McGOWAN and RICHARDS: BLUEFIN TUNA LARVAE 



(Olson et al. 1984). Gyres and slow recirculation 

 in this area could retain fish larvae for several 

 days. This means that the evidence presented 

 here does not eliminate the possibility that blue- 

 fin tuna spawn off the southeastern U.S. How- 

 ever, because larvae collected in rapidly moving 

 water were probably transported in that water, 

 we argue that the data support the contention 

 that the primary spawning area is farther up- 

 stream with perhaps some late spawning by a 

 few individuals in the Straits of Florida as sug- 

 gested by Rivas (1954). 



Oceanic Habitat of Bluefin Tuna Larvae 



Four aspects of larval fish habitat are impor- 

 tant to the survival of individuals and recniit- 

 ment to the adult stock: thermal and salinity 

 conditions, prey, predators, and patterns of 

 ocean circulation that can retain the larvae in 

 favorable areas. Our data do not permit discuss- 

 ing the predators coincident with the bluefin 

 tuna larvae found off the southeastern U.S., but 

 we can discuss the sahnity and temperatures, 

 the food potentially available, and the hkelihood 

 of retention in a favorable area. 



Surface salinity over the shelf of the south- 

 eastern U.S. is generally 35 ppt or less, with 

 peak river runoff in spring affecting the central 

 and inner shelf (Atkinson and Menzel 1985). Sa- 

 linities over the outer shelf are similar to those in 

 the Gulf Stream (35.0-36.5 ppt; Stommel 1965). 

 Salinity lower than full-strength seawater could 

 be potentially detrimental to larval bluefin tuna 

 although the adults tolerate reduced salinities 

 (Topp and Hoff 1971). All the larvae collected in 

 this study were found within a narrow range of 

 salinity near 36 ppt (Fig. 8). The larvae were 

 also found in a fairly narrow range of tempera- 

 tures near 26°C (Fig. 8). In the Gulf of Mexico in 

 1984 and 1986, bluefin tuna larvae were found 

 where sea surface temperature (SST) ranged 

 from 22.0° to 28.1°C. More than 87% of the lar- 

 vae occurred in a narrow range of temperatures 

 between 24.0° and 26.1°C (Southeast Fisheries 

 Center, National Marine Fisheries Service, 

 unpubl. data). This is similar to the temperature 

 range off the southeastern U.S. where bluefin 

 tuna larvae were found at SST's from 24.7° to 

 26.5°C However, the mean temperature of oc- 

 currence of the bluefin tuna larvae was higher 

 here than in the Gulf, 25.72° vs. 24.99°C it = 

 2.98; df = 50; P < 0.005). At higher temperature 

 metabolic requirements of the larvae would be 

 higher; larvae would require more food for 



optimal growth and survival, other conditions 

 being equal. 



A potential mechanism for producing larval 

 fish food does e.xist in this region. Onshore 

 meanders of the Gulf Stream along the south- 

 eastern U.S. can cause upwelhng of nutrient rich 

 water along the shelf edge (Yoder et al. 1981; 

 Yoder 1983). This and the compression of iso- 

 therms near the edge of the Gulf Stream might 

 produce a stable stratified region favorable to 

 the growth and to the persistence of patches of 

 larval fish food (Lasker 1981). It is true that 

 intrusions of cold, upwelled water provide pulses 

 of phytoplankton production on the shelf which 

 initiate the formation of patches of zooplankton 

 (Paffenhofer et al. 1987). However, these iso- 

 lated patches are most often produced in July, 

 when winds as well as currents are favorable for 

 upwelling. Furthermore, the zooplankton in the 

 patches consist primarily of small species of 

 copepod and gelatinous salps and doliolids which 

 are most abundant in cool water near the bottom 

 and in the thermocUne. Small copepods are not 

 ideal food for larval bluefin which eat other lar- 

 val fishes. Larval fishes were not noticeably 

 abundant in the patches; however, the sampling 

 gear used by Paffenhofer et al. (1987) was not 

 optimal to catch fish larvae. The gelatinous zoo- 

 plankters which can be predators of fish larvae 

 pose a potential hazard to the tuna larvae. 

 Therefore the patches of plankton on the shelf 

 caused by onshore meanders of the Gulf Stream 

 do not appear to be favorable habitat for the 

 feeding or survival of bluefin tuna larvae. These 

 isolated patches on the shelf probably benefit 

 benthic filter-feeders more than larval fishes. 



Not all meanders that cause upwelling may 

 result in isolated patches on the shelf. A pulse of 

 upwelled, nutrient rich shelf-break water could 

 move offshore, be entrained in the Gulf Stream, 

 and increase the local productivity of near-sur- 

 face water, thus enhancing the offshore habitat 

 for larval fishes. Longhurst (1983) suggested 

 that surplus production occurs on continental 

 shelves. Walsh et al. (1987) detected export of 

 phytoplankton from the Mid- Atlantic Bight dur- 

 ing a spring plankton bloom. Sherman et al. 

 (1984) found that peak spawning for some 

 species is related to topogi-aphic features and 

 circulation, and is synchronized with production 

 of the copepod prey of their larvae. Something 

 about the two stations with bluefin tuna larvae 

 off North Carolina in this study (Figs. 6, 7) may 

 have resembled "good" spawning habitat enough 

 to induce migrating adult bluefin tuna to spawn 



627 



