126 



The authors of the Risk Assessment place substantial emphasis on in-situ live ca^e 

 studies done by Meeldn, Turner and WeiLkamp. The authors state that concentrations were 

 as high as 126 and 128 percent and no mortality occurred. For most of the duration of these 

 tests the concentrations were nearer 120 percent. Considering the clearer water (fish tend to 

 be deeper in clear water, Dawley et al., 1975) and the duration of the tests (7 days), I would 

 not expect mortality. They do not refer to a test done in the Snake River where 

 concentrations of TDG were 127 percent for the entire duration (7 days) and 48 percent 

 mortality occurred in the volitional cage 4.5 m. deep. It appears the modelers must have 

 given more weight to in-situ experiments that showed lower mortality rates. In figure S, 

 page 42, there are several observations above the mortality line between 120 and 130 TDG. 



Depth Compensatioa 



_ The depth of a fish in the water affects the level of gas supersaturation that the fish 

 can tolerate. For example, each foot of depth compensates for approximately 3% excess 

 saturation. Thus a fish at 3 feet of depth in water supersaturated at 120% will be subjected 

 to the equivalent of a gas supersaturation level of only about 1 10%. Tests done in deep 

 tanks and in four-and-one-half meter deep live cages in the river showed that significant 

 mortality still occurred at exposure times as short as six days. Dawley et al. (1976); Ebel 

 (1971). 



NMFS has determined that "there is no evidence that fish can 'sense' TDG 

 supersaturated water and deliberately sound to compensate". (Biological Opinion at 108). I 

 agree with this conclusion. The authors of the Risk Assessment suggest that fish will detect 

 and avoid supersaturated water by either sounding or moving laterally. There is some 

 evidence that salmonids can avoid supersaturated water by moving laterally to normally 

 saturated water, but this is irrelevant when large areas are supersaturated and there is no 

 normally saturated water to escape to. 



While fish cannot avoid TDG supersaturation, the normal depth distnbution of salmon 

 does compensate for some excess gas supersaturation. This compensating effect is limited 

 by the fact that a significant portion of migrating juveniles travel in the upper 3 feet of the 

 water column. For example, Smith (1973) found approximately 30% of juvenile chinook 

 salmon in the upper three feet of the water column at Lower Monumental Dam. Dawley 

 (1986) found similar distributions of chinook in the forebay of the The Dalles Dam. 



It should also be remembered that juvenile salmon, particularly fall chinook juveniles, 

 feed as they move downstream. Most of their food supply appears to be insects and insect 

 larvae which are found at the surface, suggesting that juveniles spend a substantial fraction of 

 the time in shallow water to feed. 



According to press accounts, gas supersaturation recendy killed a large proportion of 

 juvenile salmon awaiting release in net pens in the Willamette River. Oregonian (1995). 

 ODFW reports that mortality occurred at TDG levels measured by saturometer at 114-117%, 

 and that the effective depth of the net pens was at least six feet. If juvenile salmon could 



