E. F. Melvin, L. L. Conquest, and J. K. Parrish
EXECUTIVE SUMMARY
This study compared entanglement rates of seabirds and marine mammals and catch rates of salmon among up to three experimental gear treatments and a control (nylon monofilament netting) and among three time-of-day categories in two Washington non-treaty salmon fisheries: the 1996 sockeye fishery in Management Area 7, the San Juan Islands vicinity of north Puget Sound; and the 1995 fall chum fishery in Management Area 10, south Puget Sound. Because the scope of activities and seabird interactions were greater in the sockeye fishery than in the chum fishery, research in the sockeye fishery is emphasized.
This research continued a university-industry research program begun in the 1994 non-treaty sockeye fishery and continued in a 1995 sockeye test fishery. Our goal was to develop methods that eliminate or significantly reduce the incidental capture of seabirds in salmon gillnet fisheries without significantly reducing the fishing efficiency of the nets. This work was funded by a grant from the National Marine Fisheries Service Saltonstall -Kennedy Grant Program and by Washington Sea Grant.
Experimental nets incorporated either visual or acoustic alerts into traditional nylon monofilament gear. Visual barrier nets were monofilament nets with highly visible netting replacing the upper quarter (50 Mesh) or upper eighth (20 Mesh) of the net. Acoustic alert nets were monofilament nets with low frequency sound-emitting devices (pingers) attached to the corkline. Pingers were tested in the 1996 sockeye test fishery only. Fishing time was divided into three categories: morning change of light (AM COL), daytime, and evening change of light (PM COL). Puget Sound Gillnetters' Association fishing vessels were contracted by the Washington Department of Fish and Wildlife (WDFW) to fish experimental nets in a Washington State Test Fishery under our research protocol.
During the 1996 sockeye test fishery, we caught 13,151 sockeye salmon in 642 sets during seventeen fishing trips from 28 July to 29 August. This level of effort exceeded our minimum effort target of 600 sets by 7%. During the 1995 chum test fishery, we caught 6,822 chum salmon in 107 sets in eight fishing trips from 25 October to 11 November. This level of effort met our minimum effort target of 100 sets. Both test fisheries were highly selective for the target species relative to other salmon species: sockeye salmon accounted for 98.4% of the salmon caught in the 1996 sockeye test fishery and chum salmon accounted for 99.7% of the salmon catch in the 1995 chum fishery.
In the 1996 sockeye test fishery, common murres were the most abundant seabird in the study area (30.7 sightings/set) a rate 59 times greater than what we observed in our earlier research in the 1995 test sockeye fishery. Rhinoceros auklets made up almost all other alcid sightings and were almost three times more abundant in 1996 (4.5 sightings/set) than in 1995. A total of 349 alcids were entangled: 260 common murres (75% of the total), 87 rhinoceros auklets, one pigeon guillemot, and one marbled murrelet. Murre entanglement rates were 15 times higher than in the 1995 sockeye test fishery (0.60 murres/set vs. 0.04 murres/set); rhinoceros auklet entanglement rates were 2.8 times higher than 1995 (0.20 auklets/set vs. 0.07 auklets/set in 1995). Seabird and sockeye abundance changed dramatically within the 1996 sockeye season but in opposing patterns. During the final weeks of the fishery, sockeye catch rates dropped off to trace levels (about two fish/set) and murre abundance peaked at near 100 sightings per set.
In the 1995 chum test fishery, there were fewer alcids present than in either the 1996 or 1995 sockeye fisheries in Area 7 and abundance patterns were most similar to the sockeye fishery in 1995. Rhinoceros auklets were the most abundant seabird (0.51 auklet sightings/set) but they were three times less abundant than in the 1995 sockeye and nine times less abundant than in 1996 sockeye test fisheries. Common murres were few (0.1 murre sightings/set) five times fewer than the 1995 sockeye test fishery and dramatically fewer (over 300 times) than the 1996 sockeye test fishery. Eleven rhinoceros auklets and twelve common murres were entangled. Murre and auklet entanglement rates exceeded those of the 1995 sockeye test fishery by four times and 1.6 times, respectively.
In the 1996 sockeye test fishery, entanglement rates of common murres and rhinoceros auklets and catch rates of sockeye salmon varied significantly among the experimental gears, time-of-day categories, and locations tested; however, the patterns of variation among all these factors were species specific. Pinger, 20 Mesh and 50 Mesh gears entangled alcids at rates 58%, 55%, and 50% (respectively) of the monofilament control; sockeye catch rates were 85%, 88% and 39%, respectively. In the 1995 chum test fishery, chum salmon catch rates varied significantly among the three gear types tested; however, alcid entanglement rates did not.
In the 1996 sockeye test fishery, daytime and evening change-of-light sets (COL) entangled alcids at rates of 36% and 68% respectively of those made during the morning COL, whereas sockeye catch rates were 79% and 74% respectively. In the 1995 chum test fishery, neither alcid entanglement rates nor chum catch rates varied significantly by the three time-of-day categories tested; however, the pattern and magnitude of variation was similar to that observed in the 1996 sockeye fishery.
Results of this study identify three basic tools that can be used to reduce seabird bycatch in Puget Sound salmon drift gillnet fisheries: abundance-based or ecosystem management, alternative gear, and time-of-day. The dramatic inter-annual and in-season variation of seabird abundance in Puget Sound is the most important factor determining the rate of seabird entanglements in Area 7. Inter-year and intra-season sources of variation in seabird abundance provide great opportunity for improved management of the fishery based on an ecosystem management concept.
We confirmed that visual barriers are an effective seabird bycatch reduction tool. The 20 Mesh gear met our original goal of significantly reducing seabird bycatch without significantly reducing fishing efficiency. It was tested and proved in multiple fisheries, and was conceived by and endorsed by the Puget Sound Gillnetters' Association as an acceptable tool to reduce seabird bycatch in this fishery. 50 Mesh nets, those with the deeper of the two visual barriers tested, were eliminated as possible seabird reduction tools because they did not meet goals of the research program, were impractical to fish, and entangled porpoise. Although pingers have the greatest potential as tools to reduce seabird bycatch in a wide range of gillnet fisheries, we do not recommend these devices as alternatives for Puget Sound at this time because we believe that they can be improved, results need to be duplicated, and the prototype device is not commercially available. The time of day that gillnets are fished significantly affects seabird bycatch rates. Elimination of morning change-of-light fishing is likely to reduce most rhinoceros auklet entanglements and contribute significantly to reducing common murre entanglements.
Although seabird bycatch and sockeye catch varied significantly by location, areas of high salmon catch and high seabird bycatch tended to overlap, eliminating the possibility of significantly reducing seabird bycatch without significantly reducing salmon catch through zonal or area closures within Management Area 7. Data suggest that the number of birds in the vicinity of the net is probably the most important factor influencing the entanglement rates of seabirds, but that sea state and weather also might be important.
Employing all available tools, fishing 20 Mesh nets at times of high fish abundance during openings that include either daytime and dusk or daylight-only fishing, have the potential to reduce seabird bycatch by up to 70% to 75% in years similar to 1996.
Recommendations
Based on this research, we recommend several management actions that will reduce the bycatch of alcid seabirds in Puget Sound drift gillnet fisheries and enhance seabird conservation in the shared waters of Washington and Canada. Recommendations focus on institutional change for fishery and wildlife management agencies as well as fishery practices. We recommend the following:
| A Publication of the University of Washington pursuant
to National Oceanic and Atmospheric Administration Award NA56FD0618 and
Washington Sea Grant, and in cooperation with the Puget Sound Gillnetters'
Association and the Washington Department of Fish and Wildlife
June 15, 1997 The views expressed herein are the authors' and do not necessarily reflect the views of the National Oceanic and Atmospheric Administration or any of its subagencies. WSG AS 97-01 Suggested Citation: Melvin, Edward F., Loveday L. Conquest, and Julia K. Parrish, 1997. Seabird bycatch reduction: new tools for Puget Sound drift gillnet salmon fisheries: 1996 sockeye and 1995 chum non-treaty salmon test fisheries final report. Washington Sea Grant. Project A/FP-7. Available on loan from the National Sea Grant Depository, and from publisher. WSG-AS 97-01 © University of Washington, Board of Regents. No part of this report may be reproduced except by permission of the publisher. Contact Washington Sea Grant, 3716 Brooklyn Avenue N.E., Seattle, WA 98105-6716, (206) 543-6600, or fax (206) 685-0380. |
Entered August 26, 1997