Increasingly, national hydrographic agencies are committing to routine acquisition of seabed backscatter strength estimates from multibeam echosounders (MBES) as part of national programs for seabed characterization. As part of their bathymetric survey mandate, these agencies have a long history of sounding quality control utilizing absolute and relative calibration (reference surfaces and crossover comparisons). Equivalent quality control is, however, not yet in place for managing seabed backscatter strength measurements, as the majority of the collected data is not absolutely referenced.

Herein, a new technique for cross-calibrating a MBES with a reference calibrated split beam echosounder (SBES) was implemented. Broadband reference bottom backscatter strength (45-450 kHz) from areas with different seafloor types, derived from data obtained with Simrad EK80 SBES, is used to adjust the received acoustic intensities acquired from the same areas with several multi-sector MBES (Kongsberg Maritime EM2040P, EM710 and EM712), thereby enabling the routine collection of absolutely referenced bottom backscatter strength measurements. Previous efforts to implement a similar cross-calibration only considered a simplified vertically referenced ensonification geometry, ignoring the dynamic variations due to vessel rotations. As a result, neither the rotation of the beam pattern with respect to the vertical reference nor the compensation due to active beam stabilization were accounted for. Furthermore, this method properly accounts for modern MBES which have multiple transmit sectors over multiple swaths with the associated changes in frequency and signal modulation.

The main output of this research is a set of two-dimensional arrays of correctors, derived for each transmit sector - the correction heatmap - providing estimates of the necessary calibration, as a function of across and along track sonar referenced angles. To test the repeatability of the proposed technique, correction heatmaps derived for the same system (using the same settings), but with data from different reference areas, were compared, resulting in differences generally within ± ~2 dB.

Finally, a pre-calibrated MBES was used to survey a different location and establish a reference area, enabling the subsequent calibration of sonars that use the same frequencies.