Utilizing Advanced Technology to Characterize an Unknown Pelagic Ecosystem

The implementation of the ecosystem approach has increased the use of acoustics during surveys of pelagic organisms. How can data from acoustic sensors be exploited to describe the structure and distributions of aquatic organisms in a new area? A Census of Marine Life survey on the mid-Atlantic ridge during summer 2004 was used as an opportunity to develop a practical approach. The Norwegian vessel, G. O. Sars, is an acoustically-quieted platform equipped with a five-frequency (18 kHz, 38 kHz, 70 kHz, 120 kHz, 200 kHz) echosounder, acoustic Doppler current profilers (ADCP), multibeam sonars, an acoustic deep-towbody, and two remotely operated vehicles. The paucity of information on species composition and acoustic characteristics of aquatic organisms limited the ability to conduct a traditional acoustic biomass survey. Our analytic strategy quantified acoustic structure and dynamics independent of biological sampling and then integrated acoustic characterizations using density and target strength observations with biological community composition and length frequency data. This approach differs from traditional surveys in that species were not assigned to backscatter thresholds or water column regions at the onset of analysis. Near-real-time analytic products were used to monitor biomass distributions throughout the water column: daily echograms, target strength data from the towbody, and frequency-differenced echograms. We observed persistent biological layers at different depths in the water column. Frequency differencing was used to separate biological components within layers. Layers or components of layers migrated toward the surface during dark hours. Modes of target strength frequency distributions and trawl catch compositions differed among layers and through the water column. Conversion of acoustic backscatter to numeric or biomass estimates will be based on functional groups within layers. The use of acoustic and net technologies can be integrated with near-real-time analytic results to quantitatively characterize pelagic ecosystems.