the naturally prevailing Columbia River levels dur- 

 ing the test. The shallow control tank was supplied 

 with gas equilibrated Columbia River water. The 

 following is a list of water quality parameters that 

 were monitored in each test tank during the N 2 

 bioassay studies: 



1. Dissolved oxygen-0 2 



2. Nitrogen gas-N 2 



3. Carbon dioxide-C0 2 



4. Ammonia-NHj 



5. Conductivity 



6. Alkalinity 



7. Turbidity 



8. pH 



With the exception of N 2 , all parameters re- 

 mained within safe biological ranges throughout 

 test period. 



Examining Table 2, we see that survival was 

 bette; in the deep tanks. This is what one would 

 expect knowing that as hydrostatic pressure 

 increases (with water depth) the percent of nitro- 

 gen decreases. In these tests the crappie and 

 squawfish were the most tolerant while the smelt 

 were the least tolerant. Within the salmonids 

 tested, the cutthroat were slightly more tolerant 

 than either the chinook or steelhead, while the 

 latter two species showed comparable tolerance. 

 These general conclusions apply only to the N 2 

 levels as indicated in Table 2. 



Small Tanks— 130, 120, 110, and 100% 

 N 2 Saturation 



The tests in the 175 V. tanks were designed to 

 provide added information on the effect of nitro- 

 gen on fish survival. Thirteen bioassay tests have 

 been done with nine species. Groups of from 5 to 

 20 fish were held in separate tanks at 130, 120, 

 110, and 100% N 2 saturation. Each test was con- 

 tinued for 192 hr during which mortality was 

 recorded. As with the preceding tests (deep versus 

 shallow tanks) water quality parameters remained 

 in acceptable biological ranges with the exception 

 of N 2 . Test temperatures were between 10 and 

 13°C. Table 3 summarizes the species of fish used 

 and the time (hr) to 50% mortality at 130 and 

 120% N 2 saturation. The 50% death level was not 

 reached in any of the 110% N 2 saturation tests 

 nor did any mortality occur in the 100% N 2 sat- 

 urated control tests. Bass and crappie were the 

 most tolerant of the species used in these tests 

 while smelt were the least tolerant. At the 130% 

 N 2 level the rainbow and steelhead trout were the 

 most and least tolerant, respectively, of the salm- 

 onid species. The remaining salmonids, including 

 whitefish, seemed to be grouped at around 24 hr 

 survival for 50% of the test animals subjected 



to 130% N 2 saturation. The ranking changes at the 

 120% level, but not drastically (Table 3). These 

 tests reflect general trends, and one could sur- 

 mise that the synergistic effect of various stresses 

 (temperature, disease, maturity, etc.) could alter 

 the pattern derived from these series of tests. 



TABLE 3 Species and Time, in hr, to 50% Mortality lor 



Groups of Fish Held at 130, 120, 110, and 100% N, Saturation. 



Included is a ranking from most to least tolerant. 



' = No 50% mortality level 



1 Number in parenthesis indicates ranking of tolerance 



DESCRIPTION OF VERTICAL DISTRI- 

 BUTION OF FISH USING DEPTH 



SOUNDING GEAR 



In 1972 a Benmar depth sounder was modified 

 for use in a 1.8 x 3 m redwood test tank. Pre- 

 liminary tests in the tank indicated that the sound- 

 ing gear would work satisfactorily to determine 

 depth distribution of fish. After examining the 

 resulting tapes of several 24-hr tests we found 

 that the fish were generally below 1 m water depth 

 (this would enhance their survival from that in a 

 1 m deep tank). Two transducers were used in the 

 test tank; one at the water surface and one on the 

 bottom. A printer/counter system provided a fish 

 count for each 0.6 m interval of water in the 2.5 m 

 tank. (A description of the technical aspects of the 

 system is attached to this report.) 



The next step was to test this equipment in 

 the river. The two-transducer- arrangement was 

 modified, to a 10-transducer array (Fig. 3) which 

 could be placed on the bottom of the river at gently 

 sloping beaches. This configuration was used at 

 two locations near Prescott for a total of 75 hr 

 during day and night. Fig. 4 summarizes the 

 results: of 776 fish approximately 72% were 

 detected between 0.9 to 2.1 m (3 and 7 ft). Many 

 more fish were detected during darkness than day- 

 light. While species could not be differentiated by 

 the sounder, a minimum seine effort (2 sets) on the 



14 Blahm, McConnell, Snyder 



