HYDRO 2016 Paper K2
John Huges Clarke
Multibeam backscatter has long been recognised to be a valuable complementary data set that can be derived from multibeam bathymetric surveying. Its use, however, has never reached the same widespread acceptance as the bathymetry. Part of this issue is due to an imperfect understanding about what the measurement actually represents, and part is a failing in the fidelity of that measurement.
It is now quite routine to use backscatter to delineate the broad regional distribution of sediments of strongly contrasting physical character (e.g.: rock vs mud). To be more valuable however there needs to be the ability to confidently (and reproducibly) discern sediments of less contrasting character (such and muddy-sands vs sandy-muds). Such finer discrimination places much higher demands on calibration, particular the removal of radiometric and geometric effects.
Even if this were achievable, there remain inherent ambiguities in inverting seabed backscatter. This is because, over a narrow range of grazing angles, quite different seabed physical properties can result in similar mean backscatter strength. Two promising approaches are using the full angular response (a much wider range of grazing angles) and multiple scattering frequencies (multi-spectral).
Both of these approaches, however, place even higher demands on those same radiometric and geometric effects. For angular response, the geometric issues dealing with data close to normal incidence and at the lowest grazing angles are the main problem. For multi-spectral, particularly for the case of multi-sector systems, the radiometric problems are compounded with each added frequency.
In the absence of absolute calibration, empirical approaches to addressing these issues have been developed in the field using surface mounted systems. All of these issues would be easier to resolve if the imaging platform were able to significantly change its altitude and orientation over a given seabed. For that AUVs are uniquely suited. Examples of these issues and potential new methods to better address them will be presented.
|