Acoustic Assessment of Polymetallic Nodule Abundance Using Sidescan Sonar and Altimeter

Polymetallic nodules (PMN) are potato-sized concretions containing metals, such as manganese, copper, nickel, cobalt, and rare earth elements, and are a potential valuable resource of minerals. They occur in high abundance and are unevenly distributed across the Clarion Clipperton Fracture Zone. Existing PMN abundance estimation methods using box corers, and manual assessment through seabed photographs are labor and time intensive, and can only survey small sections of seabed at a time. Compared to an underwater camera, acoustic sensors are able to survey the PMN abundance across larger tracts of seabed at a time. In this article, we present a method for PMN abundance assessment using heterogeneous acoustic data, which is a combination of bathymetry information and sidescan sonar measurements of seabed backscatter. We achieve this using an artificial neural network model that classifies a given region into a low or high PMN density region using these features. Our model will enable faster estimation of PMN abundance for future deep seabed site surveys without the need for underwater cameras. To date, our proposed method yields an average accuracy of 85.36% on a testing data set, demonstrating our method's effectiveness in estimating PMN abundance.