Underwater Separation of Juvenile Salmonids by Size
Total Page:16
File Type:pdf, Size:1020Kb
Underwater Separation of Juvenile Salmonids by Size MICHAEL H. GESSEL, WINSTON E. FARR, and CLIFFORD W. LONG Introduction ways, the fish must pass through a significantly lower than those ofchinook separator that grades them by size. The salmon held or transported with steel Improving the survival of downstream separator is necessary because of the head 1. Not only are most of the larger migrating Pacific salmon, Oncorhyn stress placed on fish when different steelhead separated from the smaller chus spp., and steelhead, Salrno gaird sizes are held together. Steelhead for ex salmon, but other large fish and float neri, is a primary goal of the research ample, are generally larger as smolts ing debris are removed from the system being conducted on the Snake and Co than chinook salmon, Oncorhynchus and returned to the river. lumbia Rivers by the National Marine tshawytscha, or other salmon species. Separation was originally accomplish Fisheries Service. Beginning in 1971, Recent data from stress studies con ed by a dry-type separator (Fig. 2). transportation of these fish around low ducted at Lower Granite and Little head hydroelectric dams has been an im Goose Dams on the Snake River in portant part ofthis effort. Transportation dicate that stress levels of chinook 'D. L. Park, G. M. Matthews, T. E. Ruehle, 1. began by collecting the migrants at Lit salmon held or transported alone were R. Smith, 1. R. Harmon, B. H. Monk, and S. tle Goose Dam on the Snake River and Achord. 1983. Evaluation of transportation and related research on Columbia and Snake Rivers, hauling them by truck to release sites 1982. Unpubl. manuscr., 47 p., on file at the below Bonneville Dam, the dam furthest The authors are with the Coastal Zone and Estua Northwest and Alaska Fisheries Center, National downstream on the Columbia (Fig. 1). rine Studies Division, Northwest and Alaska Fish Marine Fisheries Service, NOAA, 2725 Montlake eries Center, National Marine Fisheries Service, Blvd. E., Seattle, WA 98112. (Prep. for U.S. Army This program was expanded to include NOAA, m5 Montlake Boulevard East, Seattle, Corps of Engineers, Portland, Oregon, under Con Lower Granite Dam on the Snake River WA 98112. tract DACW68-78-C-0051.) in 1975, and McNary Dam on the Co lumbia River in 1978. Barging as an additional means of transportation (McCabe et aI., 1979) also began in 1978. Each ofthese collector dams is equip ped with a fish bypass system (Matthews et aI., 1977; Smith and Farr, 1975) that WASHINGTON diverts fish from turbine intakes to race ways where they are held for later trans portation. Prior to entering the race- ABSTRACT-Juvenile salmonids collected at dams on the Snake and Columbia Rivers require separation by size prior to transpor OREGON tation. A method ofseparating most ofthe smaller downstream migrating juvenile Pa cific salmon, Oncorhynchus spp., from the Release site Collection facilities larger steelhead, Salmo gairdneri, at hydro electric dams is described. The device util 1. Below Bonneville Dam 2. McNary Dam izes behavioral responses that allow fish to 3. Little Goose Dam remain in water during the separation pro 4. Lower Granite Dam cess. This system is presently being installed at collector dams on the Columbia and Snake River. Figure I.-Locations of fish collection facilities, transportation route, and release site. 38 Marine Fisheries Review Overhead View r-------.~ I I 1 Separator 1- ••••••••••• pipes I Interspaces Flow 1- ••••••••••••••--- J Water spray Section Water dissipation screen Mechanized 0 a 0 debris removal I I system 1 -1- --~_ :<L::J--~-~=~-T-==-==- ==-==-:::::::::¥===----=--=!.--~/~r:~sh-~~~ -.--------- -1-- ----- /overflow 7 I - ---------.:¢- i Separator pipes I I I Water level I \1 First Second Third Fourth compartment compartment compartment compartment Fish outlet Figure 2.-Side view of dry-type separators at collection facilities prior to 1982. These early models employed sloping ly larger fish entering the next compart system was employed to aid movement pipes which fanned out so that the ment, and so on. A fine spray of water of the large fish and debris. spaces between the pipes gradually in was directed onto the pipes from above Although effective as a separator, the creased. Fish separated simply by fall to aid movement offish. Large fish and dry-type separator was believed to cause ing through the gaps formed by the debris remained on top of the pipes to its own stress by keeping the fish out of diverging pipes, with the smallest fish ultimately fall into a channel leading water during the separation process. The entering the first compartment, slight- back to the river. A mechanized brush system also required constant attention 47(3), 1985 39 Separator pipes 0000000000 Water dissipation 3.9 m screen Water level Auxiliary water supply Fish outlets 10.16 em Figure 3.-Side view of prototype wet separator at Little Goose Dam, 1981. to prevent debris from wedging between first reaction is created by water jets end of the separator and return to the the separator pipes and possibly trap flowing from holes in plastic pipes river or, if desired, they can be diverted ping fish. To alleviate stress on the fish, placed beneath the grid of submerged into a collection raceway. we developed a device for separating separator pipes in each compartment Water flowing up from the jets tends them underwater (wet-type). Initial (Fig. 3 inset). This configuration pro to keep debris from accumulating on the model studies at Lower Granite Dam in duces an attraction flow toward the separation pipes. It also contributes to 1980 were followed by tests of a proto separator pipes. the flow required to move floating debris type at Little Goose Dam in 1981. This Fingerlings moving through the by along the surface of one compartment report describes the wet separator and pass system enter the separator compart to the next and eventually off the end the results obtained during the model ments at the surface after moving down of the separator. studies and tests. an inclined screen. Their response to Individual pipes, each with air-oper swim into the flow from the water jets, ated gate valves and a maximum head Design and Evaluation as well as their tendency to sound, of 60.96 cm (24 inches), supply water of the Wet Separator allows smaller fish to pass volitionally to the flow jets and the auxiliary water The prototype wet separator is illus through the spaces between the first set supply to the compartments. Flow jets trated in Figures 3 and 4. Two observed of separator pipes. Larger fish tend to consist of 0.476 cm (J;6-inch) holes at reactions of smolting salmonids were move along these pipes and eventually 10.16 cm (4-inch) centers in the top of considered in its design: Their normal enter the next compartment which has 5.08 cm (2-inch) diameter pipes. These response to orient into a water flow and wider pipe spacing. If the fish are too pipes are set at 15.24 cm (6-inch) inter their tendency to sound, or dive, as an large to pass through the pipes in the vals across the width of each compart avoidance reaction. The stimulus for the second compartment, they pass over the ment. 40 Marine Fisheries Review Railing Walkway Water level Separator pipe 000000000000000000000000000000000 00 0 0 0 0 0 0 0 0 o 0 0 0 0 0 0 0 00 Flow jet Perforated plate Fish outlet Figure 4.-End view of prototype wet separator at Little Goose Dam, 1981. Separator pipes are metal tubes cov and was based on counts of juvenile of water from the dissipation screens ered with plastic for smoothness (overall chinook salmon and steelhead observed and flow jets remained constant, and the outside diameter = 2.22 cm or ;Is-inch). in the two hoppers. There was effective auxiliary water supply was adjusted so Initially, the interspace was 1.59 cm (% separation in the model studies con that a uniform flow occurred along the inch) in the first compartment and 3.175 ducted at Lower Granite Dam, but sep surface of the separator and off the cm (1Y4-inches) in the second. Later the aration with the prototype at Little downstream end. interspace in the first compartment was Goose Dam was less effective. At Little Goose Dam, correct separa reduced to 1.27 cm O~ inch) to minimize At Lower Granite Dam, 90 percent of tion of chinook salmon into the first the numbers of small steelhead in the the smaller chinook salmon went into compartment ofthe prototype separator first compartment. Spacers are used to the first compartment and 88 percent of averaged only 73 percent (range 60-82 maintain gap size. Each grid of sep the steelhead into the second compart percent) (Table 1). Lack of adequate arator pipes is adjustable vertically to mentz. During these tests, the quantity control of the water flow into the proto achieve the desired slope and water type separator resulted in surges of depth at the downstream end. Submer water that carried chinook salmon be 'D. L. Park, 1. R. Smith, G. M. Matthews, T. E. gence ranged from about 10 cm at the Ruehle, 1. R. Harmon, S. Achord, B. H. Monk, yond the first compartment and ap upstream end of the separator pipes to and M. H. Gessel. 1982. Transportation opera peared to be the main reason for the tions and research on the Snake and Columbia approximately 1 cm at the downstream Rivers, 1981. Unpubl. manuscr., 34 p., on file at poor separation. The prototype separa end. the Northwest and Alaska Fisheries Center, Na tor was actually part of the fish collec Evaluation of both the model and tional Marine Fisheries Service, NOAA, 2725 tion system at Little Goose Dam, and Montlake Blvd.