Polovma et al.: Application of satellite altimetry to simulate transport dynamics of Panulirus marginatus 



139 



between years and spawning seasons and that were 

 seen, for example, in the distribution of 1-year-old lar- 

 vae in July 1995 (Fig. 7). Larvae released from Mid- 

 way were advected eastward, larvae from Maro were 

 advected east and south, larvae from Necker were ad- 

 vected east and south, and larvae from Oahu were ad- 

 vected primarily southwest ( Fig. 7). A fairly persistent 

 meander at about 26°N concentrated larvae zonally. 

 whereas persistent westward flow at the southern end 

 of the archipelago advected Oahu larvae westward (Fig. 

 7 ). At the center of the archipelago, specifically at Maro 

 and Necker, a substantial portion of larvae remained 

 close to the bank where they were spawned, even afl;er 

 12 months, whereas at the ends of the archipelago, at 

 Midway and Oahu, advection was high (Fig. 7). Fur- 

 ther, although Necker receives abundant larvae 

 spawned at Maro, relatively few larvae from Necker 

 reached Maro (Fig. 7). The results, averaged over three 

 years, showed that the percentage of larvae released 

 at a bank and that were within 140 km of that bank 

 365 days later varied considerably by bank (Table 1). 

 Differences in this percentage between winter and sum- 

 mer spawning seasons were not generally large but the 

 interannual range of this percentage often varied more 

 than threefold (Table 1). Averaged over all three years 

 and over the two spawning seasons, the percent of lar- 

 vae within 140 km of a bank 365 days after being re- 

 leased at that bank was 5.5'7r, 9.6'7f , 16.7'7f , and 16.6% 

 for Midway, Maro, Necker, and Oahu, respectively 

 (Table 1). 



From the three years of simulations, no radical 

 changes from the bank-specific larval distribution 

 patterns shown in Figure 7 were seen, only spatial 

 extensions or contractions. For example, the distri- 

 bution of larvae 365 days after spawning at Necker 

 had a northeast and southwest orientation, and lar- 

 vae spawned in 1993 at Necker Island were advected 

 east and southwest to a greater extent than larvae 

 in 1994 or 1995 (Fig. 8). 



To examine the exchange of larvae between banks, 

 the recruitment index was calculated on the basis of 

 the bank at which the larvae originated, for both sum- 

 mer and winter spawning seasons, with 5000 larvae 

 released from each bank (Table 2). The results were 

 similar for both spawning seasons. Recruitment to 

 Midway came almost entirely from larvae spawned at 

 Midway. Recruitment to Maro was based largely on 

 larvae from Maro and Midway; only 10-21% of the re- 

 cruits at Maro came from Necker and Oahu. Necker 

 had the broadest geographic source of recruits, receiv- 

 ing substantial larvae from all four banks. Oahu's re- 

 cruits come from Maro, Necker, and Oahu (Table 2). 



Discussion 



The altimetry data from T-P provide a new tool to 

 investigate recruitment questions. However, this ap- 

 proach has several limitations. First, because the 

 geoid is not known, an estimate of mean dynamic 

 topography (Levitus) was added to the T-P sea sur- 

 face height anomalies. Although the accuracy of the 



