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