traveled between stations so that it moved toward the 

 head of the bay or seaward at a rate controlled by the 

 speed of the tidal current. In this way the vessel was as- 

 sumed to remain in water representative of water at the 

 beginning of a transect run. 



Transects were usually run when the wind velocity was 

 10 to 15 knots; speeds greater than 13 knots caused waves 

 to become unstable and break into whitecaps, resulting 

 in an increase in the mixing of the surface waters (John- 

 son 1960). 



The vertical and horizontal profiles of salinity in inner 

 Bristol Bay are used to interpret distribution of river 

 waters in the inner bay where there is very little vertical 

 variation in salinity (Fig. 2). Vertical mixing is caused by 



rW°.' s -W" ''V^I/V&V/ ^ NAKNEK RIVER MOUTH 

 (a) W ^x^y AUGUST 26. 1966 



12 16 20 24 28 32 36 10 VI 

 KILOMETERS 



Figure 2. — Vertical profiles of salinity ( %o) at 16-km intervals across 

 inner Bristol Bay, August 1966. 



the high velocities of the tidal current, shallow depth, 

 and wind. Salinity in inner Bristol Bay varied horizon- 

 tally, as shown by surface and bottom salinities at low 

 and high tides (Figs. 3-6); salinity was low on the north- 

 west side and high on the east side. Dodimead et al. 

 (1963) found that this condition persisted farther off- 

 shore: low-salinity water moved seaward on the north- 

 west side of the bay, and a countercurrent of high- 

 salinity water was present on the southeast side. This 



current appeared as a wedge of high-salinity water 

 between the river waters entering each side of the bay. 



Dye Studies 



To track water from individual rivers, I used a tracer 

 technique utilizing a fluorescent organic pigment, 

 Rhodamine B (alkyl aminophenol derivative). 4 This 

 technique was first used by Pritchard and Carpenter 

 (1960) to observe the movement and dispersion of water 

 in various parts of Chesapeake Bay. 



Ideally, a tracer is released into a river when a con- 

 tinuous flow is expected for a period of time long enough 

 for the tracer to achieve equilibrium with its environ- 

 ment. However, the objectives of my study could be met 

 by instantaneous release of enough Rhodamine B to trace 

 its course over one or two tidal cycles. The dye was 

 released only on days when wind velocities were less than 

 15 knots because the wind is considered to be an impor- 

 tant agent of mixing and dispersion of a tracer in short- 

 term studies such as mine. The dye was rapidly poured 

 from plastic containers into the propeller wash of the 

 tracking vessel while traveling at a speed of 5 knots. Ap- 

 proximately 9.5 to 18.9 liters of dye were released about 1 

 h after high slack tide along a continuous line 3 m wide 

 across the main channel of the river. The channel was 

 crossed three or four times with a line of dye laid down 

 about 15 m upstream from the previous line. The dye 

 mixed rapidly with the water to a depth of 2 m; within 1 

 h after release it had mixed to a depth of 4 m. Usually 

 within 1 h, individual lines of dye had diffused horizon- 

 tally and merged with one another, producing a visible 

 patch across the channel 60 m or more wide. Dye was also 

 released in a similar manner near the end of the ebb tide 

 offshore across the course assumed by the dye previously 

 released at the beginning of the ebb tide. The course of 

 this release was followed during the following flood tide. 



Paths of Fluorescent Dye. — Dye was released into 

 Naknek, Egegik, and Ugashik rivers in July and August 

 1965 and 1966. No dye was released in the Kvichak River 

 because the course and distribution of Kvichak River 

 water in inner Bristol Bay could be logically deduced 

 from results of the Naknek River study and the surveys 

 to measure salinity. The tracks of 11 dye releases in the 

 three rivers showed that river waters followed rather dis- 

 crete and similar courses in the inner bay (Figs. 7, 8). 

 The course of dye released near the end of ebb tide at the 

 mouths of the three rivers indicated that a portion of the 

 river water was carried north by the succeeding flood tide 

 (Fig. 8). Water entering the bay from these rivers (and 

 Kvichak River) near the end of ebb tide constitutes the 

 greatest fraction of river water entering the bay during a 

 tidal cycle. In most cases, dye was tracked on the flood 



'Rhodamine B dye was purchased as a 40'c (by weight) acetic acid solu- 

 tion. Because the density of the solution was greater than that of the 

 water into which it was to be released, its specific gravity was adjusted to 

 1.00 by the addition of methyl alcohol. 



