Challenges and Possible Solutions with Multibeam Calibration Surveys in Very Shallow Waters

Multibeam calibration surveys are an integral part of hydrographic surveys and must be performed to the highest possible accuracy to avoid systematic errors in the hydrographic data. The calibration surveys are performed to calculate position latency and roll, pitch, yaw angular misalignment between the motion sensor and the sonar. In very shallow waters, the computation of calibration values is challenging, as the accumulated errors in seafloor depth and position, from all contributing factors, including horizontal and vertical controls, approach resolvability of the calibration method. The horizontal control of any multibeam survey is derived from GNSS aided inertial navigation system. For the vertical control, the ellipsoidally referenced survey (ERS) relies on the height derived from the GNSS-aided inertial navigation system, while the tidally referenced surveys (TRS) rely on the observed or predicted tides. The positioning and motion data are commonly post-processed to achieve the highest possible accuracy. Any errors in the positioning of the survey vessel affect the multibeam calibration values and introduce systematic errors in the entire data set. This paper accesses the relative accuracies of multibeam calibration values when using ERS and TRS. The paper also looks at the effect of horizontal and vertical positioning errors from different GNSS processing strategies on the multibeam calibration values through field experiments. This experiment and followed investigations aim to identify the achievable accuracies in the GNSS and motion data post-processing domain as reflected in ERS and its impact on multibeam calibration. An experiment was set up to obtain calibration values for three different multibeam calibration data sets for twelve unique cases of multibeam data processing strategies. Detailed analysis and discussion on the sensitivity of each multibeam calibration parameter to horizontal and vertical positioning error are presented. Finally, a set of suggestions are provided to obtain the best estimate of multibeam calibration values in very shallow waters.