Kerstelter and Graves Survival of Tetrapturus albidus afler release from longline gear 



439 



16 (72.7%) of the 22 PTT-100 HR tags transmitted data 

 to satellites in the ARGOS system as programmed, 

 and an average of 51% (range 4.4-86.1%) of each tag's 

 archived data were transmitted. Two PTT-100 HR tags 

 were found on shore after their transmission period 

 and were returned to us and all the archived data were 

 recovered from each tag. 



White marlin survival 



Transmitted temperature and depth data from 17 of 

 the 19 functional tags (89.5%) indicated that released 

 white marlin survived for the time periods over which 

 the tags were programmed to collect data. Of the two 

 confirmed mortalities in this study, one fish (GB-02- 

 01) died within one hour after release and sank to 

 the bottom at 145 meters depth. It remained there for 

 approximately 10 hours before the tag and presum- 

 ably the carcass were scavenged by a shark based on 

 an abrupt change in behavior (depth distribution and 

 movements) and light level (see Kerstetter et al., 2004). 

 The second mortality (MA-03-04) occurred approxi- 

 mately 24 hours following release. After tagging, the 

 animal remained between and 26.9 meters depth 

 before it too was inferred to be the victim of a shark 

 predation event based on similar changes in environ- 

 mental parameters. 



The net displacement of all reporting tags was used 

 as an additional line of evidence to assess postrelease 

 survival of white marlin. All of the tags from puta- 

 tively surviving animals demonstrated net movements 

 that cannot be explained by surface currents alone. 

 For the 14 surviving fish with PTT-100 HR tags, the 

 average minimum straight-line movement was 246.2 

 nautical miles (nmi) over the ten-day period, but there 

 was a wide range of net displacement among individuals 

 (80.4-631.5 nmi). Eight of the nine white marlin tagged 

 approximately 350 miles east of Ocean City, Maryland, 

 in summer 2004 moved generally east to northeast, 

 with the exception of one animal that traveled 304.9 

 nmi to the northwest. 



All but one of the tags employed in this study lacked 

 hardware or software that would cause the tag to sepa- 

 rate from the fish prematurely if a moribund fish de- 

 scended below a critical depth. Consequently, nontrans- 

 mitting tags could result from an animal that died and 

 sank in waters deeper than the pressure capacity of the 

 tags. All eight white marlin tagged with PSATs that did 

 not transmit data were released in or near areas with 

 depths in excess of 2000 meters, the manufacturer's 

 suggested pressure limit for the tags. 



The tags that did not transmit data may or may not 

 represented mortalities of the tagged white marlin. 

 These resulting calculated mortality rates therefore 

 depend on whether or not the tags that did not transmit 

 data are included as evidence of mortality. Combining 

 both hook types, the overall mortality rate was 10.5% 

 (95% CI: 0.0-26.3%) if nontransmitting tags were ex- 

 cluded and 37.0% (95% CI: 18.5-55.6%) if nontransmit- 

 ting tags were included as mortalities. 



Hook performance 



Two general hook types, circle and J-style, were used by 

 the crew of the longline vessel in this study. Nineteen 

 white marlin tagged with PSATs were caught on circle 

 hooks, two of which (10.5%) were lodged internally and 

 17 of which (89.5%) were lodged externally in the jaw 

 or mouth (Fig. 1). Neither of the two PSATs on animals 

 hooked internally with circle hooks transmitted data. 

 Two PSATs attached to the 17 fish caught with circle 

 hooks lodged externally failed to transmit data, and only 

 one fish caught with a circle hook lodged externally was 

 a confirmed mortality. Nine white marlin tagged with 

 PSATs were caught by J-style hooks. Two fish caught 

 with J-style hooks were hooked internally (22.2%) and 

 seven externally (77.8%). Of the two hooked internally, 

 one tag did not transmit data and the other (fish GB-02- 

 01) was a confirmed mortality. Three of the remaining 

 seven tags on fish caught externally with J-style hooks 

 did not transmit data. Comparisons of hook type and 

 postrelease survival were not significant (Fisher's exact; 

 P>0.16). For the 10° offset J-style hooks, the mortality 

 rate was 20.0% excluding nontransmitting tags, and 

 55.6% if nontransmitting tags were included as mor- 

 talities. The 0° offset circle hooks had a 7.1% mortality 

 rate if nontransmitting tags were excluded and 27.7% if 



