the movement of tagged fish between Rock Island 

 Dam forebay and Rocky Reach Dam in 1062 and 

 1903. 



TRAVEL TIME BETWEEN ROCK ISLAND 

 AND ROCKY REACH DAMS 



Fifteen lots of tagf!;o(l fish were iclcasod in the 

 forebay of Rock Island Dam in the combined tag- 

 ging seasons of 19fi2 and 19()3. The reai)i)earance 

 of these lots at Rocky Reach Dam 15 miles upstream 

 varied little; 12 peaked on the seconil day, 1 on the 

 first day, and 2 on the third day after release. 



We plotted the observations of tags at Rocky 

 Reach Dam by 4-hoiir intervals (fig. 7). Only the 



i 



n 



nii 



n 



n 



, o ii ^ 



(^ 60 

 o 50 



i 



n 



In. 



Sow- 9am- lorn Sd"" 



9gm lo» 5pm 9d" 



Isf Oflt OUT 



^om 9om Ipm 5o* 



9om Ipm 5pm 9pn> 



2d DAY OUT 



5om- 9Dn*- Ipm- 5pii 



9om 1pm 5pm 9pm 



3d DAY OUT 



pn 



bom- 90"^ IP""' ^C" 



^om lO"" 5 pm 9pm 



4lh DAY OUT 



Figure 7. — Xumbor of tagged sockeye eountoil over Rocky 

 Reach Dam in different 4-hour periods during the 4 days 

 after release in the Hock Island forebay in l'.)t)2 and KltiH. 

 The periods (shown at the bottom of the graph) were 

 5 a.m.-9 a.m., 9 a.m.-l p.m., 1 p.m.-5 p.m., and 5 p.m.- 

 9 p.m. Counting was discontinued between 9 p.m. and 5.a.m. 



tags observed during the first 4 days after release 

 have been included because tagged fi.sh became 

 scarce after the fourth day. Fish tagged on .luly 

 17 and 19-21, 1902, have not been included because 

 the precise time that these fish passed Rocky Reach 

 Dam was not recorded. The agreement of tlie data 

 for the 2 years is extremely close. For each release 

 the greatest numbers of tags were observed at 1 p.m. 

 to 5 p.m. on the second day after release. If we 

 consider 10 a.m. as the average release time and 

 3 p.m. (the midpoint of the 1 p.m. to o p.m. period) 

 on the second day out as the average time when 

 tagged fish passed Rocky Reach Dam, the modal 

 travel time was 53 hours. Diurnal consistency is 



140 



also evident: seven times in eight the numbers of 

 tags observed increased from the first 4-hour period 

 (5 a.m. to 9 a.m.) to the second (9 a.m. to 1 p.m.), 

 peaked during the third period (1 p.m. to 5 p.m.), 

 and decreased during the last (5 p.m. to 9 p.m.). 



The consistency of the data indicates that fish 

 passage was uniform and orderly at Rocky Reach 

 Dam in 1902 and 1903, ami that the much greater 

 travel time from Rock Island Dam to Zosel Dam in 

 1902 must be attributed to longer travel time above 

 Rocky Reach. Thus, it is necessary to look to the 

 stretch between Rocky Reach Dam and Zosel Dam 

 for the cau.ses of the slow travel time. 



EFFECTS OF W.\TER TEMPERATURE ON 

 THE MIGRATION OF SOCKEYE SALMON 



On .\ugiist 1, 19()2, after sockeye had failed to 

 appear at Zosel Dam despite high coimts at Rocky 

 Reacii Dam, the 133-mile migration route between 

 Rocky Reach and Zo.sel Dams was .searched by 

 plane for schools of salmon. Despite optimum 

 aerial-survey conditions, not a single sockeye 

 salmon was sighted — evidence that the run had not 

 yet entered the Okanogan River. 



On August 2, the following day, I (Major) visited 

 the area on the bank of the Columbia River im- 

 mediately adjacent to the confluence of the Okanogan 

 and Columbia Rivers — a traditional fishing site of 

 the Colville Indians. Of the 8 to 10 Indians 

 present, only 1 responded to questions. lie ans- 

 wered that "blueback (sockeye) were milling in the 

 area and fishing was getting i)etter every day." 



So('keye salmon did not reach Zo.sel Dam imtil 

 August 7 : at that time the counter leported several 

 hundred below the dam and captured 155 in the 

 traps, including 15 with tags. Tag recoveries in- 

 cluded individuals from six of the .seven lots. This 

 l)reakdown of the u.sual chronological order, and the 

 resultant mixing and accumulation of the various 

 segments of the rtni, indicated that the run had been 

 delayed. Information from the aerial search and 

 from the Indian's report pinpoints the delay at the 

 confluence of the Okanogan and Columbia Rivers. 



We hypothesized that the sockeye .salmon had 

 been blocked from the Okanogan River by unfavor- 

 ably high water temperatures until a sharj) tem- 

 perature drop on August 2-3 jiermitted them to 

 enter the stream on or about .Viigust 3 and to reach 

 Zosel Dam on August 7. 



To examine the validity of the general hypothesis 



U.S. FI.SH A.ND WILDLIFE SKRVICE 



