Neotectonics, Geodesy and Seismic Hazard in the Northern Walker Lane

The establishment of fleets of large numbers of ocean bottom seismometers (OBS) capable of recording for more than a year has made it possible to study Earth structure beneath the oceans using seismic observations in much greater resolution than previously possible. However, shallow water poses strong challenges for OBS deployments, with much higher noise levels from waves and currents. The on-going Cascadia Initiative, a major OBS community project directed at studying the Cascadia subduction zone with its megathrust earthquake potential and broad continental shelf, includes OBS deployments at depths as shallow as 50 m. Before the Cascadia deployment, there was a valid fear that the data from the shallowest sites would be useless for seismic observations. The shallow Cascadia OBS deployments feature shielding to protect the sensors from the flow of ocean floor currents (and trawling), The first year data show that the shielding can reduce horizontal component noise levels (due to currents) by more than 20dB permitting good SNR for horizontal phases even at shelf depths. Noise from deformation under ocean wave loading has been found to be very large at the shallowest sites. High amplitude ocean waves cause the differential pressure gauges (DPG) and unshielded seismic sensors at a few sites to sometimes clip, but shielded seismic sensors and pressure measurements from absolute pressure gauges (APGs) remain unclipped even at the shallowest sites. The study demonstrates pressure gauge records can be used to predict and remove the noise from the deformation under ocean wave loading in the spectral domain, potentially improving signal to noise for long period seismic phases by up to 40dB. A FIR digital filter can be created from the pressure to acceleration transfer function that when convolved with the pressure record accurately predicts the wave loading signal allowing removal of about 30dB of this noise. These results show that with proper shielding, good quality seismometer observations can be obtained from shallow water, opening up this critical part of the seafloor to observations for the first time. The data also show that without the use of pressure gauges to remove the wave loading noise, and without shielding, the shallow water, long period (>10s) data are almost useless for seismology.