Table 7.— Mercury in Snows South of Barrow 



apparently unpolluted ice recovered from Camp 

 Century (Weiss, et al., 1971) and the Dye-3 site 

 (Weiss. Bertine, Koide and Goldberg, in prep- 

 aration) in Greenland, where the average con- 

 centrations ranged from 50 to 60 ppt. 



The mercury concentration of waters gathered 

 25 miles upstream on the Sagavanirktok River 

 and the Umiat and Cubic sections of the Colville 

 River ranged narrowly between 16.8 to 18.7 ppt. 

 These values are not markedly different from the 

 average concentration in snow. The disparity is 

 attributable either to the error in average snow- 

 value derived from only eight numbers, whose 

 range extends over an order of magnitude, or to 

 inclusion of an incremental amount of mercury as 

 the snows melt and pass over the terrain. The 

 mercury content of the river waters is low as 

 compared with the Danube and Volga Rivers, in 

 which the concentrations measured are usually 

 about 1 to 2 ppb (Aidin'yan and Belavskava, 

 1963). On the other hand, river waters in rela- 

 tively unpolluted areas usually measure between 

 10 to 50 ppt (Dair Aglio. 1967). 



The mercury content of Beaufort Sea water 

 collected at various depths between 146°W and 

 151°W are shown in Table II. At one of these 

 stations (151°08'W, 71° 23'N) waters were 

 analyzed in duplicate, and the error shown rep- 

 resents the average diviation from the mean; thus 

 the average error is about 10%. 



Marked variations of mercury concentration in 

 tiie vertical are apparent. These variations do not 

 occur as a function of depth; rather, they are 

 random. This variability may be associated with 

 the combining of a large fraction of the element 

 with the particulate phase. 



Of further interest is the greater average con- 

 centration of mercury in the water recovered from 

 stations occupied at 151°W compared with all 

 stations to the east of this longitude. The tempera- 

 ture data (Hufford, personal communication) in- 

 dicate that Chukchi and Bering Sea water had 

 penetrated to precisely 151°W during 

 WEBSEC-73. The difference in mercury con- 

 centrations may. therefore, reflect varying con- 

 centrations of this element between the eastern 

 Beaufort Sea and Chukchi-Bering Sea water 

 masses. We contemplate the measurement of 

 other water secured west of 151°W on 

 WEBSEC-73 to further examine this possibility. 



By way of comparison, Beaufort Sea waters 

 contain considerably smaller quantities of mer- 

 cury than Antarctic waters, where concentrations 

 approximate 100 ppt, but are not markedly differ- 

 ent from those collected in a transect from San 

 Diego to an area north of the Hawaiian Islands 

 (Williams, Robertson, Chew and Weiss, in prep- 

 aration). In fact many water masses e.g., English 

 Channel (Burton and Leatherland. 1971), Gulf of 

 California (Weiss, in preparation), and the east- 

 ern tropical Pacific Ocean (Weiss, et al., 1972), 

 are not grossly different from the Beaufort Sea in 

 their mercury concentration. 



The mercury content of sediments from the 

 Sagavanirktok and Colville Rivers was 111.5 to 

 119.1 ppb, respectively. Marine sediments col- 

 lected on the shelf, slope, and in the basin be- 

 tween 143°W to 154°W, with the exception of an 

 area on the shelf from 143°W to 146°W, averaged 

 100 ppb, and no discrete pattern was evident with 

 respect to distance from the shore (Figure I). 

 These values are reasonably consistent with the 

 river sediments that drain into this general area. 

 The concentration was only 40 ppb on the shelf 

 between the Canning River and Barter Island. 

 Sediment was not available from the Canning 

 River; however, these lower concentrations prob- 

 ably reflect input from this river. That Canning 

 River sediment is different from that derived, for 

 example from the Colville River delta has already 

 been established (Naidu, personal communica- 

 tion). The illite-to-smectite ratio in the Canning 

 River sediment is three times greater than in the 

 Colville. The exchange capacity of illite for a 

 heavy metal such as mercury is considerably less 



233 



