Multisensor Microbathymetric Habitat Mapping with a Deep-Towed Ocean Floor Observation and Bathymetry System (OFOBS)

Abstract

To describe the seafloor topography, a number of different bathymetric methods can be applied. These methods vary greatly in coverage, resolution, and topographic uncertainty. Satellite-based gravimetry and radar altimetry can give large-scale structural estimates of the seafloor topography, yet, with a very low resolution and without real depth measurements. Ship-based swath bathymetry systems greatly improve the topographic uncertainty and increase the knowledge on geomorphology and depth of the seafloor. In shallow waters, ship mounted echosounders can produce high-resolution data on a submeter level. However, in deep-sea environments, the resolution deteriorates due to large acoustic footprints and a reduced number of measurement points with respect to the mapped area. In order to conduct high-resolution habitat mapping and to resolve small-scale topographic seafloor features, subsea survey vehicles need to be employed. Next to remotely operated or autonomous underwater vehicles, towed camera systems present a comparatively cheap method, both financially and with regards to support requirements, to collect close-range optical seafloor data. Nonetheless, optical sensors have very limited coverage capabilities in the deep sea, due to the nature of the sensors and the high attenuation of light in the water column. Acoustic sensors on the other hand can achieve much wider survey swaths, depending on their operation frequency. The Ocean Floor Observation and Bathymetry System (OFOBS), developed at the Alfred Wegener Institute for Polar and Marine Research, Germany, offers a novel survey technology for deep-towed multisensor microbathymetric habitat mapping. To augment the traditional optical sensors, the OFOBS was equipped with additional acoustic and navigational sensors. A bathymetric side scan sonar collects lateral seafloor reflection intensity and bathymetry at ranges up to 100 m to both sides of the vehicle. A forward looking sonar records acoustic imagery ahead of the system, which can be used for hazardous obstacle avoidance in rough terrain. This thesis introduces the newly developed system along with processing workflows for the acquired datasets. Underwater photogrammetric methods are utilized for the optical data, to reconstruct the three dimensional morphology of the seabed. The camera pose estimations of the employed bundle adjustment algorithms are used for local navigation corrections of the acoustic datasets, to achieve best possible data alignment. The resulting multilayer product consists of wide-swath acoustic bathymetry (submeter resolution), multi-frequency side scan mosaics (subdecimeter resolution), photogrammetric microbathymetry (subcentimeter resolution), and geometrically corrected, georeferenced photo mosaics (submillimeter resolution). These results offer a wide variety of use cases in high-resolution habitat analyses by the associated scientific working groups. The data used for developing the presented workflow was collected during the RV Polarstern expedition PS101 in the extreme environment of the volcanic seamounts along the Langseth Ridge in the high Arctic (87°N, 60°E).