436 



Fishery Bulletin 104(3) 



the most productive gear configuration. For our study, 

 two buoy drop lengths were used in each set, and these 

 drop lengths were alternated after every 30 hooks: usu- 

 ally 5- and 2.5-fathom (ca. 9.1 and 4.5 m, respectively) 

 lengths in the fall and 10- and 12-fathom (ca. 18.3 and 

 21.9 m, respectively) lengths in the spring. Electronic 

 hook-timers (Lindgren-Pitman, Inc.; Pompano Beach, 

 FL) were also used during many of the sets to record the 

 time at which an animal was hooked. Bait was usually 

 frozen squid illlex sp.), but occasionally included frozen 

 Atlantic mackerel {Scomber scombrus) or a haphazard 

 mixture of the two. 



This project consisted of both a pilot and a main 

 study. The pilot study occurred off the east coast of 

 Florida during June 2002 and included deployments of 

 five PTT-100 tags (Microwave Telemetry, Inc.; Columbia, 

 MD) and one PAT (Wildlife Computers; Redmond, WA) 

 tag. The main study was conducted between August 

 2002 and August 2004 and for this study only PTT-100 

 HR model tags were used. 



Tag models 



The physical characteristics of all PSAT tag models 

 used in this study were similar and included a micro- 

 processor, a transmitter, and various environmen- 

 tal sensors, all contained within a resin-filled carbon 

 fiber tube. The tag is made positively buoyant by a 

 spherical glass-bead-embedded float at the base of the 

 antenna. It measures approximately 38 cm in length 

 by 4 cm diameter (including antenna) and weighs 

 between 65 and 75 g (air weight). Tags were rigged 

 with approximately 16 cm of 400-pound test Momoi" 

 brand (Momoi Fishing Co.; Ako City, Japan) mono- 

 filament line attached to a large hydroscopic nylon 

 intramuscular tag head according to the method of 

 Graves et al. (2002). The earlier model PTT-100 tags 

 were identical to those used by Graves et al. (2002) and 

 Kerstetter et al. (2003) and recorded one temperature 

 data point for every two-hour period during their five- 

 day (A! = 3l or 30-day (n=2) deployments, as well as a 

 pre- and postdeployment inclinometer value. The PAT 

 tag recorded environmental data every minute during 

 its 43-day deployment (programmed to disengage from 

 the fish on 30 July 2002) but transmitted data as sum- 

 mary histograms rather than discrete data points. The 

 PAT tag possessed emergency release software as well 

 as a mechanical device (RD-1500; Wildlife Computers, 

 Redmond, WA) for an early emergency release before 

 reaching a depth at which it would be crushed by ambi- 

 ent water pressure (crush depth). 



The Microwave Telemetry, Inc. model PTT-100 HR 

 satellite tag was used for the main study and consti- 

 tuted the majority of the PSAT deployments (;!=22). 

 This tag has similar physical attributes to those of 

 the model PTT-100 tags previously described, but its 

 functionality was increased by the addition of light 

 and pressure (depth) sensors and an increased data 

 storage capacity. The manufacturer preprogrammed all 

 the PTT-100 HR model tags to detach themselves from 



the fish after ten days, and the tags were activated 

 prior to attachment to the animal by removing a small 

 magnet from the side of the tag. The tags sampled 

 environmental data at approximately four-minute or 

 two-minute intervals. 



White marlin tagging procedures 



Preparations for tagging operations were made before 

 each haulback of the gear. Tags were either activated 

 prior to haulback or during haulback immediately follow- 

 ing the tagging of a fish and during preparation for tag- 

 ging another animal. Regardless of the time of external 

 tag activation, all PSATs were allowed to cycle through 

 their full ten-minute computerized internal activation 

 process prior to being attached to a fish. The captain of 

 the vessel identified incoming white marlin on the line 

 during the morning haulback of the gear and fish were 

 evaluated as live or dead based on movement (or lack 

 thereof) alongside the vessel. All live white marlin were 

 tagged, regardless of physical condition. 



Fish were manually brought alongside the vessel just 

 aft of the hauling station along the rail and held briefly 

 by the leader until calm. The average distance between 

 the top of the rail and the fish (free-board) on the FV 

 Carol Ann was approximately one meter, requiring the 

 use of a tagging pole of approximately 2 m length to 

 reach the fish over the gunwale. The nylon anchor to 

 the PSAT tether was carefully inserted about 5-10 cm 

 below the midpoint of the anterior dorsal fin to a depth 

 of about 5 cm. This location on the fish provides an 

 opportunity for the nylon tag head to pass through the 

 pterygiophore bones without approaching the coelemic 

 cavity (Prince et al., 2002a). For most white marlin in 

 this study (93'7f ), a conventional streamer tag was also 

 attached well posterior to the PSAT. 



White marlin were released as soon as possible after 

 tagging by the standard commercial protocol of cutting 

 the leader near the hook unless the hook was readily 

 accessible for manual removal. No animals were resus- 

 citated after tagging. Prior to release, hook type was 

 noted and fish lengths and weights were estimated. 

 Disposition ("live" vs. "dead") and hook location data 

 were collected for all white marlin caught in 2003 and 

 2004. For the purposes of this study, "internal" hook 

 locations were those in which the barb of the hook 

 was lodged posterior to the esophageal sphincter, and 

 "external" hook locations were noted with more speci- 

 ficity (e.g., "upper jaw"). Hooking on the body away 

 from the mouth ("foul hooking") was considered an 

 "external" hook location. In addition to noting hooking 

 location, a rapid visual examination of each fish was 

 conducted using the five-point "ACESS" scale of activ- 

 ity, color, eye condition, stomach status, and body state 

 (see Kerstetter et al., 2003). The tagging operation, 

 from positive species identification to actual release 

 from the gear, lasted less than 10 minutes. All data, 

 including the time of day, vessel location, and surface 

 water temperature, were recorded immediately after 

 tagging. 



