INTRODUCTION 



Data derived from the analysis of Greenland 

 ice suggests that the flux of mercury from the 

 continents to the atmosphere has increased in 

 recent times as a result of input by society (Weiss, 

 etal., 1971). In addition to the obvious sources of 

 mercury, it was further suggested that a fraction of 

 the added quantity may be introduced through 

 activities that result in alteration of terrestrial 

 surfaces, thereby enhancing the natural outgas- 

 sing of mercury vapor from the earth's crust. 



With the anticipated intensification of human 

 activity in the North Slope area, it was considered 

 desirable to acquire baseline data. This report 

 describes the concentration of mercury that was 

 determined in plankton, snow, and in water and 

 sediments from several rivers and the southern 

 Beaufort Sea. 



METHOD 



Sample Collections. The plankton, and 

 ocean sediments were collected on the US Coast 

 Guard WEBSEC-72 Cruise during August 1972, 

 while river sediments and river and ocean water 

 were gathered in July-August 1973 on 

 WEBSEC-73. By means of facilities provided by 

 the Naval Arctic Research Laboratory (NARL), 

 the snows were collected in mid-March 1973. 



The accumulation of plankton was by standard 

 oceanographic procedure in a vertical tow at 71 

 12.0' N, 149° 31.8' W from a depth of 180 

 meters. The sample together with transfer dis- 

 tilled water was frozen until shortly before prep- 

 aration for analysis. 



Seawater samples were obtained from the 

 southern Beaufort Sea between 146°-151°W lon- 

 gitude in 8-liter Niskin bottles by standard hydro- 

 casting procedures. Aliquots were drawn into 

 storage containers within 15 minutes after the 

 collection. 



Samples of river water were taken from the 

 Umiat and Cubic regions of the Colville River and 

 from 25 miles upstream on the Sagavanirktok 

 River. The sample was collected about 6 in. from 

 the surface after wading into the river to a depth of 

 2.5 ft. 



Water samples were stored in polyethylene 

 containers that previously had been soaked for 16 



hours in concentrated nitric acid. The containers 

 were rinsed three times with sample and then 

 nearly filled. Immediately, 10 ml of 16 N nitric 

 acid were added and the sample was stored for 

 analysis. 



Sagavanirktok and Colville River sediments 

 were obtamed 25 miles upstream at a distance of 

 several feet from the shoreline. The marine sedi- 

 ments were acquired with a grab sampler. The 

 sediments were transferred to polyethylene stor- 

 age bags with a plastic utensil, and the samples 

 were frozen as rapidly as possible after collection. 



At distances of 30 to 135 miles south of Bar- 

 row, and in 15-mile increments, the samples of 

 snow were acquired. When these locations were 

 reached by plane, the engine was cut. After a 

 10-minute wait and at a distance of about 50 

 meters upwind from the plane, several inches of 

 snow were removed and discarded. The underly- 

 ing deposit was transferred with a clean plastic 

 scoop to a polyethylene bag, which was stored in a 

 thermally insulated box during transit. Upon re- 

 turn to NARL, the sample was melted at indoor 

 temperature in 3-liter polyethylene containers 

 previously leached with nitric acid as described 

 above and rinsed five times with double-distilled 

 water. These containers received 30 ml of con- 

 centrated nitric acid prior to melting. 



Mercury Analysis. The mercury analyses 

 were performed by a modification of a neutron 

 activation analysis procedure previously de- 

 scribed (Weiss and Crozier, 1972). The changes 

 were introduced to provide for radiochemical pur- 

 ity at an earlier time after the irradiation; accord- 

 ingly the sensitivity was appreciably enhanced. 



Sample Preparation. Mercury was isolated 

 from snow, river, and seawater samples by co- 

 precipitation with copper. To 1 literof sample was 

 added 1 mg of 99.999% copper as nitrate. 

 Reagent-grade hydrogen sulfide gas, after pass- 

 ing through a millipore membrane filter, was 

 bubbled through the solution for 30 seconds. The 

 resultant precipitate was collected by filtration. 

 (This, and subsequent filtrations involved pass- 

 age of the solution through millipore membrane 

 fdters of 0.45/1 pore size). After fdtration. the 

 walls of the sample container were washed with 

 10 ml of concentrated nitric acid and this wash 

 together with the copper sulfide placed in a 

 polyethylene irradiation vial. 



231 



