Stock Assessment Studies of Chilkat River Adult Salmon, 2018. Alaska Department of Fish and Game, Regional Operational Plan ROP.CF.1J.2018.06, Douglas

Regional Operational Plan CF.1J.2018.06

Operational Plan: Stock Assessment Studies of Chilkat River Adult Salmon, 2018

by Wyatt J. Rhea-Fournier Steven C. Heinl Sara E. Miller Julie A. Bednarski and Kyle R. Shedd

August 2018 Alaska Department of Fish and Game Divisions of Sport Fish and Commercial Fisheries

1 Symbols and Abbreviations The following symbols and abbreviations, and others approved for the Système International d'Unités (SI), are used without definition in the following reports by the Divisions of Sport Fish and of Commercial Fisheries: Fishery Manuscripts, Fishery Data Series Reports, Fishery Management Reports, and Special Publications. All others, including deviations from definitions listed below, are noted in the text at first mention, as well as in the titles or footnotes of tables, and in figure or figure captions. Weights and measures (metric) General Mathematics, statistics centimeter cm Alaska Administrative all standard mathematical deciliter dL Code AAC signs, symbols and gram g all commonly accepted abbreviations hectare ha abbreviations e.g., Mr., Mrs., alternate hypothesis HA kilogram kg AM, PM, etc. base of natural logarithm e kilometer km all commonly accepted catch per unit effort CPUE liter L professional titles e.g., Dr., Ph.D., coefficient of variation CV meter m R.N., etc. common test statistics (F, t, χ2, etc.) milliliter mL at @ confidence interval CI millimeter mm compass directions: correlation coefficient east E (multiple) R Weights and measures (English) north N correlation coefficient cubic feet per second ft3/s south S (simple) r foot ft west W covariance cov gallon gal copyright  degree (angular ) ° inch in corporate suffixes: degrees of freedom df mile mi Company Co. expected value E nautical mile nmi Corporation Corp. greater than > ounce oz Incorporated Inc. greater than or equal to ≥ pound lb Limited Ltd. harvest per unit effort HPUE quart qt District of Columbia D.C. less than < yard yd et alii (and others) et al. less than or equal to ≤ et cetera (and so forth) etc. logarithm (natural) ln Time and temperature exempli gratia logarithm (base 10) log day d (for example) e.g. logarithm (specify base) log2, etc. degrees Celsius °C Federal Information minute (angular) ' degrees Fahrenheit °F Code FIC not significant NS degrees kelvin K id est (that is) i.e. null hypothesis HO hour h latitude or longitude lat. or long. percent % minute min monetary symbols probability P second s (U.S.) $, ¢ probability of a type I error months (tables and (rejection of the null Physics and chemistry figures): first three hypothesis when true) α all atomic symbols letters Jan,...,Dec probability of a type II error alternating current AC registered trademark  (acceptance of the null ampere A trademark  hypothesis when false) β calorie cal United States second (angular) " direct current DC (adjective) U.S. standard deviation SD hertz Hz United States of standard error SE horsepower hp America (noun) USA variance hydrogen ion activity pH U.S.C. United States population Var (negative log of) Code sample var parts per million ppm U.S. state use two-letter parts per thousand ppt, abbreviations (e.g., AK, WA) ‰ volts V watts W

REGIONAL OPERATIONAL PLAN CF.1J.2018.06

OPERATIONAL PLAN: STOCK ASSESSMENT STUDIES OF CHILKAT RIVER ADULT SALMON, 2018

by Wyatt J. Rhea-Fournier Alaska Department of Fish and Game, Division of Commercial Fisheries, Haines

Steven C. Heinl Alaska Department of Fish and Game, Division of Commercial Fisheries, Ketchikan

Sara E. Miller Alaska Department of Fish and Game, Division of Commercial Fisheries, Juneau

Julie A. Bednarski Alaska Department of Fish and Game, Division of Commercial Fisheries, Douglas

and

Kyle R. Shedd Alaska Department of Fish and Game, Gene Conservation Laboratory, Anchorage

Alaska Department of Fish and Game Division of Commercial Fisheries August 2018

The Regional Operational Plan Series was established in 2012 to archive and provide public access to operational plans for fisheries projects of the Divisions of Commercial Fisheries and Sport Fish, as per joint-divisional Operational Planning Policy. Documents in this series are planning documents that may contain raw data, preliminary data analyses and results, and describe operational aspects of fisheries projects that may not actually be implemented. All documents in this series are subject to a technical review process and receive varying degrees of regional, divisional, and biometric approval, but do not generally receive editorial review. Results from the implementation of the operational plan described in this series may be subsequently finalized and published in a different department reporting series or in the formal literature. Please contact the author if you have any questions regarding the information provided in this plan. Regional Operational Plans are available on the Internet at: http://www.adfg.alaska.gov/sf/publications/

Wyatt J. Rhea-Fournier Alaska Department of Fish and Game, Division of Commercial Fisheries P.O. Box 330, Haines, AK 99827

Steven C. Heinl Alaska Department of Fish and Game, Division of Commercial Fisheries 2030 Sea Level Drive, Suite 205, Ketchikan, AK 99901

Sara E. Miller Alaska Department of Fish and Game, Division of Commercial Fisheries 1255 W. 8th Street, Juneau, AK 99801

Julie A. Bednarski Alaska Department of Fish and Game, Division of Commercial Fisheries, 802 Third Street, Douglas, Alaska 99824

and

Kyle R. Shedd Alaska Department of Fish and Game, Gene Conservation Laboratory 333 Raspberry Rd, Anchorage AK 99518

This document should be cited as follows: Rhea-Fournier, W. J., S. C. Heinl, S. E. Miller, J. A. Bednarski, and K. R. Shedd. 2018. Operational plan: stock assessment studies of Chilkat River adult salmon, 2018. Alaska Department of Fish and Game, Regional Operational Plan ROP.CF.1J.2018.06, Douglas.

The Alaska Department of Fish and Game (ADF&G) administers all programs and activities free from discrimination based on race, color, national origin, age, sex, religion, marital status, pregnancy, parenthood, or disability. The department administers all programs and activities in compliance with Title VI of the Civil Rights Act of 1964, Section 504 of the Rehabilitation Act of 1973, Title II of the Americans with Disabilities Act (ADA) of 1990, the Age Discrimination Act of 1975, and Title IX of the Education Amendments of 1972. If you believe you have been discriminated against in any program, activity, or facility please write: ADF&G ADA Coordinator, P.O. Box 115526, Juneau, AK 99811-5526 U.S. Fish and Wildlife Service, 4401 N. Fairfax Drive, MS 2042, Arlington, VA 22203 Office of Equal Opportunity, U.S. Department of the Interior, 1849 C Street NW MS 5230, Washington DC 20240 The department’s ADA Coordinator can be reached via phone at the following numbers: (VOICE) 907-465-6077, (Statewide Telecommunication Device for the Deaf) 1-800-478-3648, (Juneau TDD) 907-465-3646, or (FAX) 907-465-6078 For information on alternative formats and questions on this publication, please contact: ADF&G, Division of Sport Fish, Research and Technical Services,25 333 Raspberry Rd, Anchorage AK 99518 (907) 267-2375

SIGNATURE PAGE

Operational Plan: Stock assessment studies of Chilkat Project Title: River adult salmon, 2018

Project leader(s): Wyatt Rhea-Fournier

Division, Region, and Area Commercial Fisheries, Region 1, Haines

Project Nomenclature: FM-138 GF Chilkat Fishweel; FM-163 GF Didson

Period Covered 2018

Field Dates: 5 June–15 October

Plan Type: Category II

Approval

Title Name Signature Date

Project leader Wyatt J. Rhea-Fournier Biometrician Sara E. Miller Geneticist Kyle R. Shedd Research Coordinator Steven C. Heinl

TABLE OF CONTENTS Page LIST OF TABLES...... iii LIST OF FIGURES ...... iii LIST OF APPENDICES ...... iv PURPOSE...... 1 BACKGROUND ...... 1 STUDY SITE ...... 3 OBJECTIVES ...... 3 Primary objectives ...... 3 Commercial Fisheries Harvest: ...... 3 Fish Wheels ...... 4 Chilkat Lake Weir ...... 4 Secondary objectives ...... 4 Commercial Fisheries Harvest ...... 4 Fish Wheels ...... 4 Chilkat Lake Weir ...... 4 Chilkat Lake Limnology ...... 5 METHODS ...... 5 Age, Sex, and Length Composition ...... 5 COMMERCIAL FISHERIES HARVEST ...... 6 Drift Gillnet Fleet Observations ...... 6 CPUE and Harvest Estimates ...... 6 Commercial Sockeye Salmon Harvest Estimates ...... 7 Fishery Sampling ...... 7 Laboratory Analysis ...... 8 Statistical Analysis ...... 8 Data Reduction ...... 9 CHILKAT RIVER FISH WHEELS ...... 9 Fish Wheel Set-Up and Operation ...... 10 Data Collection ...... 10 Sockeye Salmon...... 11 Chinook Salmon ...... 11 Coho Salmon ...... 11 Chum Salmon ...... 11 Pink Salmon ...... 12 Data Reduction ...... 12 Data Archiving...... 12 CHILKAT LAKE ESCAPEMENT ...... 12 Chilkat Lake Weir ...... 13 DIDSON Installation and Settings ...... 13 DIDSON Operation ...... 13 Observer Training ...... 14 Age, Sex, Length Sampling ...... 16

TABLE OF CONTENTS (Continued) Page Species Apportionment ...... 16 Data Reduction ...... 17 Data Archiving...... 17 LIMNOLOGICAL ASSESSMENT ...... 17 Sampling Methods and Data Analysis ...... 17 Light and Temperature Sampling ...... 17 Zooplankton Sampling ...... 18 Data Reduction ...... 18 SCHEDULE AND DELIVERABLES ...... 18 Field Operations ...... 18 Reporting ...... 18 RESPONSIBILITIES ...... 18 REFERENCES CITED ...... 20 TABLES AND FIGURES ...... 23 APPENDIX ...... 29

LIST OF TABLES Table Page 1. Weekly sampling goals for sockeye salmon at the Chilkat River fish wheels and Chilkat Lake weir...... 24 2. Fish wheel catch, mark–recapture estimates, and expanded escapement estimates of Chilkat River fall chum salmon, 1990 and 2002–2017...... 25 3. Weekly sampling goals for chum salmon at the Chilkat fish wheels...... 26

LIST OF FIGURES Figure Page 1. The Chilkat and Chilkoot River watersheds and the District 15 commercial fishing subdistricts in Lynn Canal, including the Boat Harbor terminal hatchery area (Subdistrict 115-11)...... 27 2. Chilkat River drainage, with fish wheel locations, mainstem river sockeye salmon sampling sites, and Chilkat Lake weir location...... 28

iii

LIST OF APPENDICES Appendix Page A. ADF&G statistical weeks 20–45, 2018. Statistical weeks begin on Sunday at 12:01 a.m. and end the following Saturday at midnight and are numbered sequentially starting from the first week of the calendar year...... 30 B. ADF&G collection code, location, reporting group, and the number of sockeye salmon used in the genetic baseline for mixed stock analysis in District 15 commercial drift gillnet fishery...... 31 C. Escapement sampling data analysis...... 38 D1. Example of ADF&G Adult Salmon Age, Sex, Length Form Version 3.0 (ASL) filled out for sockeye salmon samples at Chilkat River fish wheels...... 39 D2. Definition of user codes on ADF&G Adult Salmon Age , Sex, and Length (ASL) Form Version 3.0, to be used for the fish wheel project...... 40 D3. Preferred scale sampling area on an adult salmon and proper orientation of scale samples on gum card. .... 41 D4. District 15 commercial drift gillnet fishery fleet observations...... 42 D5. Chilkat River Fish Wheel Sampling Period Form...... 43 D6. Chilkat River Fish Wheel Log Book...... 44 D7. Chilkat River Fish Wheel Summary Form...... 45 D8. Chilkat River Fish Wheel Office Form...... 46 D9. Chilkat Lake Hourly Fish Counts Form...... 47 D10. Chilkat Lake Daily Counts Spreadsheet...... 48 D11. Chilkat Lake Seine Recovery Form...... 49 D12. Daily DIDSON Counts Form...... 50 D13. Chilkat Lake Weir DIDSON/Seine Apportionment Form...... 51 D14. Limnology Sampling Form...... 52

PURPOSE The Chilkat Lake sockeye salmon (Oncorhynchus nerka) run, which spawns near Haines, is one of the largest in Southeast Alaska and contributes substantially to harvests in the District 15 commercial drift gillnet fishery in Lynn Canal. This operational plan outlines objectives, methods, and timelines for conducting sockeye salmon stock assessment designed to (1) estimate annual escapement and harvest, (2) provide information for inseason fishery management, and (3) reconstruct runs and assess stock status. The Chilkat Lake run is managed for a biological escapement goal of 70,000–150,000 fish, which is enumerated with a Dual-frequency Identification Sonar (DIDSON) system operated in conjunction with a standard picket weir located just downstream of the lake outlet. Genetic mixed stock analysis of weekly sockeye salmon harvests in the District 15 commercial drift gillnet fishery provides stock composition estimates that guide inseason management of the fishery. Biological sampling, along with escapement enumeration and stock-specific harvest data, allows for total run reconstruction required for escapement goal review. Two fish wheels operated in the lower Chilkat River provide timely indices of inriver abundance of combined Chilkat Lake and Chilkat River mainstem sockeye salmon populations, serve as sampling platforms for both Chinook (O. tshawytscha), and coho (O. kisutch) salmon, and provide the primary source of information on chum salmon (O. keta) abundance in the Chilkat River drainage. (Stock assessment of Chinook and coho salmon are covered in more detail in separate operational plans.) This project also supports the collection of basic limnology information at Chilkat Lake. BACKGROUND The Chilkoot and Chilkat sockeye salmon (Oncorhynchus nerka) runs in northern Southeast Alaska, near the town of Haines, are two of the largest runs in Southeast Alaska (Figure 1). Between 1900 and 1920, the annual commercial harvest of sockeye salmon in northern Southeast Alaska averaged 1.5 million fish, the majority of which were believed to be Chilkat and Chilkoot sockeye salmon (Rich and Ball 1933). Historically, Chilkat sockeye salmon were harvested in the large fish trap and purse seine fisheries in Icy and northern Chatham straits as well as in terminal drift gillnet areas of Lynn Canal. Fish traps were eliminated after Alaska statehood in 1959 and Lynn Canal was developed into a designated drift gillnet fishing area (District 15), where most of the commercial harvest of Chilkat sockeye salmon takes place (Figure 1). The annual harvest of sockeye salmon in the District 15 commercial drift gillnet fishery averaged 191,500 fish from 1984 to 2015, of which an average 78,000 fish originated from Chilkat Lake, 92,000 fish originated from Chilkoot Lake, and the remainder were of mixed stock origin (Bednarski et al. 2016). A smaller portion of the Chilkat run is harvested in the commercial purse seine fisheries that target pink salmon (O. gorbuscha) in Icy and northern Chatham straits (Ingledue 1989; Gilk-Baumer et al. 2015). Annual contributions to those fisheries are not known and likely vary annually depending on fishing effort and the strength of pink salmon runs. From 1976 to 2016, scale pattern analysis was used to determine the contribution of Chilkat Lake, Chilkoot Lake, and other stocks to the District 15 commercial drift gillnet fishery harvest (McPherson et al. 1992). From 2013 to 2016, scale pattern analysis and genetic stock identification were conducted concurrently to compare estimates using the two methods and to provide a transition period to assess the viability of switching exclusively to genetic stock identification. This information will be summarized in future reports (Serena Rogers Olive, ADF&G Fisheries Geneticist, personal communication). Beginning in 2017, stock separation estimates were determined by genetic stock identification. Chilkat Lake sockeye salmon escapements have been estimated through weir counts (1967–1993), weir counts with concurrent mark–recapture estimates (1994 and 1995, 1999–2007), mark– recapture estimates only (1996–1998), and Dual-frequency Identification Sonar (DIDSON) counts with mark–recapture estimates (2008–2016) (Eggers et al. 2010; Sogge and Bachman 2014; Bednarski et al. in press). Biological data have been collected annually at the lake to estimate the

age, size, and sex composition of the escapement. Two-event mark–recapture studies in conjunction with operation of fish wheels in the lower Chilkat River were initiated in 1994 because weir counts at Chilkat Lake were thought to underestimate escapement (Kelley and Bachman 2000; Eggers et al. 2010). Periodic flooding of the silty Tsirku River into Chilkat Lake (Bergander et al. 1988) required opening the weir, sometimes for extended periods, and increased boat traffic in and out of the lake required frequent lowering of a boat gate in the center of the weir through which fish could pass uncounted (Kelley and Bachman 2000). Sockeye salmon were marked at the fish wheels and sampled for marks at the Chilkat Lake weir and various Chilkat River mainstem and tributary spawning locations; drainagewide mark–recapture estimates were then generated and divided into Chilkat Lake and Chilkat River mainstem estimates (Kelley and Bachman 2000; Bachman and McGregor 2001; Bachman 2005, 2010). In 2008, a DIDSON was installed at the Chilkat Lake weir to improve counts (Eggers et al. 2010). Finally, concerns regarding mark– recapture as a reliable measure of abundance lead to suspension of mark–recapture studies after 2016 and recommendations to improve DIDSON operations to estimate escapement into Chilkat Lake (Bednarski et al. in press). The Chilkat Lake sockeye salmon run has been managed for at least five different escapement goals since 1976. Informal goals of 60,000–70,000 fish (1976–1980) and 70,000–90,000 fish (1981–1989) (Bergander et al. 1988) were replaced in 1990 with a biological escapement goal (BEG) of 52,000–106,000 sockeye salmon based on extensive spawner-recruit analysis by McPherson (1990). Efforts to update the escapement goal were hindered by lake stocking in the 1990s and concerns regarding accuracy of weir counts (Geiger et al. 2005). Geiger et al. (2005) converted the weir-based goal to mark–recapture units and the goal was revised to a sustainable escapement goal (SEG) of 80,000–200,000 sockeye salmon from 2006 to 2008. In 2009, the escapement goal was revised again to a BEG of 70,000–150,000 sockeye salmon based on a spawner-recruit analysis with weir counts converted to mark–recapture units (Eggers et al. 2008, 2010). More recently, escapements were modelled using historical mark–recapture estimates and weir counts as indices of escapement, while DIDSON counts (2008–2016) were considered ‘true’ counts of escapement; that review did not result in any changes to the current BEG of 70,000– 150,000 sockeye salmon counted at the Chilkat Lake weir site with DIDSON sonar (Miller and Heinl in press; Heinl et al. 2017). In addition to the sockeye salmon stocks, the Chilkat River drainage supports the largest fall chum salmon (O. keta) run in the region (Halupka et al. 2000). Chilkat River fall chum salmon harvests occur primarily after statistical week 33 (mid-August) in the District 15 commercial drift gillnet fishery. The current escapement goal for Chilkat River fall chum salmon is 75,000–250,000 chum salmon, based on stock recruit analysis and the range of escapements estimated to provide 70% of maximum sustained yield (Piston and Heinl 2014). Information regarding total escapements is based on conversion of fish wheel counts to five years of mark–recapture population estimates (Eggers and Heinl 2008); however, the escapement goal is considered a SEG rather than a BEG, due to uncertainty in estimated escapements (Piston and Heinl 2014). The escapement goal range converts to an equivalent fish wheel index catch of 1,160–3,875 chum salmon (Piston and Heinl 2014). This operational plan outlines current objectives, methods, and timelines for Alaska Department of Fish and Game (ADF&G) stock assessment studies of the Chilkat Lake sockeye salmon run. These studies are designed to (1) estimate annual escapement and harvest, (2) provide information for inseason fishery management, and (3) reconstruct runs and assess stock status. Information

provided by this project, in conjunction with sockeye salmon stock assessment projects on the adjacent Chilkoot River (Sogge 2016), is used inseason to manage the District 15 commercial drift gillnet fishery to ensure escapement goals are met, and to maximize and sustain the harvest of sockeye salmon from the two watersheds. Escapement and stock-specific harvest data, along with biological data on age at return, are essential for reconstruction of brood year returns for use in future escapement goal evaluation. Bednarski et al. (in press) provided a comprehensive review of historical sockeye salmon stock assessment studies in the Chilkat River drainage and recommendations for future improvements that are reflected in this plan. The Chilkat River fish wheel project is also conducted in conjunction with stock assessment studies of Chinook (O. tshawytscha), and coho (O. kisutch) salmon, which are covered in more detail in separate operational plans (Elliot and Power 2017a; Elliot and Power 2017b). STUDY SITE Chilkat Lake (ADF&G Anadromous Waters Catalogue No. 115-32-10250-2067-3001-0010; 59.32577° N, 135.89436° W) is located approximately 27 river miles upstream from the city of Haines, Alaska (Figure 1 and 2). It is a relatively large clear lake with a surface area of 9.8 × 106 m2 (2,432 acres), mean depth of 32.5 m, a maximum depth of 57 m, and a volume of 319 × 106 m3. The lake drains through Clear Creek, a 0.5 km long channel, which is also the location of the weir, and into the Chilkat River by way of the Tsirku River. Resident fish include sockeye and coho salmon, Dolly Varden (Salvelinus malma), cutthroat trout (Salmo clarki), threespine stickleback (Gasterosteus aculeatus), sculpin (Cottus sp.) and whitefish (Prosopium cylindraceum) (Johnson and Daigneault 2013). Very small numbers of adult pink and chum salmon have been counted through the Chilkat Lake weir, but the spawning location of these fish is not known. Chilkat Lake is a remote lake with moderate to heavy boat traffic. Access to the numerous private cabins on the lake (50 to 100 cabins) is limited to jet boat and floatplane. The Chilkat River (ADF&G Anadromous Waters Catalogue No. 115-32-10250) drains a large watershed stretching from British Columbia, Canada to the northern end of Lynn Canal, near Haines, Alaska (Figure 2). It is characterized by rugged, highly dissected mountains with steep- gradient streams, and braided rivers that flow through glaciated valleys. The watershed encompasses approximately 1,600 km2, and the main river and tributaries comprise approximately 350 km of river channels. Principle tributaries include the Tahkin, Tsirku, Klehini, Kelsall, and Tahini rivers. Chilkat River discharge rates range from 80 to 20,400 ft3/s (Bugliosi 1988). The river supports large runs of sockeye, coho, chum, Chinook, and pink salmon. The Chilkat River receives input from several glaciers, and heavy silt loads in the main river impairs visual salmon stock assessment methods. OBJECTIVES PRIMARY OBJECTIVES The primary project objectives, broken out by the major stock assessment components, are to: Commercial Fisheries Harvest: 1. Estimate the weekly commercial harvest of Chilkat Lake and Chilkat mainstem sockeye salmon in the District 15 commercial drift gillnet fishery for each of first 10 statistical weeks of the season, such that the estimates are within 7% of the true value 90% of the time.

2. Estimate the seasonal age distribution of the commercial harvest of Chilkat Lake sockeye salmon in the District 15 commercial drift gillnet fishery for major age classes, ages 1.2, 1.3, 2.2, 2.3 and other. Fish Wheels 1. Enumerate salmon by species captured at the fish wheels. 2. Estimate the weekly age, sex, and length composition of the sockeye salmon catch, such that the estimated proportion of each major age class is within 10% of the true value with at least 90% probability. 3. Estimate the seasonal age, sex, and length composition of the sockeye salmon catch, such that the estimated proportion of each major age class is within 5% of the true value with at least 95% probability. 4. Estimate the seasonal age, sex, and length composition of the total chum salmon catch, such that the estimated proportion of each major age class is within 5% of the true value with at least 95% probability. 5. Estimate the Chilkat River fall chum salmon escapement based on an expanded fish wheel count. Chilkat Lake Weir 1. Enumerate the annual escapement of sockeye salmon into Chilkat Lake from 20 June to 10 October. 2. Estimate the seasonal age, sex, and length composition of the Chilkat Lake sockeye salmon escapement, such that the estimated proportion of each major age class is within 5% of the true value with at least 95% probability. SECONDARY OBJECTIVES Commercial Fisheries Harvest 1. Estimate the CPUE and total harvest of each species of salmon in the District 15 commercial drift gillnet fishery for the first 24 hours of each weekly opening. Fish Wheels 1. Estimate the age, sex, and length composition of the Chinook and coho salmon catch, such that the estimated proportion of each major age class is within 5% of the true value with at least 95% probability. 2. Examine all adult Chinook and coho salmon captured at the fish wheels for adipose fin clips and coded wire tags. 3. Mark all adult Chinook salmon in conjunction with the Chilkat River Chinook Salmon Escapement Study conducted by ADF&G Division of Sport Fish. 4. Sample 200 pink salmon for baseline genetic tissue samples. Chilkat Lake Weir 1. Maintain standardized beach seine sampling schedule during September–October to ensure accurate weekly species apportionment between coho and sockeye salmon. 2. Perform periodic, systematic observer comparison of DIDSON counts to increase precision of the DIDSON count. Inseason disagreement between observers of more than 5% should be flagged for a detailed review.

Chilkat Lake Limnology 1. Measure water column temperature, record light penetration profiles, and estimate zooplankton species composition, size, density, and biomass in Chilkat Lake on a monthly basis from June to October. METHODS Chilkat River salmon stock assessment studies include four sections: District 15 commercial drift gillnet harvest, Chilkat River fish wheels, Chilkat Lake weir/DIDSON, and Chilkat Lake limnology sampling. These sections are covered below. AGE, SEX, AND LENGTH COMPOSITION Scale samples from select salmon species will be collected from the commercial fishery harvest, the Chilkat fish wheels, and at the Chilkat Lake weir. Sample size requirements will differ depending on species and are outlined in the appropriate sections below. The length of each sampled fish will be measured from mid eye to fork of tail (MEF) to the nearest 5 mm. Sex will be determined from examination of external dimorphic sexual maturation characteristics, such as kype development, belly shape, and trunk depth. Sex and length data will be recorded on standardized Age Sex Length (ASL) optical scan data sheets (Appendices D1 and D2). Additional ASL forms for coho salmon that do not have associated scale samples will be numbered differently. To eliminate gaps in the actual scale card and acetate series and since the ADF&G database does not allow for duplicate ASL numbers within a project code, numbering sequence series for samples taken with scales will begin with 001 and the numbering sequence for coho salmon length-only samples will begin with 201. Scale samples will be taken from the “preferred area”, two scale rows above the lateral line on the left side of the fish on a diagonal downward from the posterior insertion of the dorsal fin to the anterior insertion of the anal fin (INPFC 1963). All regenerated scales will be discarded. It is critical that all scale cards are clean and dry and all scales are properly oriented on the card. Scales need to be carefully cleaned of dirt, slime, and skin, then moistened and mounted on the gum card with the ridged side with grooves (rough outer side of the scale) facing out, and the anterior end (the end of the scale pointing toward the salmon’s head when plucked) is pointed toward the top of the scale card (Appendix D3). Scales will then be pressed down with a finger or a pencil so that they stick to the scale card. Scales will be collected from each fish and placed on gum cards at the rate of one fish per column over spaces 1, 11, 21, 31, and below space 31 on the gum card (depending on the required number of scales per fish per salmon species). Room will be left at the top middle portion of the card to accommodate the label. It is important to keep the scale cards as dry as possible to prevent the gum from running and obscuring the scale ridges. The gum card will be filled out completely including the names of the samplers, species, card number, locality, and statistical area/stream code. Scale samples will be analyzed at the Region 1 Scale Aging Laboratory in Douglas, Alaska. Scale impressions will be made in cellulose acetate and prepared for analysis as described by Clutter and Whitesel (1956). Scales will be examined under moderate (70×) magnification to determine age. Age classes will be designated by the European aging system where freshwater and saltwater years are separated by a period (e.g., 1.3 denotes a fish with one freshwater and three ocean years) (Koo 1962).

COMMERCIAL FISHERIES HARVEST District 15 commercial drift gillnet fishery season typically begins at 12:00 pm noon on the third Sunday of June. Openings are then conducted weekly starting at 12:00 pm noon on Sunday. Each week typically begins with a 48-hour opening, with the possibility of an extension depending on fishery performance. The District 15 commercial drift gillnet fishery historically targeted sockeye salmon from late June through September and fall run chum and coho salmon from mid-August to mid-October. In recent decades, fishing effort has shifted to Section 15-C to harvest substantial hatchery summer chum salmon returns to Douglas Island Pink and Chum, Inc., release sites at Boat Harbor and Amalga Harbor terminal harvest areas (THA). Section 15-B (Berners Bay) has only been opened once in the last 10 years to target coho salmon. Most of the salmon harvested in the fishery are sold to large commercial seafood processors, which deploy large vessels called tenders to purchase salmon on the fishing grounds. The harvest is then delivered to processing facilities located throughout Southeast Alaska. DRIFT GILLNET FLEET OBSERVATIONS Information gathered on the fishing grounds from the District 15 commercial drift gillnet fleet is used to estimate the number of fishing vessels, catch per unit effort (CPUE), and total harvest of all salmon species to be reported in weekly news releases. An average 125 boats fish per opening during the first six statistical weeks of the season, and an average 65 boats fish per opening during the remainder of the season. (See Appendix A for 2018 statistical week dates.) The sampling goal is to gather catch data from at least 20% of the drift gillnet fleet that participates in each opening. Commercial Fisheries staff from the Haines office will conduct a survey of the District 15 commercial drift gillnet fleet each Monday during the season to interview fishermen and tender operators to collect information on harvest and effort. Surveys will be conducted by two technicians from a 26-foot Boston Whaler. The survey will start from Portage Cove harbor in Haines and cover the entire fishing grounds south to the southern border of Section 15C. A total boat count will be made, and individual drift gillnet vessel captains will be interviewed in each of the open subdistricts. The number of interviews conducted in each subdistrict will be proportional to the amount of observed effort, with a goal of interviewing a total of 10 to 15 individual boats in the entire district. Tenders encountered while interviewing fishing vessels will be boarded, and E- ticket tender logs will be printed and retained with a goal of obtaining 3 to 5 tender logs for the entire district. After 24 hours of fishing, each tender log will typically include the catch of 5 to 15 boats. Information from tender logs and the individual vessel interviews will provide enough data to meet our goal of sampling 20% of the fleet. CPUE and Harvest Estimates CPUE is defined as the number of fish caught per boat per day (24 hours). The CPUE will be determined by averaging catch data from individual fishermen interviews and tender logs. The CPUE will then be multiplied by the total number of boats to estimate the total harvest during the first day of the opening. The harvest estimate and current escapement counts at the fish weirs will be compared to historical trends to determine if an extension is warranted for the opening. The CPUE and estimated total harvest will be reported in the weekly news release distributed each Thursday afternoon.

COMMERCIAL SOCKEYE SALMON HARVEST ESTIMATES Inseason stock composition of the sockeye salmon harvest of in the District 15 commercial drift gillnet fishery will be estimated through genetic stock identification. Sockeye salmon will be identified in seven reporting groups: Chilkat Lake, Chilkat mainstem, Chilkoot Lake, Juneau Mainland, Snettisham, Taku River/Stikine mainstem, and Other (Appendix B). Laboratory analysis, including quality control, will be performed by the ADF&G Gene Conservation Laboratory following methods outlined in Dann et al. (2012). Stock composition will be estimated inseason for each statistical week using a Bayesian mixed stock analysis (MSA) approach as implemented in the R package rubias (Moran and Anderson in prep.), which will compare fishery samples against the genetic baseline described in Rogers Olive et al. (in press). Postseason, samples will be reanalyzed with age composition data from the harvest using an MSA model that incorporates ages from matched scale samples to provide age-specific stock composition estimates. Fishery Sampling Matched sockeye salmon scale and genetic tissue samples will be collected from District 15 commercial drift gillnet fishery landings by ADF&G port sampling personnel at fish processing facilities in Excursion Inlet and Juneau (Buettner et al. 2017). Sampling will be stratified by week and sampling effort will span the first 10 weeks of the fishery, as approximately 93% of the sockeye salmon harvest occurs during that period (2008–2017 average). The target sample size for each statistical week is set at a minimum of 200 and a maximum of 300 paired tissues and scales. According to sample theory, under the worst-case scenario (stocks contributing equal proportions) a sample of this size should provide weekly estimates of relative proportions within 7% of the true value 90% of the time (Thompson 1987). Sampling protocols will ensure that weekly samples will be as representative of harvests as possible to account for fluctuations in harvest and effort over the course of a weekly fishery. Deliveries with harvests mixed from more than one gear type or fishing district will not be not sampled, no more than 40 samples will be collected from a single vessel delivery, no more than 200 samples will be collected from a single tender delivery, samples will be collected without regard to size or sex of fish, and, whenever possible, samples will be systematically collected from the entire hold as it is offloaded to ensure they are representative of the entire delivery. Beginning in the 2018 season, sockeye salmon harvested in the District 15 commercial drift gillnet fishery will be sampled regardless of the harvest type. In the past, sockeye salmon harvested in the Boat Harbor THA (Subdistrict 115-11) were not sampled, including sockeye salmon on tenders with fish mixed from traditional and terminal harvest fisheries. The Boat Harbor THA is designated to harvest hatchery chum salmon released inside Boat Harbor; however, the THA encompasses a portion of lower Lynn Canal (Figure 1) through which mixed stocks of sockeye salmon must migrate, and sockeye salmon are harvested incidentally in the fishery. Over the 10 years 2008–2017, an average 21% (range: 12–36%) of sockeye salmon harvested in lower Lynn Canal (Subdistricts 115-10 and 115-11) were harvested in the Boat Harbor THA. Beginning in 2018, all sockeye salmon samples will be identified as harvest code 11 (traditional fishery). Future stock composition analyses will need to include the entire harvest in Lynn Canal, harvest codes 11 and 12 (terminal hatchery harvest) combined. Length, sex, and scales will be collected from sockeye salmon caught in the fishery as outlined in Age, Sex, and Length Composition section. A 2.5 cm piece of the pelvic fin will be removed from

each sampled fish and placed on a Whatman filter paper card for dry preservation. Metadata for each sample, including matched age information, will be recorded. Tissue samples will be shipped on a weekly basis to the Region 1 Scale Aging Laboratory in Douglas, along with matching scale samples and associated data for inventory. Tissue samples will then be shipped to the ADF&G Gene Conservation Laboratory in Anchorage for analysis. Scale samples will be inventoried and prepared for postseason analysis Laboratory Analysis Genomic DNA will be extracted from tissue samples using a NucleoSpin® 96 Tissue Kit by Macherey-Nagel (Düren, Germany). A multiplexed preamplification PCR of 48 screened SNP markers will be used to increase the concentration of template DNA. Samples will be genotyped for 48 screened single nucleotide polymorphism (SNP) markers using two sets of Fluidigm® 192.24 Dynamic ArrayTM Integrated Fluidic Circuits (IFCs), which systematically combine up to 24 assays and 192 samples into 4,608 parallel reactions (https://www.fluidigm.com). The Dynamic Arrays will be read on a Fluidigm® EP1™ System or Biomark™ System after amplification and scored using Fluidigm® SNP Genotyping Analysis software. If necessary, SNPs may be rescreened on a QuantStudio™ 12K Flex Real-Time PCR System (Life Technologies) as a backup method for assaying genotypes. Genotypes will be imported and archived in the Gene Conservation Laboratory Oracle database, LOKI. A quality control analysis (QC) will be conducted postseason to identify laboratory errors and to measure the background discrepancy rate of the genotyping process. The QC analyses will be performed by staff not involved in the original genotyping, and the methods are described in detail in Dann et al. (2012). Briefly, the method will consist of re-extracting 8% of project fish and genotyping them for the same SNPs assayed in the original genotyping process. Discrepancy rates will be calculated as the number of conflicting genotypes, divided by the total number of genotypes compared. These rates will describe the difference between original project data and QC data for all SNPs and are capable of identifying extraction, assay plate, and genotyping errors. Assuming that discrepancies among analyses are due equally to errors during the original genotyping and during quality control, error rates in the original genotyping will be estimated as half the rate of discrepancies. Statistical Analysis Genotypes in the LOKI database will be imported into the statistical program R for analysis (R Core Team 2017). Prior to statistical analysis, three statistical quality control analyses will be performed to ensure high-quality data. First, we will identify and remove individuals missing >20% genotypic data (80% rule; Dann et al. 2012); second, we will identify and remove duplicate individuals; and third, we will identify and remove nonsockeye salmon. Stock composition for each stratum will be estimated using the R package rubias (Moran and Anderson in prep). A single Markov Chain Monte Carlo (MCMC) chain with starting values equal among all populations will form the posterior distribution that describes the stock composition of each stratum. Summary statistics will be tabulated from these distributions to describe stock compositions. Stock composition estimates of commercial harvest will be applied to observed harvest (obtained from fish ticket data) to quantify stock-specific harvests within each week. Postseason, age-specific stock composition for all major contributing age classes (>5%) will be estimated seasonally through a mark- and age-enhanced genetic mixed-stock analysis (MAGMA)

model, which is an extension of the Pella-Masuda GSI model (Pella and Masuda 2001) that incorporates paired scale-age data. Total season estimates will be provided, by age group, using an “age-enhanced GSI” method. This method requires two sets of parameters: 1) a vector of stock compositions summing to one weighted by harvest per stratum; and 2) a matrix of age composition, with a row for each stock summing to one and a column for each age class. This information will be “completed” iteratively by stochastically assigning each fish to a population, then estimating the stock proportions based on summaries of assignment from each iteration. In this process, all available information will be used to assign individuals to stock of origin based on age and genotype. To initialize the algorithm, all wild fish are given a stock assignment stochastically. The initialized algorithm will then proceed in the following steps: 1) Summarize all age data by assigned stock; 2) Estimate the stock proportions and age composition from previous summaries (accounting for sampling error); 3) Stochastically assign each fish with genotypes and ages to a stock of origin based on the product of its genotypic frequency, age frequency, and stock proportion for each population; 4) Stochastically assign each fish without genotypes (only those samples that were aged) to a stock of origin based on the product of its age frequency and stock proportion for each population; and 5) Repeat steps 1–4 while updating and recording the estimates of the stock proportions and age compositions with each iteration. This algorithm will be run for 40,000 repetitions, and the first 20,000 repetitions will be discarded to eliminate the effect of the initial state. The point estimates and credibility intervals for stock- specific age compositions will be summary statistics of the output. DATA REDUCTION Information collected on the fishing grounds by Haines staff will be reported on field sheets that separate each observation by subdistrict (Appendix D4). These data will be entered into an Excel spreadsheet to track opening estimates through the drift gillnet season. Estimated CPUE, total harvest, and effort will be reported by the Haines manager in weekly news releases. The ADF&G Gene Conservation Laboratory will provide inseason, weekly stock composition estimates to the Haines Office, including a running summary of the current season and stock composition estimates from the 2015–2017 seasons for comparison. There will be a total of 10 inseason reports spanning statistical weeks 25–34. Postseason, age-specific stock composition for all major contributing age classes (>5%) will be estimated seasonally. CHILKAT RIVER FISH WHEELS Fish wheels have been operated in the lower Chilkat River for stock assessment and management purposes as early as 1977 and annually since 1994. The primary purpose of the project is to monitor relative abundance of sockeye salmon as they enter the drainage. In addition, the fish wheel catch is used as a relative index to estimate the spawning escapement of Chilkat River fall chum salmon. Operation of the fish wheels is coordinated with Division of Sport Fish stock assessment studies of Chinook and coho salmon to estimate escapement of those species in the Chilkat River drainage.

Pink salmon tissue samples will be collected at the fish wheels to assist the department in constructing a genetic baseline. FISH WHEEL SET-UP AND OPERATION Two, three-basket-configured fish wheels will be operated in the lower Chilkat River, between mile 7.5 and 8 on the Haines Highway, where the main flow is constrained primarily to the eastern side of the floodplain (Figure 2). The best operating location will be determined at the beginning of the season based on assessment of river conditions. Fish wheels will be deployed in the first week of June and operated continuously (i.e., 24 hours each day) through mid-October (statistical weeks 23 to 42). The crew will consist of three Division of Commercial Fisheries staff, supervised by the Assistant Area Management Biologist. The fish wheels will be launched from the Haines Highway and pushed into the river with a tractor, then drifted downstream or pushed upstream with jet boats to the predetermined site. The fish wheels will be anchored to the highway guardrail with 0.95 cm steel cable and a 1-inch diameter polyethylene rope bridle, and held out from, and parallel to, the shoreline with an adjustable boom log system (Kelley and Bachman 2000). An average river depth of at least 1.5 m is required for the fish wheels to maintain revolution speeds adequate to capture migrating salmon (approximately 1.5–4 rpms; Bachman and McGregor 2001). Seasonal changes in water flow (particularly late August–early October, when water levels subside) may require minor changes in fishing location to maintain adequate rotation speed—wheels may be moved farther from shore into faster current or to a nearby (< 1 mile) alternate location. Each fish wheel consists of two aluminum pontoons, measuring approximately 12 m (length) × 6 m (width), filled with closed cell Styrofoam for flotation. The pontoons support a 6 m wide structure consisting of an adjustable height axle, three catch baskets, wooden slides, and enclosed fabric chutes, along with two live boxes per wheel that hold captured fish. A plywood deck spanning the full width between the pontoons provides a sampling platform. The aluminum baskets are 3.1 m (width) × 3.7 m (depth), covered with nylon seine mesh (5.1 × 5.1 cm openings), and bolted to a metal axle that spins in a pillow-block bearing assembly. The three catch baskets are rotated about the axle by the force of the water current, and fishing depth can be adjusted by moving the axle up or down within tower support channels. Migrating salmon will be captured in the rotating baskets as they swim under the structure. V-shaped, foam-padded wooden slides are bolted to the rib midsection of each basket to direct fish through fabric chutes into the lidded aluminum live boxes, which are bolted to the outer sides of each pontoon. The live boxes are perforated to allow constant flow of fresh river water. DATA COLLECTION Live boxes will be inspected a minimum of twice each day, once early in the morning and again in the afternoon. Fish will be removed from the live boxes with a dip net and all salmon will be enumerated by species and recorded on the Chilkat Fish Wheel field sheet (Appendices D5–D9). Length, sex, and scales will be collected from Chinook, sockeye, chum, and coho salmon caught in the fish wheels as outlined in the Age, Sex, and Length Composition section. Fish selected for sampling will be placed in a padded trough partially filled with fresh river water, then processed and quickly released back into the river. Other information recorded daily at the fish wheels will include water temperature (°C), fish wheel rotation speed (rpm), and fish wheel start and stop

times. River water level (mm) will be measured daily at an established staff gauge located near milepost 8 on the Haines Highway. Sockeye Salmon The age composition and mean length-at-age of sockeye salmon caught at the fish wheels will be estimated from a weekly sample of 140 fish at a rate of 20 fish per day (Table 1). This sample size was selected based on work by Thompson (2002) to estimate proportions of more than four major age classes. A total of 101 readable scales is needed to ensure the weekly estimated proportion of each age class will be within 10% of the true value 90% of the time. Over the past 10 years, an average 30% of scale samples have not been readable. Therefore, the sample size was increased to 140 fish to ensure the sampling goal would be met, even if 30% of the samples are unreadable. In addition, 3 scales will be sampled from each fish to increase the proportion of readable scales. This weekly sample size will also ensure that the estimated seasonal proportion of each age class will be within 5% of the true value 95% of the time, even if 30% of the samples are unreadable since a total of only 665 sample will be required for the entire season and three times that many samples will be collected (Table 1). All sockeye salmon not sampled for scales will be enumerated and released. These sample sizes will also meet seasonal sex and length composition requirements, as only 385 samples (assuming no data loss) are necessary to achieve the precision criteria (within 5% of the true value 95% of the time) for estimating sex composition (Thompson 2002). Chinook Salmon Chinook salmon will be enumerated and sampled following detailed protocols outlined in the ADF&G Division of Sport Fish operational plan for Chilkat River Chinook salmon escapement (Elliot and Power 2017a). Coho Salmon Coho salmon will be enumerated and sampled following detailed protocols outlined in the ADF&G Division of Sport Fish operational plan for Chilkat River coho salmon escapement (Elliot and Power 2017b). There are a few deviations from the operational plan sampling goals, including 13 fish will be sampled each day for scales, sex, and length and an additional 27 fish will have length and sex recorded. Chum Salmon The fish wheel catch of chum salmon is used as a relative index to estimate the spawning escapement of Chilkat River fall chum salmon. An estimate of the fall chum salmon escapement will be made from an expansion of the total fish wheel catch. In 1990 and 2002–2005 mark– recapture studies were conducted to estimate Chilkat River chum salmon escapements; total fish wheel catches in those years averaged approximately 1.55% of estimated escapements (Table 2) (Bachman 2005; Eggers and Heinl 2008). The seasonal age composition and mean length-at-age of chum salmon caught at the fish wheels will be estimated from a minimum sample of 665 fish (Table 3). This sample size was selected based on work by Thompson (2002) to estimate proportions of three major age classes (ages 0.2, 0.3, and 0.4). A sample of 510 fish is needed to ensure the seasonal estimated proportion of each age class will be within 5% of the true value 95% of the time. The sample size was increased to 665 fish to ensure the sampling goal would be met, even if 30% of the samples are unreadable. In addition, 3 scales will be sampled from each fish to increase the proportion of readable scales. To

meet the seasonal sampling goal, 12 chum salmon will be sampled each day for age, sex, and length (84 fish/week) (Table 3). These sample sizes will also meet sex and length composition requirements, as only 385 samples (assuming no data loss) are necessary to achieve the precision criteria (within 5% of the true value 95% of the time) for estimating seasonal sex composition (Thompson 2002). Pink Salmon A total of 200 genetic tissue samples will be collected from pink salmon at a rate of 20 samples per week, from statistical weeks 27 to 36. The left axillary fin will be removed from each sampled fish and placed in a bulk sampling bottle filled with ethanol and labeled with the site, year, and sampling method. At the end of the season, tissue samples will be shipped to the ADF&G Gene Conservation Laboratory in Anchorage for archiving. All pink salmon not sampled for genetic tissues will be enumerated and released. DATA REDUCTION Data collected at the fish wheels will be recorded on field forms specific to each activity (Appendices D5–D7). Data forms will be kept up to date at all times and will be inspected daily for errors and completeness. Scale, sex, and length information pertaining to Chinook, sockeye, chum, and coho salmon will be recorded on ASL forms (Appendix D1) as fish are captured and sampled. All ASL forms and scale cards will be checked daily to ensure that scales are clean and mounted correctly, labeled correctly, and match up with the corresponding ASL data form. Scales will be remounted when necessary. All data associated with sockeye and chum salmon scale samples will be sent to the Regional 1 Scale Aging Laboratory in Douglas each Monday morning for scanning, data analysis, and archiving. ASL forms and associated data sheets for Chinook and coho salmon captured at the fish wheels will be provided to ADF&G Sport Fish staff in the Haines office, who will review the data and send it to the Regional 1 Scale Aging Lab to be scanned into the ADF&G Zander Database. ADF&G personnel will enter fish counts from field forms into Excel computer spreadsheets at the Haines office. These data records will be updated daily and distributed within the department via weekly email. The weekly age distributions, the seasonal age distributions weighted by statistical week, and SE of age class by statistical week will be calculated postseason using standard equations from Cochran (1977) (Appendix C). Data Archiving ASL data collected from Chinook, sockeye, chum, and coho salmon will be archived in the ADF&G Zander Database. CHILKAT LAKE ESCAPEMENT A DIDSON (manufactured by Sound Metrics Corporation) sonar will be used in conjunction with a picket weir to enumerate the Chilkat Lake sockeye salmon escapement from at least 20 June to 10 October. The escapement count will be used to determine if the escapement goal is met. Data collected at the weir will provide inseason and postseason information on run timing, run strength, and age composition.

CHILKAT LAKE WEIR The weir will be installed in Clear Creek, approximately 0.4 km downstream of Chilkat Lake. The weir is a semi-removable steel bipod structure approximately 33 m wide. The maximum water depth at the weir site is approximately 3 m. The weir framework consists of 11 5-cm steel pipe bipods spaced between 2.4–2.7 m driven into the bed of the river, connected with perforated steel stringers of varying lengths. Iron pipe pickets, 2.5 cm outside diameter, spaced 3.8 cm apart, will be placed in the evenly spaced holes of the stringers to form a fence across the lake outlet. The maximum possible space between each picket is 4.1 cm. Each spring, divers and crew will install two removable bipods in the center of the weir to support a 3.6-m wide boat gate that will allow boat traffic to access the lake. The boat gate will be operated remotely via an electric hoist/winch. Sandbags and fencing will be placed along the upstream side of the weir to prevent fish from passing uncounted. The integrity of the weir will be verified throughout the season by regular underwater inspections and placement of additional sand bags and fencing if needed. Periodic flow reversals, caused when glacial water from the flooding Tsirku River backs into Clear Creek and into Chilkat Lake, will require keeping the boat gate open to prevent damage to the gate until the reversal subsides. Flow reversals could last from a few hours to several days. DIDSON INSTALLATION AND SETTINGS The DIDSON will be deployed just upstream of the weir approximately 3–5 m from the left bank of the river. Sandbags and wire mesh fencing will be anchored between the weir and the DIDSON transducer to block fish from passing uncounted between the left bank and transducer. The transducer will be attached to an aluminum pod and oriented perpendicular to the current. The wide axis of the beam will be oriented horizontally and positioned close to the river bottom to maximize residence time of targets in the beam. Daily visual inspections will be conducted to confirm proper transducer placement and orientation and to make adjustments to accommodate varying water levels. The DIDSON will be operated at 1.8 MHz (high frequency 96 beams) with a viewing angle of 29° × 14°. A 30 m cable will be used to transmit power and data between the DIDSON and a “topside box” located in the camp cabin next to the weir, and an Ethernet cable will be used to route data to a laptop computer. Sampling will be controlled on the laptop computer with the latest version of DIDSON software. A small gasoline powered generator will provide power for all equipment. Fish will be recorded as they move upstream through the boat gate opening at a range of approximately 5 m to 10 m from the face of the transducer, well within the 30-m effective range for a DIDSON set at the high frequency. The sample rate will be set at 6 frames per second. Data will be recorded continuously in 60-minute increments and saved on a 2-terabyte portable external hard drive. DIDSON OPERATION DIDSON operation will be begin in statistical week 24 (10–16 June). The DIDSON will be operated 24 hours per day for the first five days to observe fish passage when the boat gate is closed at night to ensure there are no openings in the weir that would allow fish to move upstream uncounted. Acoustic sampling will then begin each morning around 0600 and continue until 2200, when the boat gate will be closed for the night. Closing the gate at 2200 will allow fish to build up behind the weir where they can be sampled in the morning. In the rare event that a boat requires passage through the weir between 2200 and 0600, the period of time the gate is down will be

limited as much as possible, but some fish may pass without being counted. If the boat gate remains open after 2200 due to a flow reversal or if the gate is broken, the DIDSON will be operated 24 hours a day until the boat gate can be closed. In addition, the DIDSON will be operated 24 hours a day every Sunday to ensure the weir is fish tight. If fish are recorded moving though the weir uncounted, the weir will be inspected to rectify the problem. If the problem continues, we will consider operating the DIDSON 24 hours a day for the entire season. The DIDSON will be set to record data onto the hard drive in 60-minute increments, creating a separate date and time stamped file for each recording period. The weir crew will identify and tally fish traces from the play back of recorded files. This can be completed at the same time new files are being recorded. All fish determined to be salmon will be counted. Technicians will familiarize themselves with behaviors typical of various fish species through intensive observation. Fish which display feeding or milling behavior known to be associated with cutthroat trout or whitefish will not be counted. Fish that exhibit behavior identified with sockeye salmon (directional migration, no milling) will be assumed to be sockeye salmon. Fish will be counted manually with tally counters. The use of the DIDSON to count fish has limitations that need to be accounted for during operations or addressed preseason, including species apportionment (see below), shadowing effects, and observer bias from species nondetection or misclassification (cutthroat trout and whitefish identification versus salmon species) (Keefer et al. 2017). Observer fatigue or interruptions in viewing can also bias observations between operators (Cronkite et al. 2006). Acoustic shadowing effects can be a problem when fish are present in high densities—fish nearer to the DIDSON mask or “shadow” fish passing farther away—which leads to undercounting. In studies conducted elsewhere, problems associated with shadowing occurred when fish densities were greater than 1,000 fish an hour (Holmes et al. 2006; Maxwell and Gove 2007; Westerman and Willette 2012). Hourly fish counts at Chilkat Lake have usually been well below 1,000 fish. In the rare event that large schools of fish (estimated at >1,000 fish) are present immediately below the weir, fish passage will be restricted by closing the boat gate and pulling a few pickets to create a smaller opening for fish to move through, which should help reduce the occurrence of acoustic shadowing. Event nondetection bias or perception bias occurs in field observations studies when animals are visible but not observed, and typically result in underestimates of abundance (Nichols et al. 2000). Misclassification biases occur when species are misidentified, also inducing bias in abundance estimates (Conn et al. 2013). These biases can be reduced by training observers in the preseason and by routinely conducting inseason observer comparisons to maintain quality control and ensure accuracy. Observer Training Early Season Accurate DIDSON counts also rely on an observer’s expertise to detect individual fish (Martignac et al. 2015; Petreman et al. 2014). To standardize sockeye salmon identification and enumeration, along with training observers during the preseason, an ‘experienced’ observer will independently view, enumerate, and process a set of 4 raw 60-minute training files from prior seasons of DIDSON deployment. The ‘experienced’ observer will enumerate sockeye salmon for each 60-minute file, repeatedly, until the fish count has a CV of less than 10% between repetitions of the same file. The training files will include a variety of fish densities, along with upstream and downstream movement of fish, to ensure that observers are subjected to a full spectrum of fish densities,

movement patterns, and behaviors. The training files will be used to develop a sockeye salmon identification criterion, to standardize enumeration scoring, and to increase agreement among different observers about event identification. Observers in training will independently view and enumerate sockeye salmon detection events from the 4 raw 60-minute training files. For consistency across observers, software setting, monitor size, and playback speed will be held constant, but observers will be allowed to rewatch clips and adjust some settings such as brightness. Observers in training will rewatch and re-enumerate each training file until they achieve a final fish count within 85% of the benchmark interpretation set by the ‘experienced’ observer for each file. The ‘experienced’ observer will review files, as necessary, with each observer in training. Inseason Periodic, systematic observer comparisons of DIDSON counts will be conducted inseason to ensure accuracy of escapement counts. Every four weeks, 4 raw 60-minute files, representing the peak count for the four previous weeks, will be independently reviewed and enumerated for sockeye salmon abundance one time by each observer. These files should incorporate abundances greater than 100 fish/file when possible. An average of all individual counts ( ) will be used as a baseline to compare against the averages of each observer ( ) (Westerman and Willette 2011), 𝑥𝑥̅ 𝑖𝑖 𝑥𝑥̅ = and = . (1) ∑ 𝑓𝑓 ∑ 𝑓𝑓𝑖𝑖 where: 𝑥𝑥̅ 𝑛𝑛 𝑥𝑥̅𝑖𝑖 𝑛𝑛𝑖𝑖 = sum of all fish counts of all observers,

∑ 𝑓𝑓 = sum of fish counts of an individual observer, n∑ =𝑓𝑓 𝑖𝑖number of files counted by all observers, and = number of files counted by an individual observer.

The𝑖𝑖 difference (d) from the average will be measured for each observer, then compared to the 𝑛𝑛 average of all individual subsample counts,

= . (2)

𝑖𝑖 𝑑𝑑 𝑥𝑥̅ − 𝑥𝑥̅ If an individual observer’s count is higher than the average, the difference will be expressed as a plus (+), and if below the average, a minus (-). Mean absolute percent error (MAPE) will be calculated as the mean of the absolute values of the percent errors. MAPE yields a measure of observer performance with over- and under-estimates treated equally (e.g. Ryall 1998),

( ) MAPE = 100. (3) 𝑥𝑥̅−𝑥𝑥̅𝑖𝑖 �− 𝑥𝑥̅ � ∗

Disagreement of more than MAPE=5% between the average all the individual counts of the observers and an individual observer’s average should be flagged for a detailed review. AGE, SEX, LENGTH SAMPLING The seasonal age composition of the Chilkat Lake sockeye salmon escapement (including jack sockeye salmon) will be determined from a minimum of 665 scale samples captured on the downstream side of the weir. This sample size was based on work by Thompson (1987, 2002) to estimate proportions of four or more major age classes. A sample of 510 fish is needed to ensure the estimated proportion of each age class will be within 5% of the true value 95% of the time. The sample size was increased to 665 fish to ensure the sampling goal will be met, even if 30% of the samples are unreadable. In addition, 3 scales will be sampled from each fish to increase the proportion of readable scales. Ten sockeye salmon will be sampled each day for age, sex, and length (70 fish/week) (Table 1). If fish are present at 0600, beach seine gear will be used to capture fish below the weir. This weekly sample size will ensure that the desired precision of the estimated seasonal proportion of each age class will be met, since the total seasonal sample will be twice what is actually needed (Table 1). These sample sizes will also meet seasonal sex and length composition requirements, as only 385 samples (assuming no data loss) are necessary to achieve the precision criteria (within 5% of the true value 95% of the time) for estimating sex composition (Thompson 2002). Length and sex of each sampled fish will be collected as described in the Age, Sex, and Length Composition section. The weekly age distribution, the seasonal age distribution and SE weighted by statistical week, and SE of sex composition weighted by statistical week will be calculated postseason using standard equations from Cochran (1977) (Appendix C). SPECIES APPORTIONMENT The DIDSON cannot be used to identify salmon to species when two or more species of similar size and shape are present (Martignac et al. 2015). Although on some river systems, apportioning sonar counts by species requires separate, intensive net or fish wheel sampling programs (Bromaghin 2005; Lozori and McIntosh 2014), species identification on the Chilkat River only includes two species (coho and sockeye salmon) and is not an issue until late August or early September. The ratio of coho to sockeye salmon in beach seine sampling events during late August to the end of the season will be used to apportion the DIDSON counts by species. At least 68 fish (combination of sockeye and coho salmon) are needed to estimate the proportion of each salmon species within 10% of the true value 90% of the time. This assumes that the proportion remains relatively constant. To account for daily and weekly changes in species apportionment, a daily goal of at least 68 fish (combination of sockeye and coho salmon) will be captured in beach seine sets and enumerated by species, and fish counts will be recorded for each beach seine set conducted. If the daily goal is not met, the fewest number of adjacent days needed to reach the goal of 68 fish will be combined. To avoid duplicate sampling, all fish will be marked with a hole punch on the upper left operculum. The prior average ratio of sockeye to coho salmon will be applied to daily DIDSON counts until another beach seine set is completed. During beach seining efforts, coho salmon will simply be marked and enumerated and will not be sampled for age, sex, and length. Pink and chum salmon will be counted if captured during beach seine sampling events; however, their numbers are expected to be very low since few were historically counted through the weir (the 1981–2007 average annual count was 10 chum salmon and 1 pink salmon).

DATA REDUCTION Data collected at the fish weir will be recorded on field forms specific to each activity (Appendices D9–D13). Data forms will be kept up to date at all times and will be inspected daily for errors and completeness. Scale, sex, and length information pertaining to sockeye salmon recovery from daily beach seins will be recorded on ASL forms (Appendix D1) as fish are captured and sampled. Seine sampling events and total catch by species will be recorded on the recovery field data sheet. All ASL forms and scale cards will be checked daily to ensure that scales are clean and mounted correctly, labeled correctly, and match up with the corresponding ASL data form. Scales will be remounted when necessary. All data associated with sockeye salmon scale samples will be sent to the Regional 1 Scale Aging Laboratory in Douglas each Monday morning for scanning, data analysis, and archiving. ADF&G personnel will enter the beach seine recovery data from field forms into Excel computer spreadsheets at the field camp. Hourly DIDSON counts will be recorded on a paper data sheet during enumeration and recorded into Excel computer spreadsheets at the field camp. The summarized hourly counts will be entered directly into a daily counts Excel spreadsheet with proper apportionment when multiple species are caught in beach seine sampling events. The finalized daily counts will be called into the Haines office via cell phone daily and distributed within the department via weekly email. The weekly age distribution, the seasonal age distribution weighted by statistical week, and SE of age class by statistical week will be calculated postseason using standard equations from Cochran (1977) (Appendix C). Data Archiving The ASL data collected from sockeye salmon and the finalized daily weir counts will be archived in the ADF&G Zander Database. LIMNOLOGICAL ASSESSMENT Basic limnological data, including zooplankton, light, and temperature sampling, will be collected monthly at Chilkat Lake from June to October. Sampling will be conducted as close as possible to the first day of each month. Sampling will be conducted at two stations marked by anchored buoys in the lake: station 1A (59.3420° N, 135.9131° W) and station 2A (59.3263° N, 135.8961° W). SAMPLING METHODS AND DATA ANALYSIS Light and Temperature Sampling Light profiles will be collected at each station using an ILT 1400International Light Technologies Photometer. Photometric illuminance will be recorded as Lumens per square meter at 0.5-m intervals, from just below the lake surface to the depth at which ambient light level equals 1% of the light level just below the surface. Light penetration measurements will be used to determine vertical light extinction coefficients and algal compensation depths. The natural log (ln) of the ratio of light intensity just below the surface to light intensity at depth z, I0/Iz, will be calculated for each depth. The vertical light extinction coefficient (Kd) will be estimated as the slope of ln(I0/Iz) versus depth. The euphotic zone depth (EZD) is defined as the depth at which light (photosynthetically available radiation at 400–700 nm) is attenuated to 1 percent of the intensity just below the lake surface (Schindler 1971) and will be calculated with the equation EZD = 4.6502/Kd (Kirk 1994). Temperature (ºC) will be measured with a Yellow Springs Instruments Model 58 meter. Temperature will be recorded at 1-m intervals from the lake surface to a depth of

20 meters, and at 5-m intervals from 20 meters to a depth of 50 meters. All sampling data will be recorded directly onto Limnology Sampling Forms (Appendix D14). Temperature readings will be recorded in the “Meter” column. Zooplankton Sampling Zooplankton samples will be collected at each station using a 0.5 m diameter, 153 µm mesh conical net. Vertical zooplankton tows will be pulled from a depth of 50 meters to the surface at a constant speed of 0.5 m/sec. Once the top of the net clears the water surface, the rest of the net will be pulled slowly from the water and rinsed from the outside with lake water to wash organisms into the screened sampling container at the cod end of the net. All specimens in the sampling container will be carefully rinsed into a labeled 500 ml sampling bottle and preserved in buffered 10% formalin solution. Sampling bottles will be topped off with clean lake water. The lake name, date, sampler’s name, station, tow depth, and net diameter will be recorded on the bottle label. DATA REDUCTION Limnological data collected at the Chilkat Lake will be recorded on a field form (Appendix D14). Data forms will be kept up to date at all times and will be inspected daily for errors and completeness. Light and temperature data will be entered into specific Excel spreadsheets at the Haines office and stored on the ADF&G regional shared network drive Zooplankton samples and associated forms will be delivered to the Haines office of ADF&G Commercial Fisheries for seasonal storage. At the end of the season, all samples will be shipped to the ADF&G Kodiak Limnology Laboratory for analysis (Hopkins 2017). Monthly results will be averaged between the two stations, and seasonal estimates will be calculated as the average of the monthly values, early June–early October. SCHEDULE AND DELIVERABLES FIELD OPERATIONS Field sampling activities are scheduled as follows: 1. Chilkat River fish wheels about 5 June–15 October 2. Chilkat Lake weir and DIDSON sonar about 15 June–15 October 3. Chilkat Lake limnology monthly, June–October REPORTING Results of this study will be presented in the annual fishery management plan for the Lynn Canal drift gillnet fishery (Fishery Management Report) in April of each year and the biannual report summarizing the results of this project (Fishery Data Series Report), which will be completed in March of 2019 and in March of each following odd year. RESPONSIBILITIES Wyatt Rhea-Fournier, Fishery Biologist III, Principal Investigator. Sets up all major aspects of project, including planning, budget, sample design, permits, equipment, personnel, and training. Supervises overall project; edits, analyzes, and reports data; oversees major repairs; and expedites major purchases. Reviews schedules and operational plan and serves

as lead biologist for the project and reports project data in ADF&G Fishery Data Series report. Heather Spore, Fishery Biologist II: Responsible for overseeing fish wheel operations and directing the projects in the absence of Rhea-Fournier. Assists with the supervision of overall project; edits, analyzes, and reports data; trains the crew in safety and project procedures; creates the crew schedule; assists with fieldwork; arranges logistics with field crew; and serves as project expeditor. Writes the operational plan and assures that it is followed or modified appropriately with consultation with the Sport Fish Division biologists. Resolves personnel or administrative issues related to this project and writes crew evaluations. Shelby Flemming Fish and Wildlife Technician IV: This position functions as the lead on the fish wheel project and assists in leading the Chilkat Lake weir project. This position will spend at least one day a week at the Chilkat weir to ensure sampling and data collection are performed in accordance with methods outlined in this operational plan. Lane Taylor, Jozef Slowick, Fish and Wildlife Technician II: These positions assist in all aspects of field operations, including safe operation of the fish wheels, riverboats and all other equipment; assist with the daily preparation of forms for data entry, marking, tagging, and sampling of fish on the fish wheels, any required fish wheel or boat maintenance, completion of data forms, and all aspects of the salmon recovery portion of this project. Tim Brush, Fish and Wildlife Technician III: Crew leader on the Chilkat Lake weir and DIDSON project; responsible for employee training and data quality control. Fish and Wildlife Technicians II (2 vacant positions): Assist in all aspects of the operation and maintenance of the Chilkat Lake weir and DIDSON; assist in the limnological sampling. Faith Lorentz, Program Technician: Coordinates communication with Chilkat Lake weir crew, updates master spreadsheet with daily weir and fish wheel counts, provides administrative assistance, tracks project budgets, and provides other assistance as needed. Kyle Shedd, ADF&G Fisheries Geneticist II, Gene Conservation Laboratory: will oversee genetic stock identification analyses and provide weekly commercial harvest stock composition estimates to fishery managers. Anne Reynolds-Manney, ADF&G Southeast Alaska Regional Port Sampling Supervisor: will monitor and participate in fishery sampling and weekly sample shipment to the ADF&G Gene Conservation Laboratory. She will also present results at industry task force meetings as needed. Steven C. Heinl, Regional Research Coordinator: Assist with project operational planning and review of project report. Sara Miller, Biometrician II: Assist with sampling design, project operational planning, and data analysis.

REFERENCES CITED Bachman, R. L. 2005. Stock assessment studies of Chilkat River adult sockeye and chum salmon stocks in 2002. Alaska Department of Fish and Game, Fishery Data Series No. 05-36, Anchorage. Bachman, R. L. 2010. Stock assessment studies of Chilkat River adult sockeye and chum salmon stocks in 2003 and 2004. Alaska Department of Fish and Game, Fishery Data Series No. 10-23, Anchorage. Bachman, R. L., and A. J. McGregor. 2001. Stock assessment studies of Chilkat River adult salmon stocks in 1999. Alaska Department of Fish and Game, Division of Commercial Fisheries, Regional Information Report No. 1J01- 36, Juneau. Bednarski, J. A., M. M. Sogge, S. E. Miller. And S. C. Heinl. In press. A comprehensive review of Chilkat Lake and River sockeye salmon stock assessment studies. Alaska Department of Fish and Game, Fishery Data Series No. 17-XX, Anchorage. Bednarski, J. A., M. Sogge, and S. C. Heinl. 2016. Stock assessment study of Chilkoot Lake sockeye salmon, 2013– 2015. Alaska Department of Fish and Game, Fishery Data Series No. 16-29, Anchorage. Bergander, F. E., S. A. McPherson, and J. P. Koenings. 1988. Southeast Alaska sockeye salmon studies, 1987–88; technical report for the period July1, 1987 to June 30, 1988. Alaska Department of Fish and Game, Division of Commercial Fisheries, Regional Information Report No. 1J88-44, Juneau. Bromaghin, J. 2005. A versatile net selectivity model, with application to Pacific salmon and freshwater species of the Yukon River, Alaska. Fisheries Research 74:157– 168. Bugliosi, E. F. 1988. Hydrologic reconnaissance of the Chilkat River basin. U.S. Geological Survey, Water-Resources Investigations Report 88-4023, Anchorage. Buettner, A. R., A. M. Reynolds, and J. R. Rice. 2017. Operational Plan: Southeast Alaska and Yakutat salmon commercial port sampling 2016–2019. Alaska Department of Fish and Game, Regional Operational Plan ROP.CF.1J.17-01, Douglas. Clutter, R., and L. Whitsel. 1956. Collection and interpretation of sockeye salmon scales. Bull. Int. Pac. Salmon Fish. Comm., No. 9. Cochran, W. 1977. Sampling techniques. 3rd ed. John Wiley and Sons, Inc., New York. Conn, P. B., B. T. McClintock, M. F. Cameron, D. S. Johnson, E. E. Moreland, and P. L. Boveng. 2013. Accommodating species identification errors in transect surveys. Ecology 94:2607-2618. Cronkite, G. M. W., H. Enzenhofer, T. Ridley, J. Holmes, J. Lilja, and K. Benner. 2006. Use of high-frequency imaging sonar to estimate adult sockeye salmon escapement in the Horsefly River, British Columbia. Canadian Technical Report of Fisheries and Aquatic Sciences 2647. Dann, T. H., C. Habicht, S. D. Rogers Olive, H. L. Liller, E. K. C. Fox, J. R. Jasper, A. R. Munro, M. J. Witteveen, T. T. Baker, K. G. Howard, E. C. Volk, and W. D. Templin. 2012. Stock composition of sockeye salmon harvests in fisheries of the Western Alaska Salmon Stock Identification Program (WASSIP), 2006–2008. Alaska Department of Fish and Game, Special Publication No. 12-22, Anchorage. Eggers, D. M., and S. C. Heinl. 2008. Chum salmon stock status and escapement goals in Southeast Alaska. Alaska Department of Fish and Game, Special Publication No. 08-19, Anchorage. Eggers, D. M., J. H. Clark, R. L. Bachman, and S. C. Heinl. 2008. Sockeye salmon stock status and escapement goals in Southeast Alaska. Alaska Department of Fish and Game, Special Publication No. 08-17, Anchorage. Eggers, D. M., R. L. Bachman, and J. Stahl. 2010. Stock status and escapement goals for Chilkat Lake sockeye salmon in Southeast Alaska. Alaska Department of Fish and Game, Fishery Manuscript No. 10-05, Anchorage. Elliott, B. W., and S. J. H. Power. 2017a. Chilkat River Chinook salmon escapement studies in 2017. Alaska Department of Fish and Game, Division of Sport Fish, Regional Operational Plan ROP.SF.1J.2017.04, Anchorage.

REFERENCES CITED (Continued) Elliott, B. W., and S. J. H. Power. 2017b. Production and harvest of Chilkat River Chinook and coho salmon, 2017 - 2018. Alaska Department of Fish and Game, Regional Operational Plan No SF.1J.2017.08, Anchorage. Geiger, H. J., R. L. Bachman, S. C. Heinl, K. Jensen, T. A. Johnson, A. Piston, and R. Riffe. 2005. Sockeye salmon stock status and escapement goals in Southeast Alaska [in] Der Hovanisian, J. A. and H. J. Geiger, editors. Stock status and escapement goals for salmon stocks in Southeast Alaska 2005. Alaska Department of Fish and Game, Special Publication No. 05-22, Anchorage. Gilk-Baumer, S. E., S. D. Rogers Olive, D. K. Harris, S. C. Heinl, E. K. C. Fox, and W. D. Templin. 2015. Genetic mixed stock analysis of sockeye salmon harvests in selected northern Chatham Strait commercial fisheries, Southeast Alaska, 2012–2014. Alaska Department of Fish and Game, Fishery Data Series No. 15-03, Anchorage. Halupka, K. C., M. D. Bryant, M. F. Willson, and F. H. Everest. 2000. Biological characteristics and population status of anadromous salmon in Southeast Alaska. United States Forest Service. General Technical Report. PNW-GTR- 468. Heinl, S. C., E. L. Jones III, A. W. Piston, P. J. Richards, L. D. Shaul, B. W. Elliott, S. E. Miller, R. E. Brenner, and J. V. Nichols. 2017. Review of salmon escapement goals in Southeast Alaska, 2017. Alaska Department of Fish and Game, Fishery Manuscript Series No. 17-11, Anchorage. Holmes, J. A., G. M. W. Cronkite, H. J. Enzenhofer, and T. J. Mulligan. 2006. Accuracy and precision of fish-count data from a ‘‘dual-frequency identification sonar’’ (DIDSON) imaging system. ICES Journal of Marine Science 63:543–555. Hopkins, A. M. 2017. Kodiak Island Limnology Laboratory analysis operational plan, 2017–2019. Alaska Department of Fish and Game, Division of Commercial Fisheries, Regional Operational Plan ROP.CF.4K.2017.14, Kodiak. Ingledue, D. 1989. Hawk Inlet shore purse seine fishery, 1989. Alaska Department of Fish and Game, Division of Commercial Fisheries, Regional Information Report No. 1J89-31, Juneau. INPFC (International North Pacific Fisheries Commission). 1963. Annual report 1961. Vancouver, British Columbia. Johnson, J., and M. Daigneault. 2013. Catalog of waters important for spawning, rearing, or migration of anadromous fishes – Southeastern Region, Effective July 1, 2013. Alaska Department of Fish and Game, Special Publication No. 13-09, Anchorage. Keefer, M. L., C. C. Caudill, E. L. Johnson, T. S. Clabough, C. T. Boggs, P. N. Johnson, and W. T. Nagy. 2017. Inter-observer bias in fish classification and enumeration using Dual-frequency Identification Sonar (DIDSON): a Pacific Lamprey case study. Northwest Science 91:41–53. Kelley, M. S., and R. L. Bachman. 2000. Stock assessment studies of the Chilkat River adult salmon stocks in 1998. Alaska Department of Fish and Game, Division of Commercial Fisheries, Regional Information Report No. 1J00- 29, Juneau. Kirk, J. T. O. 1994. Light and Photosynthesis in Aquatic Ecosystems. Cambridge University Press, England Koo, T. S. Y. 1962. Age designation in salmon [In] Studies of Alaska red salmon. University of Washington Press, Seattle. Lozori, J. D., and B. C. McIntosh. 2014. Sonar estimation of salmon passage in the Yukon River near Pilot Station, 2012. Alaska Department of Fish and Game, Fishery Data Series No. 14-22, Anchorage. Martignac, F., A. Daroux, J. L. Bagliniere, D. Ombredane, and J. Guillard. 2015. The use of acoustic cameras in shallow waters: new hydroacoustic tools for monitoring migratory fish population. A review of DIDSON technology. Fish and Fisheries 16:486–510. Maxwell, S. L., and N. E. Gove. 2007. Assessing a dual-frequency identification sonars' fish-counting accuracy, precision, and turbid river range capability. Journal of the Acoustical Society of America 122:3364–3377.

REFERENCES CITED (Continued) McPherson, S. A. 1990. An inseason management system for sockeye salmon returns to Lynn Canal, southeast Alaska. M. S. Thesis, University of Alaska, Fairbanks. McPherson, S. A., and M. A. Olsen. 1992. Contribution, exploitation, and migratory timing of Lynn Canal sockeye salmon runs in 1989 based on analysis of scale patterns. Alaska Department of Fish and Game, Division of Commercial Fisheries, Technical Fishery Report No. 92-22, Juneau. Miller, S. E., and S. C. Heinl. In press. Chilkat Lake sockeye salmon escapement goal review. Alaska Department of Fish and Game, Fishery Manuscript Series, Anchorage. Moran, B. M., and E. C. Anderson. In prep. Bayesian inference from the conditional genetic stock identification model. Nichols, J. D., J. E. Hines, J. R. Sauer, F. W. Fallon, J. E. Fallon, and P. T. Heglund. 2000. A double-observer approach for estimating detection probability and abundance from point counts. The Auk 117:393–408. Pella, J., and M. Masuda. 2001. Bayesian methods for analysis of stock mixtures from genetic characters. Fishery Bulletin 99:151–167. Petreman, I. C., N. E. Jones, and S. W. Milne. 2014. Observer bias and subsampling efficiencies for estimating the number of migrating fish in rivers using Dual-frequency IDentification SONar (DIDSON). Fisheries Research 155:160-167. Piston, A. W., and S. C. Heinl. 2014. Chum salmon stock status and escapement goals in Southeast Alaska. Alaska Department of Fish and Game, Special Publication No. 14-13, Anchorage. R Development Core Team. 2017. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/. Rich, W. H., and E. M. Ball. 1933. Statistical review of the Alaska salmon fisheries. Part IV: Southeastern Alaska. Bulletin of the Bureau of Fisheries, Vol. XLVII (47), No. 13: 437–673. Rogers Olive, S. D., E. K. C. Fox, and Sara E. Gilk-Baumer In press. Genetic baseline for mixed stock analyses of sockeye salmon harvested in Southeast Alaska for Pacific Salmon Treaty applications, 2018. Alaska Department of Fish and Game, Fishery Manuscript, Anchorage. Ryall, P. 1998. Evaluation of the reliability of in-season run size estimation techniques used for Southern British Columbia chum salmon (Oncorhynchus keta) runs. N. Pac. Anadr. Fish Comm. Bull. No.1: 380-387 Schindler, D. W. 1971. Light, temperature, and oxygen regimes of selected lakes in the experimental lakes area, northwestern Ontario. Journal of the Fisheries Research Board of Canada 28: 157–169. Sogge, M. M. 2016. Operational Plan: Stock assessment studies of Chilkoot Lake adult salmon. Alaska Department of Fish and Game, Division of Commercial Fisheries, Regional Operational Plan ROP.CF.1J.2016.01, Douglas. Sogge, M. M., and R. L. Bachman. 2014. Operational Plan: Stock assessment studies of Chilkat River adult salmon. Alaska Department of Fish and Game, Regional Operational Plan ROP.CF1J.14-03, Douglas. Thompson, S. K. 1987. Sample size for estimating multinomial proportions. The American Statistician 41:41–46. Thompson, S. K. 2002. Sampling, 2nd ed. John Wiley and Sons, Inc., New York. Westerman, D. L., T. M. Willette. 2011. Upper Cook Inlet salmon escapement studies, 2010. Alaska Department of Fish and Game, Fishery Data Series No. 11-66, Anchorage.

TABLES AND FIGURES

Table 1.–Weekly sampling goals for sockeye salmon at the Chilkat River fish wheels and Chilkat Lake weir. Weir Fish wheel Statistical Start Stop Weekly Cumulative Weekly Cumulative Weeks Date Date Target Target Target Target 23 3-Jun 9-Jun 70 70 140 140 24 10-Jun 16-Jun 70 140 140 280 25 17-Jun 23-Jun 70 210 140 420 26 24-Jun 30-Jun 70 280 140 560 27 1-Jul 7-Jul 70 350 140 700 28 8-Jul 14-Jul 70 420 140 840 29 15-Jul 21-Jul 70 490 140 980 30 22-Jul 28-Jul 70 560 140 1,120 31 29-Jul 4-Aug 70 630 140 1,260 32 5-Aug 11-Aug 70 700 140 1,400 33 12-Aug 18-Aug 70 770 140 1,540 34 19-Aug 25-Aug 70 840 140 1,680 35 26-Aug 1-Sep 70 910 140 1,820 36 2-Sep 8-Sep 70 980 140 1,960 37 9-Sep 15-Sep 70 1,050 140 2,100 38 16-Sep 22-Sep 70 1,120 140 2,240 39 23-Sep 29-Sep 70 1,190 140 2,380 40 30-Sep 6-Oct 70 1,260 140 2,520 41 7-Oct 13-Oct 2170 1,330 28 2,660 Average unreadable 30% 30% Estimated readable 931 1,862

Table 2.–Fish wheel catch, mark–recapture estimates, and expanded escapement estimates of Chilkat River fall chum salmon, 1990 and 2002–2017.

Total Drainagewide Ratio Fish Wheel Fish Wheel Mark–Recapture SE of Catch to Escapement Escapement Year Catch Estimate Estimate Estimatea Estimate 1990 3,025 275,000 NA 0.0110 275,000 ------2002 2,898 206,000 22,000 0.0141 206,000 2003 3,846 166,000 17,000 0.0232 166,000 2004 4,277 329,000 25,000 0.0138 329,000 2005 3,125 202,000 23,000 0.0155 202,000 2006 10,563 NA NA NA 689,000 2007 4,967 NA NA NA 324,000 2008 6,770 NA NA NA 441,000 2009 5,049 NA NA NA 329,000 2010 1,369 NA NA NA 89,000 2011 5,517 NA NA NA 360,000 2012 4,401 NA NA NA 287,000 2013 2,551 NA NA NA 166,000 2014 2,175 NA NA NA 142,000 2015 3,171 NA NA NA 207,000 2016 3,346 NA NA NA 218,000 2017 1,991 NA NA NA 130,000

Table 3.–Weekly sampling goals for chum salmon at the Chilkat fish wheels.

Start Stop Weekly Cumulative Statistical Weeks Date Date Target Target 28 8-Jul 14-Jul 84 84 29 15-Jul 21-Jul 84 168 30 22-Jul 28-Jul 84 252 31 29-Jul 4-Aug 84 336 32 5-Aug 11-Aug 84 420 33 12-Aug 18-Aug 84 504 34 19-Aug 25-Aug 84 588 35 26-Aug 1-Sep 84 672 36 2-Sep 8-Sep 84 756 37 9-Sep 15-Sep 84 840 38 16-Sep 22-Sep 84 924 39 23-Sep 29-Sep 84 1,008 40 30-Sep 6-Oct 84 1,092 41 7-Oct 13-Oct 84 1,176 42 14-Oct 20-Oct 84 1,260 Average unreadable 30% Estimated readable 823

Figure 1.–The Chilkat and Chilkoot River watersheds and the District 15 commercial fishing subdistricts in Lynn Canal, including the Boat Harbor terminal hatchery area (Subdistrict 115-11).

Figure 2.–Chilkat River drainage, with fish wheel locations, mainstem river sockeye salmon sampling sites, and Chilkat Lake weir location.

APPENDIX

Appendix A.–ADF&G statistical weeks 20–45, 2018. Statistical weeks begin on Sunday at 12:01 a.m. and end the following Saturday at midnight and are numbered sequentially starting from the first week of the calendar year. Statistical Start End Statistical Start End Week Date Date Week Date Date 20 13-May 19-May 33 12-Aug 18-Aug 21 20-May 26-May 34 19-Aug 25-Aug 22 27-May 2-Jun 35 26-Aug 1-Sep 23 3-Jun 9-Jun 36 2-Sep 8-Sep 24 10-Jun 16-Jun 37 9-Sep 15-Sep 25 17-Jun 23-Jun 38 16-Sep 22-Sep 26 24-Jun 30-Jun 39 23-Sep 29-Sep 27 1-Jul 7-Jul 40 30-Sep 6-Oct 28 8-Jul 14-Jul 41 7-Oct 13-Oct 29 15-Jul 21-Jul 42 14-Oct 20-Oct 30 22-Jul 28-Jul 43 21-Oct 27-Oct 31 29-Jul 4-Aug 44 28-Oct 3-Nov 32 5-Aug 11-Aug 45 4-Nov 10-Nov

Appendix B.–ADF&G collection code, location, reporting group, and the number of sockeye salmon used in the genetic baseline for mixed stock analysis in District 15 commercial drift gillnet fishery.

ADF&G reporting group Location Reporting Group n SCKAT07E Chilkat Lake07 Early Chilkat Lake 95 SCKAT07L Chilkat Lake07 Late Chilkat Lake 95 SCKAT13 Chilkat Lake13 Chilkat Lake 189 SBEARFL07 Bear Flats - Chilkat Chilkat Mainstem 95 SMULE03.SMULE07 Mule Meadows - Chilkat Chilkat Mainstem 190 SMOSQ07 Mosquito Lake - Chilkat Chilkat Mainstem 95 SCHIK03 Chilkoot River Chilkoot 159 SCHILBC07 Chilkoot Lake - Bear Creek Chilkoot 233 SCHILB07 Chilkoot Lake - beaches Chilkoot 251 SLACE13 Lace River Juneau Mainland 63 SBERN03.SBERN13 Berners Bay Juneau Mainland 165 SANTGILK13 Antler-Gilkey River Juneau Mainland 53 SWIND03.SWIND07 Windfall Lake Juneau Mainland 142 SSTEE03 Steep Creek Juneau Mainland 91 SAUKE13baseline. SLAKECR14 Lake Creek (Auke Creek Weir) Juneau Mainland 318 SKUTH06 Kuthai Lake Taku River/Stikine Mainstem 171 SKSLK10.SKSLK11 King Salmon Lake Taku River/Stikine Mainstem 214 SLTRA90.SLTRA06 Little Trapper Lake Taku River/Stikine Mainstem 237 SLTAT11 Little Tatsamenie11 Taku River/Stikine Mainstem 59 STATS05.STATS06 Tatsamenie Lake Taku River/Stikine Mainstem 288 SHACK08 Hackett River Taku River/Stikine Mainstem 52 SNAHL03.SNAHL07. SNAHL12 Nahlin River Taku River/Stikine Mainstem 179 STAKU07 Taku River Taku River/Stikine Mainstem 95 Taku Mainstem – STAKWA09 Takwahoni/Sinwa Taku River/Stikine Mainstem 67 SSUSTA08.SSHUST09 Shustahini Slough Taku River/Stikine Mainstem 185 STUCH08.SCHUNK09. STUSK08.SBEARSL09. STUSKS08.STUSKS09 Tuskwa/Chunk Slough Taku River/Stikine Mainstem 356 SYELLB08.SYELLB10. SYELLB11 Yellow Bluff Slough Taku River/Stikine Mainstem 81 STULS07.STULS08. STULS09 Tulsequah River Taku River/Stikine Mainstem 156 SFISHCR09.SFISHCR10 Fish Creek Taku River/Stikine Mainstem 160 SYEHR07.SYEHR09 Yehring Creek Taku River/Stikine Mainstem 171 SCHUT08 Chutine River Taku River/Stikine Mainstem 94 SCHUTL09.SCHUT11 Chutine Lake Taku River/Stikine Mainstem 224 SFOWL07.SFOWL08. SFOWL09.SANDY07. SANDY09 Andy Smith slough Taku River/Stikine Mainstem 54 SPORCU07.SPORCU11 Porcupine Taku River/Stikine Mainstem 74 SDEVIL07.SDEVIL08 Devil's Elbow0708 Taku River/Stikine Mainstem 148 -continued-

Appendix B.–Page 2 of 7.

ADF&G reporting group Location Reporting Group n

SDEVIL09 Devil's Elbow09 Taku River/Stikine Mainstem 53 SSCUD07.SSCUD08. SSCUD09 Scud River Taku River/Stikine Mainstem 192 SISKU85.SISKU86. SISKU02.SISKU06. SISKU08.SISKU09 Iskut River Taku River/Stikine Mainstem 153

SISKU07 Iskut River (Craigson Slough) Taku River/Stikine Mainstem 42 SCRAIG06.SCRAIG07. SCRAIG08 Craig River-CAN Taku River/Stikine Mainstem 38

SBRON08.SBRON09 Bronson Slough Taku River/Stikine Mainstem 78 SSHAKS06.SSHAKES07. SSHAKS09 Shakes Slough Taku River/Stikine Mainstem 67

SCHRI11.SCHRI12 Christina Lake Taku River/Stikine Mainstem 70

SCRES03 Crescent Lake Snettisham 194

SSPEE03 Speel Lake Snettisham 95

SSNET06.SSPEE07 Snettisham Hatchery0607 Snettisham 190

SSPEE13 Snettisham Hatchery13 Snettisham 146

SVIVID93 Vivid Lake Other 48

SSECLK14.SSECLKIN14 Seclusion Lake Other 117

SNBERG91 North Berg Bay Inlet91 Other 53

SNBERG92 North Berg Bay Inlet92 Other 100

SBART13 Bartlett River Other 69

SNEVA08 Neva Lake08 Other 94

SNEVA09.SNEVA13 Neva Lake0913 Other 255

SHOKTAI04 Hoktaheen - main inlet Other 47

SHOKTAO04 Hoktaheen - outlet Other 49

SHOKTAM14 Hoktaheen - marine waters Other 47

SKLAG09 Klag Bay Stream Other 200

SFORD04 Ford Arm Lake Other 207

SFORD13 Ford Arm Creek Other 199

SREDOUBT13 Redoubt Lake Other 200

SSALML07.SSALML08 Salmon Lake Other 185

SNECKER91.SNECKER93 Benzeman Lake Other 95

SFALL03.SFALL10 Falls Lake Other 190

SREDB93 Redfish Lake Other 94

SKUTL03 Kutlaku03 Other 95

SKUTL12 Kutlaku12 Other 78

SKUTL13 Kutlaku13 Other 50

SPAVLOF12.SPAVLOFR13 Pavlof River Other 174 SKOOK07.SKOOK10L. SKOOK12L Kook Lake Late Other 194

SKOOK12E.SKOOK13 Kook Lake early Other 148 -continued-

Appendix B.–Page 3 of 7.

ADF&G reporting group Location Reporting Group n SSITK03.SSITK11. SSITK12 Sitkoh Lake Other 351 SLEVA12 Lake Eva Other 115 SHASSEL12.SHASSELR13 Hasselborg Lake Other 209 SKANA07.SKANA10. SKANAL13 Kanalku Lake Other 319 SBAIN10 Bainbridge Lake Other 95 SCOGH91.SCOG92HL.SCOG92ES.SCOGH10 Coghill Lake Other 378 SESHAR08.SESHA91 Eshamy Creek Other 185 SMAIN91 Main Bay Other 96 SMINE91.SMINE09 Miners Lake Other 191 SEYAM07 Eyak Lake - Middle Arm Other 95 SEYASB07 Eyak Lake - South beaches Other 87 SEYAK10 Eyak Lake - Hatchery Creek Other 95 SMEND08.SMEND09 Mendeltna Creek Other 188 SSWEDE08 Swede Lake Other 95 SFISHC08 East Fork Gulkana River Other 95 SGULK08EF Gulkana River - East Fork Other 75 SPAXSO09 Paxson Lake Other 75 SMENT08 Mentasta Lake Other 95 STANA05 Tanada Creek Other 94 STANAO09 Tanada Lake - lower outlet Other 95 STANAS09 Tanada Lake - shore Other 93 SKLUT08 Klutina River Other 95 SKLUTI08.SKLUTI09 Klutina Lake Other 95 SBEARH08 Bear Hole - Klutina Other 94 SBANA08 Banana Lake - Klutina Other 80 SSANN05.SSTACR08 St. Anne Creek Other 186 SMAHL08 Mahlo River Other 94 STONSL09 Tonsina Lake Other 94 SLONGLK05 Long Lake Other 95 STEBA08 Tebay River Other 93 SSTEAM08 Steamboat Lake - Bremner Other 95 SSALMC08 Salmon Creek - Bremner Other 93 SCLEAR07 Clear Creek Other 87 SMCKI07 McKinley Lake07 Other 95 SMCKI08 McKinley Lake08 Other 95 SMCKI91 McKinley Lake91 Other 95 SMCKSC07 McKinley Lake - Salmon Creek Other 93 SMART07.SMART08 Martin Lake Other 187 -continued-

Appendix B.–Page 4 of 7.

ADF&G reporting group Location Reporting Group n SMARTR08 Martin River Slough Other 95 STOKUN08.STOKUN09 Tokun Lake Other 189 SBERI91 Bering Lake Other 95 SKUSH07.SKUSH08 Kushtaka Lake Other 189 SSITU07 Mountain Stream Other 159 SSITU13 Situk Lake Other 190 SOSITU07 Old Situk River Other 163 SLOST03B Lost/Tahwah Rivers Other 93 SAHRN07 Ahrnklin River Other 90 SDANG09 Dangerous River Other 95 SAKWE09 Akwe River Other 95 SEAST03B East Alsek River Other 94 SDATLAS12 Datlasaka Creek Other 95 SGOATC07.SGOATC12 Goat Creek Other 56 SBORD07.SBORD08 Border Slough0708 Other 71 SBORD09.SBORD11 Border Slough0911 Other 70 STWEED07 Tweedsmuir07 Other 48 STWEED09 Tweedsmuir09 Other 46 SVERNR09.SVERNR10 Vern Ritchie Other 114 SNESK07 Neskataheen Lake Other 195 SKLUK06 Klukshu River06 Other 95 SKLUK07 Klukshu River07 Other 94 SKUDW09.SKUDW10. SKUDW11 Kudwat Creek Other 100 SBRIDGE11.SBRIDGE12 Tatshenshini - Bridge/Silver Other 105 SSTINKY11 Tatshenshini - Stinky Creek Other 40 SUTATS03 Upper Tatshenshini Other 95 SLTATS01.SLTATS03 Little Tatshenshini Lake Other 65 SKWAT11 Kwatini River Other 65 SBLAN07 Blanchard River07 Other 89 SBLAN09 Blanchard River09 Other 62 SLTAH90 Tahltan Lake90 Other 95 STAHL06 Tahltan Lake06 Other 196 SPETL04 Petersburg Lake Other 95 SKAHS03 Kah Sheets Lake Other 96 SMILLC07E Mill Creek Weir Early Other 94 SMILLC07L Mill Creek Weir Late Other 95 SKUNK03 Kunk Lake Other 96 STHOM04.STHOM14 Thoms Lake Other 93 SREDBL04 Red Bay Lake Other 95 -continued-

Appendix B.–Page 5 of 7.

ADF&G reporting group Location Reporting Group n SSALM04.SSALM07 Salmon Bay Lake Other 170 SSHIP03 Shipley Lake Other 94 SSARK00.SSARF05 Sarkar Lakes Other 91 SHATC03.SHATC07 Hatchery Creek Other 142 SLUCK04 Luck Lake Other 94 SBIGLK10.SBIGLA14 Big Lake Other 161 SMCDO01.SMCDO03. SMCDO07.SMCDO13 McDonald Lake Other 369 SKART92.SMCGI03. SMCGI04.SMCGI16 Karta River Other 472 SGENE07 Unuk River07 Other 95 SGENE08 Unuk River08 Other 69 SHELM05 Helm Lake Other 94 SHECK04.SHECK07 Heckman Lake Other 189 SMAHO03.SMAHO07 Mahoney Creek Other 154 SKEGA04 Kegan Lake Other 95 SFILLM05 Fillmore Lake Other 52 STHRE04.STHRE10 Klawock - Three Mile Other 181 SINCK03.SINCK08. SHALF08 Klawock - Inlet Creek Other 212 SHETT03.SHETT08. SHETT09L Hetta Lake Other 281 SHETT09M Hetta Creek - middle run Other 95 SHETT10E Hetta Creek - early run Other 95 SEEK04.SEEK07 Eek Creek Other 50 SKLAK04 Klakas Lake Other 95 SBAR04 Essowah Lake Other 95 SHSMI92.SHUGH13 Hugh Smith Other 155 SHUGH04 HS - Buschmann Other 151 SCOBB07 HS - Cobb Creek Other 99 SKWIN01.SKWIN12U Kwinageese Other 76 SBOWS01 Bowser Lake Other 94 SBONN01.SBONN12 Bonney Creek Other 164 SDAMD01 Damdochax Creek Other 93 SMERI01.SMEZIB06 Meziadin Lake Other 186 SHANNA06 Hanna Creek Other 93 STINT06 Tintina Creek Other 94 SGING97 Gingit Creek Other 94 SALAS87.SALAS06 Alastair Lake Other 118 SLAKEL06 Lakelelse Lake Other 93 SSUST01 Sustut River Other 79 SSALIX87.SSALIX88 Salix Bear Other 94 -continued-

Appendix B.–Page 6 of 7.

ADF&G reporting group Location Reporting Group n SMOTA87 Motase Lake Other 47 SSLAM06 Slamgeesh River Other 95 SUBAB06 Babine River Other 95 SFMILE06 Four Mile Creek Other 85 SPINK94.SPINK06 Pinkut Creek Other 187 SGRIZ87 Grizzly Creek Other 76 SPIER06 Pierre Creek Other 95 SFULT06 Fulton River Other 95 SMORR07 Morrison Other 92 SLTAH94 Lower Tahlo River Other 78 STAHLO07 Tahlo Creek Other 95 SMCDON02.SMCDON06 McDonell Lake (Zymoetz River) Other 131 SKALUM06 Kitsumkalum Lake06 Other 56 SKALUM12 Kitsumkalum Lake12 Other 94 SKITW12 Kitwanga River Other 92 SSTECR01 Stephens Creek Other 95 SNANG06 Nangeese River Other 40 SKISP02 Kispiox River Other 53 SSWANLK06 Swan Lake Other 93 SNANI88.SNANI07 Nanika River Other 114 SKYNO97 Trembleur - Kynock Other 94 STACH01 Tachie River Other 94 SSTEL07 Stellako River Other 94 SFRAS96 Fraser Lake Other 85 SMITCH01 Mitchell River Other 94 SLHOR01.SUHOR01. SHORSE07 Horsefly River Other 274 SNAHAT02 Nahatlatch River Other 92 SCULT02 Cultus Lake Other 91 SCHILW04 Chilliwack Lake Other 90 SCHILK01 Chilko Lake Other 87 SRAFT01 Raft River Other 84 SLADA02.SADAM07 Adams River Other 187 SMSHU02 Middle Shuswap River Other 91 SSCOT00 Scotch River Other 91 SGATES09 Gates Creek Other 90 SBIRK07 Birkenhead River Other 90 SWEAV01 Weaver Creek Other 89 SHARR07 Harrison River Other 95 SNTHOM05 North Thompson Other 95 -continued-

Appendix B.–Page 7 of 7.

ADF&G reporting group Location Reporting Group n SNADE95 Naden River Other 95 SYAKO93 QCI - Yakoun Lake Other 70 SKITIM10 Kitimat River Other 93 SBLOOM05 Bloomfield Lake Other 94 STANK03 Tankeeah River03 Other 47 STANK05 Tankeeah River05 Other 47 SAMBA04 Central Coast - Amback Creek Other 91 SKITL06 Kitlope Lake Other 95 SGCENLK02 Great Central Lake Other 95 SQUAT03 Vancouver Island - Quatse River Other 95 SOKAN02 Okanagan River Other 95 SLAKE97 Lake Pleasant Other 89 SISSA96 Issaquah Creek Other 82 SWENA98 Lake Wenatchee Other 95

Appendix C.–Escapement sampling data analysis. The weekly sockeye salmon age-sex distribution, the seasonal age-sex distribution weighted by week, and the mean length by age and sex weighted by week, will be calculated using equations from Cochran (1977). Let h = index of the stratum (week), j = index of the age class,

phj = proportion of the sample taken during stratum h that is age class j,

nh = number of fish sampled in week h, and

nhj = number observed in class j, week h. The age distribution will then be estimated for each week of the escapement in the usual manner,

pˆ hj = nhj nh . (1)

If Nh equals the number of fish in the escapement in week h, Standard errors of the weekly age class proportions will be calculated in the usual manner (Cochran 1977, page 52, equation 3.8),

(pˆ hj )(1− pˆ hj ) SE(pˆ hj )=  [1− nh N h ]. (2)  nh −1  The age distributions for the total escapement will be estimated as a weighted sum (by stratum size) of the weekly proportions. That is, = pˆ j ∑ phj (N h N ), (3) h such that N equals the total escapement. The standard error of a seasonal proportion is the square root of the weighted sum of the weekly variances (Cochran 1977, pages 107–108),

h = 2 2 . (4) SE(pˆ j ) ∑[SE(pˆ hj )] (N h N ) j The mean length, by sex and age class (weighted by week of escapement), and the variance of the weighted mean length, will be calculated using the following equations from Cochran (1977, pages 142–144) for estimating means over subpopulations. That is, let i equal the index of the individual fish in the age-sex class j, and yhij equal the length of the ith fish in class j, week h, so that,

∑(N h nh )∑ yhij ˆ h i Y j = , and (5) ∑(N h nh )nhj h

2  2 ˆ 1 N h (1− nh N h ) 2  nhj  ˆ VˆY  =  (y − y ) + n 1−  y −Y   .  j  ˆ 2 ∑ − ∑ hij hj hj   hj j  N j h nh (nh 1)  i  nh  

Appendix D1.–Example of ADF&G Adult Salmon Age, Sex, Length Form Version 3.0 (ASL) filled out for sockeye salmon samples at Chilkat River fish wheels. 39

Appendix D2.–Definition of user codes on ADF&G Adult Salmon Age , Sex, and Length (ASL) Form Version 3.0, to be used for the fish wheel project. User codea Comment 0 Seal bite 1 (leave blank) 2 Gillnet marked 3 (leave blank) 4 Fungus 5 Escaped but scale sample was taken 6 Mainstem fish (to be used by age lab); based on recapture location 7 Hook wound 8 Marked fish (adipose fin clip or operculum punched) 9 Adipose fin absent (Chinook or coho salmon CWT) a On ASL forms we will use the E (error) column for a missing, injured, or escaped (not scale sampled) fish.

Appendix D3.–Preferred scale sampling area on an adult salmon and proper orientation of scale samples on gum card.

Appendix D4.–District 15 commercial drift gillnet fishery fleet observations.

Gillnet Fleet Survey 2018 Date:______Crew______Page______Boat Count Interviews Numbers Pounds

Subdistrict Boat Tally Time Vessel Site Mesh Kings Reds Coho Pinks Chum Kings Reds Coho Pinks Chum

Total= 42

Total=

Comments: D115 Boat Count=______

Appendix D5.–Chilkat River Fish Wheel Sampling Period Form.

Chilkat Fish Wheels 2018 Date______Weather______

Water Level ______FW#1 RPM______Effort______Changes in Wheel Height______

Water Temp______FW#2 RPM______Effort______Changes in Wheel Height______

Chinook Sockeye Sockeye Coho Coho Chum Chum Dolly V. Other Time Count Chinook Chinook Count Scales Count Coho Scales Pink Count Pink Tissue Count Scales Count Count Other Count AM start/stop (total) AD-Clips Tagged (total) Collected (total) AD-Clips Collected (total) Collected (total) Collected (total) ______

FW #1 /

FW #2 /

subtotal

Chinook Sockeye Sockeye Coho Coho Chum Chum Dolly V. Other Time Count Chinook Chinook Count Scales Count Coho Scales Pink Count Pink Tissue Count Scales Count Count Other Count 43 PM start/stop (total) AD-Clips Tagged (total) Collected (total) AD-Clips Collected (total) Collected (total) Collected (total) ______

FW #1 /

FW #2 /

subtotal

Daily Total Conditions: Notes: Water temp (°C), water level (mm), weather conditions, wheel rotation speed (RPM) and correlating changes in wheel height Sockey, Chum, Pink Sockeye: ASL and 3 scales from 20 fish per day Chum: ASL and 3 scales from 12 fish per day Pink: 3 axillary fin samples per day during stat weeks 27 through 36 Chinook and Coho Data enterd by______Data QC/QA by______Check all Chinook and Coho for adipose fin clips Chinook: Follow all Sport Fish Chinook salmon tagging protocols Enter data into "Fish Wheel data" Check entry into spreadsheet and on Coho: ASL and 3 scales of first 13 fish, length and adipose fin spreadsheet record counts on office office form. presence for all after form Check Chinook forms and scale cards Count all adult salmonids including Morts, Jacks and Escapees

Appendix D6.–Chilkat River Fish Wheel Log Book.

Appendix D7.–Chilkat River Fish Wheel Summary Form.

Stat. Dolly Fish Fish Week Date Chinook Sockeye Coho Pink Chum Varden Wheel 1 Wheel 2 Wheel #1 Wheel #2 # # Mark Total Catch Total Catch Water Water Daily Cum. Tag Cum. Daily Cum. Mark Cum. Type Daily Cum. Daily Cum. Daily Cum. Daily Cum. all species all species Level Temp. RPM Effort RPM Effort Comments 45

Appendix D8.–Chilkat River Fish Wheel Office Form.

Stat. Chinook Tag Sockeye Coho Pink Chum Dolly Water Effort Comments Mark FW notes, Wk Date Daily Cum. Daily Cum. Daily Cum. Marked Cum. Type Daily Cum. Daily Cum. Daily Cum. Daily Cum. Level Temp FW1 FW2 weather etc. 23 3-Jun 4-Jun 5-Jun 6-Jun 7-Jun 8-Jun 9-Jun 24 10-Jun 11-Jun 12-Jun 13-Jun

46 14-Jun

15-Jun 16-Jun 25 17-Jun 18-Jun 19-Jun 20-Jun 21-Jun 22-Jun 23-Jun 26 24-Jun 25-Jun 26-Jun 27-Jun 28-Jun 29-Jun 30-Jun

Appendix D9.–Chilkat Lake Hourly Fish Counts Form.

Chilkat Lake 2018 Record hourly counts of fish passage from DIDSON files. Data entry by______DIDSON Hourly Counts Record gate opening and closure times. QC/QA by______Stat Week 24 Enter data into "Hourly counts" spreadsheet Check transcription to "Hourly Counts" Transfer daily totals to "Daily Counts" spreadsheet Check daily totals in "Daily Counts" Starting week 35 (or when coho present) transfer When coho present, daily totals to "Sonar Apportionment" check "Sonar Apportionment"

6/10 6/11 6/12 6/13 6/14 6/15 6/16 # Fish Gate # Fish Gate # Fish Gate # Fish Gate # Fish Gate # Fish Gate # Fish Gate

0:00

1:00

2:00

3:00

4:00

5:00

6:00

7:00

8:00

9:00

10:00

11:00

12:00

13:00

14:00

15:00

16:00

17:00

18:00

19:00

20:00

21:00

22:00

23:00 Notes:

Daily Total 0 0 0 0 0 0 0

Transfer Daily Totals into "Daily Counts" spreadsheet Starting week 35 (or when Coho present) transfer daily totals to "Sonar Apportionment" spreadsheet

Appendix D10.–Chilkat Lake Daily Counts Spreadsheet. Sockeye Coho Others Water Stat. Scales Week Date Daily Count Cum. Count collected Total Scales Daily Cum. Daily Cum. Level Temp. Visibility comments 23 6/3 0 0 0 0 23 6/4 0 0 0 0 23 6/5 0 0 0 0 23 6/6 0 0 0 0 23 6/7 0 0 0 0 23 6/8 0 0 0 0 23 6/9 0 0 0 0 24 6/10 0 0 0 0 24 6/11 0 0 0 0 24 6/12 0 0 0 0 24 6/13 0 0 0 0 24 6/14 0 0 0 0 24 6/15 0 0 0 0 24 6/16 0 0 0 0 25 6/17 0 0 0 0 25 6/18 0 0 0 0 25 6/19 0 0 0 0 25 6/20 0 0 0 0 25 6/21 0 0 0 0 25 6/22 0 0 0 0 25 6/23 0 0 0 0 48 26 6/24 0 0 0 0

26 6/25 0 0 0 0 26 6/26 0 0 0 0 26 6/27 0 0 0 0 26 6/28 0 0 0 0 26 6/29 0 0 0 0 26 6/30 0 0 0 0 27 7/1 0 0 0 0 27 7/2 0 0 0 0 27 7/3 0 0 0 0 27 7/4 0 0 0 0 27 7/5 0 0 0 0 27 7/6 0 0 0 0 27 7/7 0 0 0 0 28 7/8 0 0 0 0 28 7/9 0 0 0 0 28 7/10 0 0 0 0 28 7/11 0 0 0 0 28 7/12 0 0 0 0 28 7/13 0 0 0 0 28 7/14 0 0 0 0 29 7/15 0 0 0 0 29 7/16 0 0 0 0 29 7/17 0 0 0 0 29 7/18 0 0 0 0 29 7/19 0 0 0 0 29 7/20 0 0 0 0 29 7/21 0 0 0 0

Appendix D11.–Chilkat Lake Seine Recovery Form.

Chilkat Lake 2018 Close gate at night and seine in the AM. Daily, collect 10 sockeye salmon for scales, sex and length. 3 scales/fish. Data entry by______Seine Recovery Use one scale gum card per day. Scale data collected within the stat week recorded on one AWL QC/QA by______Stat Week 24 Punch operculum all sockeye and coho salmon. Count pink and chum if present in seine. Check this field sheet with AWL and scale cards CREW NAMES When coho are present, continue to seine until at least 68 fish are caught Check transcription to "Recovery" Enter data in "Recovery" spreadsheet, transfer #scales daily total to "Daily Counts" spreadsheet Check data transfer to "Daily Counts" Starting week 35 (or when coho present) transfer # sockeye and #coho daily totals to "Sonar Apportionment" When coho present, check "Sonar Apportionment"

Start ______Start ______Start ______Start ______Start ______Start ______Start ______Stop ______Stop ______Stop ______Stop ______Stop ______Stop ______Stop ______6/10 6/11 6/12 6/13 6/14 6/15 6/16 set #sock #scales #coho # sock #scales #coho # sock #scales #coho # sock #scales #coho # sock #scales #coho # sock #scales #coho # sock #scales #coho

AWL Card Number AWL Card Number AWL Card Number AWL Card Number AWL Card Number AWL Card Number AWL Card Number

______

AWL Sample Number(s) AWL Sample Number(s) AWL Sample Number(s) AWL Sample Number(s) AWL Sample Number(s) AWL Sample Number(s) AWL Sample Number(s)

______

Comments: Comments: Comments: Comments: Comments: Comments: Comments:

#pink #pink #pink #pink #pink #pink #pink

#chum #chum #chum #chum #chum #chum #chum

6/10 6/11 6/12 6/13 6/14 6/15 6/16

#sock #scales #coho # sock #scales #coho # sock #scales #coho # sock #scales #coho # sock #scales #coho # sock #scales #coho # sock #scales #coho Daily Total 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 transfer #scales daily total to "Daily Counts" spreadsheet Starting week 35 (or when Coho present) transfer #sock and #coho daily totals to "Sonar Apportionment" spreadsheet

Appendix D12.–Daily DIDSON Counts Form.

Sockeye Sockeye Sockeye Sockeye Coho Others Water Stat. DIDSON visua l sampled Total Esc. Boat Comments Wk Date Daily Cum Daily Cum Daily Cum Total Cum Daily Cum. Daily Cum. Level Temp Vis Gate Drops FW notes, weather etc. 50

Appendix D13.–Chilkat Lake Weir DIDSON/Seine Apportionment Form.

2018 Beach Seine Species Apportionment Apply beach sein proportion to Sonar Counts until a new beach sein sampling occurs Total Daily Stat. Sonar Daily # Daily # Total Daily Total Daily Week Date Count Sock Coho # Sock # Coho example: 100 30 10 75 25 <=== enter this in the "Daily Counts" spreadsheet 35 8/27 35 8/28 35 8/29 35 8/30 35 8/31 35 9/1 35 9/2 36 9/3 36 9/4 36 9/5 36 9/6 36 9/7 36 9/8 36 9/9 37 9/10

51 37 9/11

37 9/12 37 9/13 37 9/14 37 9/15 37 9/16 38 9/17 38 9/18 38 9/19 38 9/20 38 9/21 38 9/22 38 9/23 39 9/24 39 9/25 39 9/26 39 9/27 39 9/28 39 9/29 39 9/30 40 10/1 40 10/2 40 10/3 40 10/4 40 10/5

Appendix D14.–Limnology Sampling Form.

LAKE SURVEY - Field Form Lake ID Date: Sampling Station: Sampling Depth (m): Lake Elev (m): Weather / Surface Conditions: Observers:

Physical and Chemical Parameters Parameter profile: to______Hrs DO/temp meter used: Light intensity profile: to______Hrs Light meter used:

Dissolved Oxygen (mg/L) Temperature (C) Light Intensity % O2 Winkler Foot Candles meter Meter Depth saturation titration Depth Depth (Up looking) Multiplier Above Above Above Surface Surface Surface 5 cm 5 cm 5 cm

0.5 m 0.5 m 0.5 m 1.0 m 1.0 m 1.0 m

1.5 m 1.5 m 1.5 m

2.0 m 2.0 m 2.0 m

2.5 m 2.5 m 2.5 m

3.0 m 3.0 m 3.0 m

3.5 m 3.5 m 3.5 m

4.0 m 4.0 m 4.0 m

4.5 m 4.5 m 4.5 m

5.0 m 5.0 m 5.0 m

5.5 m 5.5 m 5.5 m

6.0 m 6.0 m 6.0 m

6.5 m 6.5 m 6.5 m

7.0 m 7.0 m 7.0 m

7.5 m 7.5 m 7.5 m

8.0 m 8.0 m 8.0 m

8.5 m 8.5 m 8.5 m

9.0 m 9.0 m 9.0 m

9.5 m 9.5 m 9.5 m 10.0 m 10.0 m 10.0 m 10.5 m 10.5 m 10.5 m 11.0 m 11.0 m 11.0 m 11.5 m 11.5 m 11.5 m 12.0 m 12.0 m 12.0 m 12.5 m 12.5 m 12.5 m 13.0 m 13.0 m 13.0 m 13.5 m 13.5 m 13.5 m 14.0 m 14.0 m 14.0 m 14.5 m 14.5 m 14.5 m 15.0 m 15.0 m 15.0 m -continued-

Appendix D14.–Page 2 of 2.

Lake ID Date: Sampling Station: Sampling Depth (m): Dissolved Oxygen (mg/L) Temperature (C) Light Intensity % O2 Winkler Foot Candles meter Meter Depth saturation titration Depth Depth (Up looking) Multiplier 16.0 m 16.0 m 16.0 m 17.0 m 17.0 m 17.0 m 18.0 m 18.0 m 18.0 m 19.0 m 19.0 m 19.0 m 20.0 m 20.0 m 20.0 m 25.0 m 25.0 m 25.0 m 30.0 m 30.0 m 30.0 m 35.0 m 35.0 m 35.0 m 40.0 m 40.0 m 40.0 m 45.0 m 45.0 m 45.0 m 50.0 m 50.0 m 50.0 m

Secchi Disk Depth disc disappears / reappears, DOWN / UP meters Time: hrs Lake surface conditions

Water Samples Collected for Lab Analysis Sample Depth Date Time (hrs) Bottle #

Biological Parameters Zooplankton Station Station Time of Sample End: : Time of Sample End: : Time of Sample Start: : Time of Sample Start: : Elapsed time (min:sec) : Elapsed time (min:sec) : Tow depth: m Tow depth: m

Station Station Time of Sample End: : Time of Sample End: : Time of Sample Start: : Time of Sample Start: : Elapsed time (min:sec) : Elapsed time (min:sec) : Tow depth: m Tow depth: m

Phytoplankton - Phaeophytin - Chlorophyll samples

Time set Date and time Date Depth (m) Chl a (hrs) filtered