FRENCH and DUNN: LOSS FROM HIGH-SEAS GILLNETTING 



Indirect Observations (1964 and 1965) 



1964. — The 1964 studies of dropouts were 

 carried out while conducting exploratory fishing 

 for salmon in the western North Pacific Ocean 

 in September and October. The method used to 

 estimate the rate of dropouts was to determine 

 the differences in numbers of salmon caught in 

 gill nets fished for a long period compared to 

 the cumulative catch of salmon caught in a 

 series of short sets fished at the same time and 

 place. Any decrease in catch of the nets fished 

 continuously would be attributed to dropouts. 

 In this procedure, gill nets were set, hauled, 

 and reset at intervals throughout the night 

 (Figure 1). 



Conventional multifilament nylon gill nets (1 

 shackle was 91.5 m long and 7.3 m deep)^ 

 were fished in individual units of shackles, with 

 each unit connected by a 55-m line. Each unit 



of gear consisted of 2 shackles of 114-mm and 

 2 shackles of 83-mm mesh — sizes that would 

 capture most sizes of salmon present. There was 

 a security snap on each gear unit for easy con- 

 nection and disconnection of the units. Lighted 

 buoys and radio buoys were attached to the net 

 string to ensure finding the gear during hours 

 of darkness. 



The time and fishing procedure during the 

 night were as follows: 



2000 Set gear units No. 1, 2, 3, and 4. 



2300 Haul unit No. 1, remove salmon, and 

 reset. 



0200 Haul units No. 1 and 2, remove sal- 

 mon, and reset 2. 



0500 Haul units No. 2 and 3, remove sal- 

 mon, and reset 3. 



0800 Haul unit No. 3 and the last unit. No. 

 4. 



^Details of net construction (and fishing procedure) 

 were given by Craddock (1969). 



Multifilament nylon gill nets were 100% nylon — 330 

 Starrlock type E (or equivalent) of plied or cable laid 

 salmon twine and dark green in color. Ply size ranged 

 from 3 to 12. depending on mesh size; numbers of body 

 meshes ranged from 57 to 123 (plus guard meshes), 

 depending on mesh size. Each shackle was 183-m long, 

 stretched measure, so that each gill net made a full 9i.5-m 

 shackle when hung in 50% on the cork line. Monofilament 

 nylon gill nets were of identical construction to multifila- 

 ment nylon gill nets, except for net material. Monofila- 

 ment nets were constructed of light green single strand 

 filament, double selvage, with body meshes of 0.30-, 0.40-, 

 0.50-, and 0.50-mm diameter for mesh sizes 64. 83, 114, 

 and 133 mm (stretched measure, knot to knot), respec- 

 tively. [Reference to trade names does not imply endorse- 

 ment by the National Marine Fisheries Service, NOAA.] 



2000 



I 



2300 



I 



0200 



I 



0500 



I 



0800 



(haul ond reset) 



(haul and reset) 



5iS 



(fishing period) 



(houl and reset) 



(continuously fishing) 



Figure 1. — Fishing plan for study of dropouts, 1964. Each 

 unit of gear is identical: length of shaded area refers to 

 fishing time. 



This fishing routine allowed comparison of 

 the different 3-h periods during the night (2000- 

 2300, 2300-0200, 0200-0500, and 0500- 

 0800). Additionally, comparisons were pos- 

 sible for catches of nets fished continuously 

 with the sum of the catches of the individual 

 gear units for various time periods. For exam- 

 ple, unit No. 4 fished for 12 h, from 2000 to 

 0800, and catches in these nets were compared 

 to the cumulative catch of the four individual 

 units hauled at the end of each 3-h period. 



1965. — Procedures used in 1965 were 

 extensions of those used in 1964 but on a larger 

 scale employing two vessels full time. Fishing 

 was conducted in the North Pacific Ocean 

 (near long. 166 °W) and in the Bering Sea (near 

 long. 160°W) in June and early July, concen- 

 trating on maturing sockeye salmon. 



Two experimental designs were used to 

 estimate dropout rates of salmon. In the first 

 design, the nets were fished at night; one unit of 

 gear (four 91.5-m shackles of 133-mm mesh 

 gill nets per unit) was fished for 6 h, a second 

 unit for the first 3 h, and a third for the second 

 3 h (Figure 2). The fishing was repeated over 

 three different time periods during the night. 

 The dropout rate was computed by comparing 

 the catch of the 6-h unit with those of similar 



847 



