found in the same locations, all PAH 

 compounds have been combined into a 

 single group for this report. 



All of these trace metals and groups of 

 organic compounds can be acutely or 

 chronically toxic to marine life and even 

 to people under some conditions. Those 

 conditions include the total concentra- 

 tion of chemical that is biologically 

 available and the ability of each species 

 to accommodate increased chemical 

 exposure. An important aspect of the 

 NS&T Program is to determine the dis- 

 tribution of locations where contamina- 

 tion is of biological consequence. 



Nationwide Distribution of 

 Contaminants in Sediment 



The nationwide results from NOAA 

 analyses of surface sediments can be 

 used to define the spatial distribution of 

 contamination. Before simply using the 

 data, however, it is important to know 

 that contaminants are associated with 

 particle surfaces. Sand-sized particles 

 have less contamination per unit weight 

 of sediment than silt or clay. To account 

 for this, the NS&T sediment data have 

 been adjusted in two ways. First, no 

 data for contaminants in sediment were 

 used forcomparisons among sites when 

 the sediment contained more than 80% 

 sand (particles with diameters less than 

 63 microns). Secondly, contaminant 

 levels in sediments containing less than 

 80% sand have been adjusted by divid- 

 ing the fraction of sediment that is fine- 

 grained (i.e., divided by a number be- 

 tween 0.20 and 1 .00). That adjustment 

 is the equivalent of considering sand to 

 be only a dilutant of sediment contami- 

 nation. The exclusion of very sandy 

 sediments acknowledges that some 

 contamination may be associated with 



sand, but cannot be accounted for in this 

 method of comparing among sites. 



The NS&T results of sediment analyses 

 have already been reported (NOAA, 

 1988). Those data, plus new data from 

 NOAA Mussel Watch sites occupied in 

 1988 and 1989 and NOAA Benthic Sur- 

 veillance data from 1986, provide infor- 

 mation on chemical concentrations in 

 sediments at 287 sites. Fine-grained 

 sediment was collected at 232 of those 

 sites. Each time sediments were 

 sampled at a site, three separate 

 samples of surface sediment were col- 

 lected. The overall total number of 

 samples per site varies som.ewhat de- 

 pending on how often a site was sampled 

 and on the availability of fine-grained 

 sediment, but in general, the mean 

 concentrations used in this report are 

 based on analyses of six samples at 

 each site. 



Data on cadmium and tPCB are used in 

 Figure 1 to show that, for both trace 

 metals and organic chemicals, concen- 

 trations are distributed in such a way 

 that there are few high concentrations 

 that stand out from the rest. When dis- 

 tributions of concentrations are highly 

 skewed toward the lower concentra- 

 tions, it is useful to examine the distribu- 

 tions of the logarithms of the data. As 

 exemplified in Figure 1 , when the loga- 

 rithms of the concentrations are plotted, 

 the distributions become bell-shaped. 

 In statistical analyses such distributions 

 are referred to as normal distributions 

 and, in this case because logarithms 

 were required, the distributions are log- 

 normal." 



The advantage of the fact that the con- 

 centrations are log-normally distributed 

 is that it allows a statistically objective 



