Novel Statistical Analyses of Longline Survey Data for Improved Indices of Atlantic Halibut Abundance

Abstract

The longline halibut survey is a joint project between the Atlantic Halibut Council (AHC) and Fisheries and Oceans Canada (DFO). This annual survey is used to monitor the status of Atlantic Halibut on the Scotian Shelf and southern Grand Banks (NAFO Divisions 3NOPs4VWX5Zc). The purpose of this thesis is to develop novel statistical analyses of Atlantic Halibut longline survey data for more accurate and precise indices of relative abundance. Reproduction of the survey indices has been completed using hook occupancy data collected between 2017 and 2019 and current hook competition model. In order to reflect spatial patterns in these data, a spatial random field is introduced into the model. Covariates for depth, temperature and area strata (along with their potential interactions) are also introduced as these may affect the survey indices. Model selection has been performed and the selected spatial model is then used to estimate the relative abundance indices for target and non-target species at each survey station for 2017 to 2019. Finally, the estimated indices are aggregated over space using three different approaches to compare the estimated changes in survey indices with those obtained with the non-spatial model currently being used by DFO. The results indicate that the relative abundance indices for halibut are higher on shallow banks in southwest Nova Scotia (NAFO 4X and 4W) and along the shelf edge throughout the management unit.