125 



(For purposes of these conclusions I have assumed that "in-river migration of saJmon" may 

 include transporting juvenile salmon downriver in barges.) 



State and tribal fishery managers prepared the Risk Assessment document to justify 

 increasing the dissolved gas concentrauon in a range of 120 to 125 percent based on 12 hour 

 averages. The risk assessment model compares the predicted mortality that will occur to 

 juvenile and adult migrants from TDG (total dissolved gas) induced by spilling against that 

 which occurs from passage through turbines. A nsk assessment model is only as accurate as 

 the values used to calculate the risk. I found several errors in interpretation of the results 

 from some of the literature cited. As a result, some of the values used are incorrect and 

 some of the conclusions drawn from some important research are either distorted or 

 incorrect. The specific deficiencies are set forth below and in the attached Exhibit A. 



Background: G«nerai EfTects of Gas Supersaturation 



I have conducted a number of studies concerning the effect of gas supersaturation on 

 juvenile salmon and other fish. Gas supersaturation arises when excess gas is dissolved in 

 water; that is, an amount of gas over what the body of water would hold normally. In the 

 Columbia and Snake Riven, the process of spilling water over dam spillways concentrates 

 atmospheric gases in the water in levels that exceed the norm. These excess levels are 

 measured by percentages. Normal saturation is 100%. In the Columbia River, values as 

 high as 148% have basn recorded. 



Gas supersaturation adversely affects fish in a number of ways. Excess nitrogen 

 enters the circulatory system of the fish and diffuses out, causing gas bubbles or emboli in 

 the circulatory system and gas bubbles under the skin. These gas bubbles have a number of 

 adverse physical effects. Gas bubbles occlude blood fiow in the gills, thus suffocating the 

 fish. Gas bubbles also occlude the mouth and throat of the fish, and can cause blindness in 

 the fish due to hemorrhaging or exopthaimia. The gas bubbles can also result in 

 overextension or rupture of the swim bladder, particularly in juveniles under 50 mm in 

 length. Collectively, these symptoms are referred to as gas bubble disease. 



Sublethal effects of gas bubble disease are not always evident as external visible 

 symptoms. For example, Schiewe (1974) and Dawley and Ebel (1976) determined that 

 sublethal effects such as decreased swimming performance and growth occurred at gas 

 supersaturation levels as low as 106%. Poor swimming performance can result in increased 

 predation by predators in the river. 



Laboratory research conducted by several researchers showed that the threshold levels 

 for supersaturation where direct mortality begins occurring is about 110 to 115% for juvenile 

 salmonids, depending on size and species. In shallow water, laboratory experiments have 

 shown that, for example, at 125% saturation, 50% mortality to Chinook occurs in 13.6 hours. 

 At 120%, 50% mortality occurs in 26.9 hours for chinook. To my knowledge, recent 

 studies do not contradict these results. I note that the U.S. Environmental Protection Agency 

 relied upon some of these studies in esublishing 110% TDG as the water quality critenon. 



92-531 0-96-5 



