Connecting Land Cover to Benthic Cover Within the Watershed: Quantifying Impacts of Terrigenous Sediment on Coral Reef Health and Morphology Using Remote Sensing and Machine Learning

Virtual

Erosion, with associated sedimentation of reefs and loss of habitat for marine and coastal species, is one of the most pressing environmental challenges in Hawaii. Identifying hotspots of soil erosion guides management on land, but limited work has been done connecting land cover to coral reef health and morphology. Current reef survey methods overgeneralize the spatial variation of benthic cover and this has been reflected in incorrect Geographic Information System (GIS) layers of the reef. We propose a novel and more holistic approach to improve accuracy by connecting sedimentation from source to sink using Unmanned Aerial Vehicles (UAVs), Unmanned Surface Vehicles (USVs), local knowledge, and machine learning. UAVs can capture images to create 3-Dorthomosaicsof the terrestrial system while USVs can be used to create 3-D coral reef models. As a result, we will be able to identify the effects of different land management practices on coral reef health.