as key beds; and third, the displacement component must be large enough 

 to be apparent at the resolution level of the seismic reflection system 

 in use. Absence of faulting on seismic reflection records which do not 

 meet these criteria is, therefore, not conclusive evidence of nonexistence. 



The reflection profiles taken during the ICONS survey provided pene- 

 tration in most places to 61.0 meters or more below the shelf floor. 

 However, in substantial areas over unit 11 and elsewhere (particularly 

 over unit III), there was little or no penetration below the first sub- 

 bottom reflector which lies usually at less than 4.6 meters below the 

 sea floor. Where good penetration was achieved there were many closely 

 spaced subbottom reflectors so it is likely that faulting with vertical 

 displacement would be apparent. In areas where both penetration and 

 subbottom reflectors were adequate enough to show shallow subshelf faults, 

 only a few possible faults were detected (see App. A, profiles B and C) . 

 On the basis of probability it seems reasonable to conclude that the 

 scarcity of evidence of faulting in areas suitable for their detection 

 would also apply to those areas where penetration was deficient, unless 

 lack of penetration itself was occasioned by fault displacement of 

 acoustically inpenetrable strata into a laterally contiguous relation- 

 ship with more penetrable strata. It is improbable, however, that such 

 a consistent pattern of displacement would occur in so many places through- 

 out the area; it appears more likely that large faults within the upper 

 61.0 meters of the subshelf strata are rare. 



Fine, well-sorted type A sand of the Holocene shelf deposits and the 

 similar types B and C sand have textural properties making them more 

 susceptible to liquefaction than other sediment bodies in the study area. 

 Plots of the size -frequency distribution of 34 representative samples 

 from these deposits were compared to a size distribution curve (Fig. 29) 

 showing the envelope of values for a number of sands which have liquefied 

 under cyclic loading in nature and in laboratory experiments (Seed, Arango_ 

 and Chan, 1976). A significant part of the curves for all 34 samples, 

 varying from 50 to 100 percent, fell within the envelope of greatest 

 susceptibility. 



Since sand types A, B, and C cover or underlie more than half the 

 survey area and extend in depth to 61.0 meters or more below the shelf 

 floor, their potential for liquefaction is an important aspect of the 

 inner shelf engineering environment. However, liquefaction potential is 

 also affected by other factors--in situ density (the most important 

 single factor), geological and seismic history, grain structure, cementa- 

 tion, and characteristics of the cyclic load. Available samples are not 

 suitable for measurement of grain structure or density. For factors 

 related to geological history, it seems reasonable to expect that con- 

 siderable differences exist between deposits of types B and C sediments 

 which have been in place for several millions of years and type A sedi- 

 ment which, at most, is a few thousand years old. For example, types B 

 and C sand may be considerably denser than type A sand because of longer 

 exposure to overburden stresses. The eroded upper surface of these 

 deposits, as indicated by truncation of secondary reflectors, suggests 



73 



