3D seismic-derived bathymetry: a quantitative comparison with multibeam data

This study compares bathymetry extracted from 3D seismic data at two Australian study sites of differing morphological complexities to two sources of collocated multibeam data: 50-m and 5-m multibeam digital bathymetric models (DBMs). Seafloor horizons are extracted from the 3D seismic data and converted to depth using sound velocity profiles collected during seismic acquisition. The resulting seismic-derived DBMs are independent of the multibeam DBMs and are shown to be highly comparable. For the morphologically simple site, the seismic-derived DBM was within ± 2% of the multibeam DBMs and, at 2σ, 95% of differences are in the range − 1.22 to 0.10% (− 1.02 to 0.48%) for the 50-m (5-m) multibeam DBM. For the morphologically complex site, > 80% (> 99%) of seismic-derived depths were within ± 2% (± 5%) of multibeam DBMs. At 2σ, 94% of differences are in the range − 3.48 to 1.69% (− 2.73% to 2.44%) for the 50-m (5-m) multibeam DBM. Increasing morphological complexity and slope angle were the most important factors affecting DBM comparisons, with seismic-derived depths typically underestimated in canyon thalwegs. Despite these differences, the higher data density, multichannel stacking and migration of the 3D seismic data resulted in seismic-derived DBMs with high resolution and improved feature relief and clarity when compared to multibeam DBMs for the conditions in this study (depths of 120–1900 m), particularly for morphological features such as individual rills and gullies. This method has the potential to expand the spatial coverage of high-resolution DBMs, for example, in Australia, by over 150,000 km 2 .