Update of recent developments in MULTIFAN-CL and related software for stock assessment [SA IP07] Simon Hoyle, et al.

By: Contributor(s): Material type: ArticleArticleLanguage: ENG Publication details: [Pohnpei, Federated States of Micronesia] Western and Central Pacific Fisheries Commission (WCPFC) 2009Description: 17 p. ill., graphs, tables 30 cmSubject(s): DDC classification:
  • 639.27783
Online resources: Summary: "MULTIFAN-CL (MFCL) is a statistical, age-structured, length-based model routinely used for stock assessments of tuna and other pelagic species. The model was originally developed by Dave Fournier of Otter Research for application to south Pacific albacore tuna. MFCL is typically fitted to total catch, size-frequency and tagging data stratified by fishery, region and time period. Recent tropical tuna assessments (e.g. Langley et al. 2007; Langley et al. 2008) encompass a time period of 1952–2007 in quarterly time steps, and model >20 separate fisheries occurring in 6 spatial regions. The main parameters estimated by the model include initial numbers-at-age in each region (constrained by an equilibrium age-structure assumption), the number in age class 1 for each quarter in each region (the recruitment), growth parameters, natural mortality-at-age (if estimated), selectivity-at-age by fishery (constrained by smoothing penalties or splines), effort deviations (random variations in the effort-fishing mortality relationship) for each fishery, initial catchability and catchability deviations (cumulative changes in catchability with time) for each fishery (if estimated). Parameters are estimated by fitting to a composite likelihood comprised of the fits to the data and prior distributions for various parameters. Each year the MFCL development team work to improve the model to accommodate changes in understanding of the fishery, to fix software errors, and to improve usability. This document records changes made since August 2007 to the model and to the other components of the MFCL project."
Item type: Meeting Paper
Tags from this library: No tags from this library for this title. Log in to add tags.
Holdings
Current library Call number Status Date due Barcode
SPC Special Collections SPC 639.277 83 WES 2009 Available

"WCPFC-SC5-2009/SA-IP-07"

Includes bibliographical references (p. 17)

"MULTIFAN-CL (MFCL) is a statistical, age-structured, length-based model routinely used for stock assessments of tuna and other pelagic species. The model was originally developed by Dave Fournier of Otter Research for application to south Pacific albacore tuna. MFCL is typically fitted to total catch, size-frequency and tagging data stratified by fishery, region and time period. Recent tropical tuna assessments (e.g. Langley et al. 2007; Langley et al. 2008) encompass a time period of 1952–2007 in quarterly time steps, and model >20 separate fisheries occurring in 6 spatial regions. The main parameters estimated by the model include initial numbers-at-age in each region (constrained by an equilibrium age-structure assumption), the number in age class 1 for each quarter in each region (the recruitment), growth parameters, natural mortality-at-age (if estimated), selectivity-at-age by fishery (constrained by smoothing penalties or splines), effort deviations (random variations in the effort-fishing mortality relationship) for each fishery, initial catchability and catchability deviations (cumulative changes in catchability with time) for each fishery (if estimated). Parameters are estimated by fitting to a composite likelihood comprised of the fits to the data and prior distributions for various parameters. Each year the MFCL development team work to improve the model to accommodate changes in understanding of the fishery, to fix software errors, and to improve usability. This document records changes made since August 2007 to the model and to the other components of the MFCL project."