Use this URL to cite or link to this record in EThOS:
Title: Modelling the Shetland sandeel stock
Author: Poloczanska, Elvira Severina
Awarding Body: University of Glasgow
Current Institution: University of Glasgow
Date of Award: 2001
Availability of Full Text:
Access through EThOS:
Access through Institution:
Young sandeels (0- and 1-group fish) form a substantial part of the total stock biomass at Shetland. As well as forming the basis for a local fishery at Shetland, they are also an extremely important prey item for many seabird species which breed around Shetland coasts. Improving our understanding of recruitment dynamics is vital for managing the Shetland fishery whilst minimizing adverse affects on breeding seabirds. Therefore, this thesis has concentrated on analyzing recruitment patterns and developing models to simulate hypothetical recruitment scenarios. Various models were developed and/or parameterised to simulate recruitment. Firstly, as there appeared to be no clearly defined stock-recruit relationship, time series models were applied to expose and improve understanding of patterns in the data. However, the relationship between stock and recruitment cannot be ignored if spawning stock biomass changes through exploitation. Therefore, Shepherd (1982) and Ricker (1954, 1975) stock-recruit models are parameterised using bootstrap analysis. For the Shepherd model, which is the more complex of the two, bootstrap analysis identified models which could be considered as "constant recruitment" models - that is no influence of spawning stock biomass on recruitment. It was found that four very low recruit years, which occurred in the late 1980s when spawning stock biomass was generally high or moderate, were strongly influencing model fit. This prompted the question: are these low recruit years typical of the range of recruitment for the given spawning stock biomasses or can they be considered outliers? In the face of uncertainty (although there is no biological reasoning) these data points were removed and the stock-recruit models re-parameterised. This enabled curves with 95% confidence intervals to be parameterised for both stock-recruit models (Shepherd & Ricker). The autoregressive model parameterised earlier was combined with the stock-recruit models so recruitment in any year was expressed as a function of SSB and of recruitment in the previous year.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: SH Aquaculture. Fisheries. Angling