MAJOR and PAULIK: ENCROACHMENT OF WANAPUM DAM RESERVOIR 



Table 4. — Analysis of variance tests of the hypothesis 

 that spring chinook salmon, tagged and released below 

 Rock Island Dam in 1965, traveled over the right fish 

 ladder at Rock Island Dam equally as fast as those 

 traveling over the left and center ladders. 



1  = Highly significant at the 0.01 level, reject hypothesis of equal 

 travel times, and conclude that travel time was significantly less through 

 the right ladder. 



We also analyzed the 1965 data on a ladder 

 by ladder basis. Although the data, even when 

 compared on a period by period basis, are too 

 limited to provide a sensitive comparison of the 

 mean passage times between fish using the dif- 

 ferent ladders, they did reveal that in every peri- 

 od fish moved over the right ladder faster than 

 over either the center or left ladder. Mean pas- 

 sage times for right ladder versus the left ladder 

 and right ladder versus the center ladder (all 

 periods and both release sites pooled) are com- 

 pared in Table 4." The data have been adjusted 

 for simultaneous tests according to the method 

 described by Dunn (1961). The diff"erences 

 shown in Table 4 are highly significant. Thus 

 the hypothesis that spring chinook salmon trav- 

 eled from the tagging sites over the right ladder 

 equally as fast as over the left ladder is strongly 

 rejected as is the same hypothesis for the right 

 versus the center ladder. In both cases the mean 

 passage times are significantly less for fish using 

 the right ladder. This means that ladder choice 

 was the only observed factor clearly affecting 

 passage time. Similar trends were noted in 

 1964. Mean passage time over the right ladder 

 was less than for either the left or center ladders. 



The overall travel time in 1965 (5.8 half-days) 

 was shown by an analysis of variance test to be 

 significantly less than that of 1964 (9.8 half- 

 days) . The F-value for this test was 17.72 with 

 1 and 165 degrees of freedom. 



* Because the passage of fish over the right-bank fish 

 ladder had been a source of controversy among fishery 

 biologists, we directed special attention to the right-bank 

 ladder in the present study. 



SOCKEYE SALMON 



Fewer sockeye salmon were tagged in 1965 

 than in 1964. Analysis of the 1964 data revealed 

 that the precision desired could still be achieved 

 if sample sizes were reduced from 100 to 75 fish 

 per release in each year. The tagging season 

 was divided into five periods. With one excep- 

 tion, each of these periods contained releases 

 from the left and right banks. During period 

 IV in 1965, both releases were from the left bank. 



Tagging was from July 15 to August 5 in 

 1964 and from July 14 to August 4 in 1965. Tag 

 release and tag observation data are presented 

 in Table 5. 



Percentage Observed 



The percentages of tagged sockeye from indi- 

 vidual releases observed passing Rock Island 

 Dam were similar for 1964 and 1965. The per- 

 centage ranged from 84.3 to 120.8 in 1964 and 

 from 81.2 to 98.8 in 1965. Overall, 94.1% of 

 the tagged fish were observed in 1964 and 91.8% 

 in 1965. Percentages observed from left-bank 

 releases were not significantly diflferent from 

 those released on the right bank in either year — 

 94.8 (left bank) versus 93.3 (right bank) in 

 1964 and 91.5 (left bank) versus 92.0 (right 

 bank) in 1965. 



Distribution by Ladder 



Distribution by ladder was similar in both 

 years. Of the tagged fish sighted, 55.6% used 

 the left ladder in 1964 and 53.1% in 1965. The 

 center ladder took 20.7 and 23.6% and the right 

 ladder 23.7 and 23.3% in the 2 years, respec- 

 tively. For the left- and right-bank releases, 

 respectively, the percentages using the left lad- 

 der were 54.1 and 57.4 in 1964. Comparable 

 percentages were 55.4 and 50.7 in 1965. Fish 

 not choosing the left ladder were fairly evenly 

 distributed between the center and right ladders 

 in both years. 



The between-period consistency of the per- 

 centage of tagged sockeye salmon using the pre- 

 ferred left ladder was less evident than for 

 chinook salmon. For sockeye salmon, the 



133 



