Chapter 1 Overview

This Gitbook documents my/our research on Alaskan sablefish (Anoplopoma fimbria). The objective is to develop and explore a spatially explicit stock assessment model for the Alaskan sablefish stock. Although this is focused on sablefish, there will be many topics encountered that are common to other stock assessments. The following section outlines chapters contained in this document.

Gitbook outline

  • Chapter 2 outlines a list of objectives that we have set or accomplished during this research project

  • Chapter 3 documents the current stock assessment model and assumptions (work in progress). This is the first step of the project. Its purpose is to help me understand the data and important process dynamics assumed in the current assessment. Most models applied in this research were written in TMB (Kristensen et al. 2015), many of the assessment models are available in the SpatialSablefishAssessment R package, which was developed during this research.

  • Chapter 4 documents a generalized spatial assessment model used as both an OM and EM.

  • Chapter 5 describes, characterizes and explores the survey longline data available for the assessment.

  • Chapter 6 describes, characterizes and explores fishery dependent data, which includes reported catch (log-book data) and observer records.

  • Chapter 7 describes the tagging data and how it can be used within a spatial stock assessment model but also used in a panmictic model for informing population dynamics such as growth.

  • Chapter 8 outlines diagnostics and decisions that were made when considering the spatial resolution of our model.

  • Chapter 9 displays outputs from preliminary spatial models under development.

  • Chapter 11 outlines future model assumptions/considerations.

  • Chapter 12 explores estimating age-based movement rates using a simple age-structured simulation.

  • Chapter 13 describes different tag likelihoods that I considered in the spatial model and does a simple simulation estimating annual movement rates with the different likelihoods.

  • Chapter 16 explores how to include sex disaggregated composition observations that can include sex ratio information. It conducts a simple simulation to investigate two different approaches.

  • Chapter 17 explores how to parameterize fishing mortality. The current approach is to estimate an annual fishing mortality parameter for each gear as a free parameter. I feel slightly uncomfortable about this approach moving towards a spatial model as the number of estimated parameters will explode when a spatial dimension is added. This chapter looks at two alternative approaches that either derive fishing mortality estimates using a Newton Raphson solver which is heavily borrowed from the Stock Synthesis “hybrid” approach (Methot Jr and Wetzel 2013) and Pope’s discrete approach (Pope 1972) which uses exploitation rates. Both these methods have been applied in the literature for decades. The aim of this chapter is to do a simulation and make sure the considered approaches are efficient and numerically stabile for the purposes of our research.

  • Chapter 18 explores how to calculate initial numbers at age for the accumulating age cohort (plus group), in a spatially explicit age-structured models that assumes markovian movement during initialization.

Future things to consider

  • When is the latest year that we can start the model at? Currently the assessment model starts in 1960 because there is early survey and catch info along with some of the largest catches recorded. I want to explore starting the model at a later period i.e., the 1980’s when there is consistent surveys and age data. This is thought to reduce the number of estimable parameters that have low information (early recruitment deviations). However, the downfall is we may loose information on stock production because the 60’s and 70’s have some of the largest recorded catches.
library(TMB)
#library(stockassessmenthelper)
library(ggplot2)
library(dplyr)
library(reshape2)
library(gridExtra)
library(knitr)
library(RColorBrewer)

References

Kristensen, Kasper, Anders Nielsen, Casper W Berg, Hans Skaug, and Brad Bell. 2015. “TMB: Automatic Differentiation and Laplace Approximation.” arXiv Preprint arXiv:1509.00660.
Methot Jr, Richard D, and Chantell R Wetzel. 2013. “Stock Synthesis: A Biological and Statistical Framework for Fish Stock Assessment and Fishery Management.” Fisheries Research 142: 86–99.
Pope, JG. 1972. “An Investigation of the Accuracy of Virtual Population Analysis Using Cohort Analysis.” ICNAF Research Bulletin 9 (10): 65–74.