Identifying and Removing Noise from the Ocean Bottom Broadband Seismic Data

David Dolenc, Barbara Romanowicz, and Bob Uhrhammer


Ocean bottom broadband seismic observations show increased noise level when compared to land recordings. The signal-generated noise following the arrival of the seismic phases is one type of the observed noise, which is due to reverberations in the soft sediments and the water layer. Increased background noise, on the other hand, is generated, among others, by infragravity waves and ocean currents. Both types of noise can at least partially be removed from the seismic signal.

The data recorded at MOBB (Monterey Ocean Bottom Broadband station) were used. MOBB was installed 40 km offshore in the Monterey Bay at a water depth of 1000 m in April 2002 in a collaboration between Berkeley Seismo Lab and Monterey Bay Aquarium Research Institute (MBARI) (McGill et al., 2002; Uhrhammer et al., 2002). It comprises a three-component broadband seismometer with a temperature sensor, a water current meter measuring current speed and direction, and a differential pressure gauge (DPG). The station is continuously recording data which are retrieved every three months using the MBARI ROV "Ventana".

Signal-Generated Noise

An example of a large deep teleseism recorded at MOBB (black) and at the nearby island station FARB (gray) is shown in Figure 23.1. A strong signal-generated noise is observed at MOBB, probably due to reverberations in the mud layer in which MOBB seismometer is installed. This type of noise may be unavoidable in shallow buried installations. The transfer function describing the signal-generated noise in the sediments layer was calculated by the spectral division of the first 50 seconds of the MOBB and FARB records. The transfer function was then removed from the longer MOBB record to eliminate the signal-generated noise (dashed).

Our future work will include analysis of other teleseismic events to obtain a robust transfer function, as well as 1D modeling of the soft sediments and water layer.

Figure 23.1: Vertical component recordings of a teleseism (Kurile Islands, 11/17/02, $M_{w}$=7.3, depth=459 km) at MOBB (black) and at the nearby island station FARB (gray). The two records were bandpass filtered between 0.03 and 0.3 Hz. Also shown is the MOBB record with the signal-generated noise removed (dashed). The arrivals of the P-, pP-, and sP-waves are indicated.
\epsfig{file=dolenc03_1_1.eps, width=11cm}\end{center}\end{figure*}

Figure 23.2: Background noise PSD observed at MOBB, FARB, SAO, and YBH vertical components. Shown are 4 hours data on a stormy (gray; 12/16/02), and on a quiet day (black; 05/23/02). The USGS high- and low-noise models for land stations are also shown (dotted).
\epsfig{file=dolenc03_1_2.eps, width=12cm}\end{center}\end{figure*}

Background Noise

A comparison of a power spectral density (PSD) of the background noise recorded on the vertical component at MOBB and at three stations of the Berkeley Digital Seismic Network is shown in Figure 2. Data for a quiet and for a stormy day (determined from the wave height recordings from a nearby weather buoy) are shown. The background noise PSD between 30 and 500 sec is significantly different at MOBB. The observed peak is probably related to ocean currents and infragravity waves (Webb et al., 1991; Webb, 1998). It is interesting that a similar peak can be observed at the island station FARB on the stormy day.

In our future work we plan to find the coherence between background noise and the ocean current, DPG data, temperature, and the ocean tides, and design filters to remove the coherent part of the noise. Previous studies successfully removed noise due to infragravity waves and ocean currents (Webb and Crawford, 1999; Crawford and Webb, 2000; Stutzmann et al., 2001).

The results from this work may suggest future installation improvements for the deployment of permanent or temporary off-shore seismic broadband stations, such as have been discussed in the framework of the Ocean Mantle Dynamics (OMD) workshop (Snowbird, Fall 2002).


Crawford, W.C., and S.C. Webb, Identifying and removing tilt noise from low-frequency (\(<\)0.1 Hz) seafloor vertical seismic data, Bull. Seism. Soc. Am., 90, 952-963, 2000.

McGill, P., D. Neuhauser, D. Stakes, B. Romanowicz, T. Ramirez, and R. Uhrhammer, Deployment of a long-term broadband seafloor observatory in Monterey Bay, EOS Trans. Amer. Geophys. Un., 83, F1008, 2002.

Stutzmann, E., J.-P. Montagner, A. Sebai, W.C. Crawford, J.-L. Thirot, P. Tarits, D. Stakes, B. Romanowicz, J.-F. Karczewski, J.-C. Koenig, J. Savary, D. Neuhauser, and S. Etchemendy, MOISE: A prototype multiparameter ocean-bottom station, Bull. Seism. Soc. Am., 91, 885-892, 2001.

Uhrhammer, R., B. Romanowicz, D. Neuhauser, D. Stakes, P. McGill, and T. Ramirez, Instrument testing and first results from the MOBB Observatory, EOS Trans. Amer. Geophys. Un., 83, F1008, 2002.

Webb, S.C., X. Zhang, and W. Crawford, Infragravity waves in deep ocean, J. Geophys. Res., 96, 2723-2736, 1991.

Webb, S.C., Broadband seismology and noise under the ocean, Rev. Geophys., 36, 105-142, 1998.

Webb, S.C., and W.C. Crawford, Long-period seafloor seismology and deformation under the ocean waves, Bull. Seism. Soc. Am., 89, 1535-1542, 1999.

Berkeley Seismological Laboratory
215 McCone Hall, UC Berkeley, Berkeley, CA 94720-4760
Questions or comments? Send e-mail:
© 2004, The Regents of the University of California