Simultaneous Localization and Mapping of Seafloor Cold Seeps with Autonomous Underwater Gliders

Conventional mapping of seafloor seeps relies on surface vessel survey operating acoustic arrays such as seismic, sub-bottom, and multibeam sonar systems. Over the past two decades large displacement (>300kg) autonomous underwater vehicles (AUVs), have demonstrated the ability to detect seafloor hydrothermal vents and cold seeps while operating lower power acoustic systems at reduced standoff distances (<100m) from the seafloor. Although these AUV-based methods generate less environmental disturbance, such as acoustic source noise than surface vessel systems, they are typically constrained to durations of tens of hours and require operation in close coordination with an attendant surface vessel, thus limiting coverage extent. We present an experimental method for unattended seep detection using autonomous underwater gliders (AUGs) equipped with automated mission replanners, adaptive controllers, and low power chemical and acoustic sensors. This in-situ adaptive survey method enables autonomous bottom following at altitudes within 50 m of the seafloor and uses automated analyses to infer environmental state from navigation and payload sensor data in order to simultaneously localize and map seafloor cold seeps. Because this survey architecture relies on low power AUG systems and does not require acoustic baseline transponder networks, it enables coverage in excess of 1,000 linear km. Results from field operations at multiple well-studied sites in the Eastern Pacific using AUGs equipped with low power sonar and in-situ mass spectrometry payloads are able to detect, classify, and map water column hydrocarbon seeps and characterize seafloor structure using this unattended adaptive survey process. We present results of field trials and compare with ground truth from conventional systems.