FISHWAY RESEARCH AT THE FISHERIES-ENGINEERING RESEARCH LABORATORY UNITED STATES DEPARTMENT OF THE INTERIOR FISH AND WILDLIFE SERVICE BUREAU OF COMMERCIAL FISHERIES CIRCULAR 98 Cover: Bonneville Dam, Bradford Island fishway in the foreground. Fisheries-Engineering Research Laboratory may be seen at the for end of the spillway. UNITED STATES DEPARTMENT OF THE INTERIOR FISH AND WILDLIFE SERVICE BUREAU OF COMMERCIAL FISHERIES FISHWAY RESEARCH AT THE FISHERIES-ENGINEERING RESEARCH LABORATORY By Gerald B. Collins and Carl H. Elling U.S. FISH AND WILDLIFE SERVICE CIRCULAR 98 Washington, D.C. November 1960 For sale by the Superintendent of Documents, U.S. Qovcmment Printing OfBce Washington 25, D. C. - Price 20 cents CONTENTS Paee Abstract ii Introduction 1 Fishway capacity 6 Fishway slope and fishway length 8 Fish swimming abilities 10 Attraction of fish 11 Fishway hydraulics 11 Effect of light on fish movement 12 General observations 14 Current research 15 Appendix 16 ABSTRACT Results of 4 years of research on fishway problems, data on rates of movement of salmonids ascending fishways, and of spatial requirements of fish are given and experi- ments to measure fishway capacity are described. The effect of fishway slope and fishway length on fish performance and biochemical state were measured in "endless" fishways. No evidence of fatigue was found when proper hydraulic conditions were obtained. One salmon ascended over 6,600 feet vertically. Experiments to measure swimming abilities of salmon indicated that the critical velocity was between 8 and 13 feet per second. Maximum observed swimming speed was 26.7 feet per second. Preferences of salmonids for water velocities and light conditions revealed marked differences between species. Effects of light and water velocity on rates of iiassage through channels and fishways are described. Experiments involving fingerling passage problems and the testing of fuUscale prototype fishway designs are illustrated. Reports and publications on laboratory research are listed. FISHWAY RESEARCH AT THE FISHERIES-ENGINEERING RESEARCH LABORATORY by Gerald B. Collins and Carl H. Elling Fishwiiy problems are many and complex on the Columbia River system where a long series of major dams interrupts the migration of several species of anadromous fishes. Adult fish returning to the Columbia from the sea may have to ascend as many as nine dams to reach their spawning areas. Young fish must pass downstream over all of these dams on their journey to the sea. Even small losses, injuries, or delays in the passage over each dam could threaten the entire fishery resource because of the cumulative effects of many dams. Similarly, the costs involved in providing adequate fishvvays to pass fish safely over a large dam must be multiplied by the increasing number of dams. It is therefore highly important that fish passage facilities be designed with both a maximum of safety for fish and also a maximum of economy in construction and operation costs. To accomphsh these goals obviously requires a sound basic knowledge of the behavior, abilities, and require- ments of migratory fish, particularly in relation to fish ways. To supply precise information on the behavior and performance of migrating fish, a special type of laboratory was constructed ' in wliich it is possible to measure the reactions of fish under controlled experimental conditions while the fish are actually migrating. The Fislieries-Engineering Research Laboratory adjoining one of the nnijor fishways at Bomieville Dam (fig. 1) is the only Note. — Gerald B. Collins and Carl H. Idling, Fishery Research Biologist.s, Bureau of Commercial Fisheries, U.S. Fish and Wildlife Service, Seattle, Wa.shiiigton. ' Financed by the U.S. Army Corps of Kngineers as a part of their Fi.sheries-Kngineering Research Program for the purpose of providing design criteria for more econom- ical and more efficient fish-passage facilities at the Corps' projects on the Columbia River. laboratory of its kind in the world. Fish diverted (fig. 2) from the Washington-shore fishway at Bonneville Dam swim into this laboratory (fig. 3) where their responses to full-scale fishway situa- tions are observed and recorded. Fish then swim out of the laboratory and re-enter the main fishway to continue their migration upstream. The laboratory basically consists of a level experimental flume (fig. 4) with a fish collection pool at the downstream end that is connected to the main fishway by a small entrance fishway ("B" in fig. 2), and with a flow introduction pool at the upstream end that is connected to the main fishway by an exit fishway ("F" in fig. 2). Vari- ous tjT^es of fishway structures are erected (fig. 5) in the experimental area while it is dry, then water is introduced and the gates to the main fishway are opened to permit the entry of fish. A water supply and discharge system is independent of the main fishway and is capable of delivering and discharging up to 200 cubic feet of water per second without disturbing the flow pattern of the main fishway outside. Light control is provided by a completely covered building and eighty 1,000-watt mercury-vapor lamps (fig. 6) that under standard operating conditions produce illumination equivalent to a cloudy bright day. The ability to control large flows, water levels, structures, and light makes it possible to create a wide variety of test coiulitions. Adult migrating fish are available to tlie laboratory for approxi- mately 6 months of the year. Migrants include Chinook salmon {Oncorhynchus tshawyt.scha) , blue- back salmon {0. nerka), silver salmon (0. kisutch), steelhead trout {Salmo (fairdneri), shad {Alosa sapidisdma), and also the Pacific lamprey (Lam- petra tridentata). $ -% -.---- Figure 2. — Sketch of Fisheries-Engineering Research Laboratory showing its relationship to the Wash- ington-shore Fishway. Fish ore diverted From the main Fishway by a picketed lead (A) and ascend the entrance Fishway (B) to a collection pool (C) in the laboratory. AFter release, they pass through an experimental area (D) to the Flow introduction pool (E) and then out the exit Fishway (F) where they return to the main Fishway. Insert shows plan view oF laboratory. Figure 1. — Fisheries-Engineering Research Laboratory adjoining Washington-shore fishway at Bonne- ville Dam. The dam is on the Columbia River 140 river-miles From the sea. Figure 3. — Entrance to the fish collection pool. Fish swim into the laborotory through a narrowing funnel that prevents them from leaving. Figure 4. — interior of Fisheries-Engineering Research Laboratory when empty and unwotered. Experimental area (center) is 104 feet long, 24 feet wide, and 1 7 feet deep. Fish collec- tion pool at far end is 50 feet long and 24 feet deep. Figure 5. — Experimental fishways under construction. Careful planning and great ingenuity are required in building temporary structures capable of sup- porting and controlling large volumes of water. '« n Figure 6. — Full-scale experimental fishways in op- eration. Standard ligf)ting conditions are created by mercury-vapor lamps. Picketed barriers and release gates for the control of fish have yet to be erected. The Fisheries-Engineering Research Laboratory is now (1960) in its fifth year of full-scale opera- tion. The following is a review intended only as a brief summary of researcli to show the scope and general progress of tlie studies at the laboratory. Listed in Appendix A are the proposals, reports, and publications tliat describe in detail the major aims and objectives of the project, the experi- mental designs and procedures followed, complete observations and results with tlie limitations and qualifications of tlic data and conclusions. Re- search at the laboratory is planned and conducted as a team effort, with the task of reporting divided among the staff by assignment. Selection of major research items to be assigned priority or studied in greatest detail follow the recommenda- tions of the Corps of Engineers with the advice of the Technical Advisory Committee for the Corps' Fisheries-Engineering Research Program composed of representatives of State and Federal fishery agencies. .5.S0816 o— 61- FISHWAY CAPACITY In a pool-type fishway, "capacity" (i.e., tlie maximum number of fish of a given size that a fishway of specified design and dimensions can pass per unit time) is controlled by the rate of fish movement from pool to pool and the space re- quired for each fish. Examples of laboratory data on rates of movement for chinook salmon, blueback salmon, and steelhead trout obtained from a wide variety of experiments are shown in table 1. Examination shows that although rates vary with species and time of year there is considerable consistency in the average rates of movement under a wide range of experimental Figure 7. — The 4-foot wide, 1-on-16-slope fishway used for fishway capacity tests. Passage of 3,000 salmonids per hour was demonstrated in this fishway. Table 1 . — Passage times per pool of individual and groups o( salmonids oscending experimental fishways ' at Bonneville laboratory Time per pool by species and source ^ of data Spring Chinook Blueback Steelhead Fall Chinook Min- utes 2.2 2.7 3.0 3.2 2.0 2.4 Source 5, table 1 6, table 1 10, No. 22 10, No. 22 10, No. 46 10, No. 46 10, No. 47 10, No. 47 16, table 1 Min- utes 1.7 1.6 1.3 2.8 1.2 1.2 Source 5, table 1 5, table 1 10, No. 11 10, No. 24 10, No. 24 10, No. 36 Alin- utes 1.6 2.2 1.0 2.0 1.6 2.4 1.7 1.4 1.3 Source 6, paee 1 6, paec 1 10, No. 11 10, No. 13 10, No. 13 24, table 3 24, table 3 24, table 2 24, table 2 Min- utes 1.6 1.5 2.0 1.1 1.1 Source 10, No. 38 10, No. 38 10, No. 40 10, No. 14 10, No. 14 1.9 1.9 3.0 ' Pool-and-overfall-typc flshways with slopes of 1 on 8 and 1 on 16, varyinR in width from 3 to 11.5 feet. No orifices in flshways, pool depth approximately 6.5 feet, rise between pools 1 foot and head on weirs 0.8 foot. Total rl.se varyins from 6 to 6,600 feet. ' See Appendix A. Figure 8. — The 16-pool, 1-on-8-slope endless fishway nearing completion. Worker stands in fish entry gate. Locking unit for lowering and recycling fish appears in center. conditions. A basis for estimating spatial require- ments is provided bj' data (table 2) from a series of capacity experiments in 1957. In these experi- ments large numbers of fish were collected over a 48-hour period and then released in a 1-hour test. During one such experiment, fish averaging 9.2 pounds in weight were passed through a l-on-16- slope fishway (a fishway that rises 1 foot for every 16 feet of fishway length) only 4 feet wide (fig. 7) at a rate of 3,000 fish per hour without any indica- tion that capacity had been reached. Experi- ments in the Washington-shore fishway at Bonneville examining possible effects of the collection-and-release technique upon fish per- formance, although not yet completed, appear to confirm the laboratory data shown. "if- »ak "K I HL^t: t^ ijy.^ Figure 9. — The two endless (ishways, 1-on-16 slope on leFt and 1-on-8 slope on right. Observers along walkways record progress of fish in the re- spective fish ways. r Diffusion Chamber Entronce Frshwoy Entry Chonnel (elewSO') To Exit Fishwoy - Direction Fish Moves l-on-l6 l-on-8 Slope Slope __-— By-poss Channel (elev 50) -Spillout - Fistiwoy Enfry Gofes -introductory Pool Gate ■ Releose Compartment " Introductory Pool "Collection Pool Figure 10. — Plan view oF the 1-on-16- and 1-on-8- slope endless Fishwo/s with auxiliary approach chan- nels and pools. Tabic 2. — Observed space utilization in two fishway capacity tests, 1957 Species Estimated weight- pounds Cubic feet per fish in first pool Source ' 14.0 9.2 2.6 2.2 16, table 3. Chinook, steelhead, biuebiick 16, table 3. J StT Appendix A. FISHWAY SLOPE AND FISHWAY LENGTH Initial experiments comparing the performance of salmonids in fishways with slopes of 1 on 8 and 1 on 16 indicated a higher rate of passage in the steeper slope fishvvay. However, the tests were conducted in short segments of fishways and the possibility that the increased rate was the result of turbulence and lack of resting area had to be con- sidered. To make a further comparison of the Figure 1 1 . — Biolosist extracting a sample of blood from a fish exercised in one of tfie endless fishways. Blood samples were analyzed for lactate and inorganic phospliate to determine if the fish was fatigued. eflfect of slope upon rates of fish movement and to measure the effect of length of fishway on fish per- formance, experiments were undertaken using a pair of "endless" fishways with slopes of 1 on 8 and 1 on 16. These endless fishways (figs. 8 and 9) were pooj-and-overfall fishways constructed so that each made a complete circuit (fig. 10), with the highest pool connected to the lowest pool by means of a lock. When a test fish had ascended to the top of one of these fishwaj's, it was rapidly lowered by lock to the lowest pool to ascend again. By repeating this procedure, fishways of any desired length could be simulated. Comparisons were made on the basis of fish performance and also on the basis of biochemical indices of fatigue such as lactate and inorganic phosphate of the blood (fig. 11) and muscle. No evidence of fatigue was found in either fish- way when the proper pool flow conditions pre- vailed. Blood lactates, the most sensitive of the biochemical measurements, showed (fig. 12) a moderate increase (lactate levels above 125 mg. percent may be lethal to fish under certain circum- stances) during active ascent and were back to the control level in both fishwaj's within 1 hour. Most of the fish were tested with an ascent of approx- imately 100 pools. However, a limited number were permitted to make extended ascents exceed- ing several hundred pools and at least four of each species were allowed to ascend more than 1,000 pools. One blueback salmon was permitted to ascend for over 5 days, climbing continuoush" over 6,600 pools before the test was terminated. This was a vertical ascent of more than a mile in a l-on-8-slope fishway. The conclusion drawn is that ascent of a properly designed fishway is only a moderate exercise for the fish, possibly similar to swimming at a "cruising" speed that can be maintained over long periods of time. The exaggerated pattern of work-and-rest that appeared in the l-on-8 endless fishway resulted in the fish spending about 70 percent of the time resting in the turn pools (fig. 13). In an actual fishway this would result in about 70 percent of the fish being in resting pools at any given time. To avoid this impractical condition, the hydraulic pattern in the l-on-8 fishway was changed by modifying the weirs. This changed the pattern of movement so that the fish rested in each pool (fig. 14), clearl^v demonstrating the importance of pool flow pattern to fishway design. The steeper 8 BLUEBflCK Rq f^^^ Controls I I l-on-8 Slope I I l-on-16 Slope I i Control Terminated Volitional Control Terminated Volitional Figure 12. — Compariscn of blood lactate levels of blueback and chinook salmon ascending 1-on-8- and 1-on-16-slope endless fishways. Note that fish actively ascending ("Terminated") show an increase in lactate levels typical for each species, and that lactate levels of fish that have stopped for 60 minutes of their own volition ("Volitional") are not significantly different from levels of fish that have not been exercised in a fishway ("Controls"). l-on-8 slope was shown to be as suitable for the passage of salmonids as the l-on-16 slope when the proper hydraulic conditions were provided. As to the effect of fishway length on the rate of fish movement, the evidence shows that rates of move- ment tend to increase slightly in the initial stages of a prolonged ascent, probably due to learning, and then become quite consistent. An example is given in figure 15. Note that the blueback sal- mon had ascended over 5,000 feet before slowing down to its initial rate of movement. This means that for all practical purposes the rate of move- ment of ascending fish will not decrease in the upper end of a long fishway and so result in crowd- ing or delay. Experiments at tlie laboratory support two fur- ther generalizations on the rate of fish movemont in fishways. The first is tliat ascending fish show a tendency to do a certain amount of work in a given amount of time regardless of the slope of £ ee ^. 0 o 65 S 64 y^ o 0 c 63 S 6? l-on -le slope (n = 91 / 0 ; 60 ° 69 ^^>/ _ _ 0 /T "" ° 58 / 0 S. 57 / o l-on-B slope (n =7 ) £ 56 ^ " 55 § 5-. ^ J 52 ^Z—-"^^ ^-- CHINOOK UJ 51 Loch e. 10 15 20 25 30 Titne in minutes Figure 13. — Pattern of ascent in the l-on-8- and 1-on- 16-$lope endless fishways when full overfall weirs and resting turn pools were in use. Note exag- gerated "work-rest" pattern in 1-on-8-slope fishway. c Ti o V E 65 64 63 62 61 60 59 »> 58 1 ^^ 0 56 S 55 t ^^ .ifri! 2 52 1 51 w Lock ^ o>^ A oV •S* l-on-8 slope (nOI) Y o f l.on-16 slope ( n:9) -/ O ft - / ?f ■°j^ CHINOOK r (Alternate Restricted Weirs) 10 16 20 25 30 Time in minutes Figure 14. — Pattern of ascent in the l-on-8- and 1-on- 16-slope endless fishways after rest areas were removed from turn pools and with restricted overfalls in the 1-on-8-slope fishway. Iiiiiiiiiluil 0^ 201 » 205 JOi • 30b 401 • d05 Dote July I July 2 Pools oscended previously 1600 July 3 3200 July 4 4800 July 6 64 00 Figura 15. — Passage time per circuit (16 pools) of a blueback salmon that ascended over 6,000 feet in the 1-on-8-slope endless fishway. the fishway. The rate of ascent in a l-on-8-slope fishway of proper hydrauUc design is approxi- mately the same as in a l-on-16-slope fislivvay. The second is tliat for practical purposes the rate of fish movement in fishways is independent of the numbers of fish. Groups of fish in the labora- tory tests moved as fast as individual fish. This behavior pattern reduces the concern that the rate of fish movement might drop suddenly if a fishway became crowded. FISH SWIMMING ABILITIES Tests with adult chinook salmon and steelliead trout have shown that all fish negotiated an 85- foot flume when velocities were approximately 8 feet per second. When the velocity was increased to 13 f.p.s. approximately 50 percent of the chinook and 9 percent of the steelliead failed to pass the flume. Thus, somewhere in the range of 8-13 f.p.s., we can expect some of the fish of these species to be blocked when the distance to be negotiated approaches 100 feet. Table 3. — Response of chinook and silver salmon and steelhead trout presented with a choice between entering a high- or a low-velocity channel Test condition Number of fish tested ' Chose high-velocity channel High- velocity channel Low- velocity channel Steel- head trout Chi- nook salmon Silver salmon Steel- head trout Chi- nook salmon Silver salmon F.p.s. F.P.S. Percent Percent Percent 8 2 258 100 12 79.4 93.0 83.3 8 4 249 80 14 59.0 67.5 85.7 8 6 266 66 13 52.2 45.4 46.1 6 2 264 139 24 73.1 87.1 83.3 4 2 253 134 5 67.2 73.4 100.0 6 4 257 69 22 63.8 62.3 68.2 Figure 16. — Experimental channel with a wafer veloc- ity oM 6 feef per second appears on right. Entrance to channel on left is screened to prevent access during swimming ability tests. I Includes all size groups. Additional tests were conducted in which sal- monids were subjected to a velocity of approxi- mately 16 f.p.s. (fig. 16). Marked declines in performance were noted. About 95 percent of the chinook salmon and approximately 50 percent of the steelhead failed to negotiate the 85-foot flume. Differences in performance with respect to size were also noted. Two size groups were considered (1) fish estimated at 24 inches and under, and (2) fish over 24 inches. Respective performance (distance negotiated) of the two groups in velo- cities of approximately 13 and 16 f.p.s. are shown in figure 17. Clearly, "large" fish were capable of greater performance than "small" fish. Measurements of the rate at which salmonids travel in a channel under a variety of water velocities suggest that a velocity of 2 f.p.s. may be most satisfactory for transportation purposes. These experiments utilized fall chinook salmon, silver salmon, and steelhead trout which were subjected to water velocities ranging from 2 to 16 f.p.s. Rate of movement in relation to the channel is shown in figure 18. Chinook salmon made their fastest progress at 2 f.p.s. and exhibited a progressive decline in rate of movement up to 13 f.p.s. Conversely, steelhead trout and silver salmon indicated an increase in performance as the water velocity increased up to 8 f.p.s. The three species tested registered approximately equal median rates of movement (5-f- f.p.s.) in a water velocity of 8 f.p.s. Any slight advantages the higher flows may have in expediting the move- ment of certain species are minimized by the fact that considerably more water would be required 10 Lorqe SteelheoO * Smoll Sieelheod ] Large Chinook Smoll Chinook 10 20 30 40 50 60 70 80 90 10 20 30 40 50 60 70 80 90 Disfonce ( in f eet) negotioled in channel Large Sleelhead Smoll Sleelhead Lafqe Chinook Small Chinook Responses of the nature indicated in lliesc tests suggest the possibility that upstream migrating salmonids may actually be diverted from fishway entrance flows into those of the spillway or even into the turbine discharges under certain hydraulic conditions. Experiments at Bonneville exploring the preference of adult migratuig salmonids for light conditions have demonstrated that pronounced species differences prevail. Sleelhead presented with a choice of entering a light or dark channel exhibited a marked preference (80 percent) for the dark channel. In contrast, chinook salmon entered the light and dark flumes in nearly equal proportions, indicating no particular preference for either condition. In both instances, fish were light adapted prior to exposure to the test condition. Figure 1 7. — Swimming performances of cfiinook sal- mon and steeltiead trout by size in wafer velocities of approximately 13 and 16 feet per second. for transport. It might also be expected that fish would have to expend more energy in passing through the higher velocities. Ma.ximum swim- ming speeds in relation to the water were 26.7 f.p.s. for steelhead trout and 22.1 f.p.s. for chinook salmon measured over a distance of 30 feet. ATTRACTION OF FISH FISHWAY HYDRAULICS The importance of controlling hydraulic condi- tions in fishway pools and channels has been brought to light in several experiments at the Bonneville laboratory. A change in fishway flow from a plunging to a streaming condition was found to halt temporarily almost all movement for a period of several minutes. Such information indicated that unstable flow conditions could seriously interfere with fish movement in fishways. Chinook and silver salmon and steelhead trout presented with a choice between entering channels carrying either a "high" or a "low" water velocity demonstrated a significant preference for tlie high-velocity channel in virtually every test. The response to the high velocity is given in table 3. Following these tests an additional experiment was conducted in which a sample of chinook salmon and steelhead trout were presented with a choice between flows of approximately 3 and 13 f.p.s. The high-velocity chaimel was chosen by 89.5 percent of the chinook and by 75.6 percent of the steelhead. The choice of the high-velocity channels by the chinook salmon is of particular interest since approximately half of the total sample of 51 fish failed to negotiate the channel after entry. Fish swept back after failing to pass the flume again selected the higher velocity on their second attempt. Demonstration of the importance of maintaining a uniform flow in fishway channels was evidenced in a series of tests in which fish were passed through an open channel approximately 2 feet deep. •5 6 o o c 3 — o o u a> V - " 4 E Z I I Steelheod Troul Chinook Soimon Silver Solmon i Woler velocity 8 134 in feel per second Figure 18. — Salmonid rote of movement in an open cf)annel with water velocities ranging from 2 to 1 6 feet per second. 11 When flow3 were uniform, evidence of interrupted passage of fish was lacking. When hydrauhc jumps were estabHshed in the channel a number of fish were observed to linger in the low-velocity areas created by these jumps. Occasionally fish remained in these areas for several minutes before continuing their movement tlirough the channel. Improper design and operation of diflfusion chambers conceivably may create similar disturbances in the flows of collection and entrance channels, giving rise to delays in fish movement comparable to those occasioned by the hydraulic jumps in these tests. Tests examining the performance of fall chinook salmon in a fishway under a uniform plunging or streaming flow indicated there was no significant difference between the rate of ascent in the two flow conditions. Respective rates of ascent under plunging and streaming flows were 37 and 34 pools per hour. Table 4 — Chinook solmon and steelhead trout rate of movement in light and dark channels. Water velocity — 4 feet/second, distance measured — 30 feet .MWf^^ Rate (feet per second) in relation to floor Cliannel condition Steelhead Chinook n Mean Median n Mean Median Dark. Light 22 37 F.p.l. 0.6 3.8 F.p... 0.6 4.3 20 20 F.p.,. 0.5 5.2 F.p.,. 0.7 5.3 Figure 1 9. — Covered fishway used in darkened passage experiments. All laboratory lights were turned off during dark tests. The influence of hydrauhc conditions in the pools on the pattern of ascent in the endless fishways was previously noted. The change in pool hydraulics brought about by restricting the length of the weirs in the l-on-8-slope fishway was sufficient to correct the pattern of movement to that desired, i.e., encourage fish to utilize each pool for resting (figs. 13 and 14). EFFECT OF LIGHT ON FISH MOVEMENT Measurement of the swimming performances of chinook salmon and steelhead trout in light and dark channels indicated that the movement of both species was significantly slower in the dark channel (table 4). In these tests the fish were light adapted before they entered the channels. Light intensities averaged approximately 750 foot- candles in the "light" channel and 0.3 foot-candle in the "dark" channel. In contrast to the foregoing channel experi- ments, a series of passage trials in light and dark fishways (fig. 19) indicated that faster movement occurred under the dark condition (table 5). These data apply principally to steelhead trout (98 percent steelhead and 2 percent chinook salmon). Before entering the fishway, fish were light adapted under the "light" condition and dark adapted under the "dark" condition. Pre- vailing light intensities in the fishways were about 800 foot-candles in the Hght and less than .01 foot-candle in the dark. Further research will be necessary to explain why the fish moved so securely and effectively through the fishway pools in the (lark, apparently oriented by the patterns of jets, eddies, and turbulences, while in a straight, level channel with a uniform laminar-type flow they iip[)('iir('(l to move only with great caution in the (lurk. 12 Figure 20. — Chinook salmon jumping over a weir. Salmon usually swim over an overfall of tliis height. Proximity to the wall is characteristic. Table 5. — Comparison of passage times in "light" and "dork' fishways ' Test number Passage time for 6 pools Light flshway Dark flshway n Median elapsed time Mean time n Median elapsed time Mean time 30 24 2.1 Minutes 6.1 7.3 11.9 8.5 Minutes 10.5 6.6 12.2 7.9 22 19 20 22 Minutes 1.3 2.2 2.8 1.8 Minutes 6.4 2 4.6 3-. 4.. S6mmary 5.9 8.0 103 8.3 9.3 83 2.1 6.2 ' Six-pool flshway, l-on-16 slope, 4 feet wide, 6.3 feet deep and between pools. -foot rise Tests involving around-the-clock passage of salmonids have indicated that rate of ascent during the night in a lighted fishway compared favorably with ascent in daytime hours. A blue- back salmon that ascended over a mile in height during a 5}2-day period averaged 52 pools per hour between 6:00 p.m. and 6:00 a.m. and 5.3 pools per hour from 6:00 a.m. to 6:00 p.m. Light- ing in the fisluvay was coiislant throughout tlie test period, ranging from 700 to 1,000 foot-candles at the surface of the pools. Similar trends in day and night movement were exhibited by steelliead trout and chinook salmon when constant light prevailed. 13 Figure 21. — Experimental arrarjgement for examining reactions of salmon fingerlings. to overfalls and orifices. Approach cfiannel is 24 feet wide and 15 feet deep. Release box may be seen at tfie far end of tfie clianncl. Inclined plane screen trap is visible in tfie foreground. GENERAL OBSERVATIONS The effect of human odors on fish behavior lias been quite conspicuous durinjij hiboratory experi- ments. Fish near the surface rapidly sounded on detecting the odor and activity was suppressed for as long as 20 niiiuites. This re-emphasizes tluit in the operation of fishways, effort should be made to avoid physical contact with equipment that will he immersed in the water. Fish movement through auxiliary pools and channels can be exjx'dited by restricting these channels to less than 4 feet of depth. The tendency for salmon to linger and accumulate in deeper pools and chaiuicls at the laboratory was efiVctively discouraged by the use of wire-mesh grills at shallow depths. Figure 22. — Ice Harbor prototype fisliway witf) 1-on- 10 slope now undergoing tests at Bonneville labora- tory. Temporary divider walls liave been inserted in eacli pool. Note paired orifices in eocli weir. Large numbers of fish use these ports in ascending the fishway. i 14 Figure 23. — Ice Harbor design fishway in operation. Transitions in spiitial relationships, water veloc- ities, and particularly light affect the movement of fish into, through, and out of the laboratory. The effect of an abrupt change in conditions is usually hesitancy and dela_y. The ability of all species of salmonids to nego- tiate overfalls up to '.i feet was frequently demon- strated in the laboratory (fig. 20). However, it was generally observed that the fish would swim easily over a 1-foot overfall but that it usually had to resort to jumping over higher overfalls. The additional energy expended, the delays and the increased probabilitj- of minor injuries that would make the fish more susceptible to disease suggests that overfalls greater than 1 foot are undesirable for standard fisliwaj"s. Research directed toward the passage problems of downstream migrants was largely postponed because of the priority given to adult passage problems. Exploratory experiments conducted to test tiie reactions of chinook fingerlings to an overfall, an orifice, and a siphon are illustrated in figure 21. A 3.5-inch orifice at a depth of 7 feet was far more effective (80 percent) in attracting and collecting fingerlings than a 7-inch overfall of comparable width. The siphon with an intake 18 inches below the surface was also more ef- fective than the overfall. CURRENT RESEARCH Experiments now in progress at the laboratory are centered around a full-scale model (figs. 22 and 23) of the 1-on-lO-slope fishway designed for the north shore of Ice Harbor Dam, now under construction on the lower Snake River, a major tributary of the Columbia. Pattern and rate of fish movement, space utilization, and capacity- potential are being examined in the 6-pool section of the fishway. The experiments this year demonstrate a special function of the Fisheries-Engineering Research liaboratory — that of providing the means by which new fish passage devices and new features of fishway design may be biologically tested and proven to have merit before being permanently cast in concrete. 15 APPENDIX A Proposals, Reports, and Publications Related fo Research at the Fisheries-Engineering Research Laboratory UNPUBLISHED REPORTS (Distributed to state and federal agencies and universities participating in the research program.) 1. Collins, Gerald B. 1950. Outline of proposed program of research on orientation in migrating fish. U.S. Fish and Wildlife Service, Bureau of Commercial Fisheries, Biological Laboratory, Seattle, Wash. 9 pp., typewritten. 2. 1952. Proposed research on fishway problems Proposal submitted to North Pacific Division Corps of Engineers, U.S. Army, Portland Oregon, by U.S. Fish and Wildlife Service' Bureau of Commercial Fisheries, Biological Laboratory, Seattle, Wash., 36 pp., type- written. 3. 1953. A special type of laboratory for research on fish orientation. L'.S. Fish and Wildlife Service, Bureau of Commercial Fisheries, Biological Laboratory, Seattle, Wash., 6 pp., ditto process. 4. 1954. Explanatory notes on the Fish and Wildlife Service proposal for research on fishway problems. U.S. Fish and Wildlife Service, Bureau of Commercial Fisheries, Biological Laboratory, Seattle, Wash., 9 pp., ditto process. 5. Collins, Geralu B., and Carl H. Elling. 1958. Performance of salmon in e.xperimental "endless" fishways with slopes of 1 on 8 and I on 16. U.S." Fish and Wildlife Service, Bureau of Commercial Fisheries, Biological Laboratory, Seattle, Wash., 19 pp., ditto process. 6. 1958. Supplement I to performance of salmon in experimental "endless" fishways with slopes of 1 on 8 and I on 16. U.S. Fish and Wild- life Service, Bureau of Commercial Fisheries, Biological Laboratory, Seattle, Wash., 3 pp., ditto process. 7. Collins, Gerald B., Carl H. Elling, and Edgar C. Black (with technical assistance of Arthur Hanslip) 1958. n. Blood lactates and performance of salmon and trout in experimental "endless" fishways. In: Summary report from Edgar C. Black on "Further studies on the effects of muscular fatigue on fish." Department of Physiology, University of British Columbia, N.R.C. Grant TR-7, November 1, 1958, p. 3, mimeographed. 8. Collins, Gerald B., Carl H. Elling, Edgar C. Black, and Anne C. Robertson (with technical assistance from Edward Trevor-Smith) 1959. II. Lactate and glycogen in relation to the performance of salmon and trout in experi- mental "endless" fishways. In: Summary re- port from Edgar C. Black on "Further studies on the effects of muscular fatigue on fish." Department of Physiology, University of Brit- ish Columbia, N.R.C. Grant TR-7, November . 1, 1959, p. 3 mimeographed. 9. Elling, Carl H. 1958. The effect of water velocity on the re- sponse and performance of adult salmonids. U.S. Fish and Wildlife Service, Bureau of Commercial Fisheries, Biological Laboratory, Seattle, Wash., 13 pp., typewritten. 10. Elling, Carl H., and others. 1955-1960. Monthly progress reports on re- search on fishway problems. Nos. 1 to 58. Prepared by U.S. Fish and Wildlife Service under Contract No. DA-35-026-25U2 for North Pacific Division, Corps of Engineers, U.S. Army, Portland, Greg. Irregular paging including tables, graphs, and photos. PUBLICATIONS 11. Black, Edgar C, Anne C. Robertson, and Robert R. Parker. Some aspects of carbohydrate metabolism in fish. University of Washington Press, Seattle, Wash. In press, expected 1960. 12. Collins, Gerald B. 1954. Research on anadromous fish passage at dams. Transactions of the Nineteenth North American Wildlife Conference, pp. 418-423. 13. Collins, Gerald B. 1956. Research on fishway problems. In: Re- port on Fisheries Engineering Research Pro- gram. North Pacific Division, Corps of Engi- neers, U.S. Army, Portland, Oreg., pp. 118-125. 16 14. CoLLI^f8, Gerald B. 1958. The measurement of performance of salmon in fishways. In: U. H. Mae.Milhui Lectures in Fisheries, a symposium held at the University of British Columbia, April 29 and 30, 1957. Edited by P. A. Larkin, Insti- tute of Fisheries, University of British Colum- bia, Vancouver, B.C., pp. 85-91. 15. Fi.i.iNc, Cari. H., and Howard L. Raymond. 1959. Fishway capacity experiment, 1 950. U.S. Fish and Wildlife Service, Special Scientific Report— Fisheries No. 299, 26 pp. 16. Eluno, Carl H. 1960. Further experiments in fishway capacity, 1957. U.S. Fish and Wildlife Service, Special Scientific Report — Fisheries No. 340, 16 pp. 17. Lander, Robert H. 1959. The problem of fishway capacity. U.S. Fish and Wildlife Service, Special Scientific Report — Fisheries No. 301, 5 pp. 18. Long, Clifford W. 1959. Passage of salmonids through a darkened fishway. U.S. Fish and Wildlife Service, Special Scientific Report — Fisheries No. 300, 9 pp. 19. vanHaagen, Richard H. 1956. Audio in salmon research. Journal of the Audio Engineering Society, vol. 4, no. 4, Octo- ber, pp. 151-158. MANUSCRIPTS (Reports by the staff of the Biological Laboratory, Seattle, Wash., except as noted, in various stages of prep- aration and review for future publication.) 20. Collins, Gerald B. Research in fish passage problems. 21. Collins, Gerald B., and Carl H. Elling. Progress in fishway research Bonneville Dam. 22. Collins, Gerald B., Carl H. Elling, Joseph R. Gauley, and Clark S. Thompson. Effect of slope on performance of salmonids in experimental "endless" fish- ways. 23. Collins, Gerald B., and Joseph R. Gauley. Ability of salmon to ascend long fishways. 24. Gauley, Joseph R. Effect of fishway slope on rate of passage of salmonids. 25. Gauley, Joseph R., and Clark S. Thompson. Further studies on fishway slope and its effect on rate of passage of salmonids. 26. Long, Clifford W. Effect of light on orientation and rate of movement of adult salmonids in a channel. 27. Thompson, Clark S. Influence of flow patterns on per- formance of Chinook salmon in pool-type fishways. 28. Trefethen, Parker S. Effect of sonic tags on fish movement. 29. Trevor-Smith, Edward. Changes in carbohydrate in salmon and trout in experimental "endless" fishways. M.D. thesis. University of British Colum- bia, Vancouver, B.C. 30. Weaver, Charles R. Influence of water velocity upon orientation and performance of adult migrat- ing salmonids. MS #994 17 U.S. GOVERNMENT PRlNTiNG OFFICF : 1961 MBL WHni I ihf^irv SPHals 5 W SE 00209