Figure 7. — Course of river waters in inner Bristol Bay on ebb tide as 

 determined from course of dye releases in July and August 1965 and 

 1966. [Solid lines marking boundaries of river water denote area of 

 actual observation; dotted lines indicate probable course; dashed 

 lines indicate direction of net (nontidal) current. The direction of the 

 net current is based on the results of previous oceanographic inves- 

 tigations (Dodimead et al. 1963) which showed a counterclockwise 

 movement of water in this region.] 



Figure 8. — Course of river waters in inner Bristol Bay on flood tide as 

 determined from course of dye releases in July and August 1965 and 

 1966. [Solid lines marking boundaries of river water denote area of 

 actual observation; dotted lines indicate probable course; dashed 

 lines indicate direction of net (nontidal) current. The direction of the 

 net current is based on the results of previous oceanographic inves- 

 tigations (Dodimead et al. 1963) which showed a counterclockwise 

 movement of water in this region.] 



tide at least 4.8 km north (above) of the mouths of Nak- 

 nek, Egegik, and Ugashik rivers. This water moved 

 toward shore and over the tideflats. 



The path of the dye was followed as the vessel traveled 

 at a speed of 3 knots. The vessel's position was ac- 

 curately maintained by reference to prominent land- 

 marks, marker buoys, and bottom topography (deter- 

 mined by comparing fathometer readings with 

 navigational charts of the area). Because of the low wind 

 velocity (less than 15 knots), the only factor affecting the 

 vessel's position was the tidal current. Dye was followed 

 visually for about 1 to 2 h after it was released; thereafter 

 a fluorometer (Turner Model IE) was used to detect it. 

 Continuous sampling was carried out by pumping water 

 from a depth of 1 m through the fluorometer as the ves- 

 sel moved through the water. When readings indicated 

 no dye was present, the course was altered 180° so that 

 the dye patch was reentered. 



Corrections for background fluorescence were applied 

 to fluorometer readings during the tracking phase of the 

 tracer studies. These were determined by running 

 samples of pure river water from each of the rivers to be 

 studied through the fluorometer and recording the in- 

 strument's dial readings. In addition, waters along the 

 course of proposed tracer releases and the offshore por- 

 tion of the study area were also sampled to determine the 

 level of possible interfering background fluorescence. 

 Only fluorometer dial readings above these background 

 levels were considered to indicate the presence of dye. 



Shear Lines. — Shear lines caused by adjoining cur- 

 rents having different speeds or directions affected the 

 distribution of dye during ebb tide. These lines, a charac- 

 teristic feature of the inner bay, were seen during 

 relatively calm weather. They were not present during 

 the flood tide but reformed, usually with some lateral 

 variation of the axis, during the following ebb tide. The 

 distribution of dye indicated that these lines marked the 

 boundaries between waters leaving Naknek and Kvichak 

 rivers and between the offshore water flowing seaward 

 and the waters entering the bay from Egegik and 

 Ugashik rivers. When surface turbulence was minimal, 

 shear lines were observed about 6.5 km seaward from the 

 mouth of the Naknek River. In Figure 7, the outer bound- 

 ary denoting the course of the rivers corresponds to the 

 general location of the shear lines. Dye released into 

 Naknek, Egegik, and Ugashik rivers never crossed well- 

 developed shear lines. Farther offshore the shear lines 

 were not well developed and the dye dispersed laterally 

 to some extent, no doubt because of decreased velocities 

 and less difference in direction of separate currents. The 

 northward extent of individual river waters during flood 

 tide could not be identified by shear lines (Fig. 8). 



The area of Kvichak Bay not occupied by Naknek 

 River water was occupied by Kvichak River water (Figs. 

 7, 8). Kvichak River had a much greater volume of dis- 

 charge than Naknek River, as indicated by the lower 

 salinity on the northwest side of Kvichak Bay (Figs. 3, 6). 

 Undoubtedly, waters of these two rivers mixed along 



