layers, chaotic fabrics, highly reducing sediments at depth, 

 methanogenic sediments) . 



The small-scale boundary roughness frequency distribution 

 showed a major mode in the 0.8 cm class interval (Figure 3-87), 

 with a mean value of 1.04 ± 0.49 cm. These values were 

 comparatively high, but inspection of the photographs showed that 

 this was related to both bioturbation and the presence of a smooth 

 yet undulating surface which was probably natural for this area. 



The mean apparent RPD depth frequency distribution had 

 a major mode at the 4 cm class interval (Figure 3-87) ; the mean 

 value was 3.71 cm ± 0.68 cm. The mapped distribution of RPD depths 

 (Figure 3-86) showed the area to be generally characterized by 

 values which were significantly higher than the new CLIS reference 

 station (p < 0.05, Mann-Whitney U-test) . However, an area where 

 apparent RPD depths were less than 3 cm was identified (Figure 3- 

 86) . Stations A2 and B3 fell within this area; both had sand- 

 over-mud stratigraphy and small-scale boundary roughness that was 

 greater than 1 centimeter. Station C4, which likewise had high 

 boundary roughness, was also located within this area. These 

 physical parameters could be indicative of the presence of relict 

 dredged material, although the lack of direct evidence in the 

 photographs and the absence of evidence of past disturbance in the 

 side-scan records makes this unlikely. 



The map of successional seres (Figure 3-88) showed a 

 large area containing only Stage I taxa extending from the 

 northeast corner to the south side of the area surveyed. This may 

 reflect natural patchiness in the distribution of Stage III 

 organisms, but the scale of such variablity is difficult to assess. 

 Obtaining several replicate photographs per station would have 

 enabled an assessment of small-scale patchiness, but in the Ghost 

 Site surveys such replication necessarily was sacrificed for 

 greater spatial coverage. It is interesting to note that the area 

 of shallow RPD values did not fall within the patch of Stage I 

 seres. 



The Organism-Sediment Index frequency distribution 

 (Figure 3-87) was bimodal, with a major mode at +11 and minor modes 

 at +10 and +7. Past work indicats that areas with OSI values of 

 +6 or less have undergone recent disturbance, due to either 

 disposal operations or natural factors such as erosion or 

 predation. The mean OSI value for this area (8.96 ± 1.89) was 

 relatively high; these values were not significantly different from 

 those at the new CLIS reference station. Stations 5C, 7E and a 

 cluster of four stations in the NE quadrant of the area surveyed 

 had OSI values of +6 or less (Figure 3-89) . Of these, stations F2 

 and E7 had sand-over-mud stratigraphy, and station G2 had high 

 boundary roughness. 



28 



