Fisheries harvest management procedures under time-varying productivity
Fish stocks often show persistent changes in fish production with hypothesized causes including time-varying recruit parameters, natural mortality of mature fish, and somatic growth. Changes in fish productivity can affect biological reference points and the expected robustness of harvest strategies...
Main Author: | |
---|---|
Format: | Text |
Language: | English |
Published: |
University of British Columbia
2020
|
Subjects: | |
Online Access: | https://dx.doi.org/10.14288/1.0389968 https://doi.library.ubc.ca/10.14288/1.0389968 |
Summary: | Fish stocks often show persistent changes in fish production with hypothesized causes including time-varying recruit parameters, natural mortality of mature fish, and somatic growth. Changes in fish productivity can affect biological reference points and the expected robustness of harvest strategies in meeting long-term conservation and economic objectives. Current assessment approaches commonly assume “stationarity” in stock-recruit parameters and implemented harvest strategies commonly lack consideration of changes in fish productivity. This dissertation covers three studies based on statistical analysis and simulation methods that aim to advance understanding of time-varying stock-recruit parameters and harvest control rules (HCRs) to cope with changes in fish production. Chapter 2 examined the performance of a Kalman filter algorithm that estimates dynamic state variables (e.g., changing stock-recruit parameters) from inaccurate observations obtained from stock assessment outputs. In Chapter 3, stochastic optimization in policy space (SOPS) was used to find an optimal HCR that is robust to changes in productivity under closed-loop simulation or management strategy evaluation (MSE). By using SOPS and a simple population dynamics model, this chapter examined optimal HCRs under different types of regime shifts (e.g., changes in carrying capacity and intrinsic growth rates). Chapter 4 examined empirical HCRs in an extreme situation of “non-stationarity” in stock-recruit parameters (i.e., a spasmodic recruitment pattern), using Atlantic redfish stocks as a case-study. The Kalman filter detected the overall trend in the stock-recruit parameters over time. Also, it was found that optimal HCRs depend on the timing of biomass estimates relative to harvest, observation error, intrinsic population growth, and regime shifts types. In the case of spasmodic stocks, it was found that simple HCRs that used data from well-designed fishery-independent resource surveys were capable of handling changes in productivity. The present dissertation shows the importance of well-designed fishery-independent surveys and simple HCRs for tackling changes in productivity and guidelines are provided for developing HCRs that are robust to the occurrence of such changes. The models and findings presented here can be applied to improve the management of other stocks that show persistent changes in fish productivity. |
---|