POWER and McCLEAVE: SIMULATION OF THE NORTH ATLANTIC DRIFT OFANGUILLA 



second phase in the simulated drift migrations. 

 Regardless of starting location, most larvae con- 

 verged upon the area and became part of this patch, 

 except those that entered the Florida Current by the 

 route south of the Bahamas. Gunn and Watts (1982) 

 found evidence in the January-February 1973 data of 

 a large anticyclonic eddy in the region of this patch 

 formation (cf their figure 3D). If this eddy is a perma- 

 nent or seasonal feature, it has substantial im- 

 plications for American eel leptocephalus drift. It 

 could be reasonably assumed that leptocephali 

 collect in the eddy each year and that this patch 

 phase is an integral part in the transport of most 

 American eel leptocephali. 



Transport of larvae from the patch and into the Gulf 

 Stream seems to have been a result of turbulent dif- 

 fusion, rather than advection, since currents east of 

 the Florida Current are weak (Fig. 2). In the model, 

 entry into the Gulf Stream was an example of large- 

 scale shear induced diffusion, where large dif- 

 ferences in adjoining northward current velocities 

 resulted in concentration gradients down which a dif- 

 fusive flux occurred. Whether this is the phe- 

 nomenon that facilitates actual entrainment of 

 leptocephali into the Gulf Stream is problematic. 

 Alternatively, Gulf Stream cold- core rings occur in 

 this area (The Ring Group 1981), and it may be that 

 Gulf Stream eddies and meanders act to capture 

 larvae. 



The leptocephalus collection data are still inade- 

 quate to confirm the presence of a leptocephalus 

 patch offshore of Florida and the Gulf Stream. 

 However, there is empirical evidence that this is in- 

 deed the region where most leptocephali enter the 

 Gulf Stream system. Kleckner and McCleave (1982) 

 studied near synoptic collections of leptocephali 

 taken on four transects of the Florida Current and 

 Gulf Stream between 26 July and 16 August 1978. 

 There were substantially higher concentrations of 

 leptocephali in the waters sampled on the northern 

 transects than there were in the southern ones (Fig. 

 10), indicating a significant input of leptocephali 

 from the western Sargasso Sea. Combined with the 

 present work, the studies of Kleckner and McCleave 

 (1982) and Gunn and Watts (1982) indicate that pro- 

 cesses important to the leptocephalus drift migration 

 occur east of Florida and that further research should 

 concentrate on this region. 



The prolonged existence of the leptocephalus patch 

 offshore of Florida and the Gulf Stream and the con- 

 tinuous capture of larvae by the Gulf Stream resulted 

 in the broad and surprisingly uniform distribution of 

 leptocephali along the North American coast. This 

 transport along the North American coast formed the 

 third phase of the drift migration. The patch can be 

 viewed as a mechanism causing a more uniform dis- 

 tribution of leptocephali than would otherwise occur. 

 Before developing the simulation model it had been 



35- 



30- 



J I I I 1 U„ 



J I L 



25- 



20- 



15- 



i — i — i — rVr 1 ! — i — r 



80 



75 



10- 



30 

 Latitude 



Longitude 



FIGURE 10.— A: Station positions on transects where American eel leptocephali were taken in the Gulf Stream system. B: Catches of lep- 

 tocephali on the transects. Data from Kleckner and McCleave (1982). 



497 



