the ice pack. Some low surface values were found 

 scattered throughout the survey area in open 

 water so that a given surface nutrient value should 

 be considered representative only of its sampled 

 spot. 



According to Antonov (1958) an average of 813 

 km^/yr (excluding the Yukon River) of river water 

 flows into the Arctic Ocean from the North Slope 

 rivers of North America; most of the flow occurs 

 between May and October. Because the broad 

 shelf area of the southern Beaufort Sea is com- 

 paratively shallow (<65 m), the shelf water has a 

 large surface area in relation to its total volume. 

 This makes the combined river discharge re- 

 ceived by the southern Beaufort Sea an important 

 factor in affecting the distribution of surface 

 temperatures, salinities, dissolved oxygen, nu- 

 trients, and ice. 



Relatively high surface temperatures (0.0 to 

 3.2°C) and low salinities (3.9 to 22.0°/oo) found 

 along the coast during WEBSEC-71 are probably 

 indicative of river runoff (figs. 9 and 15). The 

 extent of most of the river effluent plumes from the 

 Colville (151°W), Kuparik (149°W) and 

 Sagavanirktok (148°W) seems to be limited to less 

 than 4 km from the river mouths. It is interesting 

 to note that the small plumes seem to be better 

 defined by temperature than salinity. This condi- 

 tion may indicate that river water is relatively very 

 warm but low in total volumeforthis time of year. 



A river effluent plume was observed in the 

 eastern section of the WEBSEC-71 survey area 

 (figs. 9 and 15). This plume was characterized by 

 temperatures of 0.0 to 1.8°C and salinities of less 

 than 20°/oo: the lowest surface salinity being 

 3.92°/oo at station 6. This plume is probably due 

 to the MacKenzie River, the largest source of 

 freshwater drainage (430 km-'/yr, Antonov, 1958) 

 in the North Slope area. The MacKenzie River is 

 located east of the survey area at 137°W, suggest- 

 ing a westward distribution of the river plume. 



The horizontal distribution of river effluent in 

 the open sea seems to be governed by a combina- 

 tion of offshore circulation and local prevailing 

 winds (Budinger et al., 1964). These two proces- 

 ses appear to be affecting the distribution of the 

 MacKenzie River plume described above; how- 

 ever, their relative importance is unclear. The 

 existence of a net westward flowing surface cur- 

 rent along the slope of the Alaskan coast has been 



8 



discussed (Worthington, 1953; Kusunoki et al., 

 1962); however, there is a lack of direct meas- 

 urements on the velocity of the current. The pre- 

 vailing winds in the eastern survey area were from 

 the N-NE during the WEBSEC-71 (fig. 27) with 

 recorded wind speeds of generally 1 to 16 knots. 

 Although the wind observations are not synoptic, 

 a good correlation appears to exist between wind 

 direction and the apparent distribution of river 

 effluent. The combined effect of the wind and 

 current would influence the river effluent dis- 

 tribution toward the west supporting the 

 hypothesis that the plume is primarily due to the 

 MacKenzie River. 



River effluent plumes along the north slope 

 were poorly developed during WEBSEC-72. 

 Helicopter overflights with infrared photography 

 indicated that the plumes were located only 2 

 miles or less offshore and that the longest exten- 

 sion of the plumes was parallel to the shore and 

 towards the east on the Canning, Sagavanirktok, 

 Kuparuk, and Colville Rivers. Possible reasons 

 for the lack of large river effluent plumes during 

 WEBSEC-71 and 72 are: (1) low river runoff for 

 this time of year (August-September); (2) icemelt 

 masking the presence of some of the larger river 

 plumes along the coast; and (3) inadequate sam- 

 pling near the river mouths. 



Only a few measurements of nutrient concen- 

 trations were made in the plumes during 1971 and 

 1972 because of inadequate space on the RV 

 NATCHIK for instrumentation and sample stor- 

 age. In 1971 nutrient samples collected along the 

 coast indicated that the river plumes were phos- 

 phate free but contained small concentrations of 

 nitrate (1-3 ^ig-at/1) and silicate (2 /tg-at/l). 



During 1972, nutrient samples were collected 

 above the estuaries in the Canning, Sagavanirk- 

 tok, Kuparuk, and Colville Rivers. Phosphate 

 concentrations were near zero or below detection 

 limits. Nitrate and silicate concentrations were 

 higher than in the coastal waters except off the 

 mouth of the Colville River (table 1). 



In the summer, a strong stability exists in the 

 surface water layer due to an increase of temper- 

 ature (solar heating) and a decrease in salinity 

 (icemelt, river runoff). It would necessitate a very 

 powerful wind to destroy the stability and cause 

 mixing to deeper waters. Past reports indicate 

 that in the summer, surface winds rarely exceed 



