Characterization of a new stereo-video tool to survey deep water benthic fish assemblages with comparison to a remotely operated vehicle

Increasing use of ecosystem-based management strategies, which are often applied to broad geographic areas and preclude extractive activities, are creating a need for rapid, cost-effective monitoring of large areas. Visual surveys are increasingly being used to meet this need. In this thesis, I examine a new tool for surveying fish assemblages in deep-water habitat: a stereo-video lander. In Chapter 1, I evaluate methodological choices and their impact on the data collected. In Chapter 2 ,I compare the video lander with a Remotely Operated Vehicle (ROV). In characterizing the new stereo-video lander, I found a negligible effect of bait on the number offish observed or on the number of species observed. The rotating camera system yielded density estimates slightly lower than those determined by a stationary camera but the rotating camera system produced less variance with the same number of surveys. In comparing the lander and the ROV, both measured similar densities for most species. Furthermore, I found that estimates of the variance in fish density were similar for the two tools given a comparable sampling effort (i.e., number of sites surveyed). Differences in community assemblage were found to be significant between the two tools. Because of the similarity in results and ability to quickly perform surveys and move on to new areas, the lander represents a new option when considering visual tools for deep-water research. I certify that the Abstract is a correct representation of the content of this dissertation.