Abstract: Mapping marine benthic habitats in coastal Newfoundland - the issues of spatial scales and acoustic footprint

Seabed habitat mapping involves three primary components: bathymetry, geology, and biology. Combined these attributes describe the geomorphology of the seabed and the biological communities that reside there. The issue of natural variability within a hierarchy of nested spatial scales and our ability to resolve these scales using acoustic techniques remains a high research priority. The issue of the acoustic footprint is inextricably linked to our ability to detect and map seabed habitats, where most surveys are done from surface ships. Greater depths result in larger footprints and a reduced ability to discriminate small-scale features while introducing a disparity in observation sizes that complicates data processing and confuses interpretation. A practical solution is to reduce and standardize the sampling range using autonomous underwater vehicles (AUV). An acoustic seabed classification (ASC) survey was carried out in the rugged coastal terrain of Newfoundland using the ocean class MUN Explorer AUV and compared to a previous survey done from the surface. The study area ranged in depths from 19 m to > 200 m depth with areas of high bathymetric relief of 0.45 m m-1 (24° slope) associated with cobble substrates and low bathymetric relief of 0.14 m m-1 (8°) associated with gravel. Five survey lines 5 km in length and 50 m apart were sampled at a fixed range of 35 m from the seabed. Six acoustic classes of seabed were detected from the AUV survey. Surveying at a fixed range reduced the variability in the size of the acoustic footprint on the seabed by a factor of 25 times compared to sampling from the surface. There was a high degree of spatial heterogeneity in the pattern of acoustic classes compared to the surface based survey. The predominant spatial scale of spatially contiguous classes was = 10 m (99.6% of all observations). This is equivalent to the observational scale, indicating that the true spatial scale of acoustic seabed classes was smaller than could be measured from the surface-based survey.