In Gates Creek

Gates Creek Juvenile and Adult Salmonid Assessment 2014 (14.SON.01)

D’Arcy, BC

FWCP Project No. 14.SON.01

Prepared for: Fish and Wildlife Compensation Program Lillooet Tribal Council Fisheries and Oceans Canada

Prepared by: InStream Fisheries Research Inc. 1698 Platt Crescent North Vancouver, BC V7J 1Y1 Canada T: +1 (604) 837-9870

File no. GATESCHAN.0009

Report prepared with financial support of the Fish and Wildlife Compensation Program on behalf of its program partners BC Hydro, the Province of BC, First Nation and Public Stake Holders and Fisheries and Oceans Canada.

March 2015

This Fish & Wildlife Compensation Program report presents the results for the combined 2014 assessment operations in Gates Creek. The 2014 salmonid outmigration study on Gates Creek is presented in the first section and the results of the 2014 adult component of the Gates Creek salmonid out-migration study are in the second section. Not presented in this report is the data collected during the Level 1 Habitat Survey which will be included as a separate report.

Note on the diversion of riparian planting funds to Level 1 Fish Habitat Assessment:

Riparian planting on Gates Creek was originally a component of 2014 proposal. Lillooet Tribal Council and N’Quatqua Fisheries were under the impression this task was to be completed on lands upstream of the spawning channel; however, LTC was informed in summer 2014 the request for planting by FWCP was intended for around the spawning channel (Patrice Rothers pers. comm.). After discussion with representatives from FWCP, planting around the channel was determined to be an unwise use of funds: vegetation is a hindrance to required maintenance and spawning channel operations. Consultation was done with FWCP and Split Rock Environmental on FWCP properties 9523A Portage Road in September 2014 to assess the possibility of planting riparian vegetation on this property. It was determined that knowledge of fish habitat quality and limitations in Gates Creek were necessary to guide any future restoration activities. Therefore, diversion of planting funds to a Level 1 Fish Habitat Survey (FHAP) were approved in November 2014 (Trevor Oussoren email dated November 4th, 2014).

Table of Contents

Section 1: Gates Creek Juvenile Salmonid Outmigration Assessment - Spring 2014 ...... v 1.0 INTRODUCTION ...... 1 1.1 Study Objectives ...... 2 1.2 Study Area and Trapping Locations ...... 2 2.0 METHODS ...... 3 2.1 Fish Trap Operations ...... 3 2.1.1 Gates Creek Traps - Rotary Screw and Incline Plane Traps ...... 3 2.1.2 Channel Trap ...... 4 2.2 Sockeye Fry Marking and Recapture ...... 4 2.3 Coho Juvenile Marking and Recapturing ...... 5 2.4 Length and Weight ...... 5 2.4.1 Sockeye Fry ...... 5 2.4.2 Coho Smolt Data Collection ...... 6 2.5 Environmental Monitoring ...... 6 2.5.1 Temperature Monitoring ...... 6 2.5.2 Water Level data ...... 6 2.6 Population Estimate Methods ...... 7 2.6.1 Total Sockeye Fry Abundance ...... 7 2.6.2 Gates Creek Spawning Channel Sockeye Fry Abundance...... 8 2.6.3 Gates Creek Sockeye Fry Abundance ...... 8 2.6.4 Egg-to-fry Survival and Fry per Effective Female ...... 8 3.0 RESULTS ...... 9 3.1 Fish Trap Operations ...... 9 3.1.1 Incline Plane Trap/ Rotary Screw Trap...... 9 3.1.2 Channel Trap ...... 9 3.2 Length and Weight Sampling ...... 9 3.2.1 Sockeye Fry Length and Weight ...... 9 3.2.2 Coho Smolt Length and Weight ...... 10 3.3 Environmental Monitoring ...... 10 3.3.1 Temperature ...... 10 3.3.2 Water Level ...... 11

3.4 Sockeye Population Estimates ...... 11 3.4.1 Total Abundance of Sockeye Fry ...... 11 3.4.2 Gates Creek Spawning Channel Sockeye Fry Abundance...... 11 3.4.3 Gates Creek Sockeye Fry Abundance ...... 12 3.4.4 Egg-to-fry Survival and Fry per Effective Female ...... 12 3.4.5 Coho Catch Data ...... 13 4.0 DISCUSSION ...... 14 4.1 Trap Operations ...... 14 4.1.1 IPT/ RST ...... 14 4.1.2 Channel Trap ...... 14 4.2 Fry Length and Weight Sampling ...... 15 4.3 Water Temperature and ATU’s ...... 15 4.4 Sockeye Fry Migration ...... 15 4.4.1 Sockeye Fry Abundance ...... 15 4.4.2 Egg-to-Fry Survival and Fry per Effective Female ...... 17 4.4.3 Sockeye Fry Run-Timing ...... 17 5.0 SUMMARY and RECOMMENDATIONS ...... 19 TABLES ...... 22 FIGURES ...... 31 REFERENCES ...... 46 Section 2: Gates Creek Adult Sockeye Abundance, Fall 2014 ...... 50 1.0 INTRODUCTION ...... 57 2.0 METHODS ...... 59 2.1 Loading Strategy ...... 59 2.2 Temperature Monitoring ...... 59 2.3 Generation of Abundance Estimates ...... 60 2.3.1 Visual Survey Validation of Counter(s) ...... 60 2.3.2 Video Validation of Counters ...... 61 2.4 Classification of Female Percent Spawn ...... 61 2.5 Fecundity Sampling ...... 62 2.6 Percent Spawn and Total Egg Deposition ...... 62 3.0 RESULTS ...... 64 3.1 Temperature Monitoring ...... 64

iii

3.2 Video Validation ...... 64 3.3 Abundance Estimates ...... 64 3.3.1 Spawning Channel ...... 64 3.3.2 Gates Creek ...... 65 3.3.3 Anderson Lake Shore ...... 65 3.4 Fecundity and Female Size ...... 65 3.5 Percent Spawn and Egg Deposition ...... 66 3.5.1 Spawning Channel ...... 66 3.5.2 Gates Creek ...... 66 4.0 DISCUSSION ...... 67 4.1 Temperature Monitoring ...... 67 4.2 Counter Operations and Visual Survey/ Video Validation ...... 67 4.3 Migration Timing and Sockeye Abundance ...... 68 4.3.1 Migration Timing ...... 68 4.3.2 Sockeye Abundance ...... 68 4.4 Fecundity and Egg Deposition ...... 68 4.5 Percent Spawn ...... 68 5.0 Summary and Recommendations...... 69 TABLES ...... 72 FIGURES ...... 82 REFERENCES ...... 91

Section 1: Gates Creek Juvenile Salmonid Outmigration Assessment - Spring 2014

Executive Summary

This report presents the results of the 2014 salmonid outmigration study on Gates Creek. This was the third year of an anticipated 4-year study of juvenile Sockeye Salmon (Oncorhynchus nerka) abundance from Gates Creek and Gates Creek spawning channel. The enumeration of Coho Salmon (Oncorhynchus kisutch) smolts (1+) is a secondary objective of this study. A new ½’ mesh drum for the rotary screw trap permitted experimental fishing for Coho juveniles under higher flows than previously possible.

Juvenile Sockeye fry were caught during their outmigration using two types of traps, a partial river Incline Plane Trap (IPT) on Gates Creek from March 27th until May 10th, and a full channel weir on Gates Creek spawning channel from March 26th to May 26th, 2013. Sockeye fry abundance and migration timing were estimated from mark-recapture data collected during trap operations using Bayesian P-Spline models.

An estimated 15,463,941 (SD 516,796) Sockeye fry migrated out of the combined Gates Creek and Gates Creek spawning channel in the spring of 2014. Of this total, 18% (2,725,331) were from Gates Creek spawning channel and 82% (12,738,610) out-migrated from Gates Creek. An additional 119,698 fry were estimated to have left the spawning channel, using volume sampling at the spawning channel weir, after the termination of the mark-recapture program on Gates Creek on May 6th. With the additional fry from the channel a total estimated 15,583,639 fry migrated out of the system from March 26th to May 26th, 2014; however, this may be an underestimate of the true Sockeye abundance as the 2014 program looks to have missed a portion of the Sockeye run from Gates Creek. Egg-to-fry survival for Gates Creek spawning channel was 16%, with an estimated 537 fry produced per effective female in 2014. Estimated egg-to-fry survival for Gates Creek was 21%, with an estimated 720 fry produced per effective female.

Abundance of Sockeye fry was 3.25 times greater in 2014 than in 2013. This increase in fry abundance was driven mainly by the higher number of females that effectively spawned in the 2013 brood year as well as higher egg to fry survival in Gates Creek. While there was an overall increase in the number of effective female spawners in the spawning channel, egg to fry survival decreased by more than 50% from 2013 to 2014. In comparison to the 2012 brood year, the number of effective females in 2013 was 55% higher in the spawning channel and 27% higher in Gates Creek, which lead to a 391% increase egg deposition for the watershed as a whole.

High flows in all years have prevented the operation of the IPT after the beginning of May estimation of the juvenile Coho migration. During experimental operation of the Rotary Screw

Trap (RST) in 2014, a total of 413 juvenile Coho were captured in the period from May 14th to May 24th. The RST operated with a 6% recapture efficiency over the trial period. The addition of the larger mesh drum for the rotary screw trap extends the range of possible flows for fishing the RST and may make generating future estimates for Coho smolts possible.

vii

Contents Section 1: Gates Creek Juvenile Salmonid Outmigration Assessment - Spring 2014 ...... v 1.0 INTRODUCTION ...... 1 1.1 Study Objectives ...... 2 1.2 Study Area and Trapping Locations ...... 2 2.0 METHODS ...... 3 2.1 Fish Trap Operations ...... 3 2.1.1 Gates Creek Traps - Rotary Screw and Incline Plane Traps ...... 3 2.1.2 Channel Trap ...... 4 2.2 Sockeye Fry Marking and Recapture ...... 4 2.3 Coho Juvenile Marking and Recapturing ...... 5 2.4 Length and Weight ...... 5 2.4.1 Sockeye Fry ...... 5 2.4.2 Coho Smolt Data Collection ...... 6 2.5 Environmental Monitoring ...... 6 2.5.1 Temperature Monitoring ...... 6 2.5.2 Water Level data ...... 6 2.6 Population Estimate Methods ...... 7 2.6.1 Total Sockeye Fry Abundance ...... 7 2.6.2 Gates Creek Spawning Channel Sockeye Fry Abundance...... 8 2.6.3 Gates Creek Sockeye Fry Abundance ...... 8 2.6.4 Egg-to-fry Survival and Fry per Effective Female ...... 8 3.0 RESULTS ...... 9 3.1 Fish Trap Operations ...... 9 3.1.1 Incline Plane Trap/ Rotary Screw Trap...... 9 3.1.2 Channel Trap ...... 9 3.2 Length and Weight Sampling ...... 9 3.2.1 Sockeye Fry Length and Weight ...... 9 3.2.2 Coho Smolt Length and Weight ...... 10 3.3 Environmental Monitoring ...... 10 3.3.1 Temperature ...... 10 3.3.2 Water Level ...... 11

viii

LIST OF TABLES

Table 1. Start and end dates for all traps operated on Gates Creek, in spring 2014...... 23 Table 2. Summary of Gates Creek Sockeye fry fork lengths measured at both the spawning channel and IPT sites in, 2011-2014...... 24 Table 3. Summary of Gates Creek Sockeye fry weights (g) measured at both the spawning channel and IPT sites in, 2011-2014...... 25 Table 4. Summary of daily water temperatures °C (mean, standard deviation (SD) and range) in both Gates Creek and the spawning channel August 10th, 2013 to May 20th, 2014...... 26 Table 5. Weekly totals of Sockeye fry marked at the spawning channel, recaptured and unmarked fish enumerated at the IPT in spring 2014. Trap efficiency is the proportion of marked fish that were recaptured...... 27

Table 6. Modeled weekly estimates of total unmarked Sockeye fry passing the Gates Creek IPT site. Also displayed are the 2.5 and 97.5% credibility intervals, average weekly temperature (°C) and water level (m), standard deviation (SD), and relative standard deviation. Note: mean abundance does not match totals for system in table 7 as these strata totals do not include marks...... 28 Table 7. Weekly abundance of Sockeye fry leaving Gates Creek and spawning channel and total abundance of fry past the IPT site on Gates Creek in spring 2014...... 29 Table 8. Female Sockeye escapement, fecundity, survival figures for Gates Creek system 2011- 2013...... 30

LIST OF FIGURES

Figure 1. Overview of Seton-Anderson watershed in south western British Columbia...... 32 Figure 2. Overview of study area including Gates Creek, the spawning channel and the IPT/RST site near the confluence with Anderson Lake...... 33 Figure 3. Weir at Gates Creek spawning channel. Fish are funneled through black sampler structure in middle of trough to the wood box visible on the left...... 34 Figure 4. Incline Plane Trap (left) and rotary screw trap (right) in Gates Creek...... 35 Figure 5. Frequency distribution of juvenile Sockeye fork lengths (mm) leaving Gates Creek spawning channel and Gates Creek in spring 2014. Red dotted line indicates sample mean. ...36 Figure 6. Frequency distribution of juvenile Sockeye weight (g) leaving Gates Creek spawning channel and Gates Creek in spring 2014. Red dotted line indicates sample mean...... 37 Figure 7. Frequency distribution of Coho smolts (1+) fork lengths (mm) leaving Gates Creek in spring 2014. Red dotted line indicates sample mean...... 38 Figure 8. Frequency distribution of Coho smolts (1+) weight (g) leaving Gates in spring 2014.Red dotted line indicates sample mean...... 39 Figure 9. Average daily temperatures in Gates Creek spawning channel (red) and at IPT (black) site in Gates Creek in fall 2013 to spring 2014...... 40 Figure 10. Accumulated thermal units from September 5th, 2014 (peak of the spawning) to May 5th, 2014 (50% migration date) for both Gates Creek spawning channel (solid line) and Gates Creek (dashed line)...... 41 Figure 11. Average weekly water level (m) in Gates Creek (measured at the IPT site) in spring 2014...... 42 Figure 12. Run timing of the total abundance of Sockeye fry leaving Gates Creek watershed in spring 2014. Estimates include fry from both Gates Creek and Gates Creek spawning channel...... 43 Figure 13. Run timing of the total abundance of Sockeye fry leaving Gates Creek (solid line) and Gates Creek spawning channel (dashed line) in spring 2014. Estimates include fry from both Gates Creek and Gates Creek spawning channel...... 44 Figure 14. Water temperatures in the Fraser River at Hope, 2013. Source Fraser River Enviromental Watch, Fisheries and Oceans Canada (http://www.pac.dfo- mpo.gc.ca/science/habitat/frw-rfo/reports-rapports/archives-eng.html)...... 45

Acknowledgements This project was conducted with financial support of St’at’imc Government Services, the Pacific Salmon Foundation and the Fish and Wildlife Compensation Program on behalf of its program partners BC Hydro, the Province of British Columbia, First Nation and Public Stakeholders, and Fisheries and Oceans Canada.

We would like thank the following people for their cooperation and assistance on this project: Harry- O’Donaghey, N’Quatqua Fisheries, Lance O’Donaghey, N’Quatqua Fisheries, Chris Fletcher, N’Quatqua Fisheries, Harry O’Donaghey, Jr., N’Quatqua Fisheries, Spencer Thevarge, N’Quatqua Fisheries

LJ Wilson- Instream Fisheries Research, Inc., Peter Campbell - Department of Fisheries and Oceans, Doug Lofthouse - Department of Fisheries and Oceans, Dave Willis - Department of Fisheries and Oceans, Matthew Foy - Department of Fisheries and Oceans, Andrew Grant - Department of Fisheries and Oceans, Brian Leaf - Department of Fisheries and Oceans, Paul Welch - Department of Fisheries and Oceans

Stephanie Lingard, R.P.Bio. – Project Lead Jason Ladell, M.Sc., R.P.Bio. Junior Biologist, Instream Project Manager, Instream

Don McCubbing, M.Sc., R.P.Bio. Caroline Melville Senior Biologist, Instream Project Manager, Instream

xii Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 1

1.0 INTRODUCTION

Gates Creek is a major salmon bearing tributary of the Seton-Anderson Watershed and runs 12 kilometres from Gates Lake to Anderson Lake draining approximately 34,300 hectares (Komori 1997) (Figure 1 & 2). The Seton-Anderson watershed is located approximately 200 km north of Vancouver in the rain shadow of the southern Coast Mountains (Anon. 2000). There are no glaciers in the watershed; however, Gates and Anderson Lake provide storage with-in the Seton-Anderson watershed. Anderson Lake is connected to Seton Lake via the Portage River, and Seton Lake drains into the Fraser River via Seton River (Figure 1). Gates Creek supports a population of Fraser River Sockeye salmon (Onchorynchus nerka) that is important for First Nation, commercial and recreation fisheries, as well as smaller populations of Coho Salmon (O. kisutch), and Pink Salmon (O. gorbushca).

Starting in the early 1900’s Gates Creek Sockeye have been affected by several major development projects. Fraser salmon populations upstream of Hell’s Gate including Gates Creek populations were heavily impacted by the slides of 1913 and 1914 (Talbot 1950; Andrew and Green 1958). In 1956, as part of the Bridge River Hydro development, a diversion dam was constructed on the Seton River 750 m downstream of Seton Lake. The development, which also included a canal to a power house on the Fraser River, has had significant impacts on the Portage River and Gates Creek salmon stocks, through entrainment of juveniles and reduced adult escapement (Fretwell 1989; Komori 1997). In addition to these downstream impacts, salmon habitat on Gates Creek has been degraded by residential and agricultural developments (Anon. 2001).

In 1968, a Sockeye specific spawning channel was constructed by the International Pacific Salmon Fisheries Commission (IPSFC) on Gates Creek 800 m upstream of Anderson Lake (Figure 2) to enhance Sockeye escapement in the Seton-Anderson watershed. The spawning channel was originally overseen by the IPSFC and the Department of Fisheries and Oceans Canada (DFO). In 1987, responsibility for channel maintenance and monitoring was turned over to the N’Quat Qua First Nation with technical oversight from DFO. A gravel replacement project was undertaken in 2008 and 2009 by DFO and the BC Hydro Fish and Wildlife Compensation Program (FWCP) with the goal of increasing egg-to-fry survival in Gates Creek spawning channel (Anon. 2009). In addition to gravel replacement, changes were made to channel structure and gradient during this project (Anon. 2009). While a long standing time series of juvenile and adult abundances are available for Gates Creek spawning channel, detailed assessment of the gravel replacement activities had yet to be undertaken. Previous to this study, data regarding Sockeye egg-to-fry survival and abundance from Gates Creek had never been collected. Enumeration of Coho juvenile abundance in Gates Creek had also never been undertaken, although there has been a recent evaluation of habitat use by rearing juvenile Coho (Hillaby 2012).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 2

In 2011, the DFO scientific advisors for the Gates Creek spawning channel requested that Instream Fisheries Research Inc. (IFR) submit a study design to enumerate out-migrant Sockeye fry and Coho juveniles in Gates Creek to compliment ongoing work on the Gates Creek spawning channel. A proposal including the juvenile enumeration study and complimentary Sockeye adult monitoring (counts and fecundity assessments), was submitted by the Lillooet Tribal Council (LTC) and DFO to the Fish and Wildlife Compensation Program. The following report summarizes the findings of the 2014 juvenile component of the third year of the study.

1.1 Study Objectives

The main objectives for this study were to assess the following biological parameters for Sockeye and Coho salmon in Gates Creek.

1. Estimate the abundance and timing of out migrant Sockeye fry for both Gates Creek and Gates Creek spawning channel 2. Estimate egg-to-fry survival of Sockeye fry in both Gates Creek, and the Gates Creek spawning channel 3. Estimate Sockeye fry production per effective female spawner 4. Estimate the abundance and timing of out migrant Coho juveniles (1+) from Gates Creek

1.2 Study Area and Trapping Locations

The study area consists of two sites, a full weir on the spawning channel (10 U 536706 5599716, 800 m upstream of the mouth of Gates Creek at Anderson Lake) (Figure 2 & 3), and an Incline Plane Trap (IPT)/rotary screw trap (RST) (10 U 537151 5599978, 280 m upstream of the mouth of Gates Creek at Anderson Lake) downstream of the channel outlet on Gates Creek (Figure 2 & 4).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 3

2.0 METHODS

Two methods were used to enumerate out migrant juvenile Sockeye in 2014, which are similar to previous years (Lingard et al. 2013). Partial traps (e.g. rotary screw traps or Incline Plane Traps), which sample a proportion of the out migrant Sockeye fry and Coho smolts, require mark-recapture sampling methods and analyses to estimate juvenile outmigration. These traps were used to capture fish in Gates Creek. A complete channel trap was used to capture all out- migrating Sockeye fry from the spawning channel. The study design was developed to ensure sampling methods minimized fish mortality and stress.

2.1 Fish Trap Operations

2.1.1 Gates Creek Traps - Rotary Screw and Incline Plane Traps

Two downstream recapture traps were used in this study. An Incline Plane Trap (IPT) was run Monday to Friday from March 27th to May 11th. The operation of the IPT 5 days per week was an alteration to methodology used in 2011 and 2012 where the IPT was fished 7 days per week (Lingard et al. 2013). Restriction of trap operations to 5 days was done as a cost saving measure in the face of budget cuts in the spring of 2014. On May 14th the IPT was replaced with the rotary screw trap (RST) to facilitate the juvenile Coho component portion of the project. The RST was in operation for two weeks until May 24th to experiment with running the trap at higher flows. Outmigration of Coho juveniles from other systems such as the Cheakamus and Seymour Rivers (Melville and McCubbing 2012; Ramos- Espinoza and McCubbing 2012) consistently peaks during the first two weeks of May, which typically overlaps with high flow conditions in Gates Creek.

Both traps were oriented on a cableway pulley system allowing the lateral position in the river to be adjusted to optimize the sampling location in the main flow. In addition the trap could be brought to shore on either side of the creek for cleaning and sampling (Figure 4). The IPT or RST was set out fishing each day at dusk (between 6 pm and 8 pm). The trap was checked at 8 am the following day for manual counting of captured fish. Restriction of trap operations to night hours was deemed adequate because it is known from channel trap operations that relatively low numbers of fry migrate during daylight hours. During times of high water the trap was brought to shore and operations were suspended to ensure safety of crew and equipment.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 4

2.1.2 Channel Trap

The full channel trap at the downstream end of the spawning channel guides all out-migrating Sockeye fry to a single trough. A proportional sampler (Red Fish Services) (Figure 3) divided out migrants into two separate capture boxes, “sample” and “full channel”. The sample box is smaller than the full channel box and meant to capture approx. 5% of the nightly fry migration. The full channel box is meant be in place 1-2 times per week to capture the remaining 95% of the nightly migration and allow calibration of the sampler. The sampler structure is a rectangular box with a screened wall dividing it length wise in half. Fish are sampled by a small two-inch funnel that moves across an opening on the channel weir. Fish captured by the funnel are diverted into one chamber of the structure which is connected to the “sample” box. Fish not captured by the funnel enter the other chamber and are routed into the “full-channel” box, or are released into the river to continue their migration down river.

The proportional sampler was first installed in 2011 and has experienced mechanical problems on several occasions. However, after modification in the winter of 2013/ 2014 the sample operated without trouble for the entire 2014 season. To test the variability in proportion of total catch captured by the modified sampler, all Sockeye leaving the spawning channel were counted but when there was >3000 fry per night abundance was estimated by volume sampling (see section 2.6.2).

2.2 Sockeye Fry Marking and Recapture

Mark recapture methods were used to assess the capture efficiency of mainstem juvenile traps. A known number (approximately 2,500) of channel fish were marked and released from the channel (upstream of the IPT) four days a week (Monday-Thursday). A proportion of the marked fish were subsequently caught in the downstream IPT (Figure 2), which provides an estimate of capture efficiency. No fish were marked during the remaining three days (Friday-Sunday); this allowed for all marked fish from the previous marking period to pass by the IPT. This assumption was validated by the zero recaptures of marked Sockeye fry on Monday morning during our study period. All marked fry were released at dusk. This temporally stratified method was developed on the Cheakamus River and is documented in Melville and McCubbing (2001, 2002a; 2002b; 2003, 2004, 2006-2012).

On marking days, Sockeye fry collected at the channel trap were not sampled in the morning of capture but held in the trap boxes until late afternoon. A maximum of 2,500 fry were marked each day by immersion in Bismark Brown Y dye diluted to a ratio of 1:100,000 with river water. Fish were immersed for 45 minutes in 50 litres of dye solution aerated with electric air pumps. After marking fish were immediately released into Gates Creek downstream of the channel trap. This marking technique was developed to minimize stress related mortality due to the

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 5 marking/holding process on fry (Melville and McCubbing 2002b). Daily fry catch data represented a 24-hour sample period beginning at 8 am each morning.

2.3 Coho Juvenile Marking and Recapturing

Coho juveniles have varied freshwater life histories prior to migration to salt water. For the purpose of field marking and enumeration of fish it is necessary to use discrete length intervals to identify which life stage a Coho juvenile is at when captured during the spring migration period. In the absence of Gates Creek specific length frequency and ageing data; data from Melville and McCubbing (2005) were used to identify length cut-offs for the various life stages: smolts (1+ migrating): >70 mm, parr (1+ non-migrating): 60-70mm, fry (0+ non-migrating) < 60mm.

The RST functioned as both a mark and a recapture trap for Coho smolts (1+ migrating). Beginning May 14th, unmarked Coho smolts captured in the RST were marked daily with a caudal fin clip. To allow for temporal stratification of capture efficiency, mark patterns were changed on Wednesday of each week. Thus, mark groups for Coho ran from Thursday through Wednesday from May 14th to May 24th. Fish were marked in the morning after the RST catch had been enumerated. The marked Coho were then placed in a holding box in the channel flow downstream of the channel trap and released at dusk. The daily Coho catch was defined as the 24 hour period beginning at 8 am.

To reduce stress to fish and ensure successful marking, fish were anaesthetized using a water bath with clove oil diluted in ethanol. Caudal fin clips consisted of a dorso-ventral cut approximately ¼ the distance from the caudal peduncle to the tip of either the upper or lower caudal lobe. Temperature stress was minimized by frequent changes of anaesthetic baths, and by marking early in the morning in shade

2.4 Length and Weight

2.4.1 Sockeye Fry

Samples of Sockeye fry (N=25) were taken from all trap sites twice a week. Fork length was measured for each fish to the nearest millimetre and the weight of five fry was taken to the nearest hundredth of a gram. To ensure accuracy of measurements and reduce handling stress fish were anaesthetized in water baths with clove oil diluted in ethanol.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 6

2.4.2 Coho Smolt Data Collection

Coho juveniles were only sampled at the IPT/RST site as few were captured in the channel trap. To ensure accuracy of measurements and reduce handling stress fish were anaesthetized in water baths with clove oil diluted in ethanol. Fork lengths were taken to the nearest millimetre. Coho were weighed individually to the nearest tenth of a gram.

2.5 Environmental Monitoring

2.5.1 Temperature Monitoring

Temperature data was collected for both the spawning channel and Gates Creek over the period of spawning and egg incubation from August, 10, 2013 to May 20th, 2014. Three sites on each Gates Creek and Gates Creek spawning channel were monitored with pairs of Onset TidbiT v2 Water Temperature Data Loggers which have an accuracy range of 0.2°C. Temperature was logged every 15 minutes from August 10th to September 25th. For analysis data was condensed to average daily temperature (°C).

Pairs of temperature loggers were placed in the spawning channel at the upper, fifth and bottom legs. In Gates Creek the most upstream logger was placed at the Davidson’s horse farm (UTM 10535385E, 5596420N), the second was placed downstream in a heavily used spawning area (UTM 10536331E, 5598092N), and the third most downstream pair was placed on river left across from the spawning channel intake (UTM 10563462E, 5599630N).

Mean daily water temperature was used as an estimate of the daily thermal units. Accumulated thermal units (ATU) over the egg incubation period were calculated as the sum of the daily thermal units between the peak of spawn to the peak of fry migration (September 5th, 2013 and May 5th, 2014).

2.5.2 Water Level data

Water level was monitored using a staff gauge at the IPT site. Water level readings were recorded 3 times per day. Mean weekly water level over the survey period was calculated from the mean daily staff gauge readings.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 7

2.6 Population Estimate Methods

2.6.1 Total Sockeye Fry Abundance

The Bayesian P-SPLINE model developed by Bonner and Schwarz (2011) was utilized for mark- recapture data analysis. The analysis is carried out using the statistical software R (R Development Core Team, 2012), with the R2OpenBUGS (Sturtz et al. 2005) package for interfacing with Open Bugs (Lunn et al. 2009), BTSPAS (Bonner and Schwarz 2012), CODA (Plummer et al. 2006), Actuar (Dutang et al. 2008), and Lattice (Deepayan 2008) packages.

Historically, the Pooled Peterson estimate or temporally-stratified Peterson methods (e.g. Ricker 1975; e.g. Arnason et al. 1996) have been the preferred analysis method for mark-recapture data. These two methods make a number of assumptions as outlined by Seber (2002):

1) The population is closed such that there is no immigration or emigration 2) In a sample period all untagged fish have the same probability of capture 3) Marking, clipping, and releasing fish upstream does not affect, their subsequent catchability in the downstream trap 4) The sample caught in the downstream trap is a random sample, and all combinations of untagged and tagged fish have equal probabilities of occurrence 5) No marks are lost between release and recapture sites 6) All marks are reported on recovery in the downstream sample 7) Marked and unmarked fish have similar movement patterns from the release site to the downstream trap 8) Fish can pass the downstream trap once all marked fish pass the traps by the end of the study period, i.e., none of the fish remain above the downstream trap 9) There is no mortality and all fish pass the trap

Bonner and Schwarz (2011) developed an alternate method that uses Bayesian spline models for estimating population size. This modeling approach has many advantages over existing methods. The key features of this method are the use of splines to model the general shape of the run. Additionally, estimates of abundance are provided for each recapture stratum making it possible to estimate quantities such as the time at which 50% of the run has passed, or the time needed to reach a pre-specified target number of fish. The model can also deal with the common problem of not being able to sample in all strata; the spline curve for the run is used to “interpolate” for the missing data. These last two features are difficult to obtain from the previous methods. However, the spline model is not a panacea to solve all potential problems encountered in capture-mark-recapture studies. There are a number of caveats that apply to this and potentially to other stratified models which are described further in Bonner and Schwarz (2012).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 8

2.6.2 Gates Creek Spawning Channel Sockeye Fry Abundance

Volume sampling was completed between 8:00 am and 12:00 pm daily. This consisted of weighing a subsample of 500 fry each day and then weighing the total catch of fry in batches. During the peak of the migration the full channel box was sampled between 11:00 pm and 12:00 am to prevent mortality caused by overcrowding in the trap box. To convert the weights of fry to number of fry a subsample of 500 fish was weighed each day. The number of fry per gram was then calculated and multiplied by the total weight of fry.

2.6.3 Gates Creek Sockeye Fry Abundance

Gates Creek abundance was calculated as the difference between the total fry abundance estimated from the IPT mark-recapture site and Gates Creek spawning channel estimate.

2.6.4 Egg-to-fry Survival and Fry per Effective Female

Egg-to-fry survival was calculated by dividing fry abundance by the number of eggs successfully deposited in the previous year. Egg deposition for Gates Creek channel was estimated using fecundity data for the 2013 brood year (Lingard et al. 2013a). Fry per effective female1 was calculated by dividing the number of fry into the number of effective females for 2013 (see Lingard et al. (2013a) for further information regarding the calculation of the number of effective females).

1 Effective female refers to a female which successfully spawned

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 9

3.0 RESULTS

3.1 Fish Trap Operations

3.1.1 Incline Plane Trap/ Rotary Screw Trap

The 2014 mark-recapture program was 14 days longer than the 2013 program and 6 days shorter than the 2012 program. Recapture rates for the IPT ranged from 1% to 27% with a mean of 13% (SD 9%). Catches of Sockeye fry in the IPT ranged from 1,646 to 93,274 with a mean daily catch of 16,580 (SD 25,305). The rotary screw trap with the new drum fished safely at water levels ranging from 0.65 m and 0.85 m as measured at the IPT/ RST trap site.

3.1.2 Channel Trap

The channel trap was operated 100% of the study period (61 of 61 days). The proportional sampler worked 100% of days with a mean sample rate of 6% (SD 2%).

3.2 Length and Weight Sampling

3.2.1 Sockeye Fry Length and Weight

Fry caught at both the IPT and the channel traps were similar in length. No significant difference was found in mean fry length (Welch’s t-test, P=0.121). The mean length of Sockeye fry caught were the same at both sites (mean=29 mm, N= 420 (channel), 274 (IPT), SD=2 (channel), 1 (IPT)) (Table 2, Figure 5). A significant difference was found in variance of fork length (F-test, F -16 (419,273) =3.66, P=2.2 x 10 ) between the two sites (Table 2). Fry caught at the IPT had a slightly contracted range (27 to 36 mm) from that of fry caught at the channel (23 to 47 mm).

Fry weight was similar between the two capture sites. Mean of fry caught at the channel was the same as mean weight of fry caught at the IPT (Welch’s t-test: P=0.767); however, the variance -16 of the two samples did differ significantly (F-test: F(419,273) =3.09, P=2.2 x 10 ). The distribution of fry weights for both the IPT and channel were uni-modal at 0.15 g and skewed towards smaller values (Figure 6). Mean fry weight was 0.21 g for both sites (N= 419 (channel), 273 (IPT), SD= 0.07 (channel), 0.04 (IPT)) (Table 3). Fry caught in the IPT ranged from 0.13 to 0.33 g, while fry weighed at the channel ranged from 0.10 to 0.52 g.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 10

3.2.2 Coho Smolt Length and Weight

A broad range of sizes were observed in the Coho juveniles captured in the RST. Coho Smolts sampled in 2014 ranged from 75mm to 123 mm with a mean fork length of 98 mm (N= 119) (Figure 7). Wet weight of Coho smolts sampled in 2014 ranged from 4.74 g to 21.84 g with a mean of 11.14 g (N= 119) (Figure 8).

3.3 Environmental Monitoring

3.3.1 Temperature

Mean daily water temperature was similar between Gates Creek and the spawning channel. Mean daily temperature across all creek loggers ranged from 0.0°C to 15.1°C for the period of spawning, egg incubation and fry migration (August 10th, 2013 to May 20th, 2014) (Table 4, Figure 9). Mean daily water temperature across all spawning channel loggers ranged from 0.0◦C – 14.1◦C over the same period (Table 4, Figure 9). Overall, mean water temperature in both the spawning channel and Gates Creek were 5.1°C (SD 3.9) and 5.3°C (SD 3.6) respectively for the period of August 10th to May 20th (beginning of spawning period to end of juvenile migration). From January to May 2014 Gates Creek experienced more frequent and larger variations in water temperature than the spawning channel (Figure 9).

Accumulated Thermal Units (ATU’s) over the incubation period were similar between Gates Creek and the spawning channel. Between September 5th, 2013 and May 5th, 2014 (50% spawn to 50% migration date) total ATU’s for the spawning channel were 1,003. Gates Creek accumulated a total of 1,045 thermal units over the same period (Figure 10).

Over the period of the mark re-capture program March 27th to May 12th, 2014 mean weekly water temperature ranged from 1.1°C to 7.5°C .at the IPT sight and 2.2°C to 6.8°C for the spawning channel. During the peak week of migration, May 5th, the mean weekly water temperature was 6.5°C and 6.8°C in the Gates Creek and the spawning channel, respectively. The week of the warmest water temperatures at the IPT site was the week ending April 7th (Figure 9). In the spawning channel the week with the highest water temperatures was the week ending May 12th (Figure 9); although, the spawning channel also experienced warmer water temperatures (6.7°C ) in the week ending April 7th as well (Figure 9).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 11

3.3.2 Water Level

Daily river water level measured at the IPT site ranged from 0.47 m to 0.65 m from March 27th - May 12th, (Figure 11). Trapping using the IPT was discontinued due to high water levels on May 10th due to in an increase in water levels to greater than 0.65 m.

3.4 Sockeye Population Estimates

3.4.1 Total Abundance of Sockeye Fry

A total of 52,504 marked Sockeye fry were released at the channel weir across five marking groups, of which 7,033 were recaptured. The total number of Sockeye fry captured at the IPT site from March 27th to May 12th was 739,083 fish (Table 5). The number of Sockeye fry that passed the IPT site was estimated to be 15,463,941 (SD 516,796) between March 27st and May 12th, 2014 (Table 6). There were an additional 119,698 fry that left the spawning channel after May 12th. Taken together there was a total of 15,583,639 Sockeye fry produced from the Gates Creek system (Table 6).

Based on estimated weekly abundance, it appeared that migration of Sockeye fry had started prior to trapping commencing. In the first weekly strata, March 27th to March 31st, an estimated 93,451 fry (SD 7,669) passed the trap (Table 6). While the number of fry in the first strata of 2014 only represented 1% of the total abundance estimate, the figure for the first week of the program in 2014 was 21 times greater than the number of fry that passed the trap in the same period of 2013 (Lingard et al. 2013b). Over the sampling period, the run-timing of Sockeye fry passing the IPT site, was uni-modal, peaking in the week of April 29th to May 5th, 2014 at 8,583,646 fry (Figure 12, Table 6). The run reached the 50% and 90% marks in the week of April 29th to May 5th. Only 10% of the total abundance estimate was estimated to pass the trap in the week of May 6th to May 12th, 2014.

A total 4,256 mortalities (included in abundance estimates) representing 1.0% of total catch, or 0.03 % of total estimate were incurred at the IPT.

3.4.2 Gates Creek Spawning Channel Sockeye Fry Abundance

An estimated 2,845,029 Sockeye fry migrated out of the Gates Creek spawning channel between March 27th and May 26th. During the operation of the IPT (March 27th-May 12th) the

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 12 channel abundance of Sockeye fry (2,725,331) represented 18% of the total Sockeye estimated from the downstream IPT (Table7, Figure 13).

Fry migration from the channel appeared to have started when trapping commenced as only 0.3% of the total run was captured in the first week of sampling. Channel fry abundance reached 10%, 50% and 90% of the total migration during the weeks ending April 21st, May 5th and May 12th, respectively. Migration of fry out of the channel was uni-modal with a peak estimated weekly emigration of 1,341,285 fry (47.1% of total channel migration) occurring in the week ending May 5th (Table 7, Figure 13). Specifically, the channel peaked on May 2nd with a total nightly migration of 526,880 fry. In the two-week period following the removal of the Gates Creek IPT, less than 5% (119,698) of total channel abundance emigrated.

3.4.3 Gates Creek Sockeye Fry Abundance

An estimated 12,738,610 Sockeye fry migrated out of Gates Creek over the period March 27th to May 12th, 2014 (Table 7, Figure 13). Fry abundance from Gates Creek represented 82% of the estimated total Sockeye fry abundance in 2014. Similar to the spawning channel and total fry abundance from the system, Gates Creek fry abundance run timing was uni-modal (Figure 13). The peak occurred in the week of April 29th to May 5th with a total of 7,249,679 fry (57% of the total creek abundance).

Based on the weekly abundance estimates it appears the beginning and end of the Sockeye fry migration may have been missed by the mark-recapture program. In the first week of trapping 85,789 of the fry leaving Gates Creek (1% of creek total) migrated past the IPT site. Abundance was 10% of the total creek abundance in the week ending April 21st and reached the 50% and 90% of total abundance in the week ending May 5th. In the final week of trapping (May 12th) 1,211,988 fry, 10% of the total abundance, emigrated out of Gates Creek (Table 7, Figure 13).

3.4.4 Egg-to-fry Survival and Fry per Effective Female

Egg-to-fry survival and fry per effective female in the Gates Creek were 1.3 times the values observed in the spawning channel. Egg-to-fry survival for Gates Creek spawning channel and Gates Creek were estimated to be 16%, and 21% respectively for the 2013 brood year. The number of fry produced per effective female was 537 for Gates Creek spawning channel, and 720 for Gates Creek (Table 8).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 13

3.4.5 Coho Catch Data

A total of 413 Coho were captured using the RST between May 14th and May 24th, 2014. Daily catches ranged from 49 to 88 individuals. Of a total of 282 marked Coho smolts that were released, 17 were recaptured giving a preliminary recapture rate of 6% for the RST in the spring of 2014.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 14

4.0 DISCUSSION

4.1 Trap Operations

4.1.1 IPT/ RST

Stable low flow conditions in 2014 increased the length of the mark-recapture program by 2 full weekly strata from 2013 and allowed for the IPT to be fished through the peak of fry migration. In 2014 the IPT was fished 61% of the mark-recapture study period, a decrease from 69% in 2013. However, the mark-recapture program was 14 days longer in 2014 than in 2013. High water levels after May 12th (>0.65 m) left the IPT inoperable and reduced the capture efficiency of the IPT. As a result of high water levels the Sockeye portion of the program was terminated on May 12th. The inability to fish the IPT after May 12th may have resulted in an underestimate of the Sockeye population for 2014 as over 1 million fry were estimated to have migrated in the last strata of the mark-recapture program.

The addition of a 1/2’ mesh drum on the RST allowed the trap to be fished experimentally for Coho smolts at higher water levels than possible in previous years. In 2014, once water levels exceeded the limits of the IPT (0.65 m), the RST was able to fish up to a limit of 0.85 m. Several modifications would be advantageous for the RST operation in subsequent years. When a second stage trap box was added to a trap the pontoons were not lengthened which causes the trap to be tail heavy and life the drum out of the water under increased flows. Additionally, because the back of the trap extends past the pontoons the trap box drags along the creek bed which risks damage to the trap box and staff injury while brining trap to shore. The screen on the RST could also be made of wider mesh which would pass more water at higher velocities and reduce strain on the cable way.

4.1.2 Channel Trap

The repairs made to the sampler over the off-season resulted in the proportional sampler functioning 100% of the 2014 field season. The sampler averaged 6% capture rate with a standard deviation of 2%. Going forward it may be possible to enumerate only the sub-sample capture box the majority of days and allow the remainder of fry to pass downstream unhandled. Calibration of the sampler counting the entire nightly migration could be instituted once or twice per week.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 15

4.2 Fry Length and Weight Sampling

In salmonids and other species of teleost fish, survival has been positively correlated with juvenile size (West and Larkin 1987; Henerson and Cass 1991; Sogard 1997; Eimum and Fleming 2000). In this study, fry length and weight were similar between the IPT and channel samples. Mean fry length and weight were found to be the same (29 mm, 0.21 g) at both sites. There was a significant difference between sites in the variance of fry weight; however, the error associated with using wet weights of live fry leads to imprecise measures of weight and the associated error is likely larger than the difference between sites. It should also be noted that fry measured at the IPT are not necessarily representative of the size of fish emerging from Gates Creek as they are a mixture of fry from the spawning channel and Gates Creek.

4.3 Water Temperature and ATU’s

Embryo development, survival, hatching and fry emergence date in salmonids are known to be influenced by stream water temperature (Murray and McPhail 1987). The accumulation of thermal units for the spawning channel and Gates Creek followed similar temperature profiles over the 2013 portion of the incubation period. Between January and May 2014 when the spawning channel experienced fewer large fluctuations in water temperature than Gates Creek. Overall, the less variable water temperatures in the spawning channel did not result in a large difference in thermal units or mean water temperature between the two habitats, nor does it appear from Figure 10 that either habitat collected thermal units significantly faster than the other. However despite the similarity of temperatures profiles and means in the two sites, surface water temperature in the spawning channel and Gates Creek are likely not an accurate indicator of the intra-gravel water temperatures experienced by the embryos. Intra-gravel (redd) water temperatures have been found to vary by as much as 6◦C in the summer and 1◦C in the winter from surface water temperatures in similar Pacific North West streams typically as a result of ground water influence (Shepherd et al. 1986).

4.4 Sockeye Fry Migration

4.4.1 Sockeye Fry Abundance

The estimate of 15.6 million is the second highest estimate of the three years of the study, but is likely an underestimate of the true fry abundance for 2014. The estimate of total Sockeye fry abundance from the system in 2014 is 69% higher than the estimate in 2013, but may not have captured the beginning and end of the run. Nearly 100,000 fry were estimated to have past the trap in the first strata which is nearly 21 times higher than the 4,163 fry estimate for the first strata in 2013. In the last strata of the mark recapture program in 2014 more than 1 million fry

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 16 past the IPT which is 14 times greater than the last strata in 2013, and indicates the migration of fry may not have been approaching zero when the mark recapture program ended.

Constraining trapping from Monday to Friday, instead of 7 days per week as in previous years, also likely resulted in an underestimation of the Sockeye fry abundance in 2014. Due to budgetary constraints in 2014 the IPT was fished 5 days per week instead of 7. During all years the study has been undertaken (2012-2014), marking has been limited to Monday through Thursday evenings to ensure all marked fish from a given strata pass the recapture trap before the next strata begins on the following Monday. Fry in Gates Creek appear to move quickly once released from the channel. In previous years, the majority of recaptures observed Tuesday through Thursday (the days immediately following evening releases of marked fish). Therefore, reducing fishing effort on Saturdays and Sundays, when few (<10) marked fish have typically been caught, reduced the total of unmarked fish captured for the strata but did not comparably reduce the recapture rate. In hindsight, the number of marking days should also have been reduced to reflect the reduction in fishing effort. The combination of the mark-capture program failing to capture the entirety of the Sockeye fry migration period and the reduction in unmarked catch for each strata indicates that the 15.5 million estimate likely underestimates the Sockeye population in 2014.

The main factors contributing to the 69% increase in Sockeye fry abundance from 2013 to 2014 were increases in the number of effective females in the 2013 brood and egg to fry survival in Gates Creek. In 2013 there was an increase in total female escapement as well as the number that successfully spawned over 2012. Overall female escapement increased by 167% and the number of successful spawners increased by 361%. In the spawning channel 81% (5,302 females) of total female escapement was estimated to have been effective females in 2013, up from 26% (2,351 females) in 2012 (Lingard et al. 2013a). In Gates Creek the number of effective females was estimated to be 79% of total female escapement in 2013, up from 52% in 2012 (DFO Unpublished Data). Overall, estimated egg deposition for the whole system was 391% higher in the fall of 2013 than in 2012. The increase in spawning success in 2013 was unexpected given the high water temperatures in the Fraser River (>21◦C) (Figure 14) which are typically associated with thermal stress and susceptibility to disease and parasites (Crossin et al. 2008; Martins et al. 2012). As neither anomalous temperatures nor disease were found to be factors in 2012 (DFO, unpublished data) the high PSM rate for this particular brood remain a mystery. High pre-spawn mortality in Gates Sockeye is likely a result of a combination of multiple stressors experienced by the fish in both freshwater and marine environments.

The relative contributions of Sockeye fry from Gates Creek and Gates Creek spawning channel changed by ~40% from 2013 to 2014. In 2014 Gates Creek contributed 82% of total Sockeye fry abundance which is a 37% increase from 2013. Conversely, the contribution of fry from the spawning channel decreased from 35% in 2013 to 18% in 2014. The change in relative contribution of fry abundance from the two sites was likely from the overall higher escapement

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 17 of females into Gates Creek, as well as an unexplained decreased in egg to fry survival in the spawning channel.

4.4.2 Egg-to-Fry Survival and Fry per Effective Female

To date few studies have been undertaken to estimate wild egg-to-fry survival in Sockeye; however in their reviews of published studies, Bradford (1995) and Quinn (2005) found average wild egg-to-fry survival rates to vary from 7% to 12%. In reference to the averages estimated by Bradford (1995) and Quinn (2005), Gates Creek has had above average survival rates for wild Sockeye in the past two years. In addition to percent spawn and egg deposition increases between the 2012 and 2013 brood years, egg-to-fry survival also increased from 2012 to 2013 (17% and 21% respectively) in the creek.

In the spawning channel, egg-to-fry survival dropped by 20% from the 2012 brood to the 2013 brood years. The survival rates of 36% in 2012 and 16% in 2013 indicate that the Gates Creek spawning channel may require maintenance and expanded monitoring of water quality and sediment impaction. In the previous two years of the study the spawning channel had been operating at the lower end of egg-to-fry survival ranges observed at DFO operated spawning channels. Egg-to-fry survival for Nadina spawning channel ranged between 30% and 80% from 1994 to 2011, and at Weaver Creek spawning channel egg-to-fry survival rates ranged from 48% to 86% from 1988 to 2008 for Sockeye (DFO, unpublished data.). The 2008 and 2009 modifications to the gradient and substrate in Gates Creek spawning channel have helped to reduce some of the operational and maintenance requirements of the channel, but may come at a sacrifice to egg to fry survival. From 2012 to 2013 the spawning channel contributed 35-55% of total fry abundance, and is important for the preservation of this Fraser River Sockeye population. Because of the importance of the channel in protecting the Gates Creek population from stochastic events in Gates Creek and in light of lower survival rates in 2014, inspection and monitoring of channel sediment properties and water quality may be warranted going forward if egg to fry survival rates continue to decline. Machine scarification has been used annually in kokanee (O. nerka) spawning channels in Meadow Creek and Hill Creek, B.C to reduce gravel impaction. Newer techniques using suction, SandWands created by Streamside Environmental for example, have also shown promising results in removing fine sediment (Grey 2013).

4.4.3 Sockeye Fry Run-Timing

In the fall of 2013 the first 4000 fish were allowed up Gates Creek under the direction of DFO scientific advisers (Lingard et al. 2013a). This was done to assess whether early run fish had a higher pre-spawn mortality than later run fish. As percent spawn was similar in both habitats (81% spawning channel, 79% Gates Creek) it is not clear weather this initiative helped to reduce PSM in the spawning channel (Lingard et al. 2013a) in 2013. The original intention was

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 18 to let the first part of the run up the Creek and then load the Channel and Gates Creek evenly throughout the run up to a total of ~9,000 females in the spawning channel, but due to a malfunction of counter gates the channel was loaded all from the middle portion of the run.

Only allowing fish from the middle of the run to enter the spawning channel did not appear to impact the juvenile out-migration timing although it is possible the run timing in the spawning channel was constrained to a narrower window than in Gates Creek. For example, the channel migration had only just begun when the mark-recapture program began on March 27th while Gates Creek was already seeing an abundance of nearly 100,000 fry. However, both habitats peaked in the week of May 5th with the majority (>80%) of fish migrating in the 3 week period from April 15th to May 5th (Figure 13).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 19

5.0 SUMMARY and RECOMMENDATIONS

The primary objective of obtaining data on the status of the Sockeye population in Gates Creek and channel were largely met:

1. The abundance of out-migrating Sockeye fry were estimated. Total estimates were 12,738,610 emigrating Sockeye fry from Gates Creek, and 2,845,029 from Gates Creek spawning channel 2. Egg-to-fry survival was estimated. Estimated egg-to-fry survivals were 21% for Gates Creek and 16% for Gates Creek Spawning channel. 3. Fry per effective female was estimated. Sockeye fry production per effective female was estimated to be 537 for Gates Creek spawning channel and 720 for Gates Creek. 4. Biological information on Sockeye fry was taken. Lengths and weights were measured for Gates Creek and the spawning channel.

The low stable flows into the beginning of May facilitated a robust Sockeye fry estimate. Additionally, the addition of a ½’ mesh drum for the RST allowed experimentation with capture of Coho juveniles into higher flows than possible in previous years. A partial estimate for Coho juveniles was derived in 2012 but several issues hampering the collection of sufficient data to develop an annual estimate have become apparent in the three years of study:

1. It hasn’t been possible to assess the early portion (April) of the Coho migration in 2012, 2013 or 2014 since the IPT is not an effective method for Coho smolt capture.

2. It is not possible to assess both Sockeye and Coho in the RST as the RST is not efficient at capturing Sockeye fry. In contrast to the IPT the RST, which has good capture efficiency (16% on average in 2012) for Coho smolts, demonstrated an inability to capture Sockeye fry. In 2012 capture efficiency for Sockeye was low (0.1-2.8%) and in 2013 when the RST was utilized to capture Sockeye it dropped from 30,000 on April 25th (IPT) to 3,000 on April 26th (RST). Thus operating the RST from the start of the Sockeye fry migration in March would result in a poor Sockeye estimate.

3. Gates Creek freshet commences early to late May depending on weather conditions. The freshet is characterized by high velocity, debris laden flow, resulting in an inability to operate traps safely and consistently. May is the known peak migration timing for Coho juveniles from other BC watersheds (Cheakamus and Seymour Rivers) (Melville and McCubbing 2012; Ramos-Espinoza and McCubbing 2012). Therefore the inability to trap consistently during this period is likely to result in highly variable abundance estimates.

The first two issues may be resolvable by fishing both an RST and an IPT at the same time. There may be room in the current pool the IPT is fished in and a second cable way could be set up as a trial in 2015. The purchase of the ½’ mesh drum and some small modifications to floatation and screens in the capture box will make the RST more stable at higher flows and

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 20 increase the chance of fishing into May for Coho. The IPT can be fished to a water level of ~0.65 m, while the RST can be fished to a maximum water level of ~0.85 m. However, once water levels reach 0.85 m at the IPT site it becomes too turbulent to safely fish the RST.

The third issue of water velocity and debris load when water exceeds 0.85 m is not as easily resolved; however, once water levels stabilize debris levels decrease to manageable levels. It is likely that year-to-year weather and freshet conditions (particularly in May) will dictate the ability to derive accurate estimates of the number Coho juvenile out-migrating from Gates Creek.

The 2014 modifications to the proportional sampler at the channel were a success. Because of the increased reliability and consistency of the sampler only a portion of fish leaving the spawning channel will need to be exposed to handling stress going forward. For the spring of 2015 it is feasible the sample box could be calibrated (i.e. diverting full channel count into a separate trap box for comparison to sample box) 1-2 times per week, thus reducing the number of fish handled.

There are several important reasons to continue this study. Since the construction of the spawning channel, it has generally been thought that Gates Creek has limited suitable habitat for salmon spawning and the best operating regime for safe guarding the valuable Gates Creek Sockeye population against low productivity and stochastic events in river conditions has been to load the spawning channel to maximum capacity (Doug Lofthouse, pers. comm.) in the attempt to maximize fry abundance. Prior to this study, these assumptions had never been evaluated. The collection of juvenile abundance and survival data are crucial to the evaluation of the spawning channel management and developing best practices that will maximize fry abundance. The results from the 2012 – 2014 juvenile outmigration studies suggest Gates Creek produces a significant amount of fry and that best practice may be to ensure utilization of creek spawning habitat even in years of low escapement. Continuation of this study will aid DFO scientific advisory staff in making informed channel loading decisions. It is hoped that in future years Coho juvenile escapement can be assessed consistently as little data currently exists regarding Gates Creek Coho. More broadly, this study is one of only a few that estimates wild Sockeye fry abundance and egg-to-fry survival in British Columbia. The data from both the Sockeye and Coho components of this study will improve fisheries manager’s ability to develop integrated fisheries management plans and harvest forecasts.

This study is strongly linked to a number of ongoing monitoring projects throughout the Seton watershed. For example, ongoing projects are assessing the passage of adult Sockeye and Coho salmon through the Seton dam fishway and their success on the spawning grounds (BC Hydro WLR requirement, BGRMON-14, Effectiveness of Cayoosh Flow Dilution, Dam Operation, and Fishway Passage on Delay and Survival of Upstream Migration of Salmon in the Seton/Anderson Watershed), smolts are being enumerated as they migrate out of the Seton watershed, and the productivity of both Anderson and Seton lakes are being monitored

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 21

(BGRMON-6, Seton Lake Aquatic Productivity Monitoring). There also exist significant opportunity for capacity building in staff resources for the N’Quat’Qua, by continuation of these studies and through the Sta’at’imc Government Services fisheries technician program (Anon. 2012).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 22

TABLES

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 23

Table 1. Start and end dates for all traps operated on Gates Creek, in spring 2014.

Trap/Counter Name Start Date End Date Comments Channel Weir March 26 May 26 Fished 7 days per week

IPT March 27 May 11 Only fished Monday to Friday each week

RST May 12 May 24 Installed when water became too high for safe operation of IPT. Only fished Monday to Friday each week

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 24

Table 2. Summary of Gates Creek Sockeye fry fork lengths measured at both the spawning channel and IPT sites in, 2011-2014.

2012 2013 2014

Channel IPT Channel IPT Channel IPT

N 340 326 375 300 420 274

Range (mm) 25-54 26-40 23-44 22-39 23-47 27-36

Mean fork 30 29 29 29 29 29 length (mm)

SD (mm) 3 1 3 1 2 1

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 25

Table 3. Summary of Gates Creek Sockeye fry weights (g) measured at both the spawning channel and IPT sites in, 2011-2014.

2012 2013 2014

Channel IPT Channel IPT Channel IPT

N 340 326 375 275 420 274

Range (mm) 0.12-0.80 0.14-0.58 0.12-0.62 0.15-0.40 0.10-0.52 0.13-0.33

Mean fork 0.31 0.30 0.25 0.23 0.21 0.21 length (mm)

SD (mm) 0.13 0.08 0.09 0.04 0.07 0.04

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 26

Table 4. Summary of daily water temperatures °C (mean, standard deviation (SD) and range) in both Gates Creek and the spawning channel August 10th, 2013 to May 20th, 2014.

channel creek Average of average daily temp 5.1 5.3 SD of average daily temp 3.9 3.6 Min of average daily temp 0.0 0.0 Max of average daily temp 15.1 14.1

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 27

Table 5. Weekly totals of Sockeye fry marked at the spawning channel, recaptured and unmarked fish enumerated at the IPT in spring 2014. Trap efficiency is the proportion of marked fish that were recaptured.

Trap Efficiency Week Ending Marks Recaptures Unmarked Mortalities (%) March-31 877 129 5,841 59 15 April-7 5,831 785 14,006 182 14 April-14 10,000 2,148 63,451 176 22 April-21 9,036 2,402 235,102 642 27 April-28 10,000 1,160 258,481 1,084 12 May 5 7,500 305 150,306 1,976 4 May-12 9,260 104 11,896 137 1 Total 52,504 7,033 739,083 4,256 -

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 28

Week Ending Mean SD Relative 2.50% 97.50% Temp Water Level SD1 (◦C) (m) Mar-31 93,451 7,669 0.08 79,673 109,922 1.45 0.47 Apr-07 146,111 4,952 0.03 136,699 156,062 7.45 0.47 Apr-14 413,935 7,989 0.02 398,641 429,375 5.18 0.48 Apr-21 1,548,488 27,395 0.02 1,495,167 1,604,776 1.18 0.49 Apr-28 3,122,084 87,027 0.03 2,953,916 3,291,163 3.30 0.50 Mar-05 8,583,464 482,183 0.06 7,695,273 9,561,876 6.51 0.55 May-12 1,503,903 14,8354 0.10 1,239,452 1,826,553 6.96 0.65 Total 15,411,437 516,796 0.03 14,497,151 16,494,072 - - Abundance 1 Relative SD is a measure of precision. Values > than 0.30 indicate low precision.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 29

Table 7. Weekly abundance of Sockeye fry leaving Gates Creek and spawning channel and total abundance of fry past the IPT site on Gates Creek in spring 2014.

Week Ending Channel Creek Total fry passing IPT (includes marks) Mar-31 8,539 85,789 94,328 Apr-07 23,468 128,474 151,942 Apr-14 67,556 356,379 423,935 Apr-21 247,600 1,309,924 1,557,524 Apr-28 735,708 2,396,376 3,132,084 May-05 1,341,285 7,249,679 8,590,964 May-12 301,175 1,211,988 1,513,163 Sub Total 2,725,331 12,738,610 15,463,941 May-19 109,661 - - May-26 10,037 - - Total 2,845,029 12,738,610 15,583,639

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 30

Table 8. Female Sockeye escapement, fecundity, survival figures for Gates Creek system 2011- 2013.

2011 2012 2013

Creek Channel Whole Creek Channel Whole Creek Channel Whole System System System

Total female 25,907 8,302 34,209 8,336 9,514 17,850 23,376 6,510 29,877 escapement

Effective females 21,297 4,574 25,895 4,026 2,351 6,377 17,702 5,302 23,004

Mean Fecundity 3,192 3,192 3,192 3,119 3,119 3,119 3,378 3,378 3,378

Egg deposition 69,428,220 14,989,480 84,417,700 12,557,992 7,332,769 19,890,761 59,798,345 17,910,452 77,708,797

Fry produced 10,214,909 5,515,083 15,792,991 2,154,746 2,637,647 4,792,393 12,738,610 2,845,029 15,583,639

Egg-to-fry survival 15% 38% 19% 17% 36% 24% 21% 16% 20%

Fry per effective 480 1,199 610 535 1,122 751 720 537 677 female

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 31

FIGURES

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 32

Figure 1. Overview of Seton-Anderson watershed in south western British Columbia.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 33

Figure 2. Overview of study area including Gates Creek, the spawning channel and the IPT/RST site near the confluence with Anderson Lake.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 34

Figure 3. Weir at Gates Creek spawning channel. Fish are funneled through black sampler structure in middle of trough to the wood box visible on the left.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 35

Figure 4. Incline Plane Trap (left) and rotary screw trap (right) in Gates Creek.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 36

Figure 5. Frequency distribution of juvenile Sockeye fork lengths (mm) leaving Gates Creek spawning channel and Gates Creek in spring 2014. Red dotted line indicates sample mean.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 37

Figure 6. Frequency distribution of juvenile Sockeye weight (g) leaving Gates Creek spawning channel and Gates Creek in spring 2014. Red dotted line indicates sample mean.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 38

Figure 7. Frequency distribution of Coho smolts (1+) fork lengths (mm) leaving Gates Creek in spring 2014. Red dotted line indicates sample mean.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 39

Figure 8. Frequency distribution of Coho smolts (1+) weight (g) leaving Gates in spring 2014.Red dotted line indicates sample mean.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 40

Figure 9. Average daily temperatures in Gates Creek spawning channel (red) and at IPT (black) site in Gates Creek in fall 2013 to spring 2014.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 41

Figure 10. Accumulated thermal units from September 5 th, 2014 (peak of the spawning) to May 5th, 2014 (50% migration date) for both Gates Creek spawning channel (solid line) and Gates Creek (dashed line).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 42

Figure 11. Average weekly water level (m) in Gates Creek (measured at the IPT site) in spring 2014.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 43

Figure 12. Run timing of the total abundance of Sockeye fry leaving Gates Creek watershed in spring 2014. Estimates include fry from both Gates Creek and Gates Creek spawning channel.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 44

Figure 13. Run timing of the total abundance of Sockeye fry leaving Gates Creek (solid line) and Gates Creek spawning channel (dashed line) in spring 2014. Estimates include fry from both Gates Creek and Gates Creek spawning channel.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 45

Figure 14. Water temperatures in the Fraser River at Hope, 2013. Source Fraser River Enviromental Watch, Fisheries and Oceans Canada (http://www.pac.dfo- mpo.gc.ca/science/habitat/frw-rfo/reports-rapports/archives-eng.html).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 46

REFERENCES

Andrew FJ and Green GH (1958) Sockeye and pink salmon investigations at the Seton Creek hydroelectric installation. Progress Report, International Pacific Salmon Commission, New Westminister, Canada. 78 p. Anon. (2001) Gates Creek assessment project. Creekside Rescource Inc., Mount Currie, Canada. 35 p. Anon. (2009) Rehabilitation of Sockeye spawning gravel in the Gates Creek spawning channel- year 2. Northern St'at'imc Fisheries, Lillooet, Canada. 56 p. Anon. (2012) St'at'imc Draft Fisheries Implementation Plan. St'at'imc Government Service, Lillooet, Canada. Arnason AN, Kirby CW, Schwarz C, J., and Irvine JR (1996) Computer analysis of data from stratified mark-recovery experiments for the estimation of salmon escapements and other populations. Canadian Technical Report of Fisheries and Aquatic Sciences 2106: 37. Bonner SJ and Schwarz C, J. (2011) Smoothed estimates for time stratified mark-recapture experiments using Bayesian p-splines. Biometrics 67: 1498-1507. Bonner SJ and Schwarz C, J. (2012) An application of a bayesian stratified-petersen model to estimate the number of outgoing fish on the Cheakamus River, British Columbia. BC Hydro. 119 p. Available at: http://people.stat.sfu.ca/~cschwarz/Consulting/BCHydro- 2012-04-04/Report-2012-04-04.pdf . Bradford MJ (1995) Comparative review of Pacific salmon survival rates. Canadian Journal of Fisheries and Aquactic Sciences 52: 1327-1338. Crossin GT, Hinch SG, Cooke SJ, Welch DW, Patterson DA, Jones SRM, Lotto AG, Leggatt RA, Mathes MT, Shrimpton JM, Van Der Kraak G and Farrel AP (2008) Exposure to high temperature influences the behaviour, physiology, and survival of sockeye salmon during spawning migration. Canadian Journal of Zoology. 86(2):127-140. Dutang C, Goulet V and Pigeon M (2008) Actuar: an R package for actuarial science. Journal of Statistical Software 25(7): 1-37. Einum S and Flemming IA (2000) Selection against late emergence and small offspring in Atlantic Salmon (Salmno salar). Evolution. 54(2): 628-639. Fretwell MR (1989) Homing behaviour of adult Sockeye salmon in repsonse to a hydroelectric diverion of homewater at Seton Creek. International Pacific Salmon Commission Bulletin 25: 38. Grey D (2013) Fine sediment removal from streams; environmental effects, protocols and a proposed rule. Environment Canterbury Recional Council. 73p.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 47

Henderson MA and Cass AJ (1991) Effect of smolt size on smolt-to-adult survival for chilko lake Sockeye salmon (Oncorhynchus nerka). Canadian Journal of Fish and Aquatic Science. 48: 988-994. Hillaby J (2012) Late summer distribution of juvenile Coho salmon in the Gates Creek watershed. BC Hydro. 63 p. Komori V (1997) Strategic fisheries overview for the Bridge/ Seton habitat management area. Fraser River Action Plan, Department of Fisheries and Oceans, Vancouver, Canada. 83 p. Lingard S, Ladell J and McCubbing D (2013a) Gates Creek adult Sockeye abundance fall 2013. Prepared for the Lillooet Tribal Council and Department of Fisheries and Oceans, Canada, Instream Fisheries Research, Inc., North Vancouver, Canada. 20 p. Lingard S, Melville C, Braun D and McCubbing D (2013b) Gates Creek Juvenile Salmonid Outmigration Assessment, Spring 2013. Prepared for the Lillooet Tribal Council and Department of Fisheries and Ocean, Canada, Instream Fisheries Research, Inc., North Vancouver, Canada. 33 p. Lunn D, Spiegelhalter D, Thomas A and Best N (2009) The BUGS project: Evolution, critique and future directions (with discussion). Statistics in Medicine. 28: 3049-3082 Martins EG, Hinch SG, Patterson DA, Hague MJ, Cooke SJ, Miller KM, Robichaud D, English KK and Farrel AP (2012) High river temperature reduces survival of sockeye salmon (Oncorhynchus nerka) approaching spawning grounds and exacerbates female mortality. Canadian Journal of Fisheries and Aquaitc Sciences. 69(2): 330-342. Melville CC and McCubbing DJF (2001) Assessment of the 2000 juvenile salmon migration from the Cheakamus River, using rotary traps. BC Hydro WUP Report. 42 p. Melville CC and McCubbing DJF (2002a) Assessment of the 2001 juvenile salmon migration from the Cheakamus River, using rotary traps. BC Hydro WUP Report. 53 p. Melville C.C and D.J.F. McCubbing. 2002b. Assessment of the 2002 Juvenile Salmon Migration from the Cheakamus River, using Rotary Traps. BC Hydro WUP Report. 36p. Melville CC and McCubbing DJF (2003) Assessment of the 2003 juvenile salmon migration from the Cheakamus River, using rotary traps. BC Hydro WUP Report. Melville CC and McCubbing DJF (2004) Assessment of the 2004 juvenile salmon migration from the Cheakamus River, using rotary traps. BC Hydro WUP Report. 80 p. Melville CC and McCubbing DJF (2006) Assessment of the 2005 juvenile salmon migration from the Cheakamus River, using rotary traps. BC Hydro WUP Report. 76 p. Melville CC and McCubbing DJF (2007) Assessment of the 2006 juvenile salmon migration from the Cheakamus River, using rotary traps. BC Hydro WUP Report. 71 p. Melville CC and McCubbing DJF (2008) Assessment of the 2007 juvenile salmon migration from the Cheakamus River, using rotary traps. BC Hydro WUP Report. 104 p.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 48

Melville CC and McCubbing DJF (2009) Assessment of the 2008 juvenile salmon migration from the Cheakamus River, using rotary traps. BC Hydro WUP Report. 103 p. Melville CC and McCubbing DJF (2010) Assessment of the 2009 juvenile salmon migration from the Cheakamus River, using rotary traps. BC Hydro WUP Report. 114 p. Melville CC and McCubbing DJF (2011) Cheakamus River juvenile salmonid outmigration enumeration assessment, spring 2010. Cheakamus River Water Use Plan Report. 118 p. Melville CC and McCubbing DJF (2012) Cheakamus River juvenile salmonid outmigration enumeration assessment summary report 2001-2012. Cheakamus River Water Use Plan Report, North Vancouver, Canada. 92 p. Melville CM and McCubbing D (2002b) Assessment of the 2002 juvenile salmon migration in the Cheakamus River, using rotary traps., Instream Fisheries Research, Inc. 36 p. Murray CB and McPhail JD (1987) Effect of incubation temperature on the development of five species of Pacific salmon (Oncorhynchus) embryos and alevins. Canadian Journal of Zoology. 66:266-273. Plummer M, Best N, Cowles K and Vines K (2006) CODA: convergence diagnosis and output analaysis for MCMC. R News 6: 7-11. Quinn TP (2005) The Behaviour and Ecology of Pacific Salmon and Trout. University of Washington Press, Seattle, U.S.A. 378 p. Ramos- Espinoza D and McCubbing DJF (2012) Seymore River juvenile salmonid outmigration monitoring, Spring 2012. Prepared for Metro- Vancouver Poilcy and Planning Department- Utility Analysis & Environmental Management Division, Instream Fisheries Research, Inc., North Vancouver, Canada. 54 p. Ricker WE (1975) Computation and interpretation of biological statistics of fish populations. Bulletin of the Fisheries Research Board of Canada 191: 382. Seber GAF (2002) The estimation of annual animal abundance and related parameters. Blackburn Press, New Jersey, USA. 676 p. Shepherd BG, Hartman GF and Wilson, WJ (1986) Relationships between stream and intragravel temperatures in coastal drainages, and some implication for fisheries workers. Canadian Journal of Fish and Aquatic Science. 45: 1818-1822 Sogard SM (1997) Size-selective mortality in the juvenile stage of teleost fishes: a review. Bulletin of Marine Science. 60(3): 1129-1157. Sturtz S, Ligges U and Gelman A (2005) R2WinBUGS: a package for running WinBUGS from R. Journal of Statistical Software. 12(3): 1-16. Talbot GB (1950) A biological study of the effectivness of the Hell's Gate fishways. Bulletin 3, International Pacific Salmon Commission, New Westminster, Canada. 80 p. West CJ, and Larkin PA (1987) Evidence of size- selective mortality of juvenile Sockeye salmon (Oncorhynchus nerka) in Babine Lake, British Columbia. Canadian Journal of Fish and

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 49

Aquatic Science. 44: 712-721.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 50

Section 2: Gates Creek Adult Sockeye Abundance, Fall 2014

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 51

Executive Summary

This report presents the results of the 2014 adult component of the Gates Creek salmonid outmigration study. The data from this report are collected to compliment juvenile Sockeye Salmon abundance data to be collected in the spring of 2015. The abundance, fecundity and percent spawn data from this report will be used to generate estimates of egg to fry survival for both Gates Creek and Gates Creek spawning channel in the spring of 2015.

The first Sockeye arrived at Gates Creek on August 20th in 2014, 10 days later than 2013. The total estimated abundance of Sockeye returning to the Gates Creek watershed was 16,929 for the fall of 2014. Of the total abundance, a total of 10,300 fish were estimated to have entered Gates Creek, 76 were estimated to spawn on the Anderson Lake shore and 6,553 were estimated to have ascended the spawning channel.

Mean fecundity for females sampled from the spawning channel was 3,358 (SD 777). Overall percent spawn for the watershed was estimated to be 85.4%. A total of 3,211 and 5,245 females successfully completed spawning in the spawning channel and Gates Creek, respectively. Egg deposition for Gates Creek was estimated to be 10,782,538 for the spawning channel and 17,612,710 for Gates Creek in the fall of 2014.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 52

Table of Contents

Section 2: Gates Creek Adult Sockeye Abundance, Fall 2014 ...... 50 1.0 INTRODUCTION ...... 57 2.0 METHODS ...... 59 2.1 Loading Strategy ...... 59 2.2 Temperature Monitoring ...... 59 2.3 Generation of Abundance Estimates ...... 60 2.3.1 Visual Survey Validation of Counter(s) ...... 60 2.3.2 Video Validation of Counters ...... 61 2.4 Classification of Female Percent Spawn ...... 61 2.5 Fecundity Sampling ...... 62 2.6 Percent Spawn and Total Egg Deposition ...... 62 3.0 RESULTS ...... 64 3.1 Temperature Monitoring ...... 64 3.2 Video Validation ...... 64 3.3 Abundance Estimates ...... 64 3.3.1 Spawning Channel ...... 64 3.3.2 Gates Creek ...... 65 3.3.3 Anderson Lake Shore ...... 65 3.4 Fecundity and Female Size ...... 65 3.5 Percent Spawn and Egg Deposition ...... 66 3.5.1 Spawning Channel ...... 66 3.5.2 Gates Creek ...... 66 4.0 DISCUSSION ...... 67 4.1 Temperature Monitoring ...... 67 4.2 Counter Operations and Visual Survey/ Video Validation ...... 67 4.3 Migration Timing and Sockeye Abundance ...... 68 4.3.1 Migration Timing ...... 68 4.3.2 Sockeye Abundance ...... 68 4.4 Fecundity and Egg Deposition ...... 68 4.5 Percent Spawn ...... 68 5.0 Summary and Recommendations...... 69

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 53

TABLES ...... 72 FIGURES ...... 82 REFERENCES ...... 91

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 54

LIST OF TABLES

Table 9. Temperature data from TidbiT v2 Water Temperature Data Loggers in Gates Creek and Gates Creek spawning channel. Temperatures were logged every 15 minutes but are summarized as mean average daily temperature, Fall 2014 ...... 73 Table 10. Video validation data for mechanical counter in Gates Creek, Fall 2014 ...... 74 Table 11. Summary of Sockeye abundance estimates for Gates Creek spawning channel collected via visual surveys 2011-2014 ...... 75 Table 12. Summary of adult Sockeye abundance estimates for Gates Creek 2011- 2014. Abundance estimates for Gates Creek are developed using a combination of data from visual surveys and the mechanical counter ...... 76 Table 13 Summary of adult Sockeye abundance for Anderson Lake shore 2011-2014. Abundance estimates developed from visual survey data collected by DFO stock assessment 77 Table 14 Summary of adult Sockeye abundance for the Gates Creek Watershed from 2011- 2014 ...... 78 Table 15. Mean fecundity and post-orbital hypural (POH) length (cm) in Gates Creek spawning channel females Sockeye for 2011 - 2014 ...... 79 Table 16. Break down of female percent spawn in Gates Creek spawning channel from visual survey data 2011 - 2014 ...... 80 Table 17. Estimated number of effective females, mean fecundity and egg deposition for the 2011-2014 brood years for Sockeye spawning in Gates Creek and Gates Creek spawning channel (estimates of effective females for Gates Creek were provided by DFO unpublished data) ...... 81

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 55

LIST OF FIGURES

Figure 15. Overview of Seton-Anderson watershed in South Western British Columbia ...... 83 Figure 16. Map of study area including Gates Creek and spawning channel near confluence with Anderson Lake. Also shown is the trap site for the juvenile monitoring component of the study...... 84 Figure 17. Schematic diagram of enumeration equipment at confluence of Gates Creek and Gates Creek spawning channel spawning channel ...... 85 Figure 18. Average daily temperature in Gates Creek (dotted black line) and Gates Creek spawning channel (grey line) over the Sockeye spawning period from August 10th to October 6st, 2014 ...... 86 Figure 19. Scatter plot with fitted regression line (y= 159.3x+ -4459.7, R2 0.6127) of post-orbital hypural (POH) length (cm) and female fecundity (number of eggs) in Gates Creek Sockeye, Fall 2014 ...... 87 Figure 20. Distribution of female percent spawn over the fall 2014 Sockeye spawning period at Gates Creek spawning channel. Red, blue and black lines represent 0, 50 and 100% spawned females respectively...... 88 Figure 21 Temperature data for Fraser River at Mission for summer 2014 Sockeye migration window (http://www.pac.dfo-mpo.gc.ca/science/habitat/frw-rfo/reports-rapports/2014/2014-08- 28/2014-08-28-eng.html) ...... 89 Figure 22 Total annual adult Sockeye (*102) loaded into Gates Creek Spawning channel (black solid line) plotted with annual percent spawn (dotted grey line) 1996-2014 (DFO unpublished data) ...... 90

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 56

Acknowledgements This component of the study was conducted and report prepared with financial support of the Fish and Wildlife Compensation Program on behalf of its program partners BC Hydro, the Province of BC, First Nation and Public Stakeholders, and Fisheries and Oceans Canada.

We would like thank the following people for their cooperation and assistance on this project:

Harry O’Donaghey - N’Quatqua Fisheries Lance O’Donaghey - N’Quatqua Fisheries

Chris Fletcher - N’Quatqua Fisheries Harry O’Donaghey, Jr. - N’Quatqua Fisheries

Spencer Thevarge - N’Quatqua Fisheries

Peter Campbell - Fisheries and Oceans Doug Lofthouse - Fisheries and Oceans Canada Canada

Dave Willis - Fisheries and Oceans Canada Keri Benner - Fisheries and Oceans Canada

Matthew Foy - Fisheries and Oceans Canada Brian Leaf - Fisheries and Oceans Canada

Andrew Grant - Fisheries and Oceans Shona Smith - Fisheries and Oceans Canada Canada

Paul Welch - Fisheries and Oceans Canada Matt Townsend - Fisheries and Oceans Canada

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 57

1.0 INTRODUCTION

Gates Creek is a major salmon bearing tributary of the Seton-Anderson Watershed and runs 12 kilometres from Gates Lake to Anderson Lake draining approximately 34,300 hectares (Komori 1997) (Figure 15). The Seton-Anderson watershed is located approximately 200 km north of Vancouver in the rain shadow of the southern Coast Mountains (Anon. 2000) and contains no glaciers. Anderson Lake is connected to Seton Lake via the Portage Creek, and Seton Lake drains into the Fraser River via Seton River (Figure 1). Gates Creek supports a population of Fraser River Sockeye Salmon (Onchorynchus nerka) that is important for First Nation, commercial and recreation fisheries, as well as smaller populations of Coho Salmon (O. kisutch), and Pink Salmon (O. gorbushuca).

The over 100 populations of Fraser River Sockeye are assigned into four distinct run timing groups based on the time period they enter freshwater to begin their upstream migration: Early Stuart, Early Summer, Summer and Late Summer. Gates Creek Sockeye fall into the early summer run-timing group entering the Fraser River from mid-July through to mid-August. Gates Creek Sockeye along with the Portage River population form the Anderson-Seton-ES conservation unit (Anon 2012). Fraser River Sockeye face numerous challenges during their upriver migration including recreational and commercial fisheries in the lower portion of the Fraser River and turbulent flows in the Fraser Canyon. Mean summer water temperature in the Fraser River has warmed ~2.0°C since the 1950’s (Patterson et al. 2007) and is forecasted to continue to increase with climate change (Morrison et al. 2002, Ferrari et al. 2007). High temperatures have been associated with increased energetic demands, susceptibility to infection, and reduced spawning success in Pacific salmon (Lee et al. 2003, Crossin et al. 2008).

Beginning in the early 1900’s Gates Creek Sockeye have been affected by several major development projects. Fraser salmon populations upstream of Hell’s Gate including Gates Creek populations were heavily impacted by the slides of 1913 and 1914 (Talbot 1950, Andrew and Green 1958). In 1956, as part of the Bridge River Hydro development, a diversion dam was constructed on the Seton River 750 m downstream of Seton Lake. The development, which also included a canal to a powerhouse on the Fraser River, has had significant impacts on the Portage Creek and Gates Creek salmon stocks, through entrainment of juveniles and reduced adult escapement (Fretwell 1989, Komori 1997). In addition to these downstream impacts salmon habitat on Gates Creek has been degraded by residential and agricultural developments (Anon. 2001).

In 1968, a Sockeye specific spawning channel was constructed by the International Pacific Salmon Fisheries Commission (IPSFC) in Gates Creek 800 m upstream of Anderson Lake to enhance Sockeye escapement in the Seton-Anderson watershed (Figure 16). The spawning

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 58 channel was originally overseen by the IPSFC and Fisheries and Oceans Canada (DFO). In 1987, responsibility for channel maintenance and monitoring was turned over to the N’Quatqua First Nation with technical oversight from DFO. A gravel replacement project was undertaken in 2008 and 2009 by DFO and the Fish and Wildlife Compensation Program (FWCP) with the goal of increasing egg-to-fry survival in Gates Creek spawning channel (Anon. 2009). In addition to gravel replacement, changes were made to channel structure and gradient during this project (Anon. 2009). While a long standing time series of juvenile and adult abundances are available for Gates Creek spawning channel, detailed assessment of the gravel replacement activities had yet to be undertaken. Previous to this study, data regarding Sockeye egg-to-fry survival and abundance from Gates Creek had never been collected.

In 2011, the DFO scientific advisors for the Gates Creek spawning channel requested that Instream Fisheries Research Inc. (IFR) submit a study design to enumerate out-migrant Sockeye fry and coho juveniles in Gates Creek to compliment ongoing work on the Gates Creek spawning channel. A proposal including the juvenile enumeration study and complimentary Sockeye adult monitoring was submitted by the Lillooet Tribal Council (LTC) and DFO to the BC Hydro Fish and Wildlife Compensation Program. The objectives of the adult component of this study are to assess the following parameters for adult Sockeye returning to spawn in Gates Creek and the spawning channel:  Estimate adult Sockeye abundance  Estimate female percent spawn  Estimate mean female fecundity  Generate estimates of egg deposition This report summarizes the findings of the 2014 adult component of the study.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 59

2.0 METHODS

2.1 Loading Strategy

The Gates Creek spawning channel and Gates Creek are loaded with a known number of adults using a full channel weir and mechanical counters. Fish are excluded from Gates Creek by a full span diversion fence installed on the creek adjacent to the spawning channel out-flow (Figure 3). A second exclusion fence which houses the mechanical counters is installed in the spawning channel (Figure 17). The mechanical counters are similar to those used on subway entry gates which count a single up count as the gate is pushed past a trigger point. There are three counters on the exclusion fence with openings large enough for a single Sockeye salmon to enter at a time. Two of the gateways direct fish into the spawning channel. The third gate is attached to a culvert which returns fish approximately 100 m underground into Gates Creek above the diversion fence. Counter gates are lockable allowing fish to be directed into either the spawning channel or the creek based on a management strategy directed by DFO advisory staff. Ideally, the channel and creek are loaded so fish from each portion of the run are present in equal proportions in each environment.

In 2012 and 2013 the DFO scientific advisory staff intended to allow the early portion of the run fish into Gates Creek with the postulation that early run fish are more likely be affected by pre- spawn mortality (PSM). Pre-spawn mortality was too high in the spawning channel in 2012 to test this theory (Lingard et al. 2013). In 2013, a communication error resulted in the spawning channel being loaded entirely with fish from the early portion of the run (Lingard et al. 2014). In 2014 the initial plan was to load the channel with the latter portion of the run; however, returns were low and it was not possible to carry out this strategy in its entirety.

The first Sockeye arrived at Gates Creek on August 20th, 2014. Sockeye were loaded into Gates Creek from August 20th to August 27th, and again from August 31st to September 15th. Loading of the channel began on August 28th through to September 6th, and again from September 14th to September 24th.

2.2 Temperature Monitoring

Pairs of Onset TidbiT v2 Water Temperature Data Loggers (accuracy range of 0.2°C) were placed in the spawning channel and Gates Creek in the summer of 2014. The loggers in the spawning channel were placed near the middle of the channel on the fifth leg. In Gates Creek the loggers were placed on river left across from the spawning channel intake (UTM

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 60

10563462E, 5599630N). Temperature was logged every 15 minutes. For analysis data was condensed to average daily temperature (°C).

2.3 Generation of Abundance Estimates

Abundance estimates for Gates Creek and the spawning channel were generated using a combination of counter totals, visual survey data and video validation in the fall of 2014. In the spawning channel a visual survey was used to generate the total escapement for the entire channel. Visual foot surveys in Gates Creek are used to determine percent spawn and sex ratios, but do not validate the counter totals. To evaluate counter efficiency a secondary count is necessary for providing estimates of counter efficiency. In Gates Creek a combination of counter totals and video data were used to generate total abundance estimates.

2.3.1 Visual Survey Validation of Counter(s)

Spawning Channel Visual surveys of adult Sockeye, consisting of removal and enumeration of carcasses, are conducted in the spawning channel annually. Sockeye carcasses are removed from the water and placed on the channel bank. The carcasses are separated as males, females, or jacks (precocious males) prior to enumeration. Females are classified as 0%, 50% or 100% spawned, which refers to the number of eggs retained in the carcass (see section 2.4 for further discussion of percent spawn criteria). Carcasses are cut in half as they are counted and loaded into a dump truck. Generally, two people work together to process the carcasses. One technician either uses a field book or hand tally to keep count while the second technician cuts the fish. Cutting of carcasses serves to both expose the egg cavity of females which allows evaluation of percent spawn as well as prevents double counting of carcasses. The enumerated carcasses are then removed to Anderson Lake to prevent bears from becoming habituated to the channel property. Comparisons of mechanical counter generated data to channel visual enumeration data are then undertaken.

A sub sample of carcasses (180 males, 185 females, and 100 jacks) were taken from the spawning channel for biological sampling for the Pacific Salmon Commission (PSC). Fish sampled for PSC sampling are measured for standard length and post-orbital hypural length. Otoliths and scale samples are also collected for each fish.

Gates Creek Weekly visual surveys in Gates Creek are conducted by N’Quatqua technicians in conjunction with DFO stock assessment personnel. N’Quatqua technicians accompany DFO personnel on

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 61 weekly visual surveys. Visual surveys in Gates Creek follow the same protocols for classification of percent spawn as outlined above for the spawning channel. However, in Gates Creek only a sample of individuals are enumerated from the population as it would be unfeasible to recover all carcasses in an unmanaged environment. The primary objective of visual surveys in Gates Creek above the counting fence (exclusion fence) is to obtain sex ratio and percent spawn estimates, which are later applied to the total escapement estimates obtained by the mechanical counter. Downstream of the fence, weekly visual surveys include a live count as well as carcass recovery. Peak live count plus the cumulative carcass recovery to that date are used to generate an estimate of the number of spawners in the portion of Gates Creek below the fence. The visual survey data are collected by DFO stock assessment and are reported back as final estimates.

2.3.2 Video Validation of Counters

In 2014 video footage of fish passing through the mechanical counters was recorded. The counter totals were recorded at the start and end of each video clip. To achieve an estimate of counter efficiency, the total number of fish counted by the counter for all video segments was compared to the total number seen in all video segments combined.

Video was only recorded for the counter into Gates Creek as the spawning channel counters are validated via visual surveys.

2.4 Classification of Female Percent Spawn

In 2011 inconsistencies were identified in survey methods employed by N’Quatqua technicians on the channel and the criteria used to estimate percent spawn of female carcasses on the creek by DFO stock assessment personnel. In an effort to eliminate these inconsistencies the N’Quatqua technicians accompanied DFO stock assessment staff during weekly Gates Creek enumeration activities in both the fall of 2012 and 2013. In general, the method of estimating female percent spawn is highly subjective and requires technicians to classify female mortalities as 0%, 50%, or 100% spawned. By DFO stock assessment definition a female Sockeye is categorized as:

 100% spawned if only a handful of loose eggs remain in the fish (which equates to approximately 500 or less eggs).  50% spawned if roughly two handfuls or more loose eggs remain in the fish (greater than a few hundred eggs, but skeins not intact). *DFO advises this category to be rare among the Gates Creek Sockeye population.  0% spawned if there are intact skeins remaining in the fish.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 62

To allow continuity of data collection between DFO and N’Quatqua Fisheries technicians, the above listed criteria were used for assessment of percent spawn in Sockeye females in Gates Creek and spawning channel in 2014.

2.5 Fecundity Sampling

Mean fecundity of female Sockeye was estimated using intact egg skeins from 46 un-spawned (0%) females taken from the spawning channel. All sampled individuals were natural pre-spawn mortalities. Samples were distributed over the duration of the spawning period lasting from August 20th to September 10th. Post-orbital hypural (POH) lengths (cm) were taken for each fish. Total gonad weight (raw and water hardened) (g) was recorded for each female. A sub sample of 100 water hardened eggs was weighed from each fish. The sub sample weight was divided into the total gonad weight (water hardened) to arrive at the total number of eggs for each female. Full count of all eggs was carried out for 15% of the fish for verification of volume sampling technique.

2.6 Percent Spawn and Total Egg Deposition

Percent spawn refers to the proportion effective females (100% spawned) represent of the total annual female abundance. To estimate total effective females and percent spawn the total number of spawned females must be adjusted for females which only spawned partially (the 50% category). While 50% spawn is an arbitrary and subjective figure applied to a wide range of partially spawned Sockeye salmon, it was advised that standard DFO protocol for using this category in estimates of percent spawn and PSM is to split the 50% category total equally between the 0% and 100% categories (S. LePage D.F.O Stock Assessment, pers. comm.). In essence it is assumed a 50% spawned fish represents half a fully spawned fish (100%) and half an un-spawned fish (0%). The following equations were used to estimate percent spawn and pre-spawn mortality rates.

PS= (S+0.5*P)/ TF PSM= (U+0.5*P)/ TF

Where: PS= Percent spawn PSM= Pre-spawn mortality S= Spawned females (100%)

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 63

U= Un-spawned females (0%) P= Partial spawned females TF = Total number of females (summed 0%, 50% and 100% categories)

Egg deposition was estimated by multiplying the number of effective (100% spawned) females by the estimated mean fecundity.

ED= EF*MF

Where: ED= Estimated egg deposition EF= Number of effective females given by (S +0.5*P) MF= Mean fecundity

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 64

3.0 RESULTS

3.1 Temperature Monitoring

The spawning channel was on average 0.2°C warmer than Gates Creek. The spawning channel appears to warm slightly faster than Gates Creek during periods of warm weather (Figure 18). Average daily temperature in the spawning channel ranged from 7.8°C to 14.7°C over the period August 10th to October 1st. Over the same period, Gates Creek had a similar temperature range of 7.6°C to 14.0°C (Table 9). In the spawning channel the mean average daily temperature was similar over the spawning period (Figure 18, Table 9).

3.2 Video Validation

Video validation indicates the creek counter inflated the number of fish passing into Gates Creek by 13% in 2014. Total counts by the Gates Creek counter over the 23 hours of video was 952 and the total number of fish observed in all video segments was 841 (Table 10).

3.3 Abundance Estimates

In 2014 the first Sockeye arrived at the exclusion fence on August 20th, 10 days later than in 2013 (Lingard et al. 2013). From August 20th to September 24th a total of 10,300 (13,475 before adjustment) fish were counted into Gates Creek. From August 28th to September 24th a total a total of 6,543 fish were counted into the channel by the counters. An additional 76 spawners were counted on the Anderson Lake shore in the fall of 2014.

3.3.1 Spawning Channel

The total estimated abundance of the 2014 brood for Gates Creek Sockeye from visual surveys was 6,553. Of the total abundance females represented 37% (3,739), and males (including 481 jacks) represented 43% (2,814) (Table 11).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 65

3.3.2 Gates Creek

Visual survey data is collected and processed by DFO stock assessment staff for Gates Creek. DFO estimated (adjusted for counter efficiency for above the fence) a total of 10,300 Sockeye returned to spawn in Gates Creek (9,681 above the fence and 619 below the fence) in the fall of 2014. Of the total abundance, males represented 40% (4,140 including 692 jacks) and females represented 60% (6,160). The ratio of males (including jacks) to females was 1:1.49 in fall 2014 (Table 12).

3.3.3 Anderson Lake Shore

In addition to spawning in Gates Creek and the spawning channel, 76 adult Sockeye were observed spawning on the Anderson Lake shore near the mouth of Gates Creek. This is the second year in a row spawning has been observed on the lake shore. There were 30 were males (including 5 jacks), and 46 females (Tables 13 & 14).

3.4 Fecundity and Female Size

Mean POH length and fecundity for the females sampled for fecundity from spawning channel in 2014 were estimated to be 49.1 cm (SD 3.8) and 3,358 (SD 777), respectively (Table 15). The linear relationship between POH length and fecundity for Sockeye sampled from Gates Creek spawning channel in 2014 can be expressed by the equation:

y= 159.3x+ -4459.7 (R2=0.6127) where y= the number of eggs, and w= POH length in cm

This regression analysis (Figure 19) shows a positive relationship between egg number and POH (R2 = 0.6127) with 61% of variability in egg number being explained by POH in 2014.

In 2014, mean female fecundity ranged from a minimum of 685 to a maximum of 5,365 (Table 15). Female Sockeye measured at Gates Creek spawning channel for fecundity sampling ranged in POH length from 40.0 cm to 56.5 cm with a mean of 49.1 cm (SD 3.8) (Table 15).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 66

Over the four years of this study 2012 had the lowest mean fecundity and smallest mean POH length (Table 15). The mean fecundity of females in 2012 (3,119) was 5% lower than in 2011 (3,260), and 8% lower than in 2013 (3,378), but only 7% lower than 2014 (3,358). Females sampled for fecundity were also smallest in 2012 (46.5 cm) when compared to 2011 (48.0 cm), 2013 (49.0 cm) and 2014 (49.1 cm) (Table 15).

3.5 Percent Spawn and Egg Deposition

3.5.1 Spawning Channel

A total of 3,605 females were evaluated for percent spawn in the spawning channel in the fall of 2014. Female Sockeye spawning success in the channel was assessed as 506 un-spawned (0%), 66 partial spawners (50%), and 3,033 complete spawned (100%) (Table 16). With conversion of partial spawners to their effective female equivalents for all females loaded into the channel, percent spawn is estimated to be 85.9% or 3,211 females in total for the spawning channel in the fall of 2014. Estimated egg deposition for the fall of 2014 was 10,782,538 which is 40% lower than 2013 (Table 17).

The distribution of 0% and 100% spawned females changed over the spawning period. Early in the run, prior to September 7th, there was a greater proportion of 0% spawned female carcasses as compared to 100% carcasses. After September 10th, the number of 0% spawners decreased to near zero (Figure 20).

3.5.2 Gates Creek

The percent spawn estimate for Gates Creek was slightly (1%) higher than the estimate for the spawning channel. Of the 6,160 females estimated to have entered Gates Creek 85% or 5,245 are estimated to have been effective spawners. Estimated egg deposition in Gates Creek was 17,612,710 for the fall of 2014 which is a 71% reduction from 2013 (Table 17).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 67

4.0 DISCUSSION

4.1 Temperature Monitoring

Temperature data revealed less than a half degree difference between the spawning channel and Gates Creek in the fall of 2014. The slight difference in temperature is likely not large enough to influence biotic factors such as percent spawn. Mean daily temperatures of both Gates Creek and the spawning channel remained below the threshold for thermal stress (Servizi and Jensen 1977, Eliason et al. 2011) over the spawning period.

4.2 Counter Operations and Visual Survey/ Video Validation

The mechanical counters for the spawning channel worked without issue in the fall of 2014. Video validation indicated the Gates Creek mechanical counter tends to overestimate the number of fish passing through the bypass into the creek. Video footage of the creek counter shows that the mechanical counter inflates the abundance count by 13%.

There is still come concern by the N’Quatqua technicians that the design of the mechanical counter into Gates Creek may alter fish behavior and potentially cause physiological stress. This was first noticed in 2013. Fish entering Gates Creek are dropped over a slide with only a few centimetres of running water. In video footage numerous fish were observed making several attempts to re-ascend the slide, but did not ascend to the point of reaching the gate and initiating additional up-counts. The inflation of the abundance estimate on the creek counter may result from fish moving the gate multiple times as they pass by. The audio associated with the video segments suggests fish are moving with vigor as they pass through the counter. Additionally, the low flow of water results in fish being partially exposed to air as they fight to re- ascend the slide.

Air exposure has been shown to cause stress and reduce migration success in Sockeye (Nguyen et al. 2013). Alone, the Gates Creek counter most likely would not impact the spawning success of fish; however, Sockeye returning to Gates Creek experience a number of stressors including a hydro-electric dam, escape from fishing gear entanglement, and navigating through Hell’s Gate. While the effects of cumulative and co-occurring stressors are still not fully understood, recent research suggests acute stressors such as fisheries capture or migration obstacles can have exacerbated effects on survival and spawning success when combined with thermal stress or disease in salmonids (Barton et al. 1986, Dietrich et al. 2014).

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 68

4.3 Migration Timing and Sockeye Abundance

4.3.1 Migration Timing

Sockeye arrived at Gates Creek 9 days later (September 21st) in 2014 than in 2013. However, the first fish were counted at Seton Dam near Lillooet on July 25th, 2014 (D. Braun, pers. comm.). The migration time for sockeye from the Seton Dam to Gates Creek is known to take approximately 10 days. The gap in time between the Seton Counter and the Gates Creek counter in 2014 was more than double the expected migration time. Starting in the last week of July water temperatures in the Fraser River began to climb above 18○C (the known threshold for physiological stress and delayed migration in salmonids (Lee et al. 2003). Although temperatures in the Fraser and Seton Rivers were not as high in 2014 as in 2013, the effect of temperature on migration success or timing may have been more pronounced in 2014 due to a lower abundance of returning sockeye.

4.3.2 Sockeye Abundance

Returns of adult Sockeye to Gates Creek were modest in the fall of 2014. For the period of time that DFO has kept records for Gates Creek Sockeye returns, the combined abundance estimate of 16,929 for both Gates Creek and the spawning channel is the second highest on record for this cycle (1962-2014). Returns for 2014 were 75% of the 2010 brood (22,431) and the second highest on record for this cycle (DFO unpublished data). This is the second consecutive year returns have exceeded cycle averages.

4.4 Fecundity and Egg Deposition

Reduced female abundance in 2014 resulted in a reduction in egg deposition from 2013 for both Gates Creek and the spawning channel. Mean fecundity was reduced by less than 1% in 2014 over 2013. The estimates of egg deposition for Gates Creek (17,612,710) and the spawning channel (10,782,538) for the 2014 brood were 71% and 40% lower than 2013.

4.5 Percent Spawn

Despite warm water temperatures (>18○C) in the Fraser and Seton Rivers during the migration window (Figure 21) (http://www.pac.dfo-mpo.gc.ca/science/habitat/frw-rfo/reports-

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 69 rapports/2014/2014-08-28/2014-08-28-eng.html), percent spawn was above the historical average for this cycle year of Gates Creek sockeye in 2014. Percent spawn in 2014 (85.4% overall) exceeded the long term average (73.6%) by 11.8%.

The relative abundance of 0% spawned fish to 100% spawned fish decreased as the spawning period progressed. The data for the 2012- 2014 visual surveys in the spawning channel indicates that the proportion of 0% fish decreased as the run progressed. A similar trend was seen in 2012 and 2013 when 70% and 45% of 0% spawners were observed in the first 10 days of the run. There does appear to be a trend for a higher proportion of early run females to be to be unsuccessful spawners which suggests best practice would be to devise a loading plan which ensures females from all parts of the run are loaded into each spawning habitat. Loading females from each portion of the run into each habitat would ensure that spawning occurs in both habitats which would help to provide insurance against stochastic events such as flooding or freezing during the fry incubation period.

Differences in percent spawn between Gates Creek and the spawning channel have varied among the years of this study. In both 2011 and 2012 there was ~30% difference between values of percent spawn for Gates Creek and the spawning channel; however in 2013 and 2014 the difference in the values decreased to ~2% and 1% respectively. Percent spawn in the spawning channel is a direct count of all female carcasses in the channel and in Gates Creek percent spawn is an estimate generated from weekly sub-samples of available carcasses. Because of the different methods used to generate percent spawn values for each habitat, direct comparison of percent spawn between Gates Creek and the spawning channel in not advisable. The decrease in the difference between percent spawn values in from 2011-2012 and 2013-2014, however, suggests a change in some factor influencing spawning success. Some possible factors which may have influenced a reduction in the difference of percent spawn figures are: 1) a change in DFO staff conducting visual surveys in 2013 from 2012 and 2011 or 2) the lower density of females or Sockeye in general loaded into the spawning channel in 2013 and 2014 or 3) the above mentioned lower discharge in 2013 and 2014 over 2011 and 2012 4) a change in an unmeasured environmental parameter such as temperature or available spawning substrate in either Gates Creek or the spawning channel.

Fish density may, in fact, partially drive percent spawn in the constrained environment of the spawning channel. Over the period of this study 2012 to 2014, percent spawn in the spawning channel has ranged from 52% (2012, highest density of fish) to 84% (2014, lowest density of fish). Over an 18 year period (1996 - 2014) in all but in all by 4 years (1996, 2000, 2008 and 2012) peak in fish abundance in the channel coincided with low percent spawn values when compared to years with few fish loaded into the channel (Figure 8). In general over the last 18 years, the years with the lowest percent spawn have occurred in years where more than 15,000 fish have been loaded into the channel (Figure 22) (DFO unpublished data). Experimentation with fewer fish in the channel in future years may help to improve percent spawn for this habitat.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 70

6.0 Summary and Recommendations

The objectives were largely met in the fall of 2014 for the adult component of this study:

1) A total abundance estimate (adjusted for counter efficiency) of 16,929 Sockeye for the Gates Creek watershed (10,300 Gates Creek, 76 Anderson Lake shore, and 6,553 into the spawning channel) was obtained. 2) Total female abundance was estimated to be 3,739 for the spawning channel and 6,160 for Gates Creek. 3) Percent spawn was estimated to be 85.4% for the watershed overall (85.9% for the spawning channel and 85.1% for Gates Creek). 4) Mean female fecundity was estimated to be 3,358 in the fall of 2014 5) Egg deposition for Gates Creek and the spawning channel were estimated to be 10,782,538 and 17,612,710, respectively.

In addition to the main objectives of the study, several important gains were made on the project in the fall of 2014. Measurements of counter efficiency were obtained and the N’Quatqua technicians were trained to do video validation. Furthermore, communication with DFO stock assessment personnel and the N’Quatqua technicians also continued to improve in the fall of 2014. Additionally, the N’Quatqua technicians have taken on more data collection as well as some data entry. Temperature data continued to be collected for both Gates Creek and the spawning channel and revealed a slight difference in temperature between the two habitats. Temperature monitoring should continue into the future as differences in temperature between the two habitats may relate to biotic factors such as spawning success.

All three mechanical counters worked well for the entire 2014 spawning season. Although the mechanical counter on Gates Creek has been found to slightly over estimate the number of fish in both 2013 and 2014, video validation is a useful tool for measuring and monitoring this tendency. One recommendation from 2013, again suggested for 2014, is that the design of the Gates Creek counter be modified to allow fish to swim through the counter with-out being exposed to air and stress. Currently the counter is designed such that the struggle of fish over the counter gates is audible.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 71

In summary, recommendations for future years of the study are to:

 Continue temperature monitoring.  Continue video validation of counter abundance estimates and to allow estimation of Pink Salmon abundance in 2015.  Development of pre-season equipment inspection and in-season maintenance schedule for counters to reduce risk of malfunction.  Capacity building with local N’Quatqua technicians to ensure daily maintenance of counters is completed and trouble-shooting abilities are developed.  Consider modifications to diversion fence in Gates Creek to minimize stranding of fish behind carcasses.  Re-evaluation of optimum spawner density in spawning channel to improve percent spawn values.

Improve inter-organizational communication to prevent future miscommunication on management strategies.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 72

TABLES

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 73

Spawning Site Name Mean average daily SD Min Max Area temperature (°C) Spawning 5th Leg 11.5 1.3 7.8 14.7 Channel Gates Channel Intake 11.3 1.6 7.6 14.0 Creek

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 74

Table 10. Video validation data for mechanical counter in Gates Creek, Fall 2014

Date Start End Start End Time Time Counter Video Count Count Time Validated Total Total

26-Aug-14 1,502 1,570 10:00:00 11:45:00 1:45:00 68 60 31-Aug-14 4,238 4,414 6:15:00 7:15:00 1:00:00 176 155 2-Sep-14 8,440 8,492 17:57:00 18:57:00 1:00:00 52 46 3-Sep-14 9,433 9,596 7:31:00 8:31:00 1:00:00 163 141 4-Sep-14 10,007 10,036 19:05:00 20:17:00 1:12:00 29 26 5-Sep-14 10,450 10,540 7:45:00 9:15:00 1:30:00 90 85 5-Sep-14 10,683 10,707 18:05:00 19:35:00 1:30:00 24 25 7-Sep-14 10,947 10,988 7:29:00 8:30:00 1:01:00 41 33 7-Sep-14 11,105 11,112 18:00:00 19:00:00 1:00:00 7 6 8-Sep-14 11,382 11,427 7:29:00 9:09:00 1:40:00 45 43 8-Sep-14 11,566 11,602 17:58:00 19:24:00 1:26:00 36 30 9-Sep-14 11,943 12,005 7:47:00 9:30:00 1:43:00 62 55 9-Sep-14 12,097 12,126 18:05:00 19:35:00 1:30:00 29 27 9-Sep-14 12,372 12,439 7:47:00 10:05:00 2:18:00 67 56 9-Sep-14 12,503 12,562 17:58:00 19:36:00 1:38:00 59 48 9-Sep-14 12,677 12,681 7:45:00 9:28:00 1:43:00 4 5 Total 22:56:00 952 841

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 75

Table 11. Summary of Sockeye abundance estimates for Gates Creek spawning channel collected via visual surveys 2011-2014

2011 2012 2013 2014 Males 6,631 5,882 6,072 2,333 Jacks 551 1,672 700 481 Females 8,302 9,514 6,510 3,739 Total 15,484 17,068 13,282 6,553

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 76

Table 12. Summary of adult Sockeye abundance estimates for Gates Creek 2011- 2014. Abundance estimates for Gates Creek are developed using a combination of data from visual surveys and the mechanical counter

2013 (after 2014 (after 2011 2012 adjustment) adjustment) Males 12,224 4,264 19,864 3,448 Jacks 950 976 1,687 692 Females 25,907 8,336 22,376 6,160 Total 39,081 13,576 43,927 10,300

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 77

Table 13 Summary of adult Sockeye abundance for Anderson Lake shore 2011-2014. Abundance estimates developed from visual survey data collected by DFO stock assessment

2011 2012 2013 2014 Males - - 55 25 Jacks - - 0 5 Females - - 62 46 Total - - 117 76

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 78

Table 14 Summary of adult Sockeye abundance for the Gates Creek Watershed from 2011-2014

2011 2012 2013 2014 Lake Shore - - 117 76 Gates Creek 39,081 13,576 43,927 10,300 Spawning Channel 15,484 17,068 13,282 6,553 Total 54,565 30,644 57,326 16,929

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 79

Table 15. Mean fecundity and post-orbital hypural (POH) length (cm) in Gates Creek spawning channel females Sockeye for 2011 - 2014

Mean POH in Range of fecundity Sample size Mean Fecundity (SD) cm (SD) values 2011 48 48.0 (2.2) 3,260 (571) 1,725-4,287

2012 40 46.5 (2.4) 3,119 (516) 1,469- 4,035 2013 46 49.0 (3.5) 3,378 (519) 2,351- 4,767 2014 60 49.1 (3.8) 3,358 (777) 685- 5,365

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 80

Table 16. Break down of female percent spawn in Gates Creek spawning channel from visual survey data 2011 - 2014

Percent spawn 2011 2012 2013 2014 Number (%) Number (%) Number (%) Number (%) 0% 3,647 (43.9%) 6,915 (72.6%) 1,131 (17.3%) 506 (14.0%) 50% 588 163 154 66 (7.1%) (1.7%) (2.4%) (1.8%) 100% 4,067 (49.0%) 2,436 (25.6%) 5,225 (81.4%) 3,033 (84.1%)

Total 8,302 9,514 6,510 3,605

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 81

Table 17. Estimated number of effective females, mean fecundity and egg deposition for the 2011-2014 brood years for Sockeye spawning in Gates Creek and Gates Creek spawning channel (estimates of effective females for Gates Creek were provided by DFO unpublished data)

Spawning Area Year Effective Females Mean Fecundity Egg Deposition 2011 4,574 3,260 14,911,240 Spawning 2012 2,518 3,119 7,853,642 Channel 2013 5,302 3,378 17,910,452 2014 3,066 3,358 10,782,538 2011 21,297 3,260 69,428,220 2012 4,311 3,119 13,446,009 Gates Creek 2013 17,702 3,378 59,798,345 2014 3,211 3,358 17,612,710

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 82

FIGURES

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 83

Figure 15. Overview of Seton-Anderson watershed in South Western British Columbia

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 84

Figure 16. Map of study area including Gates Creek and spawning channel near confluence with Anderson Lake. Also shown is the trap site for the juvenile monitoring component of the study.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 85

Figure 17. Schematic diagram of enumeration equipment at confluence of Gates Creek and Gates Creek spawning channel spawning channel

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 86

Figure 18. Average daily temperature in Gates Creek (dotted black line) and Gates Creek spawning channel (grey line) over the Sockeye spawning period from August 10th to October 6st, 2014

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 87

Figure 19. Scatter plot with fitted regression line (y= 159.3x+ -4459.7, R2 0.6127) of post-orbital hypural (POH) length (cm) and female fecundity (number of eggs) in Gates Creek Sockeye, Fall 2014

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 88

Figure 20. Distribution of female percent spawn over the fall 2014 Sockeye spawning period at Gates Creek spawning channel. Red, blue and black lines represent 0, 50 and 100% spawned females respectively

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 89

Figure 21 Temperature data for Fraser River at Mission for summer 2014 Sockeye migration window (http://www.pac.dfo-mpo.gc.ca/science/habitat/frw-rfo/reports- rapports/2014/2014-08-28/2014-08-28-eng.html)

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 90

Figure 22 Total annual adult Sockeye (*102) loaded into Gates Creek Spawning channel (black solid line) plotted with annual percent spawn (dotted grey line) 1996- 2014 (DFO unpublished data)

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 91

REFERENCES

Andrew F.J. and G.H. Green. 1958. Sockeye and Pink salmon investigations at the Seton Creek hydroelectric installation. Progress Report, International Pacific Salmon Commission, New Westminster, Canada. 78 p. Anonymous. 2000. Bridge-Coastal fish and wildlife restoration program strategic plan. 1, Bridge Coastal Restoration Program. 56 p. Anonymous. 2001. Gates Creek assessment project. Creekside Resource Inc., Mount Currie, Canada. 35 p. Anonymous. 2009. Rehabilitation of Sockeye spawning gravel in the Gates Creek spawning channel- year 2. Northern St'at'imc Fisheries, Lillooet, Canada. 56 p. Anonymous. 2012. Integrated biological status of Fraser River Sockeye Salmon (Oncorhynchus nerka) under the Wild Salmon Policy. Fisheries and Oceans Canada- Pacific Region. Science Advisory Report 2012/056. 13p. Barton, B.A., Schreck, C.B., and Sigismondi, L.A. 1986. Multiple acute disturbances evoke cumulative physiological stress responses in juvenile Chinook salmon. Transactions of the American Fisheries Society 115(2):245-251 Crossin, G.T., S.G. Hinch, S.J. Cooke, D.W. Welch, D.A. Patterson, S.R.M. Jones, A.G. Lotto, R.A. Leggatt, M.T. Mathes, J.M. Shrimpton, G.V. Van Der Kraak, and A.P. Farrell. 2008. Exposure to high temperature influences the behavior, physiology and survival of Sockeye salmon during spawning migrations. Canadian Journal of Zoology 86(2): 127- 140. Dietrich, J.P., Van Gaest, A.L., Strickland, S.A., and Arkoosh, M.R. 2014. The impact of temperatures stress and pesticide exposure on mortality and disease susceptibility of endangered Pacific salmon. Chemosphere 108: 353-359. Elaison, E.J., T.D. Clark, M.J. Hague, L.M. Hanson, Z.S. Gallagher, K.M. Jefferies, M.K. Gale, D.A. Patterson, S.G. Hinch, and A.P. Farrell. 2011. Differences in thermal tolerance among Sockeye salmon populations. Science 332: 209-112. Ferrari, M.R., J.R. Miller, and G.L. Russell. 2007. Modeling changes in summer temperature of the Fraser River during the next Century. Journal of Hydrology 342(3-4): 336-346. Fretwell, M.R. 1989. Homing behaviour of adult Sockeye salmon in response to a hydroelectric diversion of home water at Seton Creek. International Pacific Salmon Commission Bulletin 25: 38 p. Komori, V. 1997. Strategic fisheries overview for the Bridge/ Seton habitat management area. Fraser River Action Plan, Department of Fisheries and Oceans, Vancouver, Canada. 83 p.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 92

Lee, C.G., A.P. Farrell, A. Lotto, M.J. MacNutt, S.G. Hinch, and M.C. Healey. 2003. Effect of temperature on swimming performance and oxygen consumption in adult Sockeye (Oncorynchus nerka) and coho (O. kisutch) salmon stocks. Journal of Experiment Biology 206: 3239-3251. Lingard, S., C. Melville, and D.J.F. McCubbing. 2012. Gates Creek adult Sockeye escapement 2012. Prepared for the Lillooet Tribal Council and Department of Fisheries and Oceans, Canada. Instream Fisheries Research, Inc., North Vancouver, Canada. 20 p. Lingard, S., J. Ladell, and D.J.F. McCubbing. 2013. Gates Creek adult Sockeye abundance. Prepared for the Lillooet Tribal Council and Department of Fisheries and Oceans, Canada. Instream Fisheries Research, Inc., North Vancouver, Canada. 39 p. Morrison, J, M.C. Quick, and M.G.G. Foreman. 2002. Climate change in the Fraser River watershed: flow and temperature projections. Journal of Hydrology 263(1-4): 230-244. Nguyen, V.M., E.G. Martins, D. Robichaud, G.D. Raby, M.R. Donaldson, A.G. Lotto, G.D. Willmore, D.A. Patterson, A.P. Farrell, S.G. Hinch, and S.J. Cooke. 2014. Disentangling the roles of air exposure, gill net injury, and facilitated recovery on the post capture and release mortality and behavior of adult migratory Sockeye salmon (Oncorhynchus nerka). Freshwater. Physiological and Biochemical Zoology 87(1): 125-135. Patterson, D.A., J.S Macdonald, K.M Skibo, K.M. Barnes, I. Guthrie, and J. Hills. 2007. Reconstructing the summer thermal history for the lower Fraser River, 1941 to 2006, and implications for adult sockeye salmon (Oncorhynchus nerka) spawning migration. Servizi, J.A. and J.O.T. Jensen. 1977. Resistance of adult Sockeye salmon to acute thermal shock. International Pacific Salmon Fisheries Commission. Progress Report 34: 1-51. Talbot, G.B. 1950. A biological study of the effectiveness of the Hell's Gate fishways. Bulletin 3, International Pacific Salmon Commission, New Westminster, Canada. 80 p.

Gates Creek Juvenile and Adult Salmonid Outmigration Assessment March 2015 Final Report 2014 Page 93

This page has been left intentionally blank.