reason for increased density is the complex topography located at the canyonheads and along the shear walls. In conjunction with increased niche space because of topographic complexity, increased attachment substrate in the form of consolidated sediment or rock scarps allows sessile invertebrates to colonize these areas, thereby increasing the available niches for other fauna which may associate with these colonies. It also has been postulated that submarine canyons may act like terrestrial watersheds, which concentrate water and waterborne materials to a main river channel. The canyons would concentrate the fine sediments and dissolved and particulate nutrients--such as particulate organic carbon (POC)--which flow off the shelf in the main axes of the canyons. The increased nutrient input would allow higher densities of organisms (primarily filter and deposit feeders, but also any associated organisms such as their predators). Earlier studies of submarine canyons were limited in their ability to define faunistic differences in species or abundance because of inherent deficiencies in their method of sampling and sampling gear. Most early studies used remote sampling with benthic grabs, small biological trawls, anchor dredges, and otter trawls--all of which operate with maximum efficiency over relatively smooth bottom. When these sampling devices encounter uneven or rugged topography the sampling becomes sporatic and non-quantitative, and the gear may be damaged or lost. More recent studies have used manned submersibles, camera sleds, and video tape transects to view the more uneven areas of the slope. But even these in situ methods of faunal analysis contain sampling errors. Visual transect methods are useful primarily for macrobenthic and megabenthic epifauna and cannot be used for infaunal species. Additionally, because the specimens are not collected, or only voucher specimens are collected, mis-identification of morphometrically similar species can occur (such as Nezumia bairdii and N. aequalis). An early study of the Hatteras Canyon system used bottom photography to study faunal zones (Rowe, 1971). The author proposed that submarine canyons have unique assemblages which disrupt the horizontal bands of faunal zonation along the continental shelf. Rowe (1971) further suggested that the presence of, or decreased abundance of, certain species could be designated as canyon indicators. He also stated that the presence of the suspension feeders Ceriantheomorphe braziliensis and Kophobelemnon stelliferum at the head of Hatteras Canyon indicated elevated levels of suspended particulates flowing down-canyon. Haedrich et al. (1975) used a trawl survey to delineate faunal zonation on the Middle Atlantic Slope south of New England. They concluded that the small Alvin Canyon did not contain a unique fauna. However, the presence of certain species such as the echinoderm Amphilimna olivacea, or the absence of other species such as the polychaete Hyalinoecia artifex, could be considered as canyon indicators." It should be noted that only seven samples were obtained in the canyon complex and, because of the limitations of using a trawl as the sampling gear, the more rugged head of the canyon was not sampled. Hecker et al. (1980) conducted photographic surveys of two north Atlantic and one middle Atlantic (Baltimore) canyons. They noted that the faunal density in Baltimore Canyon was highest in the 300-to-399 m depth interval primarily because of high concentrations of the polychaete Hyalinoecia artifex. The authors stated that the middle Atlantic canyon most closely resembled a slope habitat of the three canyons studied and exhibited the least substrate variability. Hecker et al. (1980) further noted that the shelf fauna of Baltimore Canyon was dominated by two crustaceans--Cancer borealis and Munida valida. Progressing downslope, two other organisms, the white sea pen and the burrowing anemone Cerianthus borealis, joined Cancer borealis as the dominants in the head of the canyon and on the canyon walls. From 300 to 500 m in depth, the polychaete Hyalinoecia artifex dominated along the canyon axis and demonstrated high concentrations and contagious distribution. Two species of anemones predominated along the axis in slightly deeper water, Bolocera tuediae and an unidentified species of cerianthid. Table III.B.2-3 lists the common taxa which were photographed by Hecker et al. (1980) in Baltimore Canyon. Hecker and Blechschmidt (1980) reported that coral populations tended to be more diverse in middle and north Atlantic canyon habitats than the slope areas. The primary reason for increased diversity in the canyons was that those species restricted to hard substrates were found only in canyons but soft substrate types were found both in the canyons and on the slope. Table III.B.2-4 lists the species of corals found by the authors in Norfolk, Carteret, and Toms Canyons. Hecker et al. (1983) studied faunal differences between canyons and slopes in the middle and north Atlantic areas. The authors reported that no consistent differences between canyons and slope faunal densities were found in the middle Atlantic area. However, their slope II area contained a small canyon (Hendrickson) which could have complicated the analysis. The report did not contain species counts except for rough graphs of selected species. Comparison of the graphs of individuals per 100 m2 in Hecker et al. (1983) indicates that similar differences between slope and canyon densities in the middle as well as the north Atlantic areas are evident (Figures III.B.2-2 and III.B.2-3). Total densities are similar between Lydonia and Baltimore Canyons except in the 700-to-1,000-m depth range where the sponge Asbestopluma sp. dominates in Lydonia Canyon. In the 300-to-500-m depth range, the maximum densities are approximately 1,500 individuals per 100 m2 in Baltimore Canyon versus 3,000 individuals per 100 m2 in Lydonia Canyon. Below a depth of 1,700 m, Baltimore Canyon demonstrate a generally equal faunal density with Lydonia Canyon. Figure III.B.2-4 demonstrated the differences in faunal densities between Hendrickson Canyon and a slope station, which were combined to provide the data for slope II in Figure III.B.2-3. As is evident, Hendrickson Canyon individuals contribute heavily to the faunal densities of slope II. 2 Generally, the available studies on middle (and north) Atlantic canyons indicate that canyons make the slope into a complex habitat by incising into the slope and shelf. Faunal densities are usually higher in canyon areas because of increased attachment substrate which is generally not available on the smoother slope. Additionally, the increased colonization by epifauna on the hard substrate allows increased population levels of associated fauna. Several studies, primarily from inference, suggest that the faunal densities of filter and deposit feeders may increase inside canyons. This could result from an increased nutrient input which may be channeled and concentrated by the dendritic canyon system. |