258 



Fishery Bulletin 104(2) 



81 OW 



32 ON - 



31 ON 



80'0'W 

 -^-l- 



South Carolina 



=^^^ 





I Savannah 

 (4-? 



25 



50 100 



Kilometers 



Georgia 



f) 



''"^'V o O o o 



' o o oj^' 



D 



J 



o o o O 



jj^^curiswick 



20 m 



• 3 



D 

 •5, 



D 



a 



^7 



30m 



50 m 



200m 



31 W 



1 



80 OW 



80 OW 



32"0'N 



- 31 "ON 



Figure 1 



Maps of the study area showing cross-shelf beam trawl stations (black dots) 

 at which juvenile fish and environmental sampling occurred quarterly from 

 April 2000-February 2002 (see Table li. Additional stations were sampled 

 inshore (circles) and offshore (squares) of Gray's Reef NMS isee Table li (A). 

 Station groups used in correspondence analysis of the inshore (Bl and offshore 

 (C) juvenile fish data sets are separated by solid lines and labeled. 



stations were sampled along three transects inshore 

 of Gray's Reef NMS and along four offshore transects 

 during three cruises (Fig. 1 and Table 1). The benthic 

 macrohabitat of the cross-shelf stations was determined 

 by using a remotely operated vehicle (ROV, Phantom 

 S2). The ROV was deployed off the starboard side of the 

 ship. Approximately 30 m of the RO'V tether was let out 

 before the tether was attached to a 45-kg weight. The 

 weight was lowered to 3 m above the bottom and the 

 ship was allowed to drift. During April 2000, two 15-min 

 drifts were made at eight of the ten cross-shelf stations. 

 Because of inclement weather, no drifts were conducted 

 at the inner- or outer-most stations. Video (HI-8 mm) 

 was recorded of each drift. 



Concurrent fish and hydrographic measurements were 

 taken to examine the relationship between juvenile 

 fish assemblages and the environment. At each sta- 



tion, temperature, salinity, density, and water depth 

 were measured from the waters surface to one meter 

 above the bottom with a Seabird CTD (SBE19, Seabird 

 Electronics, Inc., Bellevue, WA). Juvenile fish were col- 

 lected at each station by using a 2-m beam trawl with 

 a 6-mm mesh body and a 3-mm mesh tail bag. Two 

 kilogram weights were added to each skid of a standard 

 2-m beam trawl (Kuipers, 1975) to ensure that the trawl 

 stayed on the bottom. Three 5-min bottom tows were 

 made at each station. Samples were sorted on deck and 

 fish and invertebrates were preserved in 95% ethanol. 



Preparation of fish data 



All beam trawl samples were sorted and fish were iden- 

 tified to the lowest possible taxonomic level by using 

 previously published descriptions (e.g.. Able and Fahay, 



