High Resolution Seafloor Habitat Mapping – WIO

Predefined classification schemes define a number of seafloor mapping work. These classifications are often considered oversimplified. As an improvement, substrate and biological data combined with acoustics data, will be modeled to map percentage cover robustly, to produce high-resolution maps of the seafloor. Machine learning model will be used.

The seafloor habitats broadly consist of substrates and various formations of biological cover. A review showed a lack of habitat maps in the region, implying habitat information is not integrated with fisheries management, ignoring the aspect of the Ecosystem Approach.

This project will assess vulnerabilities; discriminate different seafloor habitats at optimal mapping scale; assess the impact of human activities; relate findings to management scenarios; contribute to conservation; correlate socio-economics to a habitat; and share digital maps. The project targets demersal fishing grounds in Mozambique, Kenya and Tanzania.