140 



Fishery Bulletin 97(1), 1999 



35"N 



30'N 



25'N 



20'N 



15"N 



10'N 



165"E 170"E 175"E 180" 175"W 170'W 165'W 160"W 155'W 150"W 145"W 



35"N 



30'N 



25'N 



20'N 



15'N 



10'N 



165'E 170'E 175'E 180' 175'W 170'W 165'W 160"W 155'W 150'W 145"W 



Figure 7 



Simulated spatial distribution of 5000 larvae 365 days after release on 1 July 1994 

 at lA) Midway, iBi Maro, (C) Necker, and iDi Oahu, with an eddy diffusion rate of 

 500 m'/sec. The star denotes the bank from which larvae were released. Solid circles 

 mark the other three banks. 



Levitus mean varied depending on the spatial and 

 temporal distribution of the underlying data, it was 

 likely that there were biases that may have had an 

 impact on the accuracy of the estimates of absolute 

 geostrophic current. A second limitation is that the 

 spacing between data tracks, especially in mid- and 

 low-latitude regions, may have been too broad to com- 

 pletely resolve many important fmescale and mesos- 

 cale physical features. Further, this tool is appropri- 

 ate only in situations where geostrophic transport is 

 the most significant source of larval transport, situ- 



ations, for example, where larvae are below the shal- 

 low Ekman layer. 



In our application, this approach leads to a new hy- 

 pothesis on the spatial dynamics of lobster larval re- 

 cruitment in the Hawaiian Archipelago. From the simu- 

 lations, we found that position of a spawning bank 

 within the archipelago influences both the number of 

 larvae spawned at a bank that will recruit to that bank 

 as well as the recruitment that a bank will receive from 

 other banks. Specifically, the results indicated that lar- 

 vae are transported down the ridge from the north- 



