Using Remotely Operated Vehicles and Active Acoustics to Study the Distribution of Mesopelagic Fauna

Herbert Leavitt1, Adrienne Copeland2, Amanda Nicole Netburn, PhD2 and Mike Ford3, (1)Eckerd College, St. Petersburg, United States, (2)NOAA Office of Ocean Exploration and Research, Silver Spring, United States, (3)NOAA Fisheries Marine Ecosystems Division, Washington, DC, United States
The mesopelagic zone remains one of the least explored and understood ecosystems on the planet. The mesopelagic, or midwater, is known to support populations of commercially important fish, regulate carbon sequestration, play a role in ocean climate, and support a high amount of undiscovered biodiversity. In order to fill in knowledge gaps of the midwater environment, the NOAA Office of Ocean Exploration and Research routinely integrates midwater exploration into its exploration model. This study focused on two expeditions where midwater exploration occurred, the 2017 Musicians Seamount expedition (EX1703) and the Windows to the Deep 2019 expedition (EX1903). During midwater dives, a remotely operated vehicle (ROV) was piloted to predetermined depths to perform exploratory transects. Using high-definition video and a suction sampler, midwater fauna were identified and occasionally collected for further study. Acoustic backscatter data at 18 kHz were collected during midwater exploration to determine the location of the deep scattering layer (DSL), a dense layer of midwater fauna between 300 - 1000 meters found in all parts of the ocean. The majority of midwater backscatter from the DSL is attributed to fish, but gelatinous groups such as gas-filled physonect siphonophores may play a role. Acoustic data and ROV video footage were collated to illuminate patterns in the vertical distribution and composition of midwater fauna relative to the DSL. All identified organisms were classified based on broad functional groups. Of these, all but one group, the physonect siphonophores, were statistically shown to be sighted significantly more frequently within the DSL than outside of it. The question of ROV avoidance was considered as a potential source of bias in this dataset. Acoustic backscatter was measured at ROV transect depths before and after the transect occurred. Several transects showed significant decreases in backscatter at depth after the ROV had left, suggesting that midwater fauna may be vacating areas where the ROV was present. These data provide valuable insight into the structure of the midwater ecosystem with higher resolution than is achievable with other methods. Understanding of avoidance behavior and distribution of fauna within the midwater are crucial steps in the exploration of this system.