Starr et al : Movements of Sebastes paudsplnis and S chlorostictus in Monterey submarine canyon 



325 



Few studies have described the movements of 

 commeirially caught rockfishes in water dee|)er 

 than 100 m. however, because of the difficulty 

 in achieving high sui"vival rates for fish tagged 

 at those depths. Most rockfishes have pliysoclis- 

 tous swim bladders that expand with the reduced 

 pressure as these fish are brought to the sur- 

 face. The resulting barotrauma causes death for 

 almost all fish when captured from waters deep- 

 er than 20-30 m, rendering traditional tagging 

 techniques ineffective. 



We developed techniques to surgically implant 

 sonic transmitters at depth, thus reducing tag- 

 ging mortality (Starr et al., 2000). These tech- 

 niques enabled us to estimate the movements of 

 two rockfishes with different life history charac- 

 teristics. In 1997, we placed sonic transmitters in 

 six greenspotted (S. chlorostictus) rockfish, a spe- 

 cies presumed to be relatively sedentary (Yoklav- 

 ich et al., 2000). In 1998. we tagged 16 bocaccio, a 

 more mobile species (Love, 1996). In both years, 

 we tracked the horizontal and vertical move- 

 ments of tagged fish for a three-month period. 



Materials and methods 



Field procedures 



Our study site was located on the flank of the sub- 

 merged Soquel Canyon in 100-250 m of water, 

 approximately 20 km off shore in Monterey Bay, 

 California (Fig. 1). Soquel Canyon contains steep 

 sediment slopes and rock walls interspersed with 

 50-100 m wide benches, comprising soft sediment and 

 rock outcrops (Yoklavich et al., 2000). Several of the rock 

 outcrops located at the rim of the canyon are 10-20 ni high 

 by 50-100 m long scarps with boulders at the bases of the 

 linear rock walls. 



In 1997, we caught greenspotted rockfish using long- 

 line fishing gear deployed from a research vessel. In 1998, 

 we hired a commercial fisherman to catch bocaccio using 

 modified trolling gear In both years, fishing lines were 

 retrieved at about 20 m/min. Fishes were brought to a 

 depth of about 20 m and held there for tagging. Divers 

 then surgically implanted Vemco V16 (Vemco Ltd., Nova 

 Scotia, Canada) sonic tags into the captured fish. Follow- 

 ing surgery, tagged fish were placed in a recovery-release 

 cage, then towed to the differential GPS location at which 

 they were caught, whereupon the cage was lowered to the 

 seafloor and fish were released. We used the Delta sub- 

 mersible to verify that tagged fish were alive after release. 

 A more complete description of tagging and underwater 

 tracking procedures is provided in Starr et al. (2000). 



Vemco VR-20 receivers were moored on the seafloor for 

 the duration of the study as a means of tracking tagged 

 fish. The positions of receiver deployment spanned the dis- 

 tribution of release locations of the tagged fish. To increase 

 the amount of positional information available from the 

 receiver data, we placed the tagged fish and receivers 



Figure 1 



Fish-release locations and tag numbers, receiver locations, current meter 

 location, expected signal detection range, and resulting receiving zones for 

 the 1997 study of greenspotted rockfish. 



along the side of a submarine canyon that stretched north- 

 east to southwest. On 7 October 1997, we placed three re- 

 ceivers on a ledge at a depth of about 160 m along the wall 

 of the canyon (Fig. 1). The receivers spanned a distance 

 of 1500 m and recorded signals for 6 minutes out of every 

 half-hour We retrieved them on 5 January 1998. 



In 1998, we deployed two receivers on surface buoys 

 from 17 August through 10 September because tagging 

 operations commenced one month before the submersible 

 was scheduled to deploy the underwater moorings. These 

 receivers were placed near the locations at which fish were 

 released, in the northeastern portion of the Soquel Canyon 

 study area (site 1, Fig. 2) and in the southwestern portion 

 of the study area (site 2, Fig. 2). On 16 September 1998, we 

 placed six receivers at various depths about 1000 m apart 

 (Fig. 2) and retrieved them on 30 December 1998. Two of 

 the receivers were located in about 100 m of water on the 

 relatively flat substrate above the canyon rim. The other 

 four receivers were placed in deeper water on flat benches 

 or gently sloping walls below the canyon rim. These receiv- 

 ers recorded signals for 12 minutes every hour In both 

 years, battery life of the tags exceeded the length of time 

 the receivers were moored. 



Signal detection range of the moored receivers was 

 about 800 m (Starr et al. 2000). Thus, if a signal was re- 

 corded at a particular time by only one receiver, we knew 



