64 



Fishery Bulletin 103(1) 



over much longer time scales (sometimes years) and 

 across ocean-basin geographic scales (Arnold and Dew- 

 ar, 2001; Gunn and Block, 2001). These tags can provide 

 information on both seasonal behavior and migration 

 routes. Although data collection is fishery-independent, 

 data retrieval is dependent on the recapture of the fish 

 by fishermen and on the recognition and return of the 

 archival tag. PSATs are a merger of archival and satel- 

 lite telemetry technology. Because PSATs are attached 

 externally, only environmental data can be collected. 

 The tags can be programmed to gather data for a prede- 

 termined duration and then to disengage and transmit 

 data at a determined time. The major advantage of 

 this tag is that both data acquisition and retrieval are 

 fishery-independent and the researcher knows when to 

 expect to receive data. However, data retrieval is lim- 

 ited by data compression required to compensate for low 

 data transfer rates to the Argos satellites, finite battery 

 life, and relatively high transmission errors (Arnold and 

 Dewar, 2001; Gunn and Block, 2001). PSATs provide 

 accurate endpoint locations based on Doppler shifts of 

 successive transmissions during a single satellite pass. 

 However, geolocation throughout the tagging duration 

 is based on light levels that estimate dawn and dusk. 

 By determining time of local noon and day length, lon- 

 gitude and latitude can be calculated. According to Hill 

 and Braun (2001), even with optimal geolocation analy- 

 sis, the expected variability in longitude is a constant 

 0.32° but the expected variability in latitude will never 

 be less than 0.7°. The relationship between day length 

 and latitude is strongest at high latitudes and at the 

 time of the solstices but weakens near the equator and 

 becomes nearly indeterminate at the equinoxes (Sibert 

 et al., 2003). 



An implicit assumption in using these tags is that 

 while the fish tows the tag, the tag does not affect the 

 study animal's behavior or survival. This is a reason- 

 able assumption for large pelagic fishes and is sup- 

 ported by theoretical estimates of the energetic cost of 

 towing a PSAT (Kerstetter, 2002); however, the actual 

 energy cost to a given fish has not previously been 

 quantified. The success of early studies on pelagic fishes 

 has spurred increasing interest in using these tags on a 

 large variety of species and age groups. As studies are 

 undertaken with PSATs, a logical extension is to pose 

 the question: "At what point does the energy cost of car- 

 rying a PSAT negatively affect a study animal?" Blay- 

 lock (1990) addressed a similar question regarding the 

 impact of sonic transmitters on the swimming behavior 

 of cownose rays (Rhinoptera bonasus). In his study, he 

 videotaped cownose rays for ten-minute intervals before 

 and after attachment of a mock transmitter. Energy 

 expenditure was estimated by counting wingbeats per 

 second before and after attachment of the transmitter. 

 He concluded that in the short term a transmitter-to- 

 ray mass ratio of less than 0.03 had no statistically 

 significant effect on ray swimming behavior. 



In this study, the impact of a PSAT on a study ani- 

 mal is evaluated in terms of the forces that the PSAT 

 exerts on the animal, specifically lift (i.e., buoyancy) 



and drag. Lift and drag are both vector quantities; lift 

 acts in the vertical direction and drag, as measured 

 in this study, acts in the horizontal direction. These 

 vector components are additive to give the total force 

 acting on the attachment site of a PSAT. At a recent 

 tagging workshop associated with the Pelagic Fisheries 

 Research Program, 1 the problem of premature release 

 of some PSATs from the research animal was cited 

 as a common difficulty. Premature release may be at- 

 tributed to a number of potential failures of either the 

 tag itself or the attachment device. Possible sources 

 for this problem cited at this workshop include detach- 

 ment of the anchor from the study animal, failure of 

 the tether between the PSAT and the anchor, failure of 

 the release pin on the PSAT, and failure of the release 

 software itself. The magnitude of the total force acting 

 on the attachment site chronically may provide some 

 insight into whether anchor failure is a possible source 

 for this problem. 



Drag as an isolated force is the product of four defin- 

 ing factors: 



F D = VzpS IflCn, 



(1) 



where F D = force due to drag (in newtons, N); 



p = density (kg/m 3 ) of the fluid through which 



the object is moving; 

 S = projected surface area (m 2 ) of the object; 

 U = relative velocity (m/s) between the object 

 and the fluid; and 

 C D = drag coefficient (dimensionless) which is 

 largely dependent upon the shape of the 

 object. 



Furthermore, the power required to pull the tag through 

 the water can also be related to drag mathematically: 



F D U 



V 2 pS U 3 C D , 



(2) 



where P = power (in watts, W). 



Of particular note in these relationships, drag is pro- 

 portional to velocity squared and power is related to ve- 

 locity cubed provided that all other factors are constant. 

 For example, as velocity is doubled, drag increases by 

 a factor of four, whereas power increases by a factor of 

 eight. The characteristic of the tag that most affects 

 drag in this relationship is its projected surface area 

 which, in turn, is defined by its size and shape. The 

 projected surface area of the PSAT changes as the tag 

 is pulled through the water at different velocities and 

 in turn changes the drag coefficient at each velocity. 

 On the other hand, lift is determined by the buoyancy 

 of the tag. The dry weight of the tag is not a factor 

 in either of these relationships under steady flow con- 



1 Pelagic Fisheries Research Program. 2002. PFRP PI Meet- 

 ing, December 4-6, 2002. University of Hawaii at Manoa, 

 1000 Pope Rd., MSB 312, Honolulu, HI 96822. http://www. 

 soest.hawaii.edu/PFRP/meetings.html. (Accessed 16 June 

 2004.] 



