Show simple item record

dc.contributor.advisorMinto, Dr Cóilín
dc.contributor.advisorGerritsen, Dr Hans
dc.contributor.authorBatts, Luke
dc.contributor.authorGerritsen, Dr Hans
dc.contributor.authorBrophy, Dr Deirdre
dc.date.accessioned2023-01-10T15:38:17Z
dc.date.available2023-01-10T15:38:17Z
dc.date.copyright2022
dc.date.issued2022-02
dc.identifier.citationBatts, L. Addressing cohort uncertainty through advanced length frequency and stage-based assessment models with application to anglerfish. PhD Thesis. ATU.en_US
dc.identifier.urihttps://research.thea.ie/handle/20.500.12065/4357
dc.description.abstractFisheries stock assessments are important tools for successful management of fisheries. Reliability of a stock assessment model is often determined by the data available and accounting for key uncertainties in the data is an important aspect of stock assessments and management. Central to many assessment models is the tracking of cohorts through the population, but uncertainty in the identification of cohorts, in the form of uncertainty in age-estimation and fish growth, can impact model performance. Overall, the aim of this thesis was to develop methods to address this cohort uncertainty, as well as assess the impact of biased age-composition data on the fisheries advice process. I focus on anglerfish as the main case study species, as their assessment and management is affected by the challenges touched upon earlier. To provide context, Chapter 1 gives an overview of stock assessment and management, focussing on the challenges of uncertainty in age-estimation and growth, as well as the approaches used to account for them. Anglerfish fisheries are also summarised and issues affecting their assessment and management discussed. Chapter 2 addresses the topic of cohort uncertainty by developing a new method of length frequency analysis. The model is a maximum likelihood-based procedure that uses Gaussian mixture models and the Expectation Maximisation algorithm to estimate von Bertalanffy growth parameters from length frequency data from fisheries surveys. The model was applied to length data from the white-bellied anglerfish stock in the Celtic Sea and Bay of Biscay. The basic model estimated a single set of growth parameters, whereas the hierarchical extension to the model was able to model some of the natural variability in fish growth between cohorts or years with bivariate random effects on key parameters. Chapter 3 approaches the issue of cohort uncertainty from a different perspective, implementing and and testing the performance of stage-based stock assessment models. Stage-based assessment models have less data requirements and simpler population dynamics than more complex assessment models, so are likely to be more robust to cohort uncertainty in the data. The stage-based assessment models implemented were: Catch-Survey Analysis (CSA), and a model first described in a theoretical paper by Schnute (1987). The performance of these two theoretically different stage-based assessment models was assessed with a simulation-testing framework and on a real anglerfish stock. The findings showed both models are useful stock assessment models, with CSA more robust but less precise than the Schnute model. The Schnute model was more precise than CSA but required growth and mean fish weight data unaffected by selectivity. As part of the work conducted for Chapter 3, the R package ‘sbar’ was developed. This is a fully documented R package that contains the functions to run the stage-based assessment models. Chapter 4 outlines the key assumptions and data requirements of the models, as well as demonstrating use with data from a real anglerfish stock. Versions of the Schnute model not described or tested in Chapter 3 were also detailed here. A goal of this chapter was to allow new users to begin running stage-based assessment models with relevant background information provided. Chapter 5 addresses the impact of using age-composition data generated with a biased growth function in stock assessment and management over time. A management strategy evaluation framework was used, with both stage-based and age-based management procedures tested. A method for estimating CSA reference points was also developed. Both management procedures were affected by the biased sampling data, but in different ways. Performance statistics indicated that it is important to consider the uncertainty and potential bias in growth estimates when generating age-composition data. Overall, the research presented in this thesis has developed and implemented techniques that aim to further advance the field of fisheries stock assessment and management when cohorts are uncertain. The thesis focussed on anglerfish for the majority of case studies due to the fisheries’ reported issues, however the methods implemented here are useful in a wider context and there are many species to which the techniques could be applied.en_US
dc.formatapplication/pdfen_US
dc.language.isoengen_US
dc.publisherAtlantic Technological Universityen_US
dc.rightsAttribution-NonCommercial-ShareAlike 3.0 United States*
dc.rights.urihttp://creativecommons.org/licenses/by-nc-sa/3.0/us/*
dc.titleAddressing cohort uncertainty through advanced length frequency and stage-based assessment models with application to anglerfishen_US
dc.typeinfo:eu-repo/semantics/doctoralThesisen_US
dc.contributor.sponsorThis PhD (Cullen Fellowship: CF/16/03) was carried out with the support of the Marine Institute and is funded under the Marine Research Sub-Programme by the Irish Government.en_US
dc.description.peerreviewnoen_US
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessen_US
dc.subject.departmentMarine and Freshwater Research Centreen_US


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record

Attribution-NonCommercial-ShareAlike 3.0 United States
Except where otherwise noted, this item's license is described as Attribution-NonCommercial-ShareAlike 3.0 United States