190 



Fishery Bulletin 97(1), 1999 



discarding often preclude the identification of defi- 

 nite areas and times of persistently high discards of 

 species (see also Robin, 1991; Martinez et al., 1993; 

 Liggins and Kennelly, 1996; Kennelly et al., 1997). 

 The present study shows that high discard rates for 

 scup were rarely consistent in particular areas and 

 times, although one area and time did show some 

 persistence throughout the four years: the relatively 

 small area (area 613) off Long Island, New York, be- 

 tween 30 and 40 fm in the months of November and 

 December each year. It is well known that from May 

 to August each year scup spawn in estuaries, bays, 

 and inshore areas south of Cape Cod (e.g. see Wilk 

 and Brown, 1980; Eklund and Targett, 1990), par- 

 ticularly around Long Island (Finkelstein, 1971). In 

 autumn, after spawning, they migrate south at 

 greater depths towards their wintering grounds from 

 southern New Jersey to Cape Hatteras, returning back 

 to their inshore spawning grounds in spring ( see Neville 

 and Talbot, 1964; Wilk and Brown, 1980; Jeffries and 

 Terceiro, 1985; Eklund and Targett, 1990). It is appar- 

 ent from this behavior, that the occurrence of large 

 numbers of scup as trawl discards in area 613 off Long 

 Island in the autumn of each year coincides with their 

 migration from Long Island's inshore waters to their 

 wintering grounds farther south. 



When spatial and temporal patterns in observer 

 data reveal persistent areas and times of high dis- 

 card rates, they can be used by fisheries managers 

 to identify where and when various management 

 tools can be applied to reduce discards. These tools 

 usually involve either spatial or temporal closures 

 to fishing (or both) or modifications to fishing gears 

 and practices that reduce discards. 



If fisheries managers consider the area and time 

 of high scup discard identified in this paper as a can- 

 didate for a closure, the data in Figure 5 provide some 

 information on the consequences this may have on 

 landings of squid and other species. In the identified 

 area and time, an average of over 507 lb of squid 

 were retained per hour towed, whereas an average 

 319 lb of scup were discarded per hour towed. The 

 actual numbers of fish involved were not available 

 but, because the discarded scup were those individu- 

 als considered too small to retain, discarded weights 

 often represent more individual fish per pound than 

 retained weights. One could expect, therefore, that 

 closing this area at this time would protect signifi- 

 cant numbers of small scup, thereby assisting the 

 1995 Northeast Regional Stock Assessment Work- 

 shop's recommendation to decrease their exploitation 

 (NEFSCM. Larger quantities of dogfish and lesser 

 quantities of skates, butterfish, and whiting would 

 also be protected by such a closure. The "cost" of such 

 a closure, however, would be significant reductions 



in the landings of squid from the area at that time 

 and lesser reductions in the landings of scup and 

 whiting. One way of estimating the potential effects 

 that such a closure strategy might have for the re- 

 gion is to compare the overall squid retained and scup 

 discard rates with rates adjusted by excluding all 

 tows done in the identified area and time. Through- 

 out the region, the overall discard rate for scup dur- 

 ing the four years was 37.5 Ib/h (standard error 

 (SE)=4.7), but if the tows done in the identified area 

 and time are excluded, the overall rate falls to 27.3 

 Ib/h (SE=4.5): a decrease of 27%. This should be com- 

 pared to the overall retained rate of squid in the re- 

 gion of 128.5 Ib/h (SE=6.1) falling to 114.9 Ib/h 

 (SE=6.0) if the tows done in the identified area and 

 time are excluded — a decrease of 10.6%. It is impor- 

 tant to note, however, that closing the area at this 

 time to trawling will not simply remove trawling ef- 

 fort from the region but merely redirect it to other 

 areas that may yield lower scup discards. This means 

 that the effects on landings and discards in the re- 

 gion will not simply be those protected inside the clo- 

 sure but will be tempered by increased landings and 

 discards outside the closure by the redirected vessels. 

 The other suite of management tools available to 

 reduce discards in areas and at times of high dis- 

 carding involve modifying fishing gears and practices 

 to improve selectivity. Such modifications as the 

 Nordmore Grid and square-mesh panels have proven 

 successful in reducing discards of small fish in trawl 

 fisheries (e.g. Carr, 1989; Isaksen et al., 1992; 

 Broadhurst et al., 1996; Broadhurst and Kennelly, 

 1997), and modifications like downward sorting grids 

 and horizontal panels in nets have reduced the 

 bycatches of unwanted sizes of species in groundfish 

 trawls (see Larsen and Isaksen, 1993; Engas and 

 West"^ ). For the issue of scup discarding in the north- 

 eastern United States, gear modifications would be 

 a better management alternative than the closure 

 strategy outlined above if they could reduce greater 

 quantities of scup discards (and concomitantly re- 

 duce landings of squid and other species by smaller 

 amounts) than those under a closure strategy. Such 

 a general, gear-based solution is also advantageous 

 because its adoption throughout the region would not 

 only lead to reductions in scup discard in identified 

 "hotspots" but in all areas where scup discarding 

 occurs. The information provided in the present study 

 identifies the ideal locations and times for any gear- 

 based research that aims to reduce scup discard by 

 trawlers and where and when any gear modifications 



■* Engas, A., and C. W. West. 1995. Development of a species- 

 selective trawl for demersal gadoid fisheries. Int. Coun. Explor 

 Sea Council Meeting (CM) 1995/B+G+H+J+K:1. 



