Do marine substrates 'look' and 'sound' the same? Supervised classification of multibeam acoustic data using autonomous underwater vehicle images

► Multibeam acoustic variables can act as surrogates for marine biodiversity prediction modelling. ► Seafloor acoustic imagery can be accurately quantitatively classified with visual classes from AUV. ► Multibeam backscatter and bathymetry can be simultaneously utilised with image segmentation and modelling techniques. ► Random forest models outperform other models to predict the substratum and rugosity classes of acoustic data.