Hynes, Natasha2024-06-122024-06-122024-04https://unbscholar.lib.unb.ca/handle/1882/37997Monitoring zooplankton populations, especially those of energy-rich copepods, is of interest due to their roles as prey for many commercially important and endangered marine species. Autonomous sampling could help improve fine-scale temporal and spatial monitoring, complimenting conventional methods. Here, I evaluate the performance of a prototype, glider-mounted shadowgraph camera for estimating zooplankton identifications and concentrations. The evaluation was completed via a gear-comparison with standard samplers (a MultiNet Midi and UVP6-HF). Copepods were the most abundant taxon in the zooplankton community and the shadowgraph yielded mostly definite identifications, while the UVP6 had more uncertainties. The shadowgraph and UVP underestimated concentrations compared to the MultiNet, yet correlations were significant with both imaging sensors. The shadowgraph, with a Spearman correlation of 0.73 with the MultiNet, excelled at detecting vertical copepod layers and further studies should explore the shadowgraph’s ability to detect diel-vertical migration and develop machine learning tools to aid finer taxonomic identification.xvii, 92electronicenhttp://purl.org/coar/access_right/c_abf2master thesisDavies, KimberleyBiology