Application of AUVs in the Exploration for and Characterization of Arc Volcano Seafloor Hydrothermal Systems

Friday, 19 December 2014
Cornel E J de Ronde1, Sharon L Walker2, Fabio Caratori Tontini1, Edward T Baker3, Robert W Embley4 and Dana Yoerger5, (1)GNS Science, Lower Hutt, New Zealand, (2)NOAA/PMEL, Seattle, WA, United States, (3)Joint Institute for the Study of the Atmosphere and Ocean, Seattle, WA, United States, (4)NOAA Newport, Newport, OR, United States, (5)WHOI, Woods Hole, MA, United States
The application of Autonomous Underwater Vehicles (AUVs) in the search for, and characterization of, seafloor hydrothermal systems associated with arc volcanoes has provided important information at a scale relevant to the study of these systems. That is, 1-2 m resolution bathymetric mapping of the seafloor, when combined with high-resolution magnetic and water column measurements, enables the discharge of hydrothermal vent fluids to be coupled with geological and structural features, and inferred upflow zones. Optimum altitude for the AUVs is ~70 m ensuring high resolution coverage of the area, maximum exposure to hydrothermal venting, and efficency of survey. The Brothers caldera and Clark cone volcanoes of the Kermadec arc have been surveyed by ABE and Sentry. At Brothers, bathymetric mapping shows complex features on the caldera walls including embayment’s, ridges extending orthogonal to the walls and the location of a dominant ring fault. Water column measurements made by light scattering, temperature, ORP and pH sensors confirmed the location of the known vent fields on the NW caldera wall and atop the two cones, and discovered a new field on the West caldera wall. Evidence for diffuse discharge was also seen on the rim of the NW caldera wall; conversely, there was little evidence for discharge over an inferred ancient vent site on the SE caldera wall. Magnetic measurements show a strong correlation between the boundaries of vent fields determined by water column measurements and observed from manned submersible and towed camera surveys, and donut-shaped zones of magnetic ‘lows’ that are focused along ring faults. A magnetic low was also observed to cover the SE caldera site. Similar surveys over the NW edifice of Clark volcano also show a strong correlation between active hydrothermal venting and magnetic lows. Here, the survey revealed a pattern resembling Swiss cheese of magnetic lows, indicating more widespread permeability. Moreover, the magnetic survey showed evidence for a highly magnetized ring structure ~350 m below the volcano summit considered to represent a buried (by continued growth of the cone) caldera rim. Zones of magnetic lows located inside the inferred caldera that are not associated with present-day venting are consistent with an earlier stage of hydrothermal activity.