The determination of near ocean-bottom properties is of significant importance for marine geophysical exploration. P-wave velocity, S-wave velocity and density determine the partitioning of the incident wave field into transmitted P- and S-waves at the seafloor. Furthermore, these parameters are essential inputs for decomposing multi-component seafloor data, as well as for the numerical study of wave propagation phenomena.
Several methods have been developed in recent years to estimate these near seafloor properties including laboratory measurements on cores, in-situ measurements with probes, and inference from seismic data. The most common seismic techniques are the modeling or the inversion of data recorded by streamer hydrophones in the water layer Berge et al. (1986); Bronston and Graul (1994); Chapman et al. (1986); Chapman and Rohr (1991); Schmidt and Jensen (1986). Amplitude versus offset (AVO) data contain information about the lithology contrast across an interface and thus provide the means for determining the near seafloor parameters. Recent new developments in acquisition technology make it possible to apply similar techniques to data recorded directly at the ocean bottom by ocean-bottom seismometers (OBS) as well as ocean-bottom hydrophones (OBH). Methods have been proposed to estimate the P- and S-wave velocity and density either from the acoustic/elastic reflection coefficient from point source measurements of pressure and vertical particle velocity component Amundsen and Reitan (1995), or to estimate P-wave and S-wave velocity from the vertical and radial particle velocity component measured at the seafloor Amundsen and Reitan (1994). In order to apply these techniques on real data, we must understand how factors such as layering or poroelasticity might influence the seafloor parameter inversion, and how robust the methods are under various conditions.
In this paper, we compare the two techniques of seafloor parameter estimation based on (1) AVO from ocean-bottom pressure and vertical particle velocity data and (2) AVO from ocean-bottom vertical and radial particle velocity data. We perform a sensitivity analysis of both methods for the parameters for which we are inverting. Furthermore, we construct simple synthetic seismograms and explore the effect of layering and poroelasticity on the parameter inversion. Subsequently, we draw conclusions about the robustness and preciseness of the two methods under different starting conditions.