Abstract

Any seismic survey consists of large number of profiles. Often different profiles are shot on different dates. In case of 3D marine seismic survey it is often that even adjacent profiles are shot on different dates, which can be days, weeks or even months. The two adjacent sail-lines can receive reflected rays from the same subsurface reflection point; this implies that the same reflection point will be recorded on different dates. The propagation velocity of seismic waves in sea water is a function of physical properties of sea water, especially temperature and salinity. These can vary seasonally, so can the seismic velocity in water. When two different sail-lines receive reflected rays from the same subsurface reflection point and the water velocity changes during the time gap between shooting these lines, reflection times will be affected by water velocity variations. But during seismic data processing all the traces corresponding to the same reflection point are processed with same water velocity, which is not valid in this case. This means the effects of water velocity variations are not taken into effect. If the water bottom reflection times of one line data are inverted to find velocity during the acquisition of that line and this velocity is used to generate final seismic image then the effects of water velocity variations could be removed. So in this study estimation of water velocity for each sail-line considering water-bottom reflections as an input to the Gauss-Newton tomography method is discussed.