flowing over the spillways. Because of the immi- 

 nent danper to future runs, the Bureau of Com- 

 mercial Fisheries has a researcli program under 

 way to develop methods for protecting fingerlings 

 as they pass thi'ougli turbines. 



From the outset of this program, knowledge of 

 the behavior of fish immediately upstream of tur- 

 bines was required. Information on diel movement 

 was essential for obtaining nioro precise measures 

 of total mortality and for estimating the feasibility 

 of different methods for protecting fish. Research 

 at Big Cliff Dam by the State of Washington De- 

 partment of Fisheries (footnote 1) showed that 

 mortality at two turbine loads (40 ver.sus 80 per- 

 cent wicket gate openings) differed significantly. 

 Because turbine loads fluctuate daily, it was obvi- 

 ous that mortalities also might vary daily. The 

 estimation of a\-erage daily mortality required 

 knowledge of (1) mortality for a wide range of 

 turbine loads, (2) daily fluctuations of turbine 

 loads, and (3) daily variation of fish movement 

 through turbines. If the relation of these three fac- 

 tors were better understood, perhaps mortality 



could bo minimized by manipulation of turbine 

 loads when most of the migrants are passing down- 

 stream at damsites. 



Another facet of fish behavior, important to the 

 development of protective methods, is the route 

 used by most of the fish. Studies of models showed 

 tliat flows through turbine intakes and associated 

 scroll cases of dams such as McXaiT were well 

 ordered; e.g., flows near intake ceilings pass by the 

 top of the wicket gates, and flows near intake floors 

 pass by the bottom of the wicket gates. Because, the 

 turbine blades lie only a few meters farther down- 

 stream, it is probable that distribution of flows at 

 the blades corresponds with distriI)ution of flows at 

 the wicket gates; i.e., flows from the top of the 

 wicket gates pass the blades at the hub, and flows 

 from the bottom of the wicket gates pass the Ijlades 

 at their tips.= Thus, the distribution of fish at the 

 turbine blades might be deduced with some de- 

 gree of accuracy from the distribution of fish in 

 the intakes. "Wlien fish distribution is known, meth- 



= Personal communication. Johnson, G. Dugan. AUIs-Clialmers 

 ^[aniifacturing Co., York, Pa. 



CANADA 



WASHI NGTON 



ROCKY REACH 



ROCK ISLAND 



WANAPUM/ 

 PRIEST RAPIDS 



WELLS 



<%^8 



LOWER 



IDAHO 

 I W-^E 



:; 

 I 



I 

 I 



MONUMENTAtr^^^iq'ffciLOWER GRANITE 



THE DALLES JOHN DAY 

 BONNEVILLE / 



BEN FRANKLIN\ 



LITTLE GOOSE 

 ■ICE HARBOR 



^.v)iO^ 



'mcnary 



OREGON 



APPROXIMATE SCALE 8 16 KILOMETER 



iKK) 



i^iQUBE 1.— Location of present and future low-head dams on main stem, Columbia and Snake Elvers. 



U.S. FISH AXD WILDLIFE SERVICE 



