Abstract. — Videotapes of the 

 sea floor were taken from a sub- 

 mersible during dives at two areas 

 on the continental slope off Cape 

 Hatteras and Cape Lookout, North 

 Carolina, in September 1989. We 

 counted demersal nekton, epi- 

 fauna, and environmental features 

 for 1-minute intervals from video 

 transects. Common morphospecies 

 of demersal nekton were identified, 

 and multivariate analyses were 

 performed to find environmental 

 features that related to habitat use 

 by these forms. In both areas, the 

 ocean floor was extensively sculp- 

 tured with holes and mounds, and 

 both small and large sea anemones 

 were commonly observed. Crinoids 

 were seen in Cape Hatteras dives. 

 Small sea anemones were much 

 more abundant off Cape Hatteras, 

 whereas holes and mounds were 

 more densely distributed off Cape 

 Lookout. Rattails, hake, and serge- 

 stid shrimp were common at both 

 locations. Eels were extremely 

 abundant at the Cape Lookout site, 

 whereas eelpouts, flounder, and 

 lizardfish were found only at the 

 Cape Hatteras location. At both lo- 

 cations, analyses of nekton habitat 

 choices showed that habitat selec- 

 tion was related to density of the 

 holes and mounds made by infauna 

 and to density of the epifauna, such 

 as crinoids and the different types 

 of anemones. Hake, squid, serge- 

 stid shrimp, and lizardfish showed 

 the strongest evidence of habitat 

 selection. Analysis of videotapes, 

 originally recorded for other pur- 

 poses, is a cost-effective means for 

 preliminary examination of the 

 problems that may only be ad- 

 dressed by in situ observations. 



Assessing habitat use by nekton on 

 the continental slope using archived 

 videotapes from submersibles 



James D. Felley 



Office of Information Resource Management, Room 23 1 



A&l Building, Smithsonian Institution, Washington, DC. 20560 



Michael Vecchione 



National Marine Fisheries Service Systematics Laboratory 

 National Museum of Natural History, Washington, DC 20560 



Manuscript accepted 29 August 1994. 

 Fishery Bulletin 93:262-273 (1995). 



Understanding of the ecology of the 

 deep ocean floor has improved sub- 

 stantially since underwater cam- 

 eras have begun recording life at 

 depths beyond which divers may 

 penetrate. Recently, nekton commu- 

 nities on the shelf and slope have 

 been studied by means of underwa- 

 ter cameras carried by remotely 

 operated vehicles (ROVs) and occu- 

 pied submersibles. Such studies 

 have included analyses of environ- 

 mental features (Hecker, 1990b; 

 Levin et al., 1991), spatial distribu- 

 tion of individual species (Vecchione 

 and Gaston, 1986; Wenner and 

 Barans, 1990; Schneider and Haed- 

 rich, 1991), and patterns of habitat 

 use by species assemblages (Rich- 

 ards, 1986; Felley et al., 1989; 

 Auster et al., 1991; Carey et al., 

 1990). Underwater cameras have 

 allowed questions to be addressed 

 that are intractable to conventional 

 sampling methods (Haedrich and 

 Gagnon, 1991). Though problems 

 with accurate identification of spe- 

 cies and habitat variables are inher- 

 ent to these studies, such studies 

 open an important window to poorly 

 known ecosystems. 



We used archived videotapes re- 

 corded by the submersible Johnson 

 Sea-Link to investigate patterns of 

 habitat use by demersal and bentho- 

 pelagic nekton on the continental 

 slope off Cape Hatteras and Cape 

 Lookout, North Carolina. From the 



videotapes, we identified and counted 

 nekton species, and quantified se- 

 lected environmental variables dis- 

 cernible from video images. Using 

 these environmental variables, we 

 identified the habitat where each 

 species was most likely to be found. 

 We used factor analysis to identify 

 patterns of habitat use among the 

 species (Felley and Felley, 1987; 

 Felley et al., 1989) and to determine 

 which environmental variables 

 seemed most important in structur- 

 ing the nekton assemblage of the 

 continental slope off North Caro- 

 lina. We then compared distribu- 

 tional variances of environment and 

 species occurrence to identify those 

 species selecting subsets of avail- 

 able habitats. 



Materials and methods 



Video recording 



Video transects were recorded dur- 

 ing dives by the Johnson Sea-Link 

 II submersible from the RV Edwin 

 Link. Table 1 summarizes latitudes 

 and longitudes, dates, and dive 

 times. These data and videotapes 

 are from NOAA's National Under- 

 sea Research Center at the Univer- 

 sity of North Carolina at Wilming- 

 ton. Time starting and time ending 

 are the beginning and ending points 

 of the videotape section that we 



262 



