the shelf-slope break (200-m isobath). These canyon areas (enclosed on three 

 sides by the shelf) are those most likely to be impacted by drilling on the 

 shelf, on canyon rims, and on the uppermost slope near the canyons. 



Butman offered a series of general statements based on the concept of a 

 "typical" canyon, with a length of 10 to 20 km, a depth of 500 to 1,000 m, and 

 a width at the mouth of approximately 5 km (Figures 19 through 22 illustrate a 

 variety of canyon types in the Georges Bank region). Although the axis and 

 rim areas comprise a small portion of the total area of canyons, most of the 

 biological and geological sampling has been done in these areas. In contrast, 

 many of the visual observations made by submersibles in canyons have come from 

 the canyon walls. Nearly all the information to date has been obtained from 

 the upper portion (less than 700 m) of a very few canyons. 



PHYSICAL FEATURES OF CANYONS 



Many statements presented in the workshop were largely applicable to the 

 large, deep canyons such as Oceanographer, or the generic canyon presented in 

 the preceding figures (Figures 19 to 22). These statements were progressively 

 amended during discussion to include conditions and mechanisms found in other, 

 generally smaller canyons such as Lydonia. Throughout the workshop the 

 question was posed, in a number of different forms, as to just how applicable 

 many of the broader statements were to canyons in general.* The heterogeneity 

 of canyons was widely recognized. 



There was consensus on the fact that canyons are unique habitats and 

 environments — as compared to slopes at comparable depths. 



The general categories by which canyons could be defined as unique were: 



■ Topography and features: canyon shape, substrate type, and 

 patchiness; 



'See also reviewer comment in Appendix A. 



80 



