NOTE Graves et a\ An evaluation of satellite tags for estimating postrelease survival of Makaim nigricans 139 



c 



tag number 24519 



10 20 30 40 50 60 70 80 90 100 110 12 



B tag number 24059 



10 20 30 40 50 60 70 80 90 100 110 

 tag number 24522 



50 60 70 



Time (hours) 



Figure 3 



Temperature and inferred depth for eight of nine blue marlin equipped with 

 pop-up satellite tags near Bermuda. 25 July-11 August 1999. See text for 

 explanation of inferred depth and Table 1 for release information correspond- 

 ing to each tag number Hours of darkness are shaded on the time line. 



The blue marlin whose tag did not report (tag no. 24040. 

 Table 1 ) was hooked in the jaw, caught in less than 20 min- 

 utes, did not require resuscitation, was quickly tagged, and 

 actively swam away from the boat when released. Shark 

 predation on released billfish has been reported (Holland 

 et al., 1990; Pepperell and Davis, 1999); therefore mortal- 

 ity cannot be excluded despite the apparent vigorous con- 

 dition of the fish. Failures in component subsystems could 

 account for the failure of reporting from a pop-up tag. A 

 detailed analysis of the reliability of each tag component 

 could be undertaken, but several factors external to the 

 tag could also result in a failure of reporting. Tag man- 

 ufacturer innovations and upgrades of the systems will 

 allow researchers to better identify mortalities, but they 

 will not completely solve the problem of discriminating 

 between tag failure and fish mortality. Nonreporting tags 

 would have significant consequences for efforts to make 



ocean-wide estimates of postrelease sui-vival. The ability 

 to account for all pop-up satellite tags deployed is directly 

 related to the accuracy of the resulting estimates of postre- 

 lease survival (Goodyear, in press). Nonreporting satellite 

 tags introduce uncertainty that cannot be quantified in 

 the estimates of postrelease survival, thus compromising 

 meaningful conclusions. Excluding nonreporting tags from 

 the analysis decreases precision of the estimate, and in- 

 cluding mortalities biases the survival estimate down- 

 ward. Further, any extension of the 5-day pop-up period 

 to allow study of possible delayed effects of tagging should 

 involve careful consideration of the benefits and the li- 

 abilities that longer durations might have on estimating 

 postrelease survival (Goodyear, in press) 



Successful tagging and reporting of pop-up tags from 

 four fish under 200 lb (90.9 kg) indicate that the size and 

 design of the PTT-100 tag is tolerated by smaller blue mar- 



