The mercury concentrations measured at the disposal 

 area (Table 3-1) are relatively high compared to those measured 

 both on and off of dredged material at FADS (SAIC, 1986b) . 

 Mercury concentrations were particularly high in relationship to 

 the other parameters measured. Therefore, the ratio of mercury 

 to one of the other elements might be a better tag. 



The ratios of several parameters were calculated for 

 the samples analyzed in the present study and for other data from 

 FADS (SAIC, 1986) . The most promising of these appeared to be the 

 arsenic/mercury ratio. This ratio was 33 ± 10 (range 21 - 48) for 

 samples from the present survey and 111 ± 25 (range 63 - 250) for 

 other areas both on and off of dredged material at FADS (SAIC, 

 1986b) . The disadvantage to using mercury is that for many 

 areas away from dredged material, the mercury concentrations have 

 been reported to be below detection. This, of course, may 

 greatly limit the usefulness of this method. Mercury does occur 

 at some level in all of the sediments and could be measured with 

 additional analytical effort. This, however, could increase the 

 costs of the analyses. 



In order to determine the number of replicates required 

 to detect a 50% difference between any sediment samples from the 

 mound and the ambient seafloor for the chemical contaminants 

 measured with 80% certainty at a 5% level of significance, the 

 following formula was used: 



a 2 

 n > 2 - {t a[l/] + t 2( i _ P ) [iy] } 2 

 5 



where n = number of replications 



a = true standard deviation 



5 = the smallest true difference that is desired to detect. 

 (NOTE: it is necessary to know only the ratio of a to 

 5 , not their actual values) 



v = degrees of freedom of the sample standard deviation 

 (yMS w -j^h^ n ) with a groups and n replications per group 



a = significance level (such as 0.05) 



P = desired probability that a difference will be found to 

 be significant (if it is as small as S) 



t a [^] and t 2 n _ p)[>] = values from a two tailed t-table with v 



degrees of freedom and corresponding to 

 probabilities of a and 2(1 - P) , 

 respectively 



10 



