Refraction Artifact Correction

Hughes Clarke and Kammerer

Chair Funding and NSERC

An operational reality of swath sonar survey is that there will be imperfect knowledge of the water mass. How these imperfections manifest themselves into the data depends strongly on the sonar hardware geometry. We have modeled the effect of surface sound velocity errors on the common sonar configurations to show how one can recognize these errors (which are dynamic, varying with the array instantaneous orientation).

Even with sufficient surface sound velocity information, changes in the deeper structure of the water mass will degrade the quality of the data. Such changes usually manifest themselves over longer time periods so that one can model the error as a slowly varying signature.

Software algorithms have been developed to first empirically and then objectively recognize the characteristic shape of these artifacts and, through a minimization technique estimate an optimal slowly varying spatial corrector to remove or at least reduce the magnitude of the error.