Modelling abrupt shifts of fish recruitment and growth
Regime shifts in marine ecosystems may result in abrupt changes in fish population dynamics, and not accounting for such shifts could have potentially large and far-reaching consequences for fisheries assessment and management decisions. In this work, I develop a methodology to model abrupt shifts i...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Memorial University of Newfoundland
2021
|
| Subjects: | |
| Online Access: | https://research.library.mun.ca/15204/ https://research.library.mun.ca/15204/1/thesis.pdf |
| Summary: | Regime shifts in marine ecosystems may result in abrupt changes in fish population dynamics, and not accounting for such shifts could have potentially large and far-reaching consequences for fisheries assessment and management decisions. In this work, I develop a methodology to model abrupt shifts in recruitment and somatic growth, which are two key processes in controlling fish population dynamics. In Chapter 1, I review the impacts of regime shifts on fish populations and find that abrupt shifts in productivity are very common among global fish species. In Chapter 2, I introduce the approach of modelling recruitment and regime shifts. The methodology includes a hidden Markov model for the unobserved environmental regimes, a stock-recruitment (SR) model for the regime-specific SR function, the maximum likelihood approach for evaluating the marginal likelihood, and the corrected Akaike information criterion (AICc) for the model selection. I conduct simulation tests to evaluate the performance of the method and results indicate that our method can objectively identify the unobserved environmental regimes and estimate regime-specific SR model parameters well. In Chapter 3, I extend the hidden Markov approach to model abrupt shifts in fish growth using a von Bertalanffy growth model (VBGM). Simulation results demonstrate that the method can accurately identify abrupt shifts in growth and estimate regime-specific growth parameters well. I apply both the hidden Markov stock-recruit model (HMSM) and the hidden Markov growth model (HMGM) to an Atlantic cod stock on the southern Grand Bank off Newfoundland, Canada. Results indicate that the cod stock has two distinct recruitment regimes and two distinct growth regimes, and our method identify one abrupt shift in recruitment and four abrupt shifts in somatic growth. I consider the methodology proposed in this thesis as a useful tool to model regime-like changes of fish population dynamics. In Chapter 4, I discuss the management implications of abrupt shifts in fish ... |
|---|