Organizations responsible for managing commercial fisheries conduct annual stock assessments to monitor stock and, in principle, reduce the risk of overexploitation. These are fundamental to setting the total allowable catch for the upcoming fishing year. While there have been many attempts to estimate uncertainty associated with certain components and estimates of stock assessments, to date there has been no systematic assessment of their forecasted predicted value. Using data from annual stock assessments from 65 commercial fisheries around the world, estimates were obtained of both predicted (from the previous year) and observed (in the current year) catch-at-age. When comparing observed (actual) and predicted catch-at-age for a given stock, estimates were obtained of the predictive power of next-season forecasts. Using other attributes of the fishery and the stock (biological (e.g. life history) and management (e.g. assessment model employed)), empirical models were constructed that attempt to determine variability in predictive power among stocks. It was observed that, on average, within-year predictive powers (age-series within time samples) were higher than year over year predictive powers (time-series within age samples). While focusing on time-series within age, it was observed that change over the period of record (in natural mortality rate, assessment model employed, etc.) reduced predictive power; while for age-series within time, it was shown that cumulative landings reduced predictive power. This study represents one of the first attempts to quantify systematically the predictive power of fisheries stock assessment models.
| Identifer | oai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/38107 |
| Date | 12 September 2018 |
| Creators | Docking, Kathryn A |
| Contributors | Findlay, C. Scott, Lane, Daniel |
| Publisher | Université d'Ottawa / University of Ottawa |
| Source Sets | Université d’Ottawa |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Format | application/pdf |
Page generated in 0.0018 seconds