A GIS approach to quantitative ice gouge depth mapping, analysis, and prediction

We describe and illustrate the application of a geographical information system (GIS) approach to map ice gouge locations and depths from high-resolution multibeam echo sounder (MBES) bathymetric surfaces by calculating residuals relative to spatially variable moving trend surfaces. The workflow can be used to rapidly characterize gouges over large areas and, because minimal human intervention is required, is especially attractive in heavily gouged areas where traditional manual measurement techniques would be tedious and produce highly uncertain results. The method produces maps showing gouge depth as a continuous field rather than point measurements or cross-gouge profiles, so that variations in depth along gouges can be easily visualized and analyzed. Once gouges have been delineated, gouge depth distribution statistics can be further used to estimate exceedance probabilities for gouge depths within local neighborhoods. Seafloor roughness maps can also be generated to highlight the spatial variability of seafloor disturbance and, in a relative sense, visualize the ages of different gouges if certain assumptions are satisfied. We illustrate application of the method using a sample MBES data set depicting a heavily gouged portion of seafloor.