Characteristics of Tide observed by broadband-seismometers and a tiltmeter in the seafloor and seafloor borehole

Low noise long-period seismic monitoring below 0.1 Hz in the seafloor became more practical by surficial burial of seismometer and use of deep seafloor borehole. But it is still not well understand how the seafloor and sub-sea ground is responding in tidal frequencies because we had difficulties obtaining seismic data without effect of seafloor water flow in the frequencies. In this study, we evaluated records from seafloor broadband seismometers, a seafloor borehole broadband seismometer, and a seafloor borehole tiltmeter in tidal frequencies, which are a part of cabled observation network called 'Dense Ocean-Floor Network for Earthquake and Tsunamis' (DONET), in the Nankai Trough, south of Japan. Each seafloor DONET station consists of a Guralp CMG3T broadband seismometer, an accelerometer, and pressure gauges. The seismometers are buried in the seafloor for 18 of 20 operating. Other two stations installed the seismometer on the seafloor. The borehole station installed a CMG3T broadband seismometer and an AGI LILY tiltmeter in the IODP C0002G borehole at 905 - 907 m below the seafloor. We analyzed tidal response of the seafloor broadband seismometers, the borehole broadband seismometer, and the borehole tiltmeter in May 25-29, 2013. The broadband seismometers have a velocity-flat response between 1/360 Hz and 50Hz and slopes in the tidal frequencies; therefore we corrected the instrument response over the tidal frequencies in the frequency domain to obtain tidal response in vertical and horizontal acceleration. The borehole tiltmeter was compared against the acceleration data from the borehole broadband seismometer and these were consistent as the gravitational acceleration by tilting of the instruments. Clear tidal responses were obtained from the borehole instruments and 15 (of 20 analyzed) seafloor broadband seismometers. Vertical tidal response from the borehole and seafloor broadband seismometer fitted well to the calculated tidal gravity change by GOTIC2 (Matsumoto et al., 2001) for both amplitude and phase. We consider this suggest the broadband seismometers are capable of observing tidal gravitational change in the installed seafloor environment. In contrast, tidal horizontal acceleration didn't fit well to the model calculated tidal ground tilt. The amplitudes of observed tidal responses were larger than the model. In the seafloor, the amplitudes of M2 tide were in the order of 1 micro-radian, the amplitude was even larger in the borehole. The phase of observed tide did not fit with the model, and are not consistent between stations. It was also observed that the particle motion of tidal horizontal acceleration from some of the stations showed highly linear trends. The direction of linear had tendency to be parallel to the slope of the seafloor. We consider the linear trends of tidal response are caused by deformation of the sedimentary layer due to pressure loading from ocean tide, and the deformation may be larger than the deformation of the crust from earth tide. This observation would imply that the effect of ocean tidal loading is prominent not only at coastal land observation but also in the seafloor where the surface structure is soft or heterogeneous.