Boulder Detection in the Shallow Sub-Seafloor by Diffraction Imaging With Beamforming on Ultra-High Resolution Seismic Data

Small-scale heterogeneities within the seafloor such as glacial boulders, concretions, or unexploded ordnance are of major scientific and economic interest. Because of their small size, such objects are hardly imageable by conventional seismic methods since they produce only faint diffractions. Despite the growing interest, reliable and efficient object detection methods are still not established. So, a marine acquisition system to image objects in the size range 0.3–4.1 m in the near-surface has been designed and workflows have been developed. Source signals with a central frequency of ∼1,000 Hz are necessary to generate diffractions for the object size range. Performing beam pattern analyses, a rigid tow frame with the dimensions 8 × 3 m and attached hydrophones was designed. A synthetic aperture approach is realized to improve the resolution. Reflection events are suppressed in Normal-Move-Out corrected shot gathers by the muting of high singular values to enhance the diffractions. The diffractions are imaged with a beamforming algorithm with a high efficiency, as a total swath angle of about 80° in across-track direction is covered. The processing of synthetic data sets allows an optimization and validation of the workflow and sensitivity analyses. Ground truthing is achieved with 1–2 m large boulders on the seafloor in 22 m depth, which were identified in a Multi-Beam Echo Sounder data set. Although diffractions are only weak events in seismic data sets, 3D object detection with seismic beamforming has been found to be an efficient and reliable method to perform derisking for offshore infrastructure installations, drillings, or geologic interpretation.