11 



Senator Kempthorne. Mr. Mundy, thank you for your comments. 

 Now we have Mr. James J. Anderson. 



STATEMENT OF JAMES JAY ANDERSON, UNIVERSITY OF 

 WASHINGTON, SEATTLE, WA 



Mr. Anderson. Thank you, Senator Kempthorne, for inviting me 

 to offer this testimony. 



I'd like to describe to you some of the results from a project that 

 I have been working on for 7 years, the Columbia River Salmon 

 Passage Project. We have developed some models to describe the 

 movements and survival of fish through the hydro system and 

 through their life cycle. This project has been rather extensive. 

 About 20 people are working on it, including undergraduates and 

 graduates. Several Ph.D. theses have come out of this work. We are 

 actively engaged in trying to understand in terms of the mathe- 

 matics and the underlying data how fish move through the river, 

 and the consequences of the actions we take on the fish. 



I want to emphasize today that these are a synthesis of the types 

 of information we have developed over quite a long time. A number 

 of people have been working on this. 



One of the things we found is that spill at low levels produces 

 low levels of gas. When you reach a certain level, the system 

 changes and the gas becomes lethal. I believe that fish have adapt- 

 ed to situations in nature where they can handle the levels of gas 

 below this critical level. In the Columbia River, we have the capac- 

 ity to produce very large levels of gas. I believe our primary con- 

 cern is to keep the levels lower. 



In a sense, it's like walking up a hill; you get a small benefit 

 from spill for low levels of gas. You reach some cliff and if you go 

 beyond that, then you have a real problem. There is a chance that 

 we have gone beyond that this year. 



There are several observations that came out of this year's pro- 

 gram. One is that we had high gas levels below Ice Harbor, up 130 

 percent of supersaturation. Another element is we had very high 

 mortalities in pens and fish in pens below the dam up to 100 per- 

 cent in some cases. In deep cages, they had 50 percent mortality. 



Another observation was we saw very few effects of the gas on 

 the fish downstream. How can we explain the high cage studies 

 with the few effects downstream. The model we have allows us to 

 balance these things and interpret both of these in terms of a con- 

 sistent story. With the model, I could represent the cage studies 

 and show the movements of fish through the river system. With 

 the model, we are able to get the same levels of cage mortality or 

 pen mortality, and then we project downstream. What we find is 

 a very small but negative impact of the spill which means that ac- 

 cording to our work, we're probably not affecting fish very much 

 with the current levels of spill. One reason is that the fish are 

 deeper in the water. We assume they are at average depth at 30 

 feet in which case, you maybe lose in total survival through the 

 system, 1 percent of the fish with the spill in the Snake River. 



What concerns me is if the fish are shallower in the water, such 

 as 10 feet, as Mr. Mundy suggests they might be, then the amount 

 of loss you get would go upwards of about 4 percent. The higher 



