Research Document 2018/020

Re-design of the Joint Industry-DFO Atlantic Halibut (Hippoglossus hippoglossus) Survey off the Scotian Shelf and Grand Banks

By Cox, S., Benson, A., and Doherty, B.

Abstract

The Industry-Fisheries and Oceans Canada (DFO) Atlantic Halibut Longline Fixed Station Survey is critical to the current management approach for Atlantic Halibut (Hippoglossus hippoglossus) on the Scotian Shelf and Grand Banks (Northwest Atlantic Fisheries Organization (NAFO) Divs. 3NOPs4VWX5Zc). The Atlantic Halibut Longline Survey indices are used as an index of abundance in the statistical catch-at-length stock assessment model and as a key input in the management procedure used to provide annual Total Allowable Catch (TAC) advice. The fixed-station survey design used from 1998-2016 has allocated more sampling in areas closer to port, areas along the shelf edge, and areas with historically high catch rates.

Alternative survey designs that achieve more representative and unbiased sampling coverage over the entire stock area are considered in this paper. Three alternative area-depth stratification options with a combination of 4 (4X5YZ, 4W, 4VnVs, 3NOPnPs) or 5 (4X5YZ, 4W, 4VnVs, 3PnPs, 3NO) NAFO area strata and 2 (30-250m, 251-750m) or 3 (30-130m, 131-250m, 251-750m) depth strata were evaluated.  Bootstrap replicates were used to assess survey precision under different stratification options, sample sizes, and sampling allocation options. Bootstrap estimates of the average annual Coefficients of Variation (CV) of Catch Per Unit Effort (CPUE) were similar for all 3 stratification options and indicated that the greatest gains in survey precision were from improvements to sample allocations among the different strata. Sampling that was allocated in proportion to stratum area ('area-based allocation') or stratum area multiplied by stratum standard deviation ('optimal allocation') produced higher precision than the historical allocation of samples on the fixed station survey. All three stratification options produced similar CVs that were below the target 20% CV for n=150 annual samples, with bootstrap estimates of mean annual CVs ranging between 12.3-12.5% and 11.1-11.8% using area-based and optimal sampling allocations, respectively.

An implementation plan for the new survey using a calibration period, where the fixed station and the new Stratified Random Survey (StRS) will be run concurrently, was developed. Analysis of CPUE from subsets of the fixed stations (100, 150, Freq100, and Freq150) indicated that a subset of the most frequently surveyed stations can be used to calculate interim TACs during the transition period to the StRS. The fixed-stations subset will continue to be sampled each year along with 150 randomly selected stations from the StRS during the 3-4 year calibration period.

Complete document (PDF, 4,617 KB)

Accessibility Notice

This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.