788 



Fishery Bulletin 92(4), 1994 



56' 



80° 55' 



54' 



53' 



—I — 

 52' 



51' 



I ABUNDANT GROWTH ASSOCIATED WITH EXPOSED ROCK 

 (DENSE LIVE BOTTOM AND LEDGE HABITAT) 



□ 



MODERATE GROWTH ASSOCIATED 

 WITH ROCK THINLY COVERED BY SAND 



SPARSE AND PATCHY GROWTH 

 ASSOCIATED WITH ROCK COVERED 

 BY UP TO 30 CM OF SANO 



BARE SAND, 

 ROCK ABSENT 





31* 25- 



FISH HAVEN 

 BUOY 



24' 



KILOMETERS 



23' 



22' 



Figure 1 



Sand, live-bottom, and ledge substrates at Gray's Reef National Ma- 

 rine Sanctuary (GRNMSl (after Hunt, 1974). 



Techniques include the use of traditional fisheries 

 assessment gear (nets, traps, and hook-and-line), 

 poisons, explosives, and visual observations. The 

 need for repeatable surveys and the constraints of 

 working in a national marine sanctuary necessitated 

 the use of nondestructive survey techniques. 



Diver observations are the most common method 

 used in studies of reef fish assemblages (Brock, 1954; 

 Bardach, 1959; Hobson, 1972; Chave and Eckert, 

 1974; Sale, 1975; Jones and Chase, 1975; Jones and 

 Thompson, 1978; Anderson et al., 1981; Ogden and 

 Ebersole, 1981; Sale and Douglas, 1981; Brock, 1982; 

 Kimmel, 1985; Bohnsack and Bannerot, 1986; 

 Parker, 1990). Although a variety of sampling tech- 

 niques have been employed to make quantitative 

 assessments of reef fish abundance, all rely on divers 

 to identify and record fish species observed in a pre- 

 defined area (transect and point counts) or over a 

 period of time (rapid visual assessment techniques). 

 Accuracy of the visual techniques is affected by light 

 levels, water clarity, currents, fish species diversity 

 and densities, substrate complexity, diver familiar- 

 ity with the fishes, and number and size of the sam- 

 pling units. Biases are induced by 1) a tendency to 

 undersample small, cryptic, and nocturnal species 

 (Brock. 1982); 2) identification, counting, and record- 

 ing errors (Brock, 1954; Russell et al., 1978); 3) at- 

 traction and aversion reactions of some species to 



the divers (Chapman et al., 1974); and 4) species dif- 

 ferences in territory, home range, life history pat- 

 terns, and behavior (Russell et al., 1978). 



Remote observation techniques, using movie or 

 closed-circuit television cameras deployed from ves- 

 sels or carried by divers, have been used to estimate 

 abundance of reef fish (Smith and Tyler, 1973; 

 Alevizon and Brooks, 1975; Powles and Barans, 1980; 

 Boland et al. 2 ). A major advantage is that a perma- 

 nent record of observed fishes is obtained without 

 destroying the fauna. However, remote systems that 

 are tethered to a surface vessel are severely limited 

 by sea conditions. In addition, camera resolution, 

 light levels, water clarity, depth, and lack of in situ 

 guidance limit the effectiveness of remote observa- 

 tions, and biases are imposed by the attraction or 

 avoidance of some species to the gear. 



The objectives of this study were 1 ) to develop a 

 nondestructive, repeatable procedure for assessing 

 diurnally active fishes inhabiting Gray's Reef Na- 

 tional Marine Sanctuary, and 2 1 to describe and com- 

 pare fish communities associated with ledge, live- 

 bottom, and sand habitats. 



- Boland, G., B. Galloway, J. Baker, and G. S. Lcwbel. 1984. Eco- 

 logical effects of energy development on reef fish of the Flower 

 Garden Banks. Final Rep. Contract No. Na80-GA-C-00057. U.S. 

 Natl. Mar. Fish. Serv.. Galveston, TX, 466 p. 



