720 



Longline-caught blue shark {Prionace glauca): 

 factors affecting the numbers available 

 for live release* 



Guillermo A. Diaz 



Joseph E. Serafy 



National Marine Fisheries Service 



Southeast Fisheries Science Center 



75 Virginia Beach Drive 



Miami. Florida 33149 



E-mail address (for G A Diaz): Guillermo diazffinoaa gov 



temperature, set duration, season, 

 and area (i.e., Grand Banks and U.S. 

 Atlantic east coast), the proportion of 

 blue shark released alive (P DA ) was 

 calculated. 



Only sharks explicitly recorded as 

 "discarded alive" or "discarded dead" 

 were used and only proportions de- 

 rived from at least 20 observations 

 (i.e., captured sharks) were analyzed. 

 The influence of the fish size, water 

 temperature, set duration, area, and 

 season (and all possible interactions) 

 on P DA was assessed by using the lin- 

 ear model 



The blue shark (Prionace glauca) is 

 an oceanic species that occurs in tem- 

 perate and tropical waters around 

 the globe (Robins and Ray. 1986). 

 This species is a major bycatch of 

 pelagic longline fleets that operate to 

 supply the world's growing demand 

 for tunas and swordfish (Xiphias gla- 

 dius) (Stevens, 1992; Bailey et al., 

 1996; Francis, 1998; Francis et al., 

 2001; Macias and de la Serna, 2002); 

 numerically, the blue shark is the top 

 nontarget species captured by the 

 U.S. longline pelagic Atlantic fleet 

 (Beerkircher et al. 1 ). 



Ward et al. (2004) examined the 

 effect of longline soak time (set du- 

 ration) on the catch rate of several 

 target and bycatch species, including 

 the blue shark. However, they did not 

 investigate the effects of fish size, 

 set duration, and water tempera- 

 ture on shark survival, and, there- 

 fore, numbers potentially available 

 for live release (Francis et al., 2001; 

 Campana et al. 2 ). Knowledge of such 

 relationships may be of value: 1) for 

 minimizing bycatch mortality on this 

 and other highly vulnerable pelagic 

 species through modification of fish- 

 ing strategy; and 2) for blue shark 

 stock assessments that are based on 

 commercial longline catch data. 



Materials and methods 



Data analyses were conducted on a 

 portion of the U.S. Atlantic Pelagic 

 Observers Program (POP) database. 

 The POP places trained observers 

 aboard commercial fishing vessels 



to record detailed information about 

 each fishing set, the catch and the 

 bycatch that would not otherwise 

 be collected. Recorded information 

 includes individual fish size (mea- 

 sured or estimated) and disposition 

 (alive or dead), surface water tem- 

 perature (°C) at gear deployment and 

 at haulback, and set location (lati- 

 tude and longitude). The duration of 

 each set (soak time, in hours) can 

 be obtained because time at start of 

 gear deployment and at end of gear 

 retrieval is also recorded. In the pres- 

 ent study, we restricted our analy- 

 ses to observed sets made from 1992 

 to 2002 by U.S. flag vessels north 

 of 35°N latitude (Fig. 1). This area 

 includes much of the U.S. exclusive 

 economic zone north of Chesapeake 

 Bay but also includes waters overly- 

 ing the Grand Banks. Data resulting 

 from experimental fishing conducted 

 from 2001 to 2004 over the Grand 

 Banks area (i.e., north of 35°N lat- 

 itude and west of 60°W longitude) 

 were not included because they did 

 not reflect typical fishing operations. 



For analysis purposes, blue shark 

 were placed in 25-cm fork length (FL) 

 size classes and water temperatures 

 (means) and set durations into 2 C 

 and 2-hour intervals, respectively. 



Size intervals were set at 25 cm 

 FL to increase the number of obser- 

 vations in each size category and to 

 reduce the bias that results from 

 estimating lengths versus actually 

 measuring them (e.g., observed in- 

 crease in the frequency of the esti- 

 mated lengths in 5- or 10-cm inter- 

 vals). For each combination of size, 



P, = /3 + ftT, + /SD, + /33S, 

 + IJ4 L,+ P5A,+ € r 



where P, = to the proportion of blue 

 shark discarded alive; 

 T = the temperature; 

 D = set duration; 

 S = season; 

 L = length; 

 A = set area, 

 € = the residual term of the 



i th observation; and 

 P - j35 are model parameters. 



Prior to regression, proportions were 

 arcsine-transformed according to the 

 methods of Sokal and Rohlf (1981). In 



1 Beerkircher, L. R., C. J. Brown, and D. 

 W. Lee. 2002. SEFSC pelagic obser- 

 ver program data summary for 1992- 

 2000. NOAA Tech. Memo. NMFS- 

 SEFSC-486, 23 p. Southeast Fisheries 

 Science Center, Miami, FL 33149. 



2 Campana S., P. Gonzalez, W. Joyce, and 

 L. Marks. 2002. Catch, bycatch and 

 landings of blue shark (Prionace glauca) 

 in the Canadian Atlantic. Canadian 

 Science Advisory Secretariat, Research 

 Document 2002/101, 41 p. Marine Fish 

 Division, Bedford Institute of Ocean- 

 ography. Dartmouth, Nova Scotia, B2Y 

 4A2, Canada. 



Contribution number SFD-2005-030 

 from the Sustainable Fisheries Divi- 

 sion, Southeast Fisheries Science Cen- 

 ter, NMFS, 75 Virginia Beach Drive, 

 Miami, FL 33149. 



Manuscript submitted 19 July 2004 

 to the Scientific Editor's Office. 

 Manuscript approved for publication 

 5 April 2005 by the Scientific Editor. 



Fish. Bull 103:720-724 (2005). 



