ADULT LAMPREY PASSAGE AND ENUMERATION STUDY, WELLS DAM, 2013:

The Effects of Head Differential on Entrance Efficiency, and of Picketed Leads on Count Window Enumeration Efficiency

David Robichaud LGL Limited Sidney, BC, Canada

and

Chas Kyger Public Utility District No. 1 of Douglas County East Wenatchee, WA, USA

Prepared for:

Public Utility District No. 1 of Douglas County East Wenatchee,

September 2014

Wells Dam Adult Lamprey Passage Study, 2013

Executive Summary As part of the relicensing of Wells Dam, Douglas PUD along with 5 signatories to the Aquatic Settlement Agreement developed the Pacific Lamprey Management Plan (PLMP) for the Wells Hydroelectric Project (Wells Project), with the objective of identifying and addressing any adverse Project-related impacts on passage of adult Pacific lamprey. Under the PLMP, a radio telemetry study was conducted in 2013 with three goals: to assess how adult Pacific lamprey Entrance Efficiencies varied under two different head differential treatments; to assess the effectiveness of recent fishway modifications on Count Window Enumeration Efficiency; and to evaluate Passage Efficiency, average travel times and general behavior of lamprey in the Wells Project fishways.

Adult Pacific lamprey were captured at Bonneville and Priest Rapids dams, and transported to the Wells Fish Hatchery for tagging. Radio and PIT tags were surgically implanted, and the tagged fish were released into the Wells Dam tailrace (n=92) or into the Wells Dam fishways above the adult fish trap (pool 38; n=18). To monitor movements, underwater antenna arrays were deployed throughout both Wells Dam fishways, and aerial antennas were deployed at the mouths of upstream tributaries (Okanogan and Methow Rivers).

Two head differential treatments were compared: a high condition (0.48 m or 1.5 ft) and a moderate condition (0.31 m or 1.0 ft). Treatment conditions occurred in 7-hour blocks (19:00 through 02:00) and alternated daily. Between treatments, the head differential was set at 0.48 m (1.5 ft). From 7 July to 7 October, 80 treatment conditions were tested, including 40 replicate tests of each treatment. The lamprey that were released into the tailrace approached the fishways on 89 occasions (35 during the high differential treatment, 12 moderate). Entrance Efficiency (the proportion of approach events that were followed by an entrance event) was 67% during the moderate treatment and 51% during the high treatment. Differences were not statistically significant, but statistical power was low.

Previous studies had identified an area (‘bypass’ behind the picketed lead) that allowed lamprey to move upstream through picketed leads without passing through the fishway video count window. Concerns about accurate passage count data prompted the installation of modified picketed leads with narrower spacing to help exclude lamprey from the bypass area. Count Station Passage (the proportion of tagged fish detected below the count window that were directed through the Count Station, rather than the bypass) was 88%, which was a significant improvement from before the new leads were installed (53.3%). Count Station Enumeration Efficiency (the proportion of fish known to have passed the count window that were tallied by the count video technicians) was significantly higher in the west fishway (68%) than in the east (33%), and was 51% overall. Count Station Enumeration Efficiency in 2013 was 11% higher than that estimated prior to installation of the modified picketed leads, but the statistical power of the comparison limited.

In 2004, 2007-2008, and 2013, Total Fishway Passage Efficiencies (the proportion of entry events that ended with successful passage) were 25% (3 of 12), 33% (2 of 6), and 9.5% (6 of 63), respectively. These rates were low relative to other dams. Lower Fishway passage efficiencies were 33%, 33% and 14%; and Upper Fishway Passage Efficiencies were 75%, 100% and 67%. Differences among years were not statistically significant, but statistical power was very low.

Travel times varied among years. In 2004, total fishway passage times averaged 0.3 days (7.2 hours), with largest delays in the reach leading up to the ‘below trap’ area, and in the ‘above trap’ to ‘below video’ reach. In 2007, a single fish passed from entrance to exit in 31.5 hours. In 2008, lower fishway travel times ranged from 1.7 to 5.5 hours, and upper fishway passage times ranged from 2.6 to 15.1 hours. In 2013, lower fishway passage times ranged from 3.7 to 6.9 hours (median 4.8 hours), and total fishway passage times ranged from 9.8 hours to 2.4 days (median 12.6 hours). Upper fishway passage times for tailrace -released fish ranged from 3.8 hours to 7.1 days (median 6.5 hours).

LGL Limited Page i

Wells Dam Adult Lamprey Passage Study, 2013

Table of Contents

Executive Summary ...... i List of Tables ...... v List of Figures ...... vi 1. Introduction ...... 1 2. Goals and Hypotheses ...... 3 2.1. Adult Pacific Lamprey Upstream Passage Evaluation ...... 3 2.2. Fishway Counts and Alternative Passage Routes ...... 4 3. Study Area ...... 4 3.1. Wells Hydroelectric Project ...... 4 4. Methods ...... 5 4.1. Fish Source ...... 5 4.1.1. Trapping at ...... 6 4.1.2. Trapping at ...... 6 4.2. Tagging ...... 7 4.3. Tracking ...... 10 4.3.1. Fixed Station Receiver Arrays ...... 10 4.3.2. Mobile Tracking ...... 12 4.3.3. PIT Tag Detections ...... 12 4.4. Data Processing ...... 12 4.5. Head Differential Treatments ...... 12 4.6. Data Analysis ...... 13 4.6.1. Detections ...... 13 4.6.2. Fishway Interaction Events ...... 13 4.6.3. Detection Benchmarks ...... 14 4.6.4. Detection Efficiency and Receiver Performance ...... 14 4.6.5. Movements ...... 14 4.6.6. Benchmark Movement Durations ...... 15 4.6.7. Entrance Efficiency ...... 15 4.6.8. Passage Efficiency ...... 16 4.6.9. Count Station Passage ...... 16 4.6.10. Count Station Enumeration Efficiency ...... 17 5. Results ...... 17 5.1. Tracking Success ...... 17 5.1.1. Fixed Station Receiver Arrays ...... 17 5.1.2. Mobile Tracking ...... 19 5.1.3. PIT Tag Detections ...... 19

LGL Limited Page iii Wells Dam Adult Lamprey Passage Study, 2013

5.2. Detections ...... 19 5.3. Fishway Interaction Events ...... 20 5.4. Within Fishway Movements ...... 20 5.5. Benchmark Movement Durations ...... 26 5.6. Entrance Efficiency ...... 29 5.7. Passage Efficiency ...... 30 5.8. Count Station Passage ...... 30 5.9. Count Station Enumeration Efficiency ...... 31 6. Discussion ...... 32 6.1. Entrance Efficiency and Head Differentials ...... 32 6.2. Effects of Picketed Leads ...... 33 6.2.1. Count Station Passage ...... 33 6.2.2. Count Station Enumeration Efficiency ...... 34 6.2.3. Bypass Exclusion ...... 35 6.3. Passage Efficiencies and Movement Durations ...... 35 7. Acknowledgements ...... 36 8. Literature Cited ...... 36 Appendices ...... 39

LGL Limited Page iv Wells Dam Adult Lamprey Passage Study, 2013

List of Tables Table 1. Numbers of adult Pacific lamprey radio-tagged and released, by collection location, release location, and tagging session ...... 9 Table 2. Number of adult Pacific lamprey known to have passed each detection zone, and the corresponding detection efficiency, in the west and east fishways at Wells Dam ...... 18 Table 3. Receiver performance of fixed-station receivers in the west and east fishways at Wells Dam ...... 18 Table 4. Details of mobile tracking survey effort, including survey dates in 2013, and the numbers of radio-tagged lamprey detected in each of three tracking zones in the Wells Dam tailrace...... 19 Table 5. The total numbers of unique radio-tagged adult Pacific lamprey that were detected at each detection zone in 2013, by release location ...... 20 Table 6. Numbers of ‘fishway interaction events’ in which the radio-tagged lamprey reached each progressive detection zone within the fishways at Wells Dam, the proportion of fish detected from the previous zone, and the proportion of entrances detected ...... 24 Table 7. Movements among fishway zones ...... 25 Table 8. Descriptive statistics of benchmark fishway passage times for radio-tagged adult lamprey at Wells Dam ...... 27 Table 9. Entrance and Passage Efficiencies by fishway and head-differential treatment ...... 29 Table 10. Approach and entrance counts around changes in treatment condition, by fishway ...... 30 Table 11. Count Station Passage, by fishway ...... 31 Table 12. Count Station Enumeration Efficiencies, by fishway ...... 32 Table 13. Summary of reported fishway Passage Efficiencies and Passage Durations for lower and mid- Columbia mainstem hydropower dams ...... 36

LGL Limited Page v Wells Dam Adult Lamprey Passage Study, 2013

List of Figures Figure 1. Picketed lead immediately downstream of the fishway count window ...... 2 Figure 2. Location map of the Well Project ...... 5 Figure 3. Relative frequencies of weight, girth and length measurements, and of tag burden for the radio-tagged lamprey, by source location ...... 8 Figure 4. Photo of an adult Pacific lamprey losing equilibrium in an anesthetic bath ...... 9 Figure 5. Photo of the surgical set-up, including a modified PVC pipe that cradled the lamprey during surgery ...... 9 Figure 6. Location of tailrace release site at Columbia River RM 514 ...... 10 Figure 7. Schematic showing approximate locations of antennas deployed in the Wells Dam fishways ...... 11 Figure 8. Chronology of head differential levels over the study period ...... 13 Figure 9. Relative frequency of the timing of fishway approach events ...... 21 Figure 10. Zones from which drop back occurred...... 22 Figure 11. The percent of lamprey entrants that reached each detection zone, by fishway ...... 23 Figure 12. Relative frequency plots showing the distribution of observed durations for passage through four specific fishway segments ...... 26 Figure 13. Relative frequency plots showing the distribution of observed durations for passage through the lower half of each fishway, the upper half, and the entire fishway ...... 28 Figure 14. Relative frequency plots showing the distribution of observed durations for the benchmark movements involving fish that were detected by the Count Station bypass antenna ...... 28

LGL Limited Page vi Wells Dam Adult Lamprey Passage Study, 2013

1. Introduction Pacific lamprey (Entosphenus tridentatus) have cultural and ecological significance in the Columbia River Basin. Historically, these fish were harvested by Native Americans for subsistence, and for ceremonial and medicinal purposes (Close et al. 2002). Ecologically, Pacific lamprey, which are anadromous and semelparous, probably play a role in the transport of marine-derived nutrients into freshwater systems (e.g., Nislow and Kynard 2009). But their infamy comes from their adult life history, as parasites on fish in the Pacific Ocean (Hart 1973, Wydoski and Whitney 2003).

Adult Pacific lamprey will usually spend 1 to 4 years in the ocean before returning to freshwater. Although parasitic at sea, adults are free-swimming in freshwater, travelling upriver to reach areas in low-gradient stream reaches. Lamprey generally spawn over gravel substrates in pools or riffles (Jackson et al. 1997), and die thereafter.

In the mid-Columbia River watershed, Pacific lamprey are known to occur in the Methow, Wenatchee, Entiat and Okanogan rivers (BioAnalysts 2000). Abundance declines, which have been observed for Columbia River Pacific lamprey populations over the last 40 years, have been attributed to juvenile and adult dam passage issues, reductions in spawning and rearing habitat, pollution, reduction of ocean food sources, and predation by introduced species (Close et al. 2002). However, the mechanisms of decline are poorly understood and highly speculative.

Leading up to the relicensing of the Wells Project, several studies were conducted to address Pacific lamprey passage issues at Wells Dam. Telemetry studies were conducted (in 2001-2003, 2007 and 2008) to determine migration and passage characteristics of lamprey entering the Project Area (Nass et al. 2005, Robichaud et al. 2009). These studies indicated that adult lamprey had difficulty negotiating the fishway entrances, that lamprey successfully passed through the middle and upper fishway sections, and that large proportions of the adult lamprey bypassed the visual counting windows (suggesting that visual counts were underestimating the total number of lamprey that passed Wells Dam). As a result, the researchers recommended that alternative methods be explored to increase Entrance Efficiency and accuracy of lamprey tallies from the visual count window.

Following these initial studies, and during the relicensing process, a Pacific Lamprey Management Plan (PLMP) was developed for the Wells Project. It was developed by Public Utility District No. 1 of Douglas County (Douglas PUD), in collaboration with federal, state and tribal relicensing participants, in support of a comprehensive Aquatic Settlement Agreement (ASA). The goal of the PLMP is to implement measures to monitor and address impacts, if any, on Pacific lamprey resulting from the Wells Project during the term of the new license.

One of the primary objectives of the PLMP is to identify and address any adverse Project-related impacts on passage of adult Pacific lamprey. Since Project-related impacts would decrease with improved lamprey Entrance Efficiencies, Douglas PUD decided to assess whether reductions in fishway entrance velocities (to levels within the swimming capabilities of Pacific lamprey) would improve entrance rates. To that end, two entrance passage studies were conducted in 2010 and 2011 utilizing DIDSON technology to visualize lamprey passage behavior at the fishway entrances (Johnson et al. 2010, 2011). Although sample sizes were small, the results suggested that lamprey Entrance Efficiencies were enhanced at lower head differential conditions (which are associated with lower entrance velocities).

The PLMP, also includes an objective to improve the accuracy of the annual enumeration of adult lamprey passing upstream through the fishways at Wells Dam. Prior studies have identified that a large percentage (78.7%) of the

LGL Limited Page 1 Wells Dam Adult Lamprey Passage Study, 2013

adult lamprey bypassed the count station by either moving into and through the picketed leads or by swimming through small gaps around the adult salmon and steelhead counting window1.

11 Towards increasing the accuracy of visual lamprey counts at existing counting windows, /16 inch picketed leads (Figure 1) were installed over the top of existing 1 inch wide picketed leads. These new leads were designed specifically to block adult lamprey from bypassing the counting station by entering the bypass structures. The large louvers located at the base of the count station were also replaced and many of the larger gaps around the counting window structure were sealed.

Count window

Count window bypass area

Picketed lead

Figure 1. Picketed lead immediately downstream of the fishway count window. Behind the picketed lead is the count window bypass area. Figure reproduced with permission from ‘Wells Dam Fish Passage System Overview’ by Douglas PUD, 2001.

In 2013, Douglas PUD used telemetry to implement the Wells Adult Lamprey Passage and Enumeration Study (2013 Wells Passage Study) to further assess how adult Pacific lamprey Entrance Efficiencies varied under two

1 In 2007-2008, 53.3% of radio-tagged lamprey (8 of 15) passed the Upper fishway via the video Count Station. In the same period, there were ten known instances of radio-tagged lamprey passing through the Count Area, of which 4 (40%) were recorded on the video gear (LGL and Douglas PUD 2008, Robichaud et al. 2009). Thus, 21.3% of fish passing the Upper Fishway were detected [53.3% × 40%], and 78.7% were missed.

LGL Limited Page 2 Wells Dam Adult Lamprey Passage Study, 2013

different head differential treatments. The 2013 Wells Passage Study design also allowed evaluation of Passage Efficiency, average travel times and general behavior of lamprey in the Wells Project fishways. A second objective of the 2013 Wells Passage Study was to assess the effectiveness of the changes that were made to the counting station that were intended to increase the number of lamprey that could be enumerated at the existing counting window.

This report details the design (including goals and specific hypotheses), methods, and results from the 2013 Wells Lamprey Passage Study. Specific metrics include travel times and entrance, passage, and Count Station enumeration efficiencies for adult Pacific lamprey that were radio-tagged and released at the Wells Project in the summer of 2013.

2. Goals and Hypotheses The goals of the 2013 Pacific lamprey passage and enumeration study were to evaluate adult lamprey passage and enumeration relative to changes in the structure and operation of the fishways at Wells Dam. Specific objectives of the study included:

• Adult Pacific Lamprey Upstream Passage Evaluation (PLMP section 4.1.6). o Evaluate passage efficiency of radio-tagged adult Pacific lamprey through Wells Dam fishways; with an emphasis in the lower fishway section (i.e., fishway entrance and collection gallery); o Evaluate travel time of radio-tagged adult Pacific lamprey through Wells Dam fishways; with an emphasis in the lower fishway section; o Evaluate radio-tagged adult lamprey behavior through Wells Dam fishways; with an emphasis in the lower fishway section; and o Compare adult Pacific lamprey entrance efficiency under reduced Wells Project fishway entrance velocities to entrance efficiencies at non-reduced velocities. • Upstream Fishway Counts and Alternative Passage Routes (PLMP section 4.1.3) o Compare the enumeration efficiency of adult lamprey at the fish count station at Wells Dam 11 using new, /16 inch picketed leads to results of previous studies with the old picketed leads. o Compare adult lamprey behavior at the fish count station with old picketed leads to that with 11 new, /16 inch picketed leads.

These objectives were met by capturing adult lamprey at Bonneville or Priest Rapids dams, transporting them to the Wells Project, surgically implanting radio transmitters, releasing them just downstream of Wells Dam in the tailrace (or into the Wells Project fishways) and tracking their movements within the two Project fishways.

2.1. Adult Pacific Lamprey Upstream Passage Evaluation The main study objective was to collect new information on the passage characteristics and behavior of adult lamprey migrating through the Wells Project fishways, as per Section 4.1.6 of the PLMP. To meet this objective, we tested the following null hypotheses, each presented with its alternative hypothesis:

• H0: There is no difference in passage metrics (i.e., Passage Efficiency, travel time and behavior) compared to other mainstem Columbia River projects.

H1: Passage metrics for lamprey differ compared to other mainstem Columbia River projects.

• H0: Flow differential consisting of one entrance velocity treatment has no effect on entrance success over another entrance velocity treatment.

LGL Limited Page 3 Wells Dam Adult Lamprey Passage Study, 2013

H1: Flow differential consisting of one entrance velocity treatment has an effect on improving entrance success over another entrance velocity treatment.

2.2. Fishway Counts and Alternative Passage Routes A second study objective was to evaluate the enumeration efficiency of the Wells Project fishway count windows, as per Section 4.1.3 of the PLMP. Since the last time that Pacific lamprey enumeration efficiency was evaluated 11 (see Robichaud et al. 2009), /16 inch picketed leads were newly installed over the top of the existing picketed leads, immediately downstream of the fishway count window (Figure 1), with the goal of blocking lamprey from bypassing the count window.

To meet this objective, lamprey behavior and count window enumeration efficiencies from 2013 (with the new picketed leads) were compared to those of previous studies (during which 1-inch-spaced picketed leads were present). To that end, we tested the following null hypotheses, each presented with its alternative hypothesis:

• H0: The proportion of tagged lamprey passing the count window is similar to previous studies.

H1: The proportion of tagged lamprey passing the count window is dissimilar to previous studies.

• H0: The proportion of tagged lamprey passing the count window that are enumerated by the Video Count technicians is similar to previous studies.

H1: The proportion of tagged lamprey passing the count window that are enumerated by the Video Count technicians is dissimilar to previous studies.

3. Study Area The core study area includes Wells Dam (specifically, the fishways) and the Wells Dam tailrace. Tracking also occurred in the Methow and Okanogan rivers (Figure 2), and in the Rocky Reach Reservoir upstream of Beebe Bridge.

3.1. Wells Hydroelectric Project The Wells Project is located at river mile (RM) 515.6 on the Columbia River in the State of Washington (Figure 2). Wells Dam is located approximately 30 river miles downstream from the Chief Joseph Hydroelectric Project, owned and operated by the United States Army Corps of Engineers; and 42 miles upstream from the Rocky Reach Hydroelectric Project owned and operated by Public Utility District No. 1 of Chelan County (Chelan PUD). The nearest town is Pateros, Washington, which is located approximately 8 miles upstream from the Wells Dam.

The Wells Project is the chief generating resource for Douglas PUD. It includes ten generating units with a nameplate rating of 774,300 kW and a peaking capacity of approximately 840,000 kW. The design of the Wells Project is unique in that the generating units, spillways, switchyard, and fish passage facilities were combined into a single structure referred to as the hydrocombine. Fish passage facilities reside on both sides of the hydrocombine, which is 1,130 feet long, 168 feet wide, with a top of dam elevation of 795 feet above mean sea level (msl).

The Wells Reservoir is approximately 30 miles long. The Methow and Okanogan rivers are tributaries of the Columbia River within the Wells Reservoir. The Wells Project boundary extends approximately 1.5 miles up the and approximately 15.5 miles up the Okanogan River. The surface area of the reservoir is 9,740 acres with a gross storage capacity of 331,200 acre-feet and usable storage of 97,985 acre feet at the normal maximum water surface elevation of 781 feet (Figure 2).

LGL Limited Page 4 Wells Dam Adult Lamprey Passage Study, 2013

Figure 2. Location map of the Well Project.

4. Methods

4.1. Fish Source Adult Pacific lamprey were captured at Bonneville Dam during sampling events conducted on 19 and 20 June and on 7 July 2013. In addition, a fourth trapping event occurred at Priest Rapids Dam, spanning three nights from 19- 21 August. Each of the four sampling events had the goal of capturing at least 25 fish. Despite best efforts, fish handling methods failed to meet two of the criteria that were listed in the study plan: some fish weights were less than the proposed minimum size; and some pre-tagging holding times exceeded the proposed maximum duration. In order to minimize tag burden, minimum fish weight criteria were approved in the study plan by the Aquatic Settlement Work Group, including tagging thresholds for fish collected at Bonneville Dam > 550 g and from Priest Rapids Dam > 450 g. Of the fish collected during the study, only 12.4% of Bonneville fish and 33% of Priest Rapids

LGL Limited Page 5 Wells Dam Adult Lamprey Passage Study, 2013

fish met the threshold tag burden criteria. In order meet minimum statistical objectives with the limited number of fish available, the minimum fish weight criteria were waived in favor of tagging a larger number of lamprey. A second deviation from the study plan, lamprey were not transported to Wells immediately after capture. Instead, most of the trapped fish were held for extended time periods (in some cases over a month) prior to being delivered to Wells Dam for tagging. While it was assumed that all captured fish were exhibiting upstream migratory behavior and that each would attempt to pass Wells Dam, deviations from the approved study plan may have increased the likelihood that fish were not exhibiting normal upstream migration behavior.

Collecting fish from outside the Wells Project area had four primary advantages:

• Adult lamprey counts at Wells Dam in recent years have been extremely low (i.e., ranging from 1 to 35 fish since 2006), therefore, capturing and tagging a sufficient number of fish at the Wells Project for the study was not feasible. • Past efforts to capture lamprey at Wells Dam have negatively biased the result of the studies as the lamprey traps were highly effective at preventing the upstream ladder passage of lamprey. • Past lamprey trapping activities at Wells Dam have incidentally captured ESA-listed anadromous salmonid species currently covered under the Wells Habitat Conservation Plan (HCP). • Adult Pacific lamprey captured at Bonneville Dam tend to be larger, thereby minimizing the tag burden, the potential for mortality, and impacts on behavior or swimming performance. This will allow for fish used in the study to behave and perform more similarly to untagged fish.

Most fish losses from transportation and holding stress are caused by poor water quality and improper handling (Wynne and Wurts 2011). To ensure that our study fish were in the best possible health upon delivery to the Wells Fish Hatchery, we used the following handling and transport protocols. No anesthetics were used during trapping operations, as this can produce a biological response similar to that caused by stress (Wynne and Wurts 2011). Upon capture, adult lamprey were immediately netted and placed into covered hauling tanks. We only transported fish that appeared to be in healthy condition (e.g., no signs of injury, disease, etc.).

4.1.1. Trapping at Bonneville Dam Fish captured at Bonneville were transferred to Prosser Hatchery where they were held for 21 to 46 days until the day before tagging. Before each tagging session, fish (of unknown capture date) were netted and placed in 70 gallon tanks and transported to the Wells Fish Hatchery. The water was chilled with ice packs to 10-12°C during transport. Upon arrival at Wells, tank water was slowly mixed with river water to gradually bring the tank water temperature to that of the ambient river water.

Fish were transported to Wells Fish Hatchery on three occasions (15, 22 and 30 July), including 37, 35, and 34 lamprey per trip. Lamprey from each of these sessions were used as the first through third release groups. For details of releases, see Section 4.2.

4.1.2. Trapping at Priest Rapids Dam On the night of 19 August, an LGL technician travelled to Priest Rapids Dam, deployed four traps in two fish ladders and let them fish overnight. The technician returned in the morning and collected the trapped lamprey. The total number of trapped lamprey was six, which did not meet the capture goal, so the lamprey were released back into the Columbia River upstream of Priest Rapids Dam.

Trapping was attempted again on the night of 20 Aug. Although only four lamprey were collected, the fish were nevertheless transported to Wells Fish Hatchery in aerated coolers. The four fish were tagged on 21 August and released into the Wells tailrace as the first part of the fourth release group (see Section 4.2).

LGL Limited Page 6 Wells Dam Adult Lamprey Passage Study, 2013

The final trapping attempt occurred on the night of 21 August. Five fish were caught and transported to Wells Fish Hatchery. These five fish were tagged on 22 August and released into the Wells tailrace as the second part of the fourth release group (see Section 4.2).

4.2. Tagging Tagging procedures followed methods described in previous lamprey radio-telemetry studies (Moser et al. 2002, Robichaud et al. 2009) and considered recent advances in knowledge and understanding of fish health and condition (e.g., Cooke et al. 2011a, b). An effort was made to minimize impacts to the biological and physiological condition of the study fish (as outlined in Liedtke et al. 2012). Specific attention was made to minimize incision length, possibility of infection, handling time, water temperature stressors, and air exposure. Fish size and holding duration criteria were waived during the study in favor of meeting sample size requirements.

Study fish were tagged with model NTC-4-2L Nano Tags (Lotek, Newmarket, Ontario). All transmitters were uniquely coded to allow for the identification of individual fish, and had a battery life of at least 162 days at a pulse rate interval (PRI) of 5.0 seconds. Tag dimensions were 16 mm (length) by 4 mm (height) by 6 mm (width) and weighed 1.10 g in air. In addition, each fish was given a full-duplex passive integrated transponder (PIT) tag with tag dimensions of 12 mm by 2.12 mm, weighing 0.1 g. Total combined weight of both tags, 2.2 g, represented a tag burden of <0.65% of the total body mass (Figure 3). All tags were disinfected in a diluted chlorhexidine solution. The tags remained in disinfectant for up to an hour and were rinsed with distilled water prior to insertion.

All lamprey were anesthetized using 110 mg/L buffered tricaine methanesulfonate (MS-222) for 5-8 minutes (Figure 4). Once the fish lost equilibrium, the surgeon screened the lamprey for fungus, disease, injury, or any obvious abnormalities. Healthy anesthetized fish were transferred to a scale and weighed to the nearest 0.5 g. All lamprey were then measured for total length and girth (to the nearest 0.5 mm; Figure 3). The girth measurement was taken anterior to the first dorsal fin.

Lamprey were then placed head first into a 5 cm (2 inch) diameter polyvinyl chloride (PVC) pipe, sealed on one end with a removable cap. A portion of the pipe was cut away to allow access to the ventral surface of the lamprey for surgery. During surgery, the head and gills of the fish were submerged in (and/or flushed with) a 35 mg/L concentration of MS-222 (Figure 5).

To implant the tags, we made a 3 cm incision approximately 1 cm off the ventral midline, with the posterior end of the incision in line with the origin of the first dorsal fin. The PIT tag was inserted into the body cavity using a MK- 25 implant gun (Biomark Boise, ID). Then, the radio tag was inserted. The radio tag antenna was threaded through the body wall approximately 3 cm posterior to the incision using a catheter (Moser et al. 2002). The incision was closed using a 17 mm tapered needle to make at least two individual stitches of 5-0 absorbable surgical suture. The antenna was secured to the side of the fish with a single suture.

After surgery, fish were transferred to a covered tank (a 113 L cooler) with flow-through river water for recovery. To reduce post-tagging handling, the containers used for post-release holding were the same ones used for transport and release. Douglas PUD staff transported all recovered fish to the release sites. During transport, the fish remained in their covered tanks, and air stones (diffusers) were used to maintain oxygen levels. The majority (83) of the radio-tagged lamprey were released in the Wells tailrace (Table 1), 1.5 miles below Wells Dam on the west bank of the Columbia River at RM 514 (Figure 6).

LGL Limited Page 7 Wells Dam Adult Lamprey Passage Study, 2013

50% 50% 45% 45% 40% All 40% All 35% Bonneville 35% Bonneville 30% Priest Rapids 30% Priest Rapids 25% 25% 20% 20% 15% 15% Relative Frequency Relative Relative Frequency Relative 10% 10% 5% 5% 0% 0% 9-9.5 9.5-10 12-12.5 12.5-13 13-13.5 13.5-14 14-14.5 14.5-15 15-15.5 10-10.5 10.5-11 11-11.5 11.5-12 200-250 500-550 550-600 600-650 650-700 700-750 750-800 800-850 250-300 300-350 350-400 400-450 450-500

Weight bin (g) Girth bin (mm)

50% 50% 45% 45% 40% All 40% All 35% Bonneville 35% Bonneville 30% Priest Rapids 30% Priest Rapids 25% 25% 20% 20% 15% 15%

10% Frequency Relative 10% Relative Frequency Relative 5% 5% 0% 0% 60-62 62-64 64-66 66-68 68-70 70-72 72-74 74-76 76-78 78-80 80-82 82-84 84-86

Total Length bin (mm) 0.25%-0.30% 0.30%-0.35% 0.35%-0.40% 0.40%-0.45% 0.45%-0.50% 0.50%-0.55% 0.55%-0.60% 0.60%-0.65% 0.65%-0.70% 0.70%-0.75% 0.75%-0.80% 0.80%-0.85% 0.85%-0.90%

Tag Burden (Tag weight as percent of body weight) Figure 3. Relative frequencies of weight, girth and length measurements, and of tag burden (tag weight as percent of body weight) for the radio-tagged lamprey, by source location.

LGL Limited Page 8 Wells Dam Adult Lamprey Passage Study, 2013

Figure 4. Photo of an adult Pacific lamprey losing equilibrium in an anesthetic bath.

Figure 5. Photo of the surgical set-up, including a modified PVC pipe that cradled the lamprey during surgery.

Table 1. Numbers of adult Pacific lamprey radio-tagged and released, by collection location, release location, and tagging session, 2013.

Collection Tagging Session, 2013 Location Release Location 16 Jul 23 Jul 30 Jul 21-22 Aug Total Bonneville Wells Tailrace 29 27 27 0 83 Bonneville Wells East Fishway 3 3 3 0 9 Bonneville Wells West Fishway 3 3 3 0 9 Priest Rapids Wells Tailrace 0 0 0 9 9 Total 35 33 33 9 110

LGL Limited Page 9 Wells Dam Adult Lamprey Passage Study, 2013

Figure 6. Location of tailrace release site at Columbia River RM 514.

For each of the first three release groups, a group of six radio-tagged lamprey were released directly into the Wells fishways (Table 1). These fish, released into the pool located directly above the adult fish trap (pool 38), were used in support of the count window enumeration efficiency objectives.

An additional five lamprey from the Bonneville Dam release groups were tagged only with PIT tags. These were fish delivered in excess of the radio-tagging target, and were typically smaller individuals or those that exhibited signs of injury or physical abnormalities. The numbers of PIT-tagged (but not radio-tagged) lamprey that were released on 16, 23 and 30 July were 2, 2, and1 fish, respectively.

4.3. Tracking

4.3.1. Fixed Station Receiver Arrays The movement and passage of radio-tagged lamprey at Wells Dam were documented by combining detection data collected using both stationary underwater and aerial antenna arrays. Fixed-station telemetry receivers and associated antenna arrays were deployed in a manner similar to previous lamprey studies at Wells Dam (see Robichaud et al. 2009). The arrays were designed to detect movements of radio-tagged lamprey from the Wells tailrace into the fishway entrances, at select locations throughout the fishways, at the fishway exits, and at the mouths of the Okanogan and Methow Rivers.

LGL Limited Page 10 Wells Dam Adult Lamprey Passage Study, 2013

A total of eight Lotek telemetry receivers (Model SRX 400) were deployed to monitor 26 underwater dipole or stripped coax antennas in the Wells Project fishways. In all, there were four receivers and 13 antennas in each fishway (Figure 7), deployed at the following locations:

• First Lower Fishway Receiver : . Outside fishway entrance; . Inside fishway entrance; . Collection gallery center pier nose; . Weir 1; • Second Lower Fishway Receiver: . Weir 7; . The lower part of the auxiliary water supply (AWS); . The upper part of the auxiliary water supply (AWS); • First Upper Fishway Receiver: . Pool below the adult fish trap; . Pool above the adult fish trap; . Below the video count window (lower portion of Pool 64 below count window); • Second Upper Fishway Receiver: . Within the count window bypass area (behind the picketed lead); . Above the video count window (upper portion of Pool 64 above count window); and . Fishway exit.

Figure 7. Schematic showing approximate locations of antennas deployed in the Wells Dam fishways, 2013. Antenna locations are shown as red dots.

LGL Limited Page 11 Wells Dam Adult Lamprey Passage Study, 2013

All underwater antennas (dipole, stripped coax) were constructed by LGL Limited from new materials. With the assistance of Douglas PUD staff, LGL staff deployed underwater antennas while the fish-ladders were dewatered (east ladder: 10,14,15 January; west ladder 11-13 February). All of these receivers were operational by 27 June.

A Lotek telemetry receiver (Model SRX 400) was deployed along with an aerial (yagi) antenna at each of the two ‘remote’ stations: one each at the mouths of the Methow and Okanogan rivers. These two receiver sites were set- up, tested, and became operational on 17 July.

Initially, receivers were maintained and downloaded weekly by LGL technicians, but downloads became biweekly when it became apparent that the receiver memory banks were always less than half-full after one week of operation. Receivers were operational from mid-July to 21 November.

4.3.2. Mobile Tracking Douglas PUD staff performed nine mobile tracking surveys in the area extending from the Wells Dam tailrace to Beebe Bridge (8, 14 and 23 August; 4, 13 and 18 September; 4 and 23 October; 8 November). The surveys were conducted from a Douglas PUD boat that had a mounted post which secured a single three-element aerial antenna. During boat-tracking, coordinates of each detection were recorded (latitude, longitude). Nevertheless, the tailrace was partitioned into zones based on the distance from the dam (immediate tailrace; area near release site; farther downstream) and all detections were assigned to one of these tailrace zones in situ.

4.3.3. PIT Tag Detections PTAGIS was queried to determine whether any of the tagged lamprey were detected on in-stream PIT arrays in the Entiat or Methow rivers, or at any other mainstem Columbia River hydropower dam. PIT tag detections were included in the Telemetry Manager database (see Section 4.4) by treating them as mobile detections.

4.4. Data Processing Data downloaded from fixed-station receivers and all mobile-tracking records (including PIT tag detections from PTAGIS) were processed using custom database software, Telemetry Manager. Individual antennas were all treated as "zones", each defining pivotal areas of interest, such as individual fishway entrances or exits.

Telemetry Manager facilitates data organization, record validation and analysis through the systematic application of user-defined criteria. Temporal or spatial resolution, and noise filtering criteria can be changed by the user at any time without altering the raw data. An important aspect of telemetry is the removal of false records in receiver files; for example, those that arise from electronic noise. Other ‘false’ records included detections prior to release or in a sequence which is not possible. Once false records are removed, Telemetry Manager creates a compressed database of sequential detections for each fish. Each record includes the tag number, detection location, and the first and last date and time of any sequential detections in that location. The compressed database was used for analyses required to address project goals and to test the study hypotheses.

4.5. Head Differential Treatments During the study period, two fishway entrance head differential treatments were compared: existing high condition (0.48 m, or 1.5 ft) and a moderate condition (0.31 m, or 1.0 ft). These treatments correspond to fishway entrance velocities of 3.0 m/sec (9.8 ft/s) and 2.4 m/sec (7.9 ft/s), respectively. It’s important to note that the fishway head differential treatments only affected the entrances as far as the collection galleries (the remainder of the ladders operated at a constant flow (1.4 m3/s, or 49.4 cfs) throughout.

Treatment conditions occurred over a 7-hour block (19:00 through 02:00; i.e., during peak periods of lamprey activity; Robichaud et al. 2009) and alternated daily (i.e., high condition one night and moderate condition the next

LGL Limited Page 12 Wells Dam Adult Lamprey Passage Study, 2013

night). In between treatments (i.e., each day from 02:00 to 19:00), the head differential was set at 0.48 m (1.5 ft, Figure 8). From 7 July to 7 October, 80 treatment conditions were tested, including 40 replicate tests of each of the two treatments.

0.6 Inter- Inter- Inter- Inter- 0.5 Treatment Treatment Treatment Treatment Period Period Period Period 0.4

0.3 Moderate High Moderate High 0.2 Differential Differential Differential Differential Treatment Treatment Treatment Treatment Head Head Differential (m) 0.1

0 0:00 6:00 12:00 18:00 0:00 6:00 12:00 18:00 0:00 6:00 12:00 18:00 0:00 6:00 12:00 18:00 0:00 6:00 12:00 18:00 0:00 Time of Day Figure 8. Chronology of head differential levels over the study period. Treatment periods (shaded) occurred every night from 19:00 to 2:00, alternating between moderate and high conditions. Between each treatment period (from 2:00 to 19:00 each day), the high differential condition was in effect.

4.6. Data Analysis

4.6.1. Detections The numbers of fish detected at each zone were summarized as tallies of individuals by zone. Individuals that were detected multiple times at any given zone were not counted more than once.

4.6.2. Fishway Interaction Events The sequential detection history was examined for each radio-tagged lamprey individually. Each time a fish interacted with a fishway, the event was tallied as a ‘Fishway Interaction Event’. Fishway Interaction Events included an ‘approach event’ as a minimum. An ‘approach event’ occurred when a fish (that was not already within the fishway) was detected on the outside fishway entrance antenna. If the ‘approach event’ was followed within 24 hours by a detection of the fish at any of the within-fishway detection zones, we tallied a successful ‘entry event’. All subsequent movements of that fish within the fishway were considered as part of the same Fishway Interaction Event until either: 1) the fish exited the fishway into the forebay (tallied as a successful passage event); or 2) the fish dropped down far enough to be detected on the outside fishway entrance antenna (or anywhere farther outside of the fishway). Many of the individual lamprey had multiple Fishway Interaction Events over the course of the study period. Each event was treated as independent.

When more than 24 h elapsed between the last detections on the outside entry antenna, and the first detection inside the fishway, these were not considered to be part of a single Fishway Interaction Event: the outside detections were considered to be part of a failed approach; and the inside detections were considered to be part of a subsequent Fishway Interaction Event for which the approach was not detected.

Note that a subset of the tagged fish were released directly into the fishways – these were not included as successful ‘entry events’. However, if any of these fish dropped down into the tailrace, any subsequent Fishway Interaction Events were tallied as described above.

LGL Limited Page 13 Wells Dam Adult Lamprey Passage Study, 2013

4.6.3. Detection Benchmarks For each Fishway Interaction Event, we determined the timing of eleven benchmark detection events, listed below.

1. ‘Approach’: first detection at the outside antenna at the fishway entrance; 2. ‘Entry’: last detection at inside antenna at the fishway entrance; 3. ‘Beyond Entrance’: first detection at ‘collection gallery’ side gate zone; 4. 'Upper Collection Gallery': first detection at weir no. 1; 5. 'Lower Fishway': first detection at weir no. 7; 6. ‘Below Trap’: first detection in the pool below the adult fish trap; 7. ‘Above Trap’: first detection at the ‘above trap’ zone; 8. ‘Below Video’: first detection in the lower portion of Pool 64, below count window; 9. ‘Above Video’: first detection in the upper portion of Pool 64, above count window; 10. ‘Begin Bypass’: first detection in the count window bypass zone; 11. ‘End Bypass’: last detection in the count window bypass zone; 12. ‘Exit Approach’: first detection at the fishway exit; and 13. ‘Passage Success’: last detection at the fishway exit (without subsequent fishway detections).

4.6.4. Detection Efficiency and Receiver Performance Detection efficiencies of the receivers were not 100%. It was possible, for example, for a tailrace fish to ‘appear’ in the fishway without being detected on the outside entrance antenna. In such cases, an ‘approach’ event must have occurred but was missed (the fish must have approached the fishway before entering it). Thus the Benchmark event was known to have occurred, but the timing of the Benchmark Detection was unknown. Similarly, many of the other Benchmark Detection events could have been missed as result of imperfect detection efficiencies.

For each zone, the detection efficiency was calculated as 1 minus the proportion of the total number of ‘movements past the zone’ that were missed. The total number of ‘movements past the zone’ was equal to the number of Fishway Interaction Events in which the fish reached or passed the detection zone in question. A fish that was detected upstream of the zone in question, but not at the zone itself, was known to have been missed. The number of fish detected at each zone included only fish moving in the upstream direction (detection efficiencies would be artificially inflated if they included fish that were missed as they passed an antenna in the upstream direction, but which were subsequently detected as they dropped back past the antenna in a downstream direction). The detection efficiency of certain zones cannot be calculated where there are no appropriate upstream detection zones.

Receiver performance was assessed as the percentage of deployment time during which the receiver was actively recording data. Activity and deployment were assessed on a per-hour basis, and then summed over weeks to calculate weekly activity percentages. Receivers were considered active in a given hour if at least one fish detection, battery check, beacon-tag hit, or noise event was recorded during the hour. Any detection (frequency- code) that did not correspond to a valid fish was considered as noise.

4.6.5. Movements Movements of fish among fishway zones were summarized by tallying movement counts. For every combination of among-zone movements, we tallied the number of times each fish performed that movement. Movements among Fishway Interaction Events were ignored. Tallies are presented in a table organized with columns and row headers called ‘from zone’ and ‘to zone’, respectively.

LGL Limited Page 14 Wells Dam Adult Lamprey Passage Study, 2013

In addition, the median durations (the time required to move from one zone to the next) were calculated for all possible among-zone movements. Durations were calculated as the difference between the last detection at the ‘from zone’ and the first detection at the ‘to zone’. Note that these durations do not include any ‘residency time’ (i.e., time spent at one zone or the next are not included – only the time spent between zones is calculated).

4.6.6. Benchmark Movement Durations Based on the Detection Benchmarks, we calculated Benchmark Movement Durations for a variety of specific fish movements, including:

• ‘Entrance passage time’: Benchmark Detection 1 to 2 • ‘Collection gallery passage time’: Benchmark Detection 2 to 3 • ‘Lower fishway passage time’: Benchmark Detection 2 to 7 • ‘Count Station passage time’: Benchmark Detection 8 to 9 • ‘Passage from below count window to exit’: Benchmark Detection 8 to 13 • ‘Upper fishway passage time’: Benchmark Detection 7 to 13 • ‘Total fishway passage time’: Benchmark Detection 1 to 13

To evaluate the use of the count window bypass area, Benchmark Movement Durations were calculated for the following segments:

• ‘Below count window to count window bypass’: Benchmark Detection 8 to 10 • ‘Residence time in count window bypass area’: Benchmark Detection 10 to 11 • ‘Count window bypass to exit’: Benchmark Detection 10 to 13

‘Entrance passage times’ were compared between head differential treatments using a binomial logistic model with Treatment as the sole explanatory variable. Fishway Interaction Events were used as replicates, each scored either as a success or a failure. If the model results showed high residual deviance relative to the number of degrees of freedom, then the data were considered to be overdispersed, and the model was re-run using a quasibinomial error distribution. The Treatment model was compared to a Null model (i.e., one without Treatment effects) using ANOVA. Where P values were less than 0.05, the result was considered to indicate a statistically significant effect of Treatment on Entrance passage time.

Upper fishway passage times were compared between release location (fish released in the tailrace vs. those released in the fishway) using a logistic model, following the same methods as described above.

4.6.7. Entrance Efficiency Radio-telemetry data from entrance locations (i.e., outside and inside fishway entrance arrays) were used to evaluate Entrance Efficiency with respect to head differential treatment conditions. Entrance Efficiency was calculated as the proportion of ‘approach events’ that were immediately followed by an ‘entrance event’. Detection efficiency was not 100% at the entrances (see Section 4.6.4). Events in which a fish was missed at the entrance but known to have moved into the fishway were counted as successful entrance events in terms of the Entrance Efficiency .

During the study period, two head differential treatments were compared. Each treatment condition occurred from 19:00 through 02:00, and alternated daily (i.e., high condition one day and moderate condition the next day, etc.). In order to be included as part of one treatment condition versus the other, entry timing must be known (i.e., the tagged fish must have been detected at the fishway entrance). Entry events that occurred between treatments (i.e., from 02:00 and 19:00, not part of either treatment condition) were not included in analyses of

LGL Limited Page 15 Wells Dam Adult Lamprey Passage Study, 2013 treatment effects. Also excluded were entry events with unknown timing (e.g., events for which the detection at the entrance was missed).

The effect of Treatment was examined using a logistic model. The effect of Fishway (east vs west) and Release Group (there were four groups of fish released over the study period; Table 1) were not included in the model due to the limitations imposed by small sample sizes.

As a power analysis, the Entrance Efficiency data were randomized using bootstrapping. In each step of the randomization, the Treatment-specific Entrance Efficiency data were sampled with replacement. The number of samples drawn was varied from n (the number of samples in the actual observational data from 2013) to 50n, always keeping the ratio of moderate to high treatment data-rows consistent (e.g., 35:12, 70:24, 210:72, etc.). For each randomly-generated dataset, the logistic model (described above) was used to test for an effect of Treatment. For each ‘sample size’ from n to 50n, 200 randomizations were run, and the proportion of the logistic tests that had P values < 0.05 was calculated. The end result of the power analysis was a curve describing the probability of statistical significance as a function of sample size. The point where 95% of randomizations produced significant results was used to determine the least significant number.

We examined the possibility that a treatment condition was deterring lamprey by identifying approach events that immediately followed a change in flow condition. For example, if moderate flows deterred approach, then there should have been elevated approach counts in the period shortly after the high flows were reinstated. Four three- hour periods were examined in each 48 hour period: the first three hours of each treatment were labeled as ‘Start of Moderate’ and ‘Start of High’; and the first three hours after each treatment were labeled as ‘After Moderate’ and ‘After High’. Recall that only two flow changes occurred during each 48 hour period (Figure 8): the ‘Start of High’ and ‘After High’ periods were not associated with a change in flow. Approach counts were tallied within each of the four periods and within each replicate. A generalized linear model (GLM) with Poisson error structure was run to test for differences among the periods.

4.6.8. Passage Efficiency Radio-telemetry data were used to evaluate Total Fishway Passage Efficiency with respect to head differential treatment conditions. Total Fishway Passage Efficiency was calculated as the proportion of Fishway Interaction Events that ended with successful passage into the Wells forebay. As before, entry events that were not part of either treatment condition were not utilized in the Passage Efficiency calculation. The effect of Treatment was tested with the same analytical method as described for Entrance Efficiency.

Also, Total Fishway, Lower Fishway and Upper Fishway Passage Efficiencies in 2013 were compared to those reported in 2004, 2007 and 2008 using a binomial logistic model. Lower Fishway Passage Efficiency was calculated as the proportion of Fishway Interaction Events in which the fish reached the ‘above trap’ detection zone. Upper Fishway Passage Efficiency was calculated as ρ/υ, where υ is the number of Fishway Interaction Events for which a fish reached the ‘above trap’ zone, and ρ is the subset of υ that subsequently passed into the Wells forebay.

4.6.9. Count Station Passage Count Station Passage was evaluated by examining telemetry detections at the ‘below video’, ‘above video’, and ‘bypass’ detection zones. Count Station Passage was calculated as the proportion of tagged fish detected below the count window that moved upstream through the count station rather than through the bypass area. Due to sample size limitations, data were not segregated by release group. No fish in the moderate head treatment were detected below the count window, thus it was not possible to test for effects of Treatment.

LGL Limited Page 16 Wells Dam Adult Lamprey Passage Study, 2013

All Fishway Interaction Events were treated as replicates. The total sample size was augmented by including data from the radio-tagged lamprey that were released directly into the fishways. Together, these data were used to calculate Count Station Passage for each fishway. The effect of Fishway was tested with a logistic model.

The true test of the effectiveness of the picketed leads was to compare Count Station Passage from 2013 to that from 2007 and 2008, using a logistic model.

4.6.10. Count Station Enumeration Efficiency The efficiency of enumerating lamprey using the existing counting station was evaluated by comparing the video- based enumeration data to the radio-telemetry data. A lamprey was known to be available for viewing if a radio- tagged fish was detected on the “below video count window” station, followed by a detection on the “above video count window” station. The event was scored as a ‘success’ if the video showed a lamprey passing during the approximate timing of the known lamprey ‘availability’. If no lamprey were observed during the period, the event was scored as a ‘failure’.

The Count Station Enumeration Efficiency was calculated as the proportion of total events that were successful. The efficiency data were segregated by fish ladder, but sample size limitation precluded further stratification (e.g., head differential treatment, or release group). The effect of Fishway was tested with a logistic model.

To test the effectiveness of the picketed leads, a logistic model was used to compare Count Station Enumeration Efficiency from 2013 with those from 2007 and 2008.

5. Results

5.1. Tracking Success

5.1.1. Fixed Station Receiver Arrays Detection efficiencies in 2013 (Table 2) ranged from 62% to 87% for the antennas at the fishway entrances, and from 57% to 91% in the gallery and weir areas. Detection efficiencies were highest (ranging from 90-100%) for the antennas at the adult fish trap and in the upper fishways.

Most receivers were operational throughout the study period (Table 3). There were three events which impacted receiver performance. The first event occurred in the lower west fishway, where a pair of receivers (‘Lower 1’ and ‘Lower 2’) had been accidentally unplugged. It is unknown when the receivers were unplugged (as they operated on their internal batteries for some time afterwards), but they stopped recording data at ~21:00 on 22 September, and remained offline until visited by an LGL technician for download on 25 September. The second event occurred in the upper west fishway. The ‘Upper 2’ receiver was found on 8 November to be frozen; its last recorded data was from 3 November. The third event occurred in the upper east fishway. The ‘Upper 1’ receiver could not be downloaded on its final visit on 21 November; thus all data recorded since the previous download (8 November) was lost. The cause of this hardware malfunction is not known. It is unlikely that any of the receiver failures resulted in reduced detection efficiency, as no fish were detected in either fishway after 16 September. PIT tag detection data (see Section 5.1.3) also confirmed that it was very unlikely that any tagged lamprey were missed: the last PIT tag detection on any of the fish ladder arrays occurred on 9 September.

LGL Limited Page 17 Wells Dam Adult Lamprey Passage Study, 2013

Table 2. Number of adult Pacific lamprey known to have passed each detection zone, and the corresponding detection efficiency (DE; see text for details), in the west and east fishways at Wells Dam, 2013.

West Fishway East Fishway Both Zone Passed DE Passed DE Passed DE Outside Entrance 52 75% 37 62% 89 70% Inside Entrance 31 87% 32 78% 63 83% Gallery 24 71% 14 57% 38 66% Weir 1 16 69% 11 91% 27 78% Weir 7 11 73% 3 67% 14 71% Below Trap 8 100% 1 100% 9 100% Above Trap * 17 100% 10 90% 27 96% Below Video 17 100% 9 100% 26 100% Above Video † 16 100% 7 100% 23 100% Exit 16 100% 8 100% 24 100% * Note 1: most of the fish detected above the trap on each fishway were fish released into the fishway at a location just above the trap. † Note 2: the DE for this location excludes fish that exclsively passed via the bypass.

Table 3. Receiver performance (percent operational) of fixed-station receivers in the west and east fishways at Wells Dam, in 2013, by week. Percentages were calculated as the number of hours of recorded receiver activity divided by the number of hours for which it was deployed, summed by week; Receivers were considered active in a given hour if at least one fish detection, battery check, beacon-tag hit, or noise event was recorded during the hour.

West Fishway East Fishway Week Lower 1 Lower 2 Upper 1 Upper 2 Lower 1 Lower 2 Upper 1 Upper 2 Methow Okanogan 6/23 - 6/29 100 100 100 100 100 100 100 100 nd nd 6/30 - 7/6 100 100 100 100 100 100 100 100 nd nd 7/7 - 7/13 100 100 100 100 100 100 100 100 nd nd 7/14 - 7/20 100 100 100 100 100 100 100 100 100 100 7/21 - 7/27 100 100 100 100 100 100 100 100 100 100 7/28 - 8/3 100 100 100 100 100 100 100 100 100 100 8/4 - 8/10 100 100 100 100 100 100 100 100 100 100 8/11 - 8/17 100 100 100 100 100 100 100 100 100 100 8/18 - 8/24 100 100 100 100 100 100 100 100 100 100 8/25 - 8/31 100 100 100 100 100 100 100 100 100 100 9/1 - 9/7 100 100 100 100 100 100 100 100 100 100 9/8 - 9/14 100 100 100 100 100 100 100 100 100 100 9/15 - 9/21 100 100 100 100 100 100 100 100 100 100 9/22 - 9/28 61 61 100 100 100 100 100 100 100 100 9/29 - 10/5 100 100 100 100 100 100 100 100 100 100 10/6 - 10/12 100 100 100 100 100 100 100 100 100 100 10/13 - 10/19 100 100 100 100 100 100 100 100 100 100 10/20 - 10/26 100 100 100 100 100 100 100 100 100 100 10/27 - 11/2 100 100 100 100 100 100 100 100 100 100 11/3 - 11/9 100 100 100 32 100 100 79 100 100 100 11/10 - 11/16 100 100 100 100 100 100 0 100 100 100 11/17 - 11/23 100 100 100 100 100 100 0 100 100 100 Red: Receiver malfunction- would not allow download; Yellow: Low power/dead battery - receivers had been unplugged.; Blue - Receiver frozen.

LGL Limited Page 18 Wells Dam Adult Lamprey Passage Study, 2013

5.1.2. Mobile Tracking Douglas PUD staff performed nine mobile tracking surveys in the area extending from the Wells Dam tailrace to Beebe Bridge, located 10 miles downstream (Table 4). These surveys detected a total of 18 of the 110 individual lamprey realeased between July 16 and August 22. Twelve of the fish were never detected on any of the fishway arrays. Four of the fish were only ever detected in the tailrace and on outside entrance arrays (i.e., these fish had approached the fishways, but did not enter). Two of the 18 fish entered the fishways, and subsequently backed out into the tailrace where they were detected in mobile surveys (one of these fish made a subsequent fishway approach but did not enter). Of the 18 fish detected, none were detected in depths greater than 10 m, suggesting that the mobile tracking equipment and methods may have been ineffective for detecting fish at greater depths. While detection efficiency of mobile tracking is unknown, it is unlikely that all the fish present in the mobile tracking area were detected since depths exceed 10 m in most of the areas surveyed in the upper Rocky Reach Reservoir.

Table 4. Details of mobile tracking survey effort, including survey dates in 2013, and the numbers of radio- tagged lamprey detected in each of three tracking zones in the Wells Dam tailrace.

Wells Tailrace Area Total Number Survey Immediately Near Release Farther of Fish Date Below Dam Site Downstream Detected 8 Aug 1 2 1 4 14 Aug 1 1 3 5 23 Aug 4 2 3 9 4 Sep 1 0 2 3 13 Sep 1 1 3 5 18 Sep 0 1 3 4 4 Oct 1 1 1 3 23 Oct 1 1 3 5 8 Nov 3 1 4 8

5.1.3. PIT Tag Detections A query of PTAGIS, performed in late January 2014, showed 40 unique detections of our adult lamprey. Thirty of these detections were recorded from within the Wells fishways. All of these detections occurred when the fish in question were known (based on radio detections) to be in the fishways. The remaining ten detections, including one at and nine in the Methow River watershed, were imported into the Telemetry Manager database. All but one of the Methow PIT detections were of fish that were already known (based on radio detections) to be in the Methow Watershed, and all had been detected exiting Well Dam.

Of the five lamprey that were PIT-tagged (and not radio tagged) and released into the Wells Dam tailrace, none were detected at any PIT tag interrogation site over the course of the study.

5.2. Detections Of the 110 radio-tagged lamprey, 33 (30%) were never detected at any point subsequent to their release. The remaining 77 fish were detected a total of 704 separate times at fixed stations or during mobile surveys. The duration of each detection ranged from a few hits over a couple seconds to thousands of hits over several days. The earliest fixed-station detection occurred on 16 July at 7 PM (when the first group of tagged lamprey was released into the fishways). For fish released in the tailrace, the earliest fixed-station detection occurred on 16 July at 11 PM (on the inside antenna of the west fishway entrance). The last fixed-station detection occurred on 26

LGL Limited Page 19 Wells Dam Adult Lamprey Passage Study, 2013

September at 12 AM on the outside antenna of the west fishway entrance. The period of detections coincides approximately with the migratory activity of lamprey in the immediate area (lamprey observations at the fish counting window ranged from 11 July to 5 October in previous years; Robichaud et al. 2009). Further, in 2013 less than 1% of 1626 lamprey observed at Rocky Reach Dam passed after September 30th, 2013.

Of all the detection locations, tailrace-released fish were most likely to encounter fishway entrances (32 and 18 on the west and east entrances, respectively). Of these, fewer were detected on the inside antennas, and the numbers of fish detected tended to decline with increasing distance up the fishways (Table 5). Fish released in the fishways tended to move upstream, with few fish detected at zones downstream of the release site (near the ‘above trap’ zone).

Table 5. The total numbers of unique radio-tagged adult Pacific lamprey that were detected at each detection zone in 2013, by release location (tailrace vs. fishways).

Tailrace Releases Fishway Releases All Releases Detection Zone West Fishway East Fishway West Fishway East Fishway West Fishway East Fishway Outside Entrance 32 18 1 2 33 20 Inside Entrance 17 14 1 1 18 15 Gallery 15 10 1 2 16 12 Weir 1 7 8 1 1 8 9 Weir 7 4 2 1 0 5 2 AWS Lower 6 1 2 1 8 2 AWS Upper 5 4 1 0 6 4 Below Trap 6 1 5 5 11 6 Above Trap 6 1 10 8 16 9 Below Video 6 1 10 8 16 9 Above Video 6 1 9 6 15 7 Count Bypass 3 0 2 5 5 5 Exit 5 1 10 7 15 8 Methow 5 9 14

5.3. Fishway Interaction Events From 7 July to 7 October, 80 treatment conditions were tested, including 40 replicate tests of each of the two treatments. The lamprey released into the tailrace exhibited 89 Fishway Interaction Events. Of these, 35 occurred during the high differential treatment (12 at the east fishway and 23 at the west fishway), and 12 occurred during the moderate differential treatment (six in each of the two fishways). The remaining Fishway Interaction Events occurred either before the treatment condition alternation began (n=5), during the inter-treatment period (2 AM to 7 PM; n=27), or with unknown approach/entry timing (n=10).

Fishway approach events (first detections at the outside antenna) were strongly skewed toward nighttime hours (Figure 9).

5.4. Within Fishway Movements For the 89 Fishway Interaction Events, we calculated the zone at which no further upstream movements were detected. Figure 10 shows the proportion of fish passing each zone that went no further upstream. Between 29% and 48% of fish dropped downstream after reaching each of the lower fishway zones up to and including Weir 7. All fish that reached the ‘below trap’, ‘above trap’ and ‘below count’ zones continued further upstream. Specifically, 60% of entrants reached the gallery, 71% of these fish reached Weir 1, 52% of the Weir 1 fish moved

LGL Limited Page 20 Wells Dam Adult Lamprey Passage Study, 2013

up to Weir 7, and 64% of the Weir 7 fish reached the ‘below trap’ zone (Table 6). Overall, 9 of 63 entry events resulted in passage into the upper fishway (14% Lower Fishway Passage Efficiency), and six of nine fish in the upper fishway successfully exited into the forebay (67% Upper Fishway Passage Efficiency). Total Fishway Passage Efficiency for the tailrace-released fish was 9.5% (Figure 11).

Movements among fishway zones are summarized in Table 7. In the west fishway, there was a total of 216 movements in the upstream direction, and 120 in the downstream. In the east fishway, there was a total of 130 and 89 upstream and downstream-directed movements, respectively. For all fish combined, movements between the trap and Count Stations were, in general, the longest in duration (median 6.6 h and 5.1 h in the west and east fishways, respectively) , but this was partly an artifact of the fishway-released lamprey taking time to recover post release. When analysis was restricted to ascents that started in the tailrace, median trap-to-count durations were 2.5 h in both fishways. Other notably long durations were between the western gallery and below trap zones (5.4 h, n=1), between the western lower AWS and below trap zones (median 2.8 h, n=7), and between the eastern lower AWS and below trap zones (3.2 h, n=1).

9 8 7 6 5 4 3 Relative Frequency 2 1 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Hour of Day

Figure 9. Relative frequency of the timing of fishway approach events.

LGL Limited Page 21 Wells Dam Adult Lamprey Passage Study, 2013

East (n=37) West (n=52) 100% All (n=89)

80%

60%

40%

20% Percent of Fish thatMoved

Upstreamafer Detection at this Zone 0% Entry Weir 1 Weir 7 Gallery Approach Below Trap Above Trap Below Count Above Count Detection Zone Count Bypass Figure 10. Zones from which drop back occurred. Bars show the proportions of fish that continued to move upstream after reaching each zone.

LGL Limited Page 22 Wells Dam Adult Lamprey Passage Study, 2013

Entrance 100% 63

9.5% 14% 80% 60%

Gallery 60% 38 67% 71% Weir1 27 40% 52% Below Below Above Trap Count Count 67% 20% 14 64% 9 9 9 9 Percent of Entrants that Reached Each Zone Each Reached that Entrants of Percent Weir7 100% 100% 100% 7 6 86% a) Both Fishways Above Bypass Exit 0% Trap 78% -0.05 0.95 1.95 2.95 3.95 4.95 5.95 6.95 7.95 8.95 9.95 10.95 11.95

Entrance 100% 31

16% 77% 26% 80% 24 Gallery 63% 60% 67% Weir1 16

40% Weir7 Below Below Above 69% Count 11 Trap Count 63% 73% 8 8 8 8 100% 20% 100% 100% 6 Above 5

Percent of Entrants that Reached Each Zone Each Reached that Entrants of Percent 83% Trap b) West Fishway 75% Bypass Exit 0% -0.05 0.95 1.95 2.95 3.95 4.95 5.95 6.95 7.95 8.95 9.95 10.95 11.95

Entrance 100% 32

3.1% 3.1% 80%

44% 60%

Gallery Weir1 14 100% 40% 79% 11

20% 27% Below Below Above Percent of Entrants that Reached Each Zone Each Reached that Entrants of Percent 3 33% Trap 100% 100% Count Count c) East Fishway 1 1 1 1 1 Weir7 0% 100% 100% Exit -0.05 0.95 1.95 2.95 3.95 4.95 5.95 Above6.95 Trap 7.95 8.95 9.95 10.95 11.95 Figure 11. The percent of lamprey entrants that reached each detection zone, by fishway. The vertical position of each circle represents cumulative passage success (i.e., the percent of entrants that reached that zone). Circle size is scaled to the number of fish observed (value shown in circle). Black arrows indicate the percent of fish in a given zone that reached the subsequent zone. Blue, red and magenta arrows indicate Lower Fishway, Upper Fishway and Total Fishway Passage Efficiencies, respectively.

LGL Limited Page 23 Wells Dam Adult Lamprey Passage Study, 2013

Table 6. Numbers of ‘fishway interaction events’ in which the radio-tagged lamprey reached each progressive detection zone within the fishways at Wells Dam, the proportion of fish detected from the previous zone, and the proportion of entrances detected, 2013.

Number of Percent Detected Percent of Fishway Zone Name Events of Zone Below Entrance Events West Approach 52 - - Entry 31 60% - Gallery 24 77% 77% Weir 1 16 67% 52% Weir 7 11 69% 35% Below Trap 8 73% 26% Above Trap 8 100% 26% Below Count 8 100% 26% Above Count 8 100% 26% Count Bypass 6 75% 19% Exit 5 83% 16% East Approach 37 - - Entry 32 86% - Gallery 14 44% 44% Weir 1 11 79% 34% Weir 7 3 27% 9% Below Trap 1 33% 3% Above Trap 1 100% 3% Below Count 1 100% 3% Above Count 1 100% 3% Count Bypass 1 100% 3% Exit 1 100% 3% Combined Approach 89 - - Entry 63 71% - Gallery 38 60% 60% Weir 1 27 71% 43% Weir 7 14 52% 22% Below Trap 9 64% 14% Above Trap 9 100% 14% Below Count 9 100% 14% Above Count 9 100% 14% Count Bypass 7 78% 11% Exit 6 86% 10%

LGL Limited Page 24 Wells Dam Adult Lamprey Passage Study, 2013

Table 7. Movements among fishway zones. Upper panels show counts of movements; lower panels show median movement durations (i.e., time required to move between zones) in days. Data are shown separately for each of the two Wells Dam fishways. Within each panel, values shown above the diagonal (bold, black) represent upstream movements, and those below the diagonal (red) represent downstream movements.

West Fishway - Movement Counts East Fishway - Movement Counts To Zone To Zone Entrance Outside Entrance Inside Gallery Weir 1 AWS Lower Weir 7 AWS Upper Below Trap Above Trap Below Count Above Count Bypass Count Exit Entrance Outside Entrance Inside Gallery Weir 1 AWS Lower Weir 7 AWS Upper Below Trap Above Trap Below Count Above Count Bypass Count Exit Entrance Outside 17 1 0 0 0 0 0 0 0 0 0 0 Entrance Outside 0 18 0 0 0 0 0 0 0 0 0 0 0 Entrance Inside 5 0 19 5 1 1 3 0 0 0 0 0 0 Entrance Inside 3 0 7 3 0 0 0 0 0 0 0 0 0 Gallery 1 10 0 11 4 2 8 1 0 0 0 0 0 Gallery 1 0 0 9 1 1 1 0 0 0 0 0 0 Weir 1 1 3 9 0 3 2 0 0 0 0 0 0 0 Weir 1 1 4 5 0 0 1 0 0 0 0 0 0 0 AWS Lower 0 0 1 0 0 0 0 7 1 0 1 0 0 AWS Lower 0 0 1 0 0 0 0 1 0 0 0 0 0 Weir 7 0 1 3 2 0 0 14 0 0 0 0 0 0 Weir 7 0 0 1 0 0 0 1 0 0 0 0 0 0 AWS Upper 1 2 9 0 0 14 0 0 0 0 0 0 0 AWS Upper 1 1 0 0 0 0 0 0 0 0 0 0 0 Below Trap 0 0 0 0 0 0 0 0 13 0 1 0 0 Below Trap 0 0 1 0 0 0 0 0 7 1 0 0 0 From Zone From Zone Above Trap 0 0 0 0 0 0 1 2 0 19 0 1 0 Above Trap 0 0 0 0 1 0 0 5 0 8 0 0 0 Below Count 0 0 0 0 0 0 0 1 1 0 55 6 2 Below Count 0 0 0 0 0 0 0 1 0 0 43 17 3 Above Count 0 0 0 0 2 1 0 0 1 37 0 4 13 Above Count 0 0 0 0 0 0 0 0 0 45 0 3 3 Bypass Count 0 0 0 1 0 0 0 0 1 7 1 0 1 Bypass Count 0 0 0 0 0 0 0 0 0 10 8 0 2 Exit 0 0 0 0 0 0 0 0 0 2 0 0 0 Exit 0 0 0 0 0 0 0 0 0 0 0 0 0

West Fishway - Median Movement Durations East Fishway - Median Movement Durations To Zone To Zone Entrance Outside Entrance Inside Gallery Weir 1 AWS Lower Weir 7 AWS Upper Below Trap Above Trap Below Count Above Count Bypass Count Exit Entrance Outside Entrance Inside Gallery Weir 1 AWS Lower Weir 7 AWS Upper Below Trap Above Trap Below Count Above Count Bypass Count Exit Entrance Outside 0.01 0.02 Entrance Outside 0.02 Entrance Inside 0.03 0.01 0.03 0.05 0.2 0.06 Entrance Inside 0.03 0.02 0.03 Gallery 0.03 0.02 0.01 0.03 0.04 0.05 5.4 Gallery 49 0.02 0.07 0.07 0.03 Weir 1 60 0.02 0.01 0.05 0.04 Weir 1 0.03 0.07 0.02 0.2 AWS Lower 0.02 2.8 0.5 0.02 AWS Lower 0.4 3.2 Weir 7 0.05 0.06 0.04 0.03 Weir 7 0.08 0.1 AWS Upper 1.3 0.06 0.03 0.03 AWS Upper 0.2 0.1 Below Trap 0.06 0.02 Below Trap 0.7 0.05 41 From Zone From Zone Above Trap 11 0.03 6.6 0.00 Above Trap 2.3 0.02 5.1 Below Count 0.3 0.08 0.02 0.01 1.3 Below Count 0.7 0.01 0.06 3.2 Above Count 7.8 13 2.7 0.01 0.06 1.2 Above Count 0.01 0.09 1.0 Bypass Count 0.01 0.01 0.03 0.00 1.6 Bypass Count 0.4 0.2 2.0 Exit 1.4 Exit

LGL Limited Page 25 Wells Dam Adult Lamprey Passage Study, 2013

5.5. Benchmark Movement Durations Entrance passage times were skewed in favor of shorter durations, with all but 2 of the 35 durations taking less than 2 hours (Figure 12). Entrance passage times ranged from 3 minutes to 17.2 hours, with a median value of 21 minutes (Table 8). Five Entrance passage times were recorded during the moderate head differential treatment, with values ranging from 8 to 21 minutes. Ten Entrance passage times were recorded during the high head differential treatment, with values ranging from 3 minutes to 2.2 hours. Treatment differences were not statistically significant (Generalized linear model (GLM) with Poisson error structure: Dev = 1.3, df = 1, P = 0.25).

Collection gallery passage times (Figure 12) were relatively fast, never exceeding 20 minutes (Table 8).

Count Station passage times were skewed in favor of shorter durations, with 18 of the durations taking less than 2 hours (Figure 12). The five slowest fish had Count Station passage times ranging from 2.5 to 23 hours.

The distribution of ‘Below count window to exit’ times were strongly skewed (Figure 12) with a median of 6 hours (Table 8). Thirteen of 22 durations were under 10 hours, and another 6 were under 20 hours. The single slowest fish was an obvious outlier, taking 149 h to move through the area. The slow fish initially moved through the Count Station in about 1 day, then waffled back and forth in this area for another 2.2 days before dropping back to below the adult fish trap. From there, the fish proceeded back up the fishway, reaching the exit after another 2.9 days.

Figure 12. Relative frequency plots showing the distribution of observed durations for passage through four specific fishway segments. Observations from the two fishways are shown with different shading. Note that x and y scales vary among panels.

LGL Limited Page 26 Wells Dam Adult Lamprey Passage Study, 2013

Table 8. Descriptive statistics of benchmark fishway passage times (in hours) for radio-tagged adult lamprey at Wells Dam.

Benchmark Duration n Minimum Median Maximum Entrance passage time 35 0.1 0.4 17.2 Collection gallery passage time 21 0.002 0.02 0.3 Lower fishway passage time 6 3.7 4.8 6.9 Count area passage time 23 0.1 0.7 23.1 Passage from below count window to exit 22 1.2 6.0 149.1 Upper fishway passage time 21 3.8 16.9 281.6 Lamprey released in tailrace 6 3.8 6.5 170.0 Lamprey released in fishway 15 6.4 43.9 281.6 Total fishway passage time 4 9.8 12.6 57.9 Below count window to count window bypass 9 0.2 0.7 26.1 Residence time in count window bypass area 9 0.01 0.6 32.0 Count window bypass to exit 7 1.0 10.3 22.0

To calculate Lower fishway passage times, we needed fish to be detected at both the entrance and the ‘above trap’ zones. Only six fish met this criterion (Figure 13), with times ranging from 3.7 to 6.9 hours (Table 8). Similarly, only four fish (all from the west fishway) met the criteria to permit calculation of Total fishway passage times (Figure 13), with values ranging from 9.8 hours to 2.4 days.

To calculate Upper fishway (‘above trap’ to exit) passage times, we needed fish to be detected both at the ‘above trap’ and exit zones. Only 6 fish that were released in the tailrace met this criterion (Figure 13), with times ranging from 3.8 hours to 7.1 days (Table 8). Of the lamprey released into the fishways, 15 met the criterion for calculation of Upper fishway passage times, with times ranging from 6.4 hours to 11.7 days (Table 8). It is possible that some of the fishway-released lamprey may have incorporated a period of post-release recovery time into their passage durations, since 8 of the 10 slowest fish were released in the fishways. In contrast, several of the faster passage times were performed by fishway-released fish, and there were certainly tailrace-released lamprey among the slowest fish (Figure 13). Moreover, Upper fishway passage times did not differ significantly between release locations (GLM with overdispersed Poisson error structure: F = 0.89, df = 1, P = 0.34).

A total of nine fish were detected by the Count Station bypass antennae. Only three of these were not detected at the ‘above count’ zone (i.e., three used the bypass area for upstream passage). The other six fish were detected at the ‘below count’ zone, then the ‘above count’ zone and then at the bypass zone. One of these fish then exited into the forebay. The other five fish waffled back and forth in the count/bypass area with repeated detections at the ‘below count’, ‘above count’ and bypass zones. Two of the five fish subsequently dropped out of the fishway. The other three eventually moved up to the fishway exit, each using the Count Station (rather than the bypass) during their final ascent. See the “Count Station Enumeration Efficiency” section (Section 4.6.10) for an assessment of the overall percentage of fish that were successfully counted when they ascended through the upper ladder.

Time required to bypass the Count Station ranged from 12 minutes to 26 hours (Table 8; Figure 14). Residence times in the bypass (including only the initial upstream movement) ranged from 23 seconds to 32 hours (Table 8; Figure 14). The three fish that exclusively used the bypass had passage times of 12 minutes, 29 minutes and 7 hours; and residence times of 23 seconds, 7 minutes and 5.2 hours. Seven fish were measured from the ‘count bypass’ to ‘fishway exit’ zones. Passage times ranged from 58 minutes to 22 hours (Table 8; Figure 14). The three fish that exclusively used the bypass had ‘count to exit’ times of 58 minutes, 2 hours, and 10.9 hours.

LGL Limited Page 27 Wells Dam Adult Lamprey Passage Study, 2013

Figure 13. Relative frequency plots showing the distribution of observed durations for passage through the lower half of each fishway, the upper half, and the entire fishway. Observations from the two fishways are shown with different shading. In the middle panel, data are shown separately for fishway-released vs. tailrace-released fish. Note that x scales vary among panels.

Figure 14. Relative frequency plots showing the distribution of observed durations for the benchmark movements involving fish that were detected by the Count Station bypass antenna. Observations from the two fishways are shown with different shading. Note that x scales vary among panels.

LGL Limited Page 28 Wells Dam Adult Lamprey Passage Study, 2013

5.6. Entrance Efficiency From 7 July to 7 October, 35 Fishway Interaction Events occurred during the high differential treatment (12 at the east fishway and 23 at the west fishway), and 12 occurred during the moderate differential treatment (six in each of the two fishways; Table 9). During the high and moderate treatments, 18 and 8 approach events were followed by successful entrance, respectively. Entrance Efficiency during the moderate differential treatment was 67% (33% at the west fishway and 100% at the east fishway). During the high differential treatment, Entrance Efficiency was 51% (43% at the west fishway and 67% at the east fishway). Overall (i.e., regardless of treatment condition), Entrance Efficiency was 55% (41% at the west fishway and 78% at the east fishway).

Table 9. Entrance and Passage Efficiencies by fishway and head-differential treatment. Also shown are the numbers of lamprey that approached, entered and passed each fishway, by treatment.

Head Diff. Entrance Passage Treatment Fishway Approached Entered Passed Efficiency Efficiency Moderate West 6 2 0 33% 0% Moderate East 6 6 0 100% 0% Moderate Both 12 8 0 67% 0%

High West 23 10 2 43% 20% High East 12 8 1 67% 13% High Both 35 18 3 51% 17%

Both West 29 12 2 41% 17% Both East 18 14 1 78% 7% Both Both 47 26 3 55% 12%

The binomial logistic model had residual deviance that was large relative to the degrees of freedom (63.8 vs 45), thus the model was re-run using a quasibinomial error distribution (i.e., a binomial distribution that allows for overdispersion). The Treatment model was not significantly different from the null model (F = 0.81, df = 1, P = 0.37), indicating that there was no significant effect of Treatment on Entrance Efficiency. The fish that made successful entries did not differ significantly in weight (F1,33 = 0.02, P = 0.89), girth (F1,33 = 0.13, P = 0.72) or length

(F1,33 = 0.68, P = 0.42) from those that were unsuccessful.

We examined the possibility that a treatment condition could have deterred lamprey by identifying approach events that immediately followed a change in flow condition. For example, if high flows deterred approach, then there should have been elevated approach counts in the period shortly after the moderate flows began. In 40 replicates over the study period, only two approaches were recorded in the first three hours of the moderate flow treatments (“Start of Moderate’, Table 10). Conversely, if the low flows deterred approach, then there should have been elevated approach counts in the period shortly after the moderate flows ended, but only seven approaches were recorded in the first three hours after the moderate flow treatment ended (“After Moderate’, Table 10). In comparison, during the ‘Start of High’ and ‘After High’ periods, which were not associated with a change in flow (the inter-treatment period had the same head differential as the high flow condition), seven and fourteen approach events were observed. The GLM model (which had residual deviance that was large relative to the degrees of freedom, 127 vs 112, and was re-run using a ‘quasipoisson’ error distribution) showed that there were no statistically significant differences among the four periods in approach rates (F = 1.3, df = 3, P = 0.28).

LGL Limited Page 29 Wells Dam Adult Lamprey Passage Study, 2013

Table 10. Approach and entrance counts around changes in treatment condition, by fishway. The first three hours of each treatment condition are labeled with the ‘Start of’ prefix. The first three hours after each treatment condition are labeled with the ‘After’ prefix.

Entrance Δ in Treatment Fishway Approached Entered Efficiency Start of Moderate West 1 1 100% Start of Moderate East 1 1 100% Start of Moderate Both 2 2 100% Start of High West 7 2 29% Start of High East 0 0 - Start of High Both 7 2 29%

Entrance Δ in Treatment Fishway Approached Entered Efficiency After Moderate West 3 1 33% After Moderate East 4 4 100% After Moderate Both 7 5 71% After High West 9 8 89% After High East 5 5 100% After High Both 14 13 93%

In between treatments, the head differential was set to 0.48 m, which was the same as that during the high differential treatment (Figure 8). To bolster sample sizes, we included the Inter-treatment Periods as part of the high differential treatment and re-examined Entrance Efficiency during this condition. In all, 62 approach events were recorded (40 west, 22 east), of which 42 entered (24 west, 18 east), thus Entrance Efficiency was 68% (60% west, 82% east).

5.7. Passage Efficiency Although the head differential treatment only affected the flows in the entrance and collection gallery, we nevertheless tested for indirect effect on Passage Efficiency. Passage Efficiency during the moderate differential treatment was 0% (0 of 8), whereas that during the high differential treatment was 17% (20% at the west fishway and 13% (3 of 18) at the east fishway; Table 9). The binomial logistic model that included a ‘Treatment’ effect was not significantly different from the null model (Dev = 2.4, df = 1, P = 0.12), indicating that there was no significant effect of Treatment on Passage Efficiency.

Total Fishway Passage Efficiency in 2004-2008 was 20.8%. Ignoring treatment conditions, season-wide Total Fishway Passage Efficiency in 2013 was 9.5%. The 11.3% difference was not statistically significant (Dev = 1.8, df = 1, P = 0.17). Lower Fishway Passage Efficiency in 2004-2008 was 25%. Ignoring treatment conditions, season-wide Lower Fishway Passage Efficiency in 2013 was 14.3%. The 10.7% difference was not statistically significant (Dev = 1.3, df = 1, P = 0.25). Upper Fishway Passage Efficiency in 2004-2008 was 88.9%, whereas that in 2013 was 66.7%. The 22.2% difference was not statistically significant (Dev = 1.3, df = 1, P = 0.25).

5.8. Count Station Passage In all, 26 fish were detected in the ‘below Count Station’ zone, of which 23 were next detected in the ‘above Count Station’ zone. Only three fish were detected in the bypass and not in the ‘above Count Station’ zone. It should be

LGL Limited Page 30 Wells Dam Adult Lamprey Passage Study, 2013

noted that subsequent to detections in the ‘above Count Station’ zone, six of the 23 fish were detected in the bypass zone. Nevertheless, these fish were counted as having passed through the Count Station successfully (and were hence available for enumeration) regardless of subsequent interactions with the bypass antenna. Thus, Count Station Passage was 94% in the west fishway, 78% in the east fishway, and 88% overall (Table 11). The binomial logistic model that included a ‘Fishway’ effect was not significantly different from the null model (Dev = 1.5, df = 1, P = 0.23), indicating that there was no significant difference in Count Station Passage rates between the east and west ladders.

Table 11. Count Station Passage, by fishway. Also shown are the numbers of lamprey detected at the ‘below count’ detection zone, and the numbers subsequently detected at the ‘above count’ zone (i.e., fish that were successfully herded through the video-enumeration area) and in the ‘bypass’ zone (i.e., fish that moved upstream without entering the enumeration area).

Bypassed Count Area Fishway Below Count Above Count Count Area Passage

West 17 16 1 94% East 9 7 2 78% Both 26 23 3 88%

In 2007, 11 fish reached the ‘below count’ zone, of which 7 passed through the Count Station, and four fish appeared to have used the bypass exclusively (Douglas PUD, unpublished data). In 2008, four fish moved upstream, one used the Count Station and three used the bypass (Robichaud et al. 2009). The Count Station Passage rate prior to the installation of the new picketed leads was thus 53.3% (8 of 15), or 34.7% lower than that in 2013. The binomial logistic model testing for the effect of the new picketed leads was significantly different from the null model (Dev = 6.2, df = 1, P = 0.013), indicating that the picketed leads significantly improve the probability that lamprey will move through the Count Station instead of the bypass.

5.9. Count Station Enumeration Efficiency In all, 26 radio-tagged lamprey passed through the upper fishway, 23 of which used the video Count Station. Because some fish dropped downstream after passing, and made repeated movement through the count window, there was a total of 37 events in which a radio-tagged lamprey was known to move upstream past the video window. Of these, 19 occurred at the time when a lamprey was tallied by count video technicians (Table 12). One of the 23 fish passed the count window in the upstream direction on 12 occasions, two of which were tallied by the video technicians. A second fish passed upstream on four occasions, and was counted twice. A third fish passed twice without being detected. The remaining 19 fish each passed the count window once in the upstream direction, of which 15 were counted. There was one fish tallied on the video that did not correspond with the presence of a radio-tagged fish.

The Count Station Enumeration Efficiency was 68% in the west fishway, 33% in the east fishway, and 51% overall (Table 12). The binomial logistic model had residual deviance that was large relative to the degrees of freedom (46.6 vs 35), thus the model was re-run using a quasibinomial error distribution (allowing for overdispersion). The Fishway model was significantly different from the null model (F = 4.4, df = 1, P = 0.043), indicating that the between-fishway difference was statistically significant.

In 2007, nine fish passed through the Count Station (LGL and Douglas PUD 2008), of which 3 were tallied by video count technicians. In 2008, 1 fish passed and was tallied (Robichaud et al. 2009). The Count Station Enumeration

LGL Limited Page 31 Wells Dam Adult Lamprey Passage Study, 2013

Table 12. Count Station Enumeration Efficiencies, by fishway. Also shown are the numbers of times a radio- tagged lamprey was tracked moving upstream through the Count Station, and the number of these that were detected on the enumeration video.

Number of times that Number of radio-tagged radio-tagged lamprey lamprey passage events were tracked moving during which a lamprey Count Area upstream past the was recorded in the video Enumeration Fishway Count Area enumeration data Efficiency West 19 13 68% East 18 6 33% Both 37 19 51%

Efficiency prior to the installation of the new picketed leads was thus 40% (4 of 10), or 11% lower than that in 2013. The binomial logistic model testing for the effect of the new picketed leads had residual deviance that was large relative to the degrees of freedom (64.7 vs 45), thus the model was re-run using a quasibinomial error distribution. The full model was not significantly different from the null model (F = 0.4, df = 1, P = 0.53), indicating that the picketed leads did not significantly improve Count Station Enumeration Efficiency.

6. Discussion

6.1. Entrance Efficiency and Head Differentials At Columbia River hydropower dams, elevated flow velocities are created at fishway entrances in order to attract adult salmonids into the entryways. However, these elevated flows are thought to impede Pacific lamprey migrations (e.g., Daigle et al. 2005), especially if water velocities greater than 1.2 m/s (3.9 ft/s) are encountered (Keefer et al. 2008). Johnson et al. (2009a) found evidence that high velocities at Bonneville Dam created difficulties for lamprey that were attempting to enter the fishway, so, although more lamprey were attracted to the entrances during high velocities treatments, disproportionately fewer lamprey entered under those conditions.

At Wells Dam, there is interest in finding a means of increasing lamprey entrance efficiency without interfering with salmon migration. Since the majority of Wells Dam lamprey movements occur at night (Robichaud et al. 2009), and to address one objective of the PLMP, it was decided to test whether nocturnal reductions in fishway entrance velocities would improve lamprey Entrance Efficiencies. Typically, Douglas PUD maintains fishway entrance velocities at 3.0 m/sec (9.8 ft/s), which produces a head differential equal to that of our ‘high differential’ treatment (i.e., 0.48 m or 1.5 ft head differential). Every second night for seven hours the flows were reduced to 2.4 m/s (7.9 ft/s), producing our ‘moderate differential’ treatment condition (0.31 m or 1.0 ft head differential). Over the study period, 80 treatment conditions were tested, including 40 replicate tests of each of the two treatments, and a total of 47 fishway approaches were recorded. Data were used to test the following null hypothesis:

• H0: Flow differential consisting of one entrance velocity treatment has no effect on entrance success over another entrance velocity treatment.

• H1: Flow differential consisting of one entrance velocity treatment has an effect on improving entrance success over another entrance velocity treatment.

LGL Limited Page 32 Wells Dam Adult Lamprey Passage Study, 2013

Entrance Efficiencies were higher during the lower flow treatment (67%) than in the higher flow treatment (51%), but the difference was not statistically significant. The power of the test was low due to sample size limitations and presumably due to variability within each treatment (these tests perform better when efficiencies are closer to 0 or 100%). Power analysis showed that more than 550 ‘approach events’ would have been required for the model to declare this 16% difference as statistically significant. Thus, we cannot reject the null hypothesis, but given low statistical power, the ‘non-significant’ result should not be viewed as evidence for a lack of true difference in Entrance Efficiency between treatments. The result should therefore be interpreted somewhat cautiously.

The total number of approach events was 12 during the moderate flow treatments and 35 during high flow treatments. There was no way to test whether this difference was a result of the treatment conditions, as we did not track the lamprey abundances in the tailrace. However, the result is consistent with the notion that the fish were more attracted to the fishway entrances under higher flow conditions.

Previous radio telemetry studies at Wells Dam presumably took place under the high head differential condition, the baseline operational condition during the adult lamprey migration period. If so, then data from our ‘high head differential’ treatment are the most appropriate for comparison. In 2013, the ‘high head differential’ entrance efficiency (51%), was notably higher than that previously reported. In 2004, Wells Dam Entrance Efficiency was 30% (Nass et al. 2005). In 2007 and 2008, average Entrance Efficiency over a two-year study period was 27% (LGL and Douglas PUD 2008, Robichaud et al. 2009).

Previous studies have attempted to examine the effect of different head differential treatments on Wells Dam lamprey Entrance Efficiency. In 2009 and 2010, a DIDSON study (Johnson et al. 2010, 2011) produced entrance efficiency estimates of 50% for the high head differential (n=2), 100% for the moderate head differential (n=3), and 50% for a lower head differential (0.15 m or 0.5 ft; n=2). Sample sizes were low (7 over two years), and the low accuracy of the estimates preclude them from being taken at face value. Regardless, the overall Entrance Efficiency (71%, or 5 of 7 fish) was higher than that measured in studies conducted between 2004 and 2008. By contrast, the 2009-2010 Entrance Efficiency results were more on par with those observed in 2013. That said, Johnson et al. (2011) indicated that their DIDSON results are not truly comparable to those from radio-telemetry studies, since DIDSON cannot distinguish among individual fish.

The effect of flows on adult lamprey Entrance Efficiency has been examined at Bonneville Dam. In 2009, nocturnal reductions in fishway flows produced improved entrance efficiencies which were statistically significant at the north entrance (26% vs 2%) but not at the south entrance (32% vs. 24%; Johnson et al. 2009a). Moser et al. (2002) found no significant differences in the proportion of successful lamprey entries during high (2.4 m/s, or 7.9 ft/s) and low (1.2 m/s, or 3.9 ft/s) velocity conditions. These results suggest that flow might not be the only factor in determining lamprey entrance efficiency at some hydro projects.

6.2. Effects of Picketed Leads

6.2.1. Count Station Passage 11 One of the main objectives of the study was to evaluate the efficacy of new /16 inch picketed leads that were installed in order to block lamprey from bypassing the count window. The new leads appeared to perform according to plan. In 2013, 26 fish were detected in the ‘below Count Station’ zone, of which 23 appeared to have passed through the Count Station, and three through the bypass. Count Station Passage was calculated to be 88% (94% in the west fishway, 78% in the east fishway). In order to evaluate the efficacy of the picketed leads, the 2013 results were compared to results from similar previous studies (LGL and Douglas PUD 2008, Robichaud et al. 2009), when the old picketed leads were in place. The goal was to test the following null hypothesis:

LGL Limited Page 33 Wells Dam Adult Lamprey Passage Study, 2013

• H0: The proportion of tagged lamprey passing the count window is similar to previous studies.

• H1: The proportion of tagged lamprey passing the count window is dissimilar to previous studies.

LGL and Douglas PUD (2008) reported on a lamprey passage study conducted at Wells Dam in 2007. No tabularization of Count Station passage data were presented, but the text indicated that 11 fish were detected in the ‘below Count Station’ zone, of which all but two were detected in the ‘above Count Station’ zone. To clarify matters, the raw detection data (Douglas PUD, unpublished data) were re-examined, and it was determined that 7 of the 11 fish passed through the Count Station (even though some later encountered the bypass) and four used the bypass. Another lamprey passage study was conducted at Wells Dam in 2008 (Robichaud et al. 2009), during which four fish moved through the upper fishway. One fish passed through the video Count Station, and the other three fish used the bypass (although one showed a subsequent single hit in the ‘above Count Station’ zone). Thus, in 2008 the Count Station Passage rate was 25%. Combined, the Count Station Passage rate prior to installation of the new picketed leads was 53.3% (8 of 15), or 34.7% lower than the post-installation rate. This difference was statistically significant, thus, we reject the null hypothesis in favor of the alternative.

6.2.2. Count Station Enumeration Efficiency To make informed management decisions, Douglas PUD attempts to gather accurate annual estimates of the numbers of lamprey passing through the Wells Project. However, lamprey do not get tallied if they use the bypass 11 instead of passing the video count window. To improve video count accuracy, the new /16 inch picketed leads were installed to block lamprey from bypassing the count window. In addition to testing for improved Count Station Passage rates, it was also necessary to test if the fish moving past the window were being detected and counted by the video technicians. In 2013, 23 radio-tagged lamprey passed through the video Count Station, some moving past several times, for a total of 37 upstream movement events. Of these, 19 were detected (51%). Count Station Enumeration Efficiency was significantly higher in the west fishway (68%) than in the east fishway (33%).

In testing the effects of the new picketed leads, the following null hypothesis was tested:

• H0: The proportion of tagged lamprey passing the count window that are enumerated by the Video Count technicians is similar to previous studies.

• H1: The proportion of tagged lamprey passing the count window that are enumerated by the Video Count technicians is dissimilar to previous studies.

In 2007, 11 fish were known to pass through the upper fishway, but only 9 were detected in the ‘above count’ zone (the remaining two must have used the bypass exclusively). The video counters tallied only 3 lamprey at times that corresponded with the presence of a radio-tagged fish. No tabularization of Count Station passage data were presented, but from the text (LGL and Douglas PUD 2008) we calculated that Count Station Enumeration Efficiency in 2007 was 33% (3 of 9). In 2008, only one radio-tagged lamprey passed the video count window, and it was tallied by the video technicians (Robichaud et al. 2009), hence Count Station Enumeration Efficiency in 2008 was (100%). Combined, the Count Station Enumeration Efficiency prior to installation of the new picketed leads was 40% (4 of 10), or 11% lower than the post-installation efficiency. Although the difference was not statistically significant, the test was hampered by a severe lack of power (power analysis showed that ~1500 lamprey would have to have passed for the model to declare this 11% difference as statistically significant). Thus, we cannot reject the null hypothesis, but given low statistical power, the ‘non-significant’ result should not be viewed as evidence for a lack of improvement resulting from the installation of the new, smaller picketed leads.

While we cannot say with certainty how fish get missed (the video monitoring is recorded 24 hours per day), it is likely that lamprey are finding a passage route that is out of the field of view of the camera. For example, gaps

LGL Limited Page 34 Wells Dam Adult Lamprey Passage Study, 2013

exist around the adjustable crowder (e.g., where it meets the bottom plate of the count area) that extend downstream beyond the view of the camera, creating a potential passage route that would be invisible to the counters. In fact, video footage from 2013 showed a lamprey swimming beneath the bottom plate and out of view of the camera.

6.2.3. Bypass Exclusion When planning to install the new picketed leads, there was some concern that excluding fish from the bypass might reduce Passage Efficiency, or increase holding time below the window, or increase the frequency of dropback. Upper fishway Passage Efficiencies were lower in 2013 (67%) than in 2004 (75%; Nass et al. 2005) or 2007-2008 (100%; Robichaud et al. 2009). Count Station passage times were longer in 2013 (8 minutes to 23 h, median 44 minutes, n = 23) than previously (15 s to 16.9 h, median 1.85 minutes, n = 15 in 2007-2008; unpublished data). Dropback from the Count Stations was seen in 4 of 23 cases in 2013, but was not recorded in 2007-2008. None of these differences were statistically significant, but together they may be suggestive of a passage issue in the Count Station.

6.3. Passage Efficiencies and Movement Durations While the main goals of the study were to test the head differential treatments and the efficacy of the new picketed leads, this study also added to the growing body of adult lamprey fishway passage data for Wells Dam.

Passage Efficiencies have varied among years. In 2004, Nass et al. (2005) reported that 10 fish performed 12 entrance events, of which eight passed the gallery, four reached the ‘below trap’ zone, and three eventually passed into the forebay. Thus, 67% of entrance events reached as far as the gallery, 50% of which reached the trap, and 75% of these passed the dam. The Total Fishway Passage Efficiency was 25%. In 2007, Entrance Efficiency was 14%, and only one fish passed into the forebay (LGL and Chelan PUD 2008). In 2008, 15 approaches lead to 5 entrances, and 1 passage into the forebay (Robichaud et al. 2009). The 2007-2008 data had 33% Lower Fishway Passage Efficiency, 100% Upper Fishway Passage Efficiency, and 33% Total Fishway Passage Efficiency. In 2013, 60% of entrants reached the gallery, 71% of these fish reached Weir 1, 52% of the Weir 1 fish proceeded to Weir 7, and 64% of the Weir 7 fish reached the ‘below trap’ zone. Overall, 9 of 63 entrants reached the above trap zone (14% Lower Fishway Passage Efficiency). Six of nine fish successfully passed from the ‘above trap’ zone to the fishway exit (67% Upper Fishway Passage Efficiency). Total Fishway Passage Efficiency was 10% in 2013. None of the three fishway passage efficiency metrics in 2013 differed significantly from those in previous studies.

Passage durations have varied among years. In 2004, total fishway passage time averaged 0.3 days. Travel times were longest in the reach leading up to the ‘below trap’ area, and in the ‘above trap’ to ‘below video’ reach (Nass et al. 2005). In 2007, a single fish passed from entrance to exit, doing so in 31.5 hours, including 6.1 hours in the lower fishway, and 12.5 hours in the adult fish trap. Upper fishway passage times in 2007 had a median value of 7.9 h (LGL and Douglas PUD 2008). In 2008, movement from the gallery to the ‘below trap’ zone took 1.7 to 5.5 hours. Upper fishway passage times ranged from 2.6 to 15.1 hours. The ‘above trap’ to ‘below video’ reach was again one of the slowest, taking 2 to 14 hours (77-92% of the total upper fishway passage time; Robichaud et al. 2009). Similarly in 2013, lower fishway passage times ranged from 3.7 to 6.9 hours (median 4.8 hours). Some upper fishway passage times in 2013 were similar to those from previous years, but a few very slow fish resulted in passage times ranging as high as 11.7 days. The slowest individuals were released into the fishways, near the ‘above trap’ zone. These fish must have delayed movement post-release (possibly recovering from surgery or other handling), as their travel times from ‘above trap’ to the ‘below video’ zone ranged from 47% to 99% of their overall upper fishway passage times. In 2013, total fishway passage times ranged from 9.8 hours to 2.4 days (median 12.6 hours).

LGL Limited Page 35 Wells Dam Adult Lamprey Passage Study, 2013

One of the study goals was to test the hypothesis:

• H0: There is no difference in passage metrics compared to other mainstem Columbia River projects.

• H1: Passage metrics for lamprey differ compared to other mainstem Columbia River projects.

The Total Fishway (entry to exit) Passage Efficiency and Passage Duration metrics are standard, and have been reported for most of the mainstem Columbia River dams (Table 13). Wells passage efficiencies (<21%) have clearly been low relative to other mid-Columbia dams (55.5% to 86.7%). Bonneville Dam had the lowest passage efficiencies of the lower-Columbia dams, ranging from 41.3% to 57.7%. Passage Durations were highly variable among dams and among years (Table 13). Nevertheless, median passage times through mid-Columbia fishways appear to be shorter, for the most part than those in the lower Columbia.

Table 13. Summary of reported fishway Passage Efficiencies (proportion of fishway entrants that successfully pass the exit into the forebay) and Passage Durations (days) for lower and mid-Columbia mainstem hydropower dams. Passage Durations were reported as medians or ranges.

Bonneville The Dalles John Day McNary 1997-2002 2007-2010 1997-2002 2007-2010 2000-2002 d 2009 2000 2005-2010

Passage Efficiency 47.4%-57.7% n 41.3%-54.0% n 72.4%-83.3% n 72.9%-88.3% n 46.7%-54.3% n 56.4% n 91.7% n 69.2%-88.7% n a,c,d h,i a,c,d h,i d k Passage Duration (d) med= 4.4-11.1 med= 3.0-3.1 med= 2.0-4.0 med= 0.8-1.1 med= 1.3-4.2 med= 0.6-3.0

Priest Rapids Wanapum Rocky Reach Wells 2002 b 2010-11 ℓ 2002 b 2010-11 ℓ 2004 f 2004-2008 e,g,j 2013 m

Passage Efficiency 74.2% 66.5% 86.7% 67.0% 55.5% 20.8% 9.5% Passage Duration (d) med= 1.0-1.2 med= 2.2 med= 1.1-1.8 med= 2.1 range: 0.3-1.3 med= 0.5 a: Moser et al. 2002; b: Nass et al. 2002; c: Moser et al. 2003; d: Moser et al. 2005; e: Nass et al. 2005; f: Stevenson et al. 2005; g: LGL and Douglas PUD 2008; h: Johnson et al. 2009a; i: Johnson et al. 2009b; j: Robichaud et al. 2009; k: Keefer et al. 2011; ℓ: B.Nass, LGL, unpublished preliminary data; m: this report; n: Keefer et al. 2012.

7. Acknowledgements We thank the many Douglas PUD employees at Wells Dam for their support throughout the study. Chas Kyger is thanked in particular for his support in project implementation. Bryan Nass (LGL) and Elmar Plate (LGL) deployed and tested the monitoring stations, and Lynda Andrews (LGL) was responsible for data downloads and maintenance of the stations throughout the project. Anita Blakley (LGL) managed and performed all tagging operations. We thank Ralph Lampman and Patrick Luke (Prosser Hatchery) and Kyle Hatch (Blue Leaf Environmental) for collecting and transporting lamprey from Bonneville and Priest Rapids dams, respectively. Douglas PUD Fish Enumerators, Betty Walters, Ginger Easley-Zacher, and Sylvia Robertson were supportive in providing detailed accounts of lamprey passage. Mary Mayo (Douglas PUD) and Marion McIntosh (LGL) managed the study’s budgets. We thank Jaimie Imrie (LGL) for help with implementation of Telemetry Manager. Karl English (LGL) and Anita Blakley provided comments on an earlier version of this report.

8. Literature Cited BioAnalysts. 2000. A status of Pacific lamprey in the Mid-Columbia Region. Report for Public Utility District No. 1 of Chelan County. Wenatchee, Washington.

LGL Limited Page 36 Wells Dam Adult Lamprey Passage Study, 2013

Close, D.A., M.S. Fitzpatrick, and H.W. Li. 2002. The ecological and cultural importance of a species at risk of extinction, Pacific lamprey. Fisheries 27(7):19-25.

Cooke, S.J., G.N. Wagner, R.S. Brown, and K.A. Deters. 2011a. Training considerations for the intracoelomic implantation of electronic tags in fish with a summary of common surgical errors. Reviews in Fish Biology and Fisheries 21:11-24.

Cooke, S.J., C.M. Woodley, M.B. Eppard, R.S. Brown, and J.L. Nielsen. 2011b. Advancing the surgical implantation of electronic tags in fish: a gap analysis and research agenda based on a review of trends in intracoelomic tagging effects studies. Reviews in Fish Biology and Fisheries 21:127-151.

Daigle. W.R., C.A. Peery, S.R. Lee, and M.L. Moser. 2005. Evaluation of adult Pacific lamprey passage and behavior in an experimental fishway at Bonneville Dam. Idaho Cooperative Fish and Wildlife Research Unit, Technical Report 2005-1.

Hart, J.L. 1973. Pacific fishes of Canada. Bulletin of the Fisheries Research Board of Canada 180:740 p.

Jackson, A.D., D.R. Hatch, B.L. Parker, M.S. Fitzpatrick, D.A. Close, and H.W. Li. 1997. Pacific lamprey research and restoration: Annual report, 1997. Report for Bonneville Power Administration, Division of Fish and Wildlife.

Johnson, E.L., C.A. Peery, M.L. Keefer, C.C. Caudill, and M.L. Moser. 2009a. Effects of lowered nighttime velocities on fishway entrance success by Pacific lamprey at Bonneville Dam and fishway use summaries for lamprey at Bonneville and The Dalles dams, 2007. Idaho Cooperative Fish and Wildlife Research Unit Technical Report 2009-2.

Johnson, E.L., T.S. Clabough, M.L. Keefer, C.C. Caudill, C.A. Peery, and M.L. Moser. 2009b. Effects of lowered nighttime velocities on fishway entrance success by Pacific lamprey at Bonneville Dam and fishway use summaries for lamprey at Bonneville and The Dalles dams, 2008. Idaho Cooperative Fish and Wildlife Research Unit Technical Report 2009-10.

Johnson, P.N., B. Le, and J.G. Murauskas. 2010. Assessment of adult Pacific lamprey response to velocity reductions at Wells Dam fishway entrances. Report for Public Utility District No. 1 of Douglas County, East Wenatchee, WA.

Johnson, P.N., B. Le, and B. Patterson. 2011. Assessment of adult Pacific lamprey response to velocity reductions at Wells Dam fishway entrances. Report for Public Utility District No. 1 of Douglas County, East Wenatchee, WA.

Keefer, M.L., W.R. Daigle, C.A. Peery, and M.L. Moser. 2008. Adult Pacific lamprey bypass structure development: tests in an experimental fishway, 2004-2006. Idaho Cooperative Fish and Wildlife Research Unit Technical Report 2008-10.

Keefer, M.L., C.T. Boggs, C.C. Caudill, and M.L. Moser. 2011. Adult Pacific lamprey migrations and behaviour at McNary Dam 2005-2010. Idaho Cooperative Fish and Wildlife Research Unit Technical Report 2011-9.

Keefer, M.L., T.C. Clabough, M.A. Jepson, E.L. Johnson, C.T. Boggs and C.C. Caudill . 2012. Adult Pacific lamprey passage: Data synthesis and fishway improvement prioritization tools. Idaho Cooperative Fish and Wildlife Research Unit, Technical Report 2012-8-DRAFT.

LGL and Douglas County PUD. 2008. Adult Pacific lamprey passage and behavior study for Wells Hydroelectric Project. Report for Public Utility District No. 1 of Douglas County, East Wenatchee, Washington.

LGL Limited Page 37 Wells Dam Adult Lamprey Passage Study, 2013

Liedtke, T.L., J.W. Beeman, and L.P. Gee. 2012. A standard operating procedure for the surgical implantation of transmitters in juvenile salmonids: U.S. Geological Survey Open-File Report 2012-1267.

Moser, M.L., P.A. Ocker, L.C. Stuehrenberg, and T.C. Bjornn. 2002. Passage efficiency of adult Pacific lampreys at hydropower dams on the lower Columbia River, U.S.A. Transactions of the American Fisheries Society 131:956- 965.

Moser, M.L, D.A. Ogden, S.G. McCarthy, and T.C. Bjornn. 2003. Migration behavior of adult Pacific lamprey in the lower Columbia River and evaluation of Bonneville Dam modifications to improve passage, 2001. Report for U.S. Army Corps of Engineers, Portland, Oregon.

Moser, M.L., D.A. Ogden, and C.A. Peery. 2005. Migration behavior of adult Pacific lamprey in the lower Columbia River and evaluation of Bonneville Dam modifications to improve passage, 2002. Report for U.S. Army Corps of Engineers, Portland District, Portland, Oregon.

Nass, B., C. Sliwinski, K.K. English, L. Porto, and L. Hildebrand. 2002. Assessment of adult lamprey migratory behavior at Wanapum and Priest Rapids Dams using radio-telemetry techniques, 2001-2002. Report for Public Utility District No. 2 of Grant County, Ephrata, Washington.

Nass, B.L., C. Sliwinski, and D. Robichaud. 2005. Assessment of adult Pacific lamprey migratory behavior at Wells Dam using radio-telemetry techniques, 2004. Report for Public Utility District No. 1 of Douglas County, East Wenatchee, Washington.

Nislow, K.H., and B.E. Kynard. 2009. The role of anadromous sea lamprey in nutrient and material transport between marine and freshwater environments. American Fisheries Society Symposium 69:485–494.

Robichaud, D., B. Nass, and Douglas PUD. 2009. Adult Pacific lamprey passage and behavior study (adult lamprey passage study). Second Year Final Report. Report for Public Utility District No. 1 of Douglas County, East Wenatchee, Washington.

Stevenson, J.R., P. Westhagen, D.J. Snyder, J.R. Skalski, and A.E. Giorgi. 2005. Evaluation of adult Pacific lamprey passage at Rocky Reach Dam using radiotelemetry techniques, 2004. Report for Public Utility District No. 1 of Chelan County, Wenatchee, Washington.

Wydoski, R.S., and R.L. Whitney. 2003. Inland Fishes of Washington. 2nd ed., rev. and expanded. American Fisheries Society. Bethesda, Maryland.

Wynne, F.S., and W.A. Wurts. 2011. Transportation of warmwater fish: equipment and guidelines. SRAC Publication No. 390.

LGL Limited Page 38 Wells Dam Adult Lamprey Passage Study, 2013

Appendices

LGL Limited Page 39 Wells Dam Adult Lamprey Passage Study, 2013

Appendix Table A. Biometric, surgical timing, and release data for the Pacific lamprey tagged as part of the Wells Adult Lamprey Passage and Enumeration Study.

Tag Information Tagging Procedure Benchmarks (hh:mm) Fish Measurements Tag Release Information Fish # Session Date Fish Source Freq Code PIT Tag ID Start MS-222 Start Surgery End Surgery Weight (g) Girth (cm) Length (cm) Burden Location Time Comments 1 S1 16 Jul Bonneville 151.3 115 3DD.003BB8F21B 12:40 12:48 12:51 422.5 11 67 0.52% Tailrace 15:35 2 S1 16 Jul Bonneville 151.3 114 3DD.003BB8F231 12:54 12:58 13:01 505 11.5 75 0.44% Tailrace 15:35 3 S1 16 Jul Bonneville 151.3 113 3DD.003BB8F205 13:05 13:09 13:14 805.5 14 82 0.27% Tailrace 15:35 4 S1 16 Jul Bonneville 151.3 112 3DD.003BB8F211 13:13 13:17 13:20 553 12.5 72 0.40% Tailrace 15:35 Unable to anchor antenna 5 S1 16 Jul Bonneville 151.3 111 3DD.003BB8F230 13:18 13:23 13:26 415 12 68 0.53% Tailrace 15:35 6 S1 16 Jul Bonneville 151.3 116 3DD.003BB8F228 13:59 14:05 14:10 493.5 12.5 72.5 0.45% Tailrace 15:35 7 S1 16 Jul Bonneville 151.3 117 3DD.003BB8F21D 14:09 14:15 14:18 460.5 12.5 71 0.48% Tailrace 15:35 8 S1 16 Jul Bonneville 151.3 118 3DD.003BB8F20C 14:25 14:29 14:32 459 11.5 71 0.48% Tailrace 15:35 9 S1 16 Jul Bonneville 151.3 119 3DD.003BB8F20D 14:43 14:47 14:52 398.5 11 66 0.55% Tailrace 15:35 10 S1 16 Jul Bonneville 151.3 120 3DD.003BB8F239 14:50 14:54 14:57 392.5 11 66 0.56% Tailrace 15:35 Bled during surgery 11 S1 16 Jul Bonneville 151.3 121 3DD.003BB8F22F 14:56 15:00 15:04 494 12 69 0.45% Tailrace 15:35 Scar on side 12 S1 16 Jul Bonneville 151.3 122 3DD.003BB8F219 15:03 15:07 15:11 550 12 68 0.40% Tailrace 15:35 13 S1 16 Jul Bonneville 151.3 123 3DD.003BB8F1EB 15:08 15:13 15:17 504.5 11.5 70 0.44% Tailrace 15:35 14 S1 16 Jul Bonneville 151.3 124 3DD.003BB8F1F0 15:15 15:19 15:23 464 11 71 0.47% Tailrace 15:35 15 S1 16 Jul Bonneville 151.3 125 3DD.003BB8F1F5 16:06 16:14 16:17 545 12.5 70 0.40% Tailrace 18:48 Long time to sedate 16 S1 16 Jul Bonneville 151.3 126 3DD.003BB8F22D 16:15 16:19 16:22 509.5 12 71 0.43% Tailrace 18:48 17 S1 16 Jul Bonneville 151.3 127 3DD.003BB8F1E8 16:21 16:25 16:29 506 12 71 0.43% Tailrace 18:48 18 S1 16 Jul Bonneville 151.3 128 3DD.003BB8F22B 16:26 16:30 16:33 464 12 69 0.47% Tailrace 18:48 19 S1 16 Jul Bonneville 151.3 129 - 16:31 16:35 16:39 458 12 66 0.48% Tailrace 18:48 PIT not recorded properly 20 S1 16 Jul Bonneville 151.3 130 3DD.003BB8F1EE 16:40 16:45 16:48 540.5 12 73 0.41% Tailrace 18:48 21 S1 16 Jul Bonneville 151.3 131 3DD.003BB8F217 16:46 16:53 16:56 488.5 12.5 68 0.45% Tailrace 18:48 22 S1 16 Jul Bonneville 151.3 132 3DD.003BB8F207 16:54 16:56 17:00 406.5 11.5 66 0.54% Tailrace 18:48 23 S1 16 Jul Bonneville 151.3 133 3DD.003BB8F20E 16:57 17:02 17:05 395.5 11 68 0.56% Tailrace 18:48 24 S1 16 Jul Bonneville 151.3 134 3DD.003BB8F200 17:03 17:08 17:12 407.5 11 69 0.54% Tailrace 18:48 25 S1 16 Jul Bonneville 151.3 135 3DD.003BB8F1F8 17:09 17:14 17:17 432.5 11.5 67 0.51% Tailrace 18:48 26 S1 16 Jul Bonneville 151.3 65 3DD.003BB8F20A 17:15 17:21 17:25 514.5 12 70 0.43% Tailrace 18:48 27 S1 16 Jul Bonneville 151.3 64 3DD.003BB8F1E0 17:22 17:26 17:29 453 11.5 70 0.49% Tailrace 18:48 28 S1 16 Jul Bonneville 151.3 63 3DD.003BB8F20B 17:34 17:38 17:42 500.5 12.5 71 0.44% Tailrace 18:48 29 S1 16 Jul Bonneville 151.3 62 3DD.003BB8F218 17:40 17:45 17:49 475.5 12.5 68 0.46% Tailrace 18:48 30 S1 16 Jul Bonneville 151.3 61 3DD.003BB8F206 17:46 17:52 17:55 411.5 11.5 66 0.53% East Fishway 19:15 31 S1 16 Jul Bonneville 151.3 70 3DD.003BB8F21F 17:53 17:58 18:02 499.5 12 77 0.44% West Fishway 19:21 32 S1 16 Jul Bonneville 151.3 69 3DD.003BB8F106 17:59 18:05 18:09 363 10.5 63 0.61% West Fishway 19:21 33 S1 16 Jul Bonneville 151.3 68 3DD.003BB8F213 18:07 18:12 18:16 516 12 73 0.43% East Fishway 19:15 34 S1 16 Jul Bonneville 151.3 67 3DD.003BB8F210 18:13 18:17 18:22 496 12 71 0.44% West Fishway 19:21 35 S1 16 Jul Bonneville 151.3 66 3DD.003BB8F222 18:18 18:23 18:27 428 12 65 0.51% East Fishway 19:15 36 S1 16 Jul Bonneville 3DD.003BB8F21E 358 10.5 68 0.61% Tailrace 18:48 Scar injury, PIT Tag only 37 S1 16 Jul Bonneville 3DD.003BB8F216 417.5 11.5 67 0.53% Tailrace 18:48 Scar injury, PIT Tag only 38 S2 23 Jul Bonneville 151.3 75 3DD.003BB8F214 12:21 12:26 12:30 578.5 12.5 72.5 0.38% Tailrace 14:44 39 S2 23 Jul Bonneville 151.3 74 3DD.003BB8F1FE 12:42 12:48 12:51 384.5 11.5 64 0.57% Tailrace 14:44 40 S2 23 Jul Bonneville 151.3 73 3DD.003BB8F1DB 12:51 12:56 12:59 623 12.5 75 0.35% Tailrace 14:44 41 S2 23 Jul Bonneville 151.3 72 3DD.003BB8F234 12:57 13:02 13:06 344 10.5 63 0.64% Tailrace 14:44 Bled during surgery 42 S2 23 Jul Bonneville 151.3 71 3DD.003BB8F1DE 13:05 13:08 13:11 452.5 11.5 69.5 0.49% Tailrace 14:44 43 S2 23 Jul Bonneville 151.3 76 3DD.003BB8F232 13:10 13:14 13:17 542 12.5 73 0.41% Tailrace 14:44 44 S2 23 Jul Bonneville 151.3 77 3DD.003BB8F1E3 13:17 13:22 13:25 556 12 72 0.40% Tailrace 14:44 45 S2 23 Jul Bonneville 151.3 78 3DD.003BB8F1E9 13:23 13:27 13:30 414.5 11 67 0.53% Tailrace 14:44 …continued

LGL Limited Page 40 Wells Dam Adult Lamprey Passage Study, 2013

Appendix Table A continued.

Tag Information Tagging Procedure Benchmarks (hh:mm) Fish Measurements Tag Release Information Fish # Session Date Fish Source Freq Code PIT Tag ID Start MS-222 Start Surgery End Surgery Weight (g) Girth (cm) Length (cm) Burden Location Time Comments 46 S2 23 Jul Bonneville 151.3 79 3DD.003BB8F1DD 13:28 13:33 13:36 412 11.5 68 0.53% Tailrace 14:44 47 S2 23 Jul Bonneville 151.3 80 3DD.003BB8F237 13:33 13:39 13:42 355 10 62.5 0.62% Tailrace 14:44 48 S2 23 Jul Bonneville 151.3 85 3DD.003BB8F1EF 13:43 13:47 13:51 477 11.5 70.5 0.46% Tailrace 14:44 49 S2 23 Jul Bonneville 151.3 84 3DD.003BB8F215 13:49 13:52 13:55 358 10.5 63 0.61% Tailrace 14:44 50 S2 23 Jul Bonneville 151.3 83 3DD.003BB8F1EC 13:54 13:58 14:00 521 12 72 0.42% Tailrace 14:44 51 S2 23 Jul Bonneville 151.3 82 3DD.003BB8F208 13:58 14:04 14:07 411.5 11 69 0.53% Tailrace 14:44 52 S2 23 Jul Bonneville 151.3 81 3DD.003BB8F1E4 14:05 14:11 14:13 520.5 12 75 0.42% Tailrace 14:44 Star-shaped mark on tail 53 S2 23 Jul Bonneville 151.3 90 3DD.003BB8F204 14:12 14:17 14:20 460.5 11.5 70 0.48% Tailrace 14:44 54 S2 23 Jul Bonneville 151.3 89 3DD.003BB8F220 14:54 15:00 15:02 452 12 70 0.49% Tailrace 14:44 55 S2 23 Jul Bonneville 151.3 88 3DD.003BB8F1EA 15:00 15:04 15:07 504 11.5 74 0.44% Tailrace 16:55 56 S2 23 Jul Bonneville 151.3 87 3DD.003BB8F1E5 15:06 15:10 15:13 488.5 11.5 70 0.45% Tailrace 16:55 57 S2 23 Jul Bonneville 151.3 86 3DD.003BB8F203 15:11 15:14 15:18 464 11.5 70.5 0.47% Tailrace 16:55 58 S2 23 Jul Bonneville 151.3 95 3DD.003BB8F225 15:16 15:21 15:24 500 12 70 0.44% Tailrace 16:55 59 S2 23 Jul Bonneville 151.3 94 3DD.003BB8F1F1 15:21 15:25 15:28 437.5 11.5 69 0.50% Tailrace 16:55 60 S2 23 Jul Bonneville 151.3 93 3DD.003BB8F1FB 15:26 15:31 15:34 605 13 76 0.36% Tailrace 16:55 61 S2 23 Jul Bonneville 151.3 92 3DD.003BB8F1F2 15:38 15:41 15:45 445.5 12 65.5 0.49% Tailrace 16:55 62 S2 23 Jul Bonneville 151.3 91 3DD.003BB8F236 15:42 15:48 15:51 464 11.5 71.5 0.47% Tailrace 16:55 63 S2 23 Jul Bonneville 151.3 100 3DD.003BB8F1F7 15:48 15:53 15:56 441.5 10.5 68 0.50% Tailrace 16:55 64 S2 23 Jul Bonneville 151.3 99 3DD.003BB8F227 15:53 16:00 16:03 582 12.5 74 0.38% Tailrace 16:55 Bled during surgery 65 S2 23 Jul Bonneville 151.3 98 3DD.003BB8F21A 16:00 16:05 16:09 447 10.5 68 0.49% East Fishway 17:15 66 S2 23 Jul Bonneville 151.3 97 3DD.003BB8F201 16:06 16:10 16:13 551 12.5 72.5 0.40% East Fishway 17:15 67 S2 23 Jul Bonneville 151.3 96 3DD.003BB8F1ED 16:11 16:15 16:18 443 11 68 0.50% East Fishway 17:15 68 S2 23 Jul Bonneville 151.3 101 3DD.003BB8F1F4 16:15 16:19 16:22 458.5 11.5 67 0.48% West Fishway 17:06 69 S2 23 Jul Bonneville 151.3 102 3DD.003BB8F1D8 16:33 16:37 16:40 544 12 71 0.40% West Fishway 17:06 Scar on back 70 S2 23 Jul Bonneville 151.3 103 3DD.003BB8F22E 16:38 16:41 16:44 476 11.5 68.5 0.46% West Fishway 17:06 71 S2 23 Jul Bonneville 151.3 3DD.003BB8F235 16:21 299 9.5 61 0.74% Tailrace 16:55 PIT Tag only 72 S2 23 Jul Bonneville 151.3 3DD.003BB8F1F9 16:29 457 11.5 69 0.48% Tailrace 16:55 PIT Tag only 73 S3 30 Jul Bonneville 151.3 104 3DD.003BB8F1E6 12:04 12:12 12:15 456 11.5 67.5 0.48% Tailrace 13:51 Long time to sedate 74 S3 30 Jul Bonneville 151.3 105 3DD.003BB8F229 12:12 12:17 12:20 363.5 11 62.5 0.61% Tailrace 13:51 75 S3 30 Jul Bonneville 151.3 110 3DD.003BB8F224 12:17 12:22 12:25 444.5 11.5 67.5 0.49% Tailrace 13:51 76 S3 30 Jul Bonneville 151.3 109 3DD.003BB8F233 12:23 12:26 12:30 479.5 12 70 0.46% Tailrace 13:51 77 S3 30 Jul Bonneville 151.3 108 3DD.003BB8F202 12:27 12:31 12:34 471.5 12 64 0.47% Tailrace 13:51 78 S3 30 Jul Bonneville 151.3 107 3DD.003BB8F1DC 12:31 12:36 12:39 503.5 12 73 0.44% Tailrace 13:51 Scar near gills 79 S3 30 Jul Bonneville 151.3 106 3DD.003BB8F1DF 12:36 12:39 12:42 406 11 66.5 0.54% Tailrace 13:51 shorten tail (chopped) 80 S3 30 Jul Bonneville 151.3 11 3DD.003BB8F209 12:40 12:45 12:48 604 13 77 0.36% Tailrace 13:51 81 S3 30 Jul Bonneville 151.3 12 3DD.003BB8F226 12:46 12:50 12:53 530 12.5 70.5 0.42% Tailrace 13:51 82 S3 30 Jul Bonneville 151.3 13 3DD.003BB8F20F 12:50 12:54 12:58 466.5 11.5 71.5 0.47% Tailrace 13:51 83 S3 30 Jul Bonneville 151.3 14 3DD.003BB8F1E2 13:00 13:02 13:06 459 12 70 0.48% Tailrace 13:51 84 S3 30 Jul Bonneville 151.3 15 3DD.003BB8F1E7 13:03 13:07 13:11 409 10.5 65 0.54% Tailrace 13:51 85 S3 30 Jul Bonneville 151.3 16 3DD.003BB8F21C 13:07 13:12 13:15 357 10.5 65.5 0.62% Tailrace 13:51 86 S3 30 Jul Bonneville 151.3 17 3DD.003BB8F1FA 13:13 13:16 13:19 475 11.5 68 0.46% Tailrace 13:51 87 S3 30 Jul Bonneville 151.3 18 3DD.003BB8F1D7 13:17 13:20 13:23 455 11.5 70 0.48% Tailrace 13:51 88 S3 30 Jul Bonneville 151.3 19 3DD.003BB8F1FF 13:21 13:24 13:28 457.5 11.5 67.5 0.48% Tailrace 13:51 89 S3 30 Jul Bonneville 151.3 20 3DD.003BB8F22C 14:02 14:05 14:08 524.5 13 72 0.42% Tailrace 15:52 90 S3 30 Jul Bonneville 151.3 21 3DD.003BB8F212 14:06 14:08 14:11 602.5 13 75 0.37% Tailrace 15:52 …continued

LGL Limited Page 41 Wells Dam Adult Lamprey Passage Study, 2013

Appendix Table A continued.

Tag Information Tagging Procedure Benchmarks (hh:mm) Fish Measurements Tag Release Information Fish # Session Date Fish Source Freq Code PIT Tag ID Start MS-222 Start Surgery End Surgery Weight (g) Girth (cm) Length (cm) Burden Location Time Comments 91 S3 30 Jul Bonneville 151.3 22 3DD.003BB8F1D9 14:10 14:13 14:16 481.5 12 67 0.46% Tailrace 15:52 Dorsal fin wound 92 S3 30 Jul Bonneville 151.3 23 3DD.003BB8F1DA 14:14 14:17 14:21 611.5 13 76.5 0.36% Tailrace 15:52 93 S3 30 Jul Bonneville 151.3 24 3DD.003BB8F22A 14:18 14:22 14:25 712.5 14 77 0.31% Tailrace 15:52 94 S3 30 Jul Bonneville 151.3 25 3DD.003BB8F238 14:22 14:25 14:29 553 12.5 73 0.40% Tailrace 15:52 Bled during surgery 95 S3 30 Jul Bonneville 151.3 26 3DD.003BB8F1F3 14:26 14:30 14:34 504.5 12 75 0.44% Tailrace 15:52 Wound on tail 96 S3 30 Jul Bonneville 151.3 28 3DD.003BB8F223 14:31 14:34 14:38 431 11.5 68.5 0.51% Tailrace 15:52 97 S3 30 Jul Bonneville 151.3 29 3DD.003BB8F1F6 14:35 14:38 14:41 418.5 11 68.5 0.53% Tailrace 15:52 98 S3 30 Jul Bonneville 151.3 30 3DD.003BB8F221 14:42 14:47 14:50 499 12 70.5 0.44% Tailrace 15:52 99 S3 30 Jul Bonneville 151.3 31 3DD.003BB8F1FC 14:48 14:52 14:55 434.5 11.5 65 0.51% Tailrace 15:52 100 S3 30 Jul Bonneville 151.3 32 3DD.003BB8F1FD 14:53 14:58 15:01 515.5 12 73 0.43% East Fishway 16:03 101 S3 30 Jul Bonneville 151.3 33 3DD.003BB8F313 14:59 15:02 15:05 530 12.5 73.5 0.42% East Fishway 16:03 102 S3 30 Jul Bonneville 151.3 34 3DD.003BB8F357 15:04 15:08 15:11 464.5 12 72 0.47% East Fishway 16:03 103 S3 30 Jul Bonneville 151.3 35 3DD.003BB8F35B 15:08 15:14 15:17 512 12 73 0.43% West Fishway 16:13 104 S3 30 Jul Bonneville 151.3 36 3DD.003BB8F35F 15:15 15:18 15:21 429 11.5 70 0.51% West Fishway 16:13 Three sutures 105 S3 30 Jul Bonneville 151.3 37 3DD.003BB8F353 15:18 15:23 15:26 411.5 11.5 64 0.53% West Fishway 16:13 Bled during surgery 106 S3 30 Jul Bonneville 151.3 3DD.003BB8F359 15:25 15:28 15:29 359.5 10 66 0.61% Tailrace 15:52 PIT Tag only 107 S4 21 Aug Priest Rapids 151.3 40 3DD.003BB8F34A 11:04 11:08 11:12 409 11.5 67 0.54% Tailrace 11:40 108 S4 21 Aug Priest Rapids 151.3 39 3DD.003BB8F365 11:10 11:14 11:18 485 12 69 0.45% Tailrace 11:40 109 S4 21 Aug Priest Rapids 151.3 38 3DD.003BB8F33C 11:15 11:22 11:25 406.5 11.5 65 0.54% Tailrace 11:40 110 S4 21 Aug Priest Rapids 151.3 45 3DD.003BB8F31E 11:25 11:30 11:33 364.5 10.5 65 0.60% Tailrace 11:40 111 S4 22 Aug Priest Rapids 151.3 41 3DD.003BB8F339 10:55 10:59 11:02 466.5 11.5 67 0.47% Tailrace 11:45 112 S4 22 Aug Priest Rapids 151.3 43 3DD.003BB8F332 11:00 11:03 11:06 421.5 11 67.5 0.52% Tailrace 11:45 113 S4 22 Aug Priest Rapids 151.3 44 3DD.003BB8F324 11:04 11:09 11:12 398 11 67 0.55% Tailrace 11:45 114 S4 22 Aug Priest Rapids 151.3 42 3DD.003BB8F314 11:10 11:13 11:16 414.5 11 69.5 0.53% Tailrace 11:45 115 S4 22 Aug Priest Rapids 151.3 50 3DD.003BB8F35E 11:14 11:18 11:21 482 11.5 73.5 0.46% Tailrace 11:45 Flesh wound

Average size of radio-tagged fish by collection location ("Fish Source") and release site (fishway vs. tailrace).

Release Mean Mean Mean Fish Source Location Weight (g) Girth (cm) Length (cm) Bonneville Fishway 472.0 11.7 69.8 Bonneville Tailrace 481.0 11.8 69.8 Priest Rapids Tailrace 427.5 11.3 67.8

LGL Limited Page 42