127 



sound to avoid gas supersaturation, the Willamette net pens provided ample depth to 

 compensate for the recorded gas levels and no fish should have died. Indeed, at the bottom 

 of the net pen, juvenile salmon could have compensated for approximately 18% (3% times 

 six) excess gas supersaturation. 



SpiU Passage vs. Turbine Passage 



At reasonable levels, spill results in lower mortality to migrating juvenile salmon than 

 passage by turbines or bypass facilities. As set forth below, however, conditions are 

 different at every project. The available data do not suggest that the benefits of passage by 

 spill are sufficiently clear or large to gamble with the known adverse effects of gas 

 supersaturation in the way proposed by NMFS. 



There is little information available concerning the comparative risk of passage by 

 turbines at Bonneville Dam and passage by spill. Turbine mortality at Bonneville is assumed 

 to be on the order of 12-15% based on an unpublished study in 1952. Holmes (1952). Since 

 1952, a second powerhouse has been constructed at Bonneville Dam with a more fish- 

 friendly design. Direct turbine mortality at this powerhouse has consistently been estimated 

 at 1-4% - not terribly different than direct spill mortality ( i.e. . non-GBT mortality). 

 Ledgerwood et al. (1990); Ledgerwood et al. (1994). Those studies did show very high 

 predation mortality in the river below the second powerhouse which more than offset the 

 benefits of the lower turbine mortality, but the consistent disuse of the second powerhouse 

 has created slackwater conditions favoring resident predators; presumably, consistent use of 

 the second pxDwerhouse would avoid this problem and was in fact recommended by the study 

 authors. 



Spill mortality at Bonneville is uncertain. The Ledgerwood studies did show lower 

 mortality for salmon passed by spill at Bonneville than when passed through either 

 powerhouse. However, that spill program was specially designed to avoid use of the bays 

 that spill onto large concrete tetrahedrons designed to dissipate the kinetic energy of the 

 water, and involved spill levels that did not generate dangerous levels of total dissolved gas. 



Dau are available concerning spiU mortality and turbine mortality at other projects. 

 Probably the most serious erron in tne Risk Assessment are in the values used for turbine 

 mortality at these other projects. They use 32 percent for turbine mortality at Ice Harbor 

 Dam. The correct value if 14.5 percent. Thirty two percent mortality was recorded by 

 Long for releases in the backroll. Fifteen percent mortality was cited for Bonneville I. The 

 study by Holmes gave a range of 12-15 percent. Thus 13.5 percent would be a more 

 appropriate value. Eighteen percent turbine mortality is incorrect for Bonneville n. The 

 correct value is 2-4 percent. Not recorded is a turbine mortality of 3.5-9.2 percent for 

 Rocky Reach Dam and 5 percent recorded for Lower Granite Dam. A 27 percent spillway 

 mortality (sieelhead) for Lower Monumental Dam for Bonneville is not mentioned. 



