast the horizontal hum can only be detected at barely a handful of them (Kurrle and Widmer-Schnidrig, 2008).…”

“…This special feature in the vertical noise PSD allows the clear detection of the Earth's background free oscillations ("Hum") in the vertical component records of many stations (e. g. Nawa et al, 1998;Suda et al 1998;Tanimoto et al 1998;Nishida et al 2002;Ekström, 2001;Fukao et al, 2002;Rhie and Romanowicz, 2004;Kurrle and Widmer-Schnidrig, 2006), while in contrast the horizontal hum can only be detected at barely a handful of them (Kurrle and Widmer-Schnidrig, 2008). Zürn and Wielandt (2007) were not able to directly demonstrate the reversal in the sign of the pressure admittance for vertical acceleration in real records. They show two examples in which the experimentally found admittances of 2.65 (their Fig.…”

“…Zürn and Wielandt (2007 , Table 1) extend the list to other phenomena (e. g. tides) recorded by vertical accelerometers and add Newtonian attractions of the sensor mass arising from changes of the density distribution in the source region as an additional acceleration. The relative magnitude of these contributions depends on frequency and on the characteristics of the phenomenon which is observed.…”

“…However, he took only the inertial effect into account in his theoretical derivations. Zürn and Wielandt (2007) considered gravitational, free air and inertial effects together and showed in a "Gedankenexperiment" and with very simple models of the atmospheric influence on vertical acceleration sensors that the air pressure admittance as a function of frequency should be zero near a few mHz. This results in a notch of the magnitude of the transfer function between air pressure and sensor acceleration and a sign-reversal of the A c c e p t e d M a n u s c r i p t 3 admittance associated with it.…”

Earth's free oscillations excited by the 2011 TohokuMw9.0 earthquake detected with a groundwater level array in mainland China

“…For 27 wells with barometric records, the local barometric effects were corrected (Zürn & Widmer 1995;Zürn & Wielandt 2007;). The barometric admittance ranges from 0.9 to 16.5 mm hPa −1 , with a mean of 7.3 mm hPa −1 and a median of 6.7 mm hPa −1 .…”