| In British Columbia, Canada, salmon purse seiners line | | | | effort level of 100 vessels (whereas the maximum |
| up at fishing access points, forming well defined | | | | effort observed was 363 boats) and indicated that (at |
| queues. These queues were measured over time, | | | | saturation) the fleet caught 80% to 90% of the |
| using a one-dimensional recording scale. Sixty-one | | | | vulnerable migrating salmon present in Johnstone and |
| overflights of Johnstone Strait and Queen Charlotte | | | | Queen Charlotte Straits during what were commonly |
| Strait were attempted; 51 flights were completed. | | | | 48- or 72-hour fishing openings. (Note: Salmon |
| Two models were presented for exploitation rates in | | | | successfully migrating through the strait on days that |
| relation to queuing patterns. The overflight model was | | | | were closed to seiners and salmon that were not |
| fit to the line-up distributions. One underlying assumption | | | | vulnerable to the gear--e.g., below the depth of the |
| was that the skippers possessed fairly accurate | | | | nets--escaped the purse-seine fleet.) |
| information regarding the distribution of catches | | | | In general, traditional assumptions were rejected. |
| (analysis of variance methods utilizing skippers' logbook | | | | Vessels did not operate independently. Boats were not |
| data showed that line-up lengths reflected catch rates). | | | | distributed in a random fashion. The overflight model |
| The model fit well and the parameter estimates | | | | provided predicted exploitation rates. The exploitation |
| reflected anecdotal and statistical information about | | | | response to effort was qualitatively distinct from the |
| fish behavior. The exploitation rates saturated at an | | | | forms incorporated in traditional models. |