Development of stereo camera methodologies to improve pelagic fish biomass estimates and inform ecosystem management in marine waters

To understand and manage marine ecosystems, long-term monitoring of fish biomass is needed. A challenge in estimating fish biomass using sampling nets is varying catchability with habitat, weather, or vessel traffic conditions. Underwater stereo cameras have shown promise in providing a non-lethal, efficient, and cost-effective method to observe and measure fish in areas that cannot be sampled otherwise. These methods, however, have yet to be demonstrated for mid-water pelagic or semi-pelagic fishes. We designed, built, and tested a stereo camera for its potential to augment survey assessments of pelagic fish biomass in areas where trawl net samples cannot be collected. In a pilot test, the stereo camera was used to identify fish species and to measure fish length, depth, tilt, and yaw. Five paired stereo camera deployments and pelagic midwater trawl hauls were compared during an acoustic survey in the Strait of Georgia, British Columbia. Fish sometimes moved in response to the stereo camera deployment, but acclimated quickly to its presence at depth. Pacific hake, Merluccius productus, and walleye pollock, Gadus chalcogrammus, were the two dominant species in both midwater trawl haul catches and stereo camera images, but fish sizes were significantly larger in most cases in stereo camera images compared to trawl haul catches. Fish length measurements were most accurate when yaw angles were <30°. Higher abundances of fish in the stereo camera images were associated with higher midwater trawl catch-per-unit-effort values. Fish orientation was close to horizontal for pollock, but slightly downward for hake (15.56°), which could have implications for acoustics-based biomass estimates. The challenges of using stereo cameras in acoustic surveys include smaller sample sizes than those of midwater trawl catches, time required for processing of images, and identification of small fish. However, stereo cameras can be viable tools for acoustic target verification of fish species and measurements of fish lengths, with the advantages of additional information on specific fish depth, tilt, and yaw. Cameras can also sample non-lethally in areas where trawling is not logistically possible, such as in shipping lanes, or permitted, such as marine protected areas. Our results suggest that stereo camera technology is a useful tool for studying fish in the water column.