5 2' a: S x Em. . $555,? fine? 9. n. 2.. Aw I: a... I. what... ~11») i .. i‘l . 6 3.4... we? x. i. ans S 1!: .ix 1’11?) :1 6 I This is to certify that the thesis entitled Relative Contribution and Comparative Life History Characteristics of Hatchery and Wild Steelhead Trout in the Betsie River, Michigan presented by James R. Harbeck has been accepted towards fulfillment of the requirements for M.S. degree in Fish. & Wildl. Major professor Date June 30, 1999 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution ._ _ _ RELATIVE CONTRIBUTION AND COMPARATIVE LIFE HISTORY CHARACTERISTICS OF HATCHERY AND WILD STEELHEAD TROUT IN THE BETSIE RIVER, MICHIGAN By James R. Harbeck A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1 999 Dr. Thomas G. Coon ABSTRACT RELATIVE CONTRIBUTION AND COMPARATIVE LIFE HISTORY CHARACTERISTICS OF HATCHERY AND WILD STEELHEAD TROUT IN THE BETSIE RIVER, MICHIGAN By James R. Harbeck Spawning runs of wild, naturalized steelhead trout occur in northern tributaries of Lake Michigan including the Betsie River. Management of Betsie River steelhead has focused on supplementing wild stocks with hatchery fish though little is known about natural production. Scales from adult migrants were collected by volunteer anglers and river guides from 1994 through 1996. Relative contribution and life history characteristics were determined through scale pattern analysis. Management agencies surrounding Lake Michigan were contacted to determine the presence of uniquely marked Betsie River hatchery steelhead in their waters. The spawning runs of 1994-1996 were composed of 46, 40, and 30 percent wild steelhead respectively. The most common life history pattern in the wild population was 2 years of growth in the stream followed by 3 years in the lake. Eighteen percent had spawned previously. The most common pattern in the hatchery population was 1 year in the hatchery/stream and 3 lake years. Ten percent had spawned previously. Wild and hatchery steelhead had similar sex ratios, length-at-age and migration timing. Betsie River hatchery steelhead were widely distributed in the southern two thirds of Lake Michigan. Straying was evident in rivers on both sides of the lake as far south as the St. Joseph River in Indiana and the Root River in Wisconsin. To my wife Judy, for your patience, encouragement, and help throughout my graduate studies. iii ACKNOWLEDGMENTS I sincerely thank my advisor and major professor, Thomas Coon, for his wise insight, guidance, and encouragement. My deepest appreciation is owed to him for seeing me through the initiation and completion of this project. I also thank the other members of my committee, James Bence and Gary Mittelbach for their valuable suggestions that enhanced the integrity of my thesis. Gratitude is due to Paul Seelbach of the Institute of Fisheries Research. His professional instruction in steelhead scale reading greatly hastened my own learning curve. I am indebted to the scale reading efforts and expertise of Steve Vanderlaan and Sally Markham, MDNR fisheries technicians and to Doug Workman, my fellow lab partner. I have also benefited from the friendships, discussions, and assistance of my other lab mates, Tammy Newcomb and John Skubinna. A research project such as this would not be possible without the help of numerous Betsie River volunteer anglers and river guides. My special thanks to John Westley who not only provided the majority of the steelhead scales but also introduced me to the Betsie and her fishery via his drift boat. I’m also grateful to the biologists of the Illinois Department of Natural Resources, the Illinois Natural History Survey, the Indiana Department of Natural Resources, the Michigan Department of Natural Resources, the Wisconsin Department of Natural iv Resources and the Chippewa/Ottawa Treaty Fishery Management Authority for providing me with their fin clip data. Funding and logistic support for this endeavor was provided by the Department of Fisheries and Wildlife at Michigan State University, the Michigan Department of Natural Resources Fisheries Division, the Fred Waara Chapter of Trout Unlimited, and the Michigan Polar-Equator Club. TABLE OF CONTENTS LIST OF TABLES ....................................................................................................... viii LIST OF FIGURES ............................... x INTRODUCTION ....................................................................................................... l Ecological Background ..................................................................................... 1 Relevant Literature ........................................................................................... 2 Steelhead Life History ...................................................................................... 4 Steelhead in the Betsie River ............................................................................. 6 Goals and Objectives ........................................................................................ 9 Study Site Description ..................................................................................... lO METHODS ................................................................................................................. . 16 Data Collection ................................................................................................ 16 Scale Analysis .................................................................................................. 17 Origins ............................................................................................................. 18 Life History Characteristics .............................................................................. 21 Smolt Size Influence ......................................................................................... 24 RESULTS ................................................................................................................... 27 Origins ............................................................................................................. 27 Life History Characteristics .............................................................................. 30 Aging Precision ..................................................................................... 30 Age Structure and Composition ............................................................ 30 Repeat Spawning Frequency ................................................................. 38 Sex Ratio .............................................................................................. 42 Length and Growth ............................................................................... 42 Migration Timing .................................................................................. 52 Smolt Size Influence ......................................................................................... 56 Influence On Return .............................................................................. 56 Influence On Age At Maturity ............................................................... 63 vi Lake Distribution And River Straying ............................................................... . 66 Lake Distribution .................................................................................. 66 River Straying ....................................................................................... 67 DISCUSSION ............................................................................................................. . 71 Origins ............................................................................................................ . 71 Accuracy ............................................................................................... 71 Composition of Spawning Runs ............................................................ 72 Life History Characteristics ............................................................................... 74 Age Structure and Composition ............................................................ 76 Repeat Spawning Frequency ................................................................. 78 Sex Ratio .............................................................................................. . 78 Length and Growth .............................................................................. . 79 Migration Timing .................................................................................. 83 Aging Precision ..................................................................................... 83 Smolt Size Influence ....................................................................................... . 84 Influence On Return .............................................................................. 84 Influence On Age At Maturity ............................................................... 87. Lake Distribution And River Straying ............................................................... 89 Lake Distribution .................................................................................. . 89 River Straying ....................................................................................... . 90 CONCLUSIONS AND RECOMMENDATIONS ........................................................ 92 APPENDICES ............................................................................................................. . 96 LITERATURE CITED ............................................................................................... . 1 12 vii LIST OF TABLES Table 1. Twenty year stocking record of steelhead in the Betsie River, 1975-1995 ............................................................................................ 7 Table 2. Distribution of sampled sport catch by river section, 1994 - 1996. The kelt sample is not included .................................................. 28 Table 3. Estimated stream ages and associated APE, CV, and D from three independent readers. The fifty scale subsample taken from the 1994 -1996 Betsie River collection .................................................................. 31 Table 4. Estimated lake ages and associated APE, CV, and D from three independent readers. The fifty scale subsample taken from the 1994 -1996 Betsie River collection .................................................................. 32 Table 5. Estimated total ages and associated APE, CV, and D from three independent readers. The fifty scale subsample taken from the 1994 -l996 Betsie River collection ................................................................ 33 Table 6. Lake age structure of maiden adult steelhead returning to the Betsie River, 1994 - 1996 .............................................................................. 39 Table 7. Distribution of age at time of smolting for wild adult steelhead according to year of return ............................................................................. 40 Table 8. Frequency of repeat spawning incidence in wild and hatchery steelhead returning to the Betsie River, 1994—1996 ......................................... 41 Table 9. Sex ratios for adult wild and hatchery steelhead according to return year. Asterisk indicates a significant difference (P < 0.05) from a 1M: lF sex ratio ............................................................................................... 43 Table 10. Analysis of variance tests for year and origin effects on steelhead length, 1994-1996 ......................................................................................... 47 viii Table 11. Mean total length (cm) by age class and sex for wild and hatchery adult steelhead from the Betsie River, 1994 - 1996 ....................................... 49 Table 12. Comparison of mean total length (cm) by sex for maiden wild and hatchery steelhead. Significant length differences denoted by asterisk (* P< 0.05; **P< 0.01; ***P< 0.001; NS = no significant difference) .................................................................................................... 53 Table 13. Comparison of mean total length (cm) by spawning history for wild and hatchery steelhead. Significant length differences denoted by asterisk (* P< 0.05; **P< 0.01; ***P< 0.001; NS = no significant difference) .................................................................................................... 54 Table 14. Comparison between adult smolt check radius (mm) and smolt scale radius (mm) of wild and hatchery steelhead according to smolt year. Significant scale radius differences denoted by asterisk (* P< 0.05; **P< 0.01; ***P< 0.001; NS = no significant difference) .............................. 60 Table 15. Observed smolt lengths (OL) and back-calculated smolt lengths (BL) from adults in relation to the smolt cohort year, the stream and lake age, the sex, and the year of return with all cohorts pooled .................... 64 Table. 16. Summary of Similarities and distinctions in life history traits of Betsie River wild and hatchery steelhead ..................................................... 75 Table 17. Marked hatchery strays and probable origin from the Betsie River fishery 1994-1996 and the 1996 kelt sample ........................................ 97 Table 18. Relative composition in 1996 of wild and hatchery steelhead by river section and gear ................................................................................... 98 Table 19a Preliminary data of Betsie River steelhead collected from the 1994 sampled fish ................................................................................................ 99 Table 19b Preliminary data of Betsie River steelhead collected from the 1995 sampled fish ................................................................................................ 104 Table 19c Preliminary data of Betsie River steelhead collected from the 1996 sampled fish ................................................................................................ 107 ix Figure 1. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Figure 8. Figure 9. LIST OF FIGURES The Betsie River watershed located in the northwestern lower peninsula. Key cultural features are detailed in the text ................. Stocking record for steelhead planted in the Betsie River. Orsini Hatchery began operations in 1991 ..................................... Scale from a wild steelhead showing measurement area for Ratio 23 determination .............................................................................. Image of a steelhead scale from a 4 year old fish (2.2) collected in the spring of 1995. The wild steelhead spent 2 years in a stream environment and 2 years in Lake Michigan .............................................. Relative contribution of hatchery and wild steelhead to the Betsie River fishery, 1994-1996. The 1996 kelt population was sampled with blocking nets and electrofishing gear. The 1984 sample is included for comparison ............................................................ Age class structure of wild and hatchery steelhead from the Betsie River, 1994 ..................................................................................... Age class structure of wild and hatchery steelhead from the Betsie River, 1995 ..................................................................................... Age class structure of wild and hatchery steelhead from the Betsie River, 1996 ..................................................................................... Proportions of hatchery and wild maiden and repeat spawners according to lake age, 1994-1996 ............................................................. Figure 10. Sex ratios by lake age in samples of wild and hatchery steelhead returning to the Betsie River during 1994-1996 ............................ 44 ...... ll 15 19 22 29 34 35 36 37 Figure 11. Figure 12. Figure 13. Figure 14. Figure 15. Figure 16. Figure 17. Figure 18. Figure 19. Figure 20. Figure 21. Length frequency distributions of wild and hatchery steelhead by return year. Length increments are grouped in two centimeter intervals ...... 45 Mean total length at lake age for wild and hatchery steelhead by return year, 1994 - 1996 .......................................................................... Mean length at maturity (cm) of maiden wild and hatchery steelhead according to sex returning to the river during 1994 - 1996. Stream ages pooled for wild steelhead at each lake age ............... Mean length at maturity (cm) according to stream age of wild adult steelhead returning to the Betsie River during 1994 -l996 ............ Walfort Plots for wild and hatchery steelhead and their calculated growth coefficients (K) and asymptotic lengths (La) ............ Migration timing based on the weekly catch at Homestead weir from March through May ................................................................ Migration timing by sex of wild and hatchery steelhead based on average weekly catch at the Homestead weir from March through May, 1994-1996 ....................................................................................... Frequency distributions of observed lengths from migrating hatchery smolts measured at the Homestead weir and back- calculated smolt lengths of surviving adult steelhead from the same cohort. Smolts were also measured at the hatchery prior to release in 1994 ............................................................................ Frequency distributions of observed lengths from migrating wild smolts measured at the Homestead weir and back- calculated smolt lengths of surviving wild adult steelhead from the same cohort ........................................................................................ Relationship between mean back—calculated length at smolting and lake age of adult steelhead from the Betsie River, 1994-1997 .......................................................................... Distribution in Lake Michigan of steelhead stocked as smolts in the Betsie River during 1993 -l995. RV clipped steelhead recovered in 1994 - 1996 .......................................................... xi 48 ...... 50 ...... 51 ...... 55 57 58 61 62 65 69 Figure 22. River Straying by steelhead Stocked as smolts in the Betsie River during 1993 - 1995. RV clipped adults recovered in 1995 and 1996 ............................................................................................ 70 xii INTRODUCTION The steelhead trout, Oncorhynchus mykiss, is an anadromous form of the rainbow trout. It is also an iteroparous salmonid in contrast to the other species within the genus which are semelparous (Leider 1985). Life history traits vary considerably among populations of this species. Diverse age structure, growth, sex ratios, age at first maturity, percent repeat Spawn, spawning and migration timing, and homing fidelity have been documented in discrete stocks of steelhead (Withler 1966; Shapovalov and Taft 1954). Although native to the Pacific Coast, the steelhead has been introduced worldwide including into the Great Lakes (MacCrimmon 1971). Its successful adaptation is due, in part, to the steelhead’s variable life history characteristics. Biette et a1. (1981) believe that steelhead populations in the Great Lakes retained their variable characteristics from their native range that allowed for the development of discrete stocks peculiar to regions and perhaps specific watersheds. Krueger et a1. (1994) found separate, genetically identifiable steelhead populations occurring in the tributaries along the Minnesota Shoreline of Lake Superior. Ferguson et al. (1993) documented both life history and genetic differences among steelhead populations in Lake Ontario. This ecologically adaptive variation in life history minimizes impacts on a population by spreading risks to multiple year classes in an unpredictable environment. Successful colonization of different watersheds thus becomes a possibility. 2 The State of Michigan obtained its first rainbow trout eggs from California in 1880 and began stocking many of its Great Lakes tributaries (Latta 1974). Populations became wild and reproduced naturally. The steelhead trout now occupies a unique ecological niche in Michigan waters (Jude et a1. 1987). Runs of naturalized steelhead presently occur in numerous northern tributaries of Lake Michigan including the Betsie River in northwest lower Michigan. This introduction has also formed the basis of a popular and economically valuable sport fishery in Michigan. Recreational angling in Michigan accounts for an estimated 1.3 billion dollars in expenditures annually (Garling and Dann 1995). Michigan anglers spend more money per day trip pursuing steelhead than any other gamefish (Mahoney et a1. 1991). Therefore, for both ecological and economical reasons, the state has a vested interested in the proper management of the steelhead trout. Relevant Literature According to Allendorf et a1. (1997), the variation in salmonid life history traits helps to maintain genetic viability within a species. They recommend that stocks be managed as discrete identities and given priority when traits and the underlying genetics are unique. MacLean and Evans (1981) also advocate the identification and preservation of each individual stock and their characteristics as a primary management goal. Various methods of stock identification from mixed samples were reviewed by Ihssen et a1. (1981). They identified the analysis of scales as a valid technique for distinguishing stocks. Scale pattern analysis is currently used to discriminate 3 commercially valuable stocks of fish originating from different environmental conditions. Stocks of Alaskan sockeye, Oncorhynchus nerka, are separated using techniques developed by Bethe and Krasnowski (1977). Pattern classification was also used to identify wild and hatchery stocks of Columbia River chinook salmon, Oncorhynchus tshawytscha (Schwartzberg and Fryer 1989) and wild and hatchery chinook from the Rakaia River, New Zealand (Unwin and Lucas 1993). Discrimination of Norwegian farmed, ranched and wild Atlantic salmon, Salmo salar, was successful using circuli spacing and scale texture data (Friedland et. a1. 1994). Techniques that used life history information resulted in higher classification accuracy than those that did not (Davis 1987). Knudsen and Davis (1985) also found that classification models that compared circuli band widths (ratio data) had higher accuracy than models with no ratio data. Seelbach and Whelan (1988) developed a scale analysis technique to identify wild and hatchery steellhead trout that relied on both life history and ratio information. The authors recognized a difference in circuli spacing between wild and hatchery steelhead from the Great Lakes. Circuli on scales from hatchery fish were evenly spaced across the portion of scale laid down during the time the fish resided in the hatchery. The pattern reflected the constant environmental conditions of the hatchery. Circuli from wild fish scales were closely spaced prior to the first stream annulus and widely spaced after the annulus. This pattern paralleled the slow growth of winter and faster growth of spring experienced by the wild fish. Based on these differences, the two researchers developed an objective assignment rule for distinguishing wild and hatchery steelhead. 4 Estimating smolt-to-adult survival rates of Vancouver Island steelhead was accomplished by Ward and Slaney (1988) and Ward et al. (1989) by using scale backcalculation procedures. Smolt lengths back calculated from adult scales were compared to observed smolt lengths. Survival based on smolt size was then determined. A similar scale technique was used by Seelbach et al. (1994) for steelhead from two Michigan rivers. Scale radii frequencies were compared between smolts and adults of the same cohort to determine the effect of smolt size on return. Steelhead Life History Steelhead generally make spawning runs in the spring to their Stream of origin. Successful reproduction depends on suitable habitat. Redd sites with gravel diameters of 1.3 cm to 11.4 cm and well oxygenated flows with current velocities of 23-155 cm/sec are chosen by the female (Pauley et al. 1986). The majority of steelhead live to spawn only once. Those fish that do spawn more than once are predominantly female, even though male/female ratios of maiden spawners are close to 1:1 (Leider 1985). The disproportionate survival of females after spawning is explained by McKeown (1984) as having a behavioral and hormonal basis. Male steelhead tend to enter the spawning stream earlier than females and remain longer. Defending the redd has its energy costs and little food is consumed once the males enter the stream (Shapovalov and Taft 1954). Fecundity varies with size. Females lay and bury between 1,000-12,000 eggs (Moyle and Cech 1988; Pauley et al. 1986). Eggs hatch in 4-7 weeks depending on water 5 temperature. When the yolk sac is absorbed, fry emerge from the gravel and develop bars or parr marks on their sides. The “parr” then spend 1-3 years in their natal stream (Biette et al. 1981; Seelbach 1993). In preparation for a change in habitat, parr begin a transformation in morphology, physiology, and behavior (Hoar 1976) until they become smolts. The new smolts are silvery, more elongated than parr, and are migratory in nature. Smoltiflcation is size dependent and its timing is influenced by temperature, photoperiod, and discharge (Damsgard 1991; Seelbach 1987; Stauffer 1972). Upon smolting, steelhead emigrate out to a lake or ocean where they grow and mature for 2 or 3 years in preparation for their own spawning runs. Juveniles primarily select aquatic and terrestrial invertebrates as food items. In Lake Michigan adult steelhead prey on alewife Alosa pseudoharengeus and rainbow smelt Osmerus mordax, although not to the degree of other salmonids (Rand et al. 1993; Jude et al. 1987). Invertebrates are also a major diet component for adults. Populations exhibit a wide range of variation within this general life history pattern. Biette et al. (1981) identified 18 different age categories in Great Lakes spawning populations. Withler ( 1966) also found high variability in traits of steelhead along the Pacific Coast. Percentage of repeat spawning and age at first maturity differ with strain and latitude of the natal stream. Kwian (1981), Seelbach (1993), and Biette et al. (1981) found the duration of stream residency to range from 0-4 years among populations. Migration tendency and homing accuracy vary between strains of both wild and hatchery steelhead (Steward and Bjomn 1990). In a Lake Superior stream Hansen and Stauffer (1971) found evidence of extensive straying of hatchery fish as did Seelbach and Miller 6 (1993). But when comparing hatchery strains stocked in southern Michigan streams, Seelbach et al. (1994) found good homing fidelity in both strains when stocked at upstream sites. Steelhead in the Betsie River As with many rivers in Michigan, management of steelhead in the Betsie River historically focused on supplementing naturalized stocks with hatchery fish (Wicklund and Dean 1958; Hansen and Stauffer 1971). According to contemporary researchers, the past contribution of hatchery fish to the fishery was probably negligible (Seelbach 1987). The adult Steelhead run in the Betsie was sampled in 1984. From a sample of 58 specimens, 93 percent were judged to be wild, naturally produced fish. Only 7 percent of the run were thought to be from hatchery steelhead (Seelbach and Wheland 1988). However, two major events with the potential to alter the relative contributions of both hatchery and wild fish, have occurred within the watershed. First, the stocking program for the Betsie has changed (Table 1). The location at which fish are stocked was moved from the mouth of the river near Frankfort to an upstream location. Initially the release site was moved upstream to River Road (1989-1991). Currently, fish are released directly from the Orsini Hatchery which is even further upstream. Stocking location has implications for smolt survival and homing ability. Steelhead smolts stocked at an upstream site develop higher homing fidelity than smolts planted lower in the watershed (Steward and Bjomn 1990), but they are exposed to river predators and other unfavorable environmental conditions (Ward and Slaney 1990). Therefore, smolts stocked 7 Table 1. Twenty year stocking record of steelhead in the Betsie River, 1975-1995. Mean Total Year Planting Site Strain Number Length (mm) Clip 1975 Frankfort L.Manistee 10,044 (558 per #) None 1976 Frankfort L.Manistee 10,260 (540 per #) None 1977 Frankfort L.Manistee 12,170 (307 per #) None 1978 Frankfort L.Manistee 15,206 (1,152 per #) None 1979 Frankfort L.Manistee 0 - - 1980 Frankfort L.Manistee 20,000 104 None 198 1 Frankfort L.Manistee 20,004 145 None 1982 Frankfort L.Manistee 15,000 81 None 1983 Frankfort L.Manistee 23,359 1 12 None 1984 Frankfort L. Manistee 15,000 154 None Frankfort Rogue 8 .000 l 66 LPRV 1985 Frankfort L. Manistee 13,000 160 None 1986 Frankfort L. Manistee 20,001 154 AD 1987 Frankfort Skamania 17,500 196 DOAD 1988 Frankfort Skamania 15,000 198 MTAD 1989 Black Bridge L. Manistee 100,875 fry 24 None River Road L. Manistee 15,000 YS 175 MTAD Black Bridge L. Manistee 15,000 FF 81 None 1990 River Road L. Manistee 10,000 175 None River Road Skamania 10,000 199 None 1991 Orsini Hatchery L. Manistee 29,17 1 174 None River Road Skamania 10,000 195 None 1992 Orsini Hatchery L. Manistee 32,141 204 None 1993 Orsini Hatchery L. Manistee 44,125 152 RV 1994 Orsini Hatchery L. Manistee 48,561 137 RV 1995 Orsini Hatchery L. Manistee 50,036 150 RV 8 upstream experience lower riverine survival but return to the natal river with less straying (Seelbach et al. 1994). Numbers of steelhead stocked annually have increased from an average of 12,000 per year( 1975- 1984) to 50,000 in 1995 (Anonymous 1975 - 1995). Intuitively, increased smolt numbers would contribute more to the adult population. Yearling smolts are now stocked instead of the formerly planted parr. In other systems, large sized hatchery smolts were proven to have greater survival rates and yield better adult returns than hatchery parr (Seelbach 1987). The second factor affecting the steelhead population is a physical change within the watershed itself. In 1989, the Thompsonville Dam failed and was not rebuilt. This event opened additional spawning and rearing habitat for anadromous fish including a previously unavailable coldwater tributary, the Little Betsie River. However, some habitat downstream of Thompsonville was compromised by sediments from the washed out impoundment. Wicklund and Dean (1958) identified the Thompsonville Dam as a major cause of warm, summer temperatures found in the mainstream that exceed steelhead tolerances. Since 1989, therefore, the potential for production of wild, naturalized steelhead has increased in the Betsie watershed. These two factors, one a habitat change, the other a change in hatchery inputs, probably changed the structure of Betsie River steelhead population. As a result the relative contributions of hatchery and wild steelhead to the fishery were unknown as were their respective life history traits. Knowledge of steelhead run composition (hatchery/wild) is important information when determining proper stocking recommendations and prescriptions. Life history 9 information is also invaluable. Because local wild population characteristics are a result of local environmental pressures and genetics, hatchery fish characteristics should resemble those of the wild population in order to be successful (Steward and Bjomn 1990). In addition, the more different the life histories of the two populations, the more likely the wild stock will suffer genetic loss of fitness when spawning conjointly with hatchery fish (Helle 1981). As an important element of resource management, the Michigan Department of Natural Resources’ Management Plan (1997) addressed the need for determining the contribution of naturally produced salmonids to the Great Lakes fisheries. Furthermore, the Great Lakes Fishery Commission (GLFC 1992) also recommends that management practices should be directed towards a naturally sustainable fishery and the preservation of wild populations. Hatchery programs should be tailored to avoid erosion of wild stock integrety. Field evaluation of hatchery stock performance is also an essential component of resource management (MDNR 1994). Therefore, these types of data are essential to the future management of the Betsie River steelhead population. Goals and Objectives The goal of this study was to evaluate the Betsie River adult steelhead population as it now exists in response to a major habitat change and a change in the hatchery stocking regime. The resulting information is intended to aid in the management of the steelhead and its fishery by providing evidence of population composition and structure. 10 The specific research objectives include: 1. Estimate the relative contribution of hatchery and wild adults to the spawning population from the 1994-1996 runs and compare this percentage with that of the 1984 study. 2. Describe and compare wild and hatchery stock characteristics (age structure, growth, sex ratios, age at first maturity, migration timing, and percent repeat spawners). 3. Evaluate smolt size influences on probability of adult return. 4. Document the lake distribution and river straying of hatchery fish. For the purposes of this thesis, the term “wild” steelhead refers to those fish which have resulted from natural reproduction regardless of ancestry (i.e. hatchery, wild, or interbred). “Hatchery” steelhead refers to those fish which have been spawned and reared for some part of their life cycle in a hatchery regardless of lineage. Study Site Description The Watershed - The Betsie River is a Lake Michigan tributary located in northwestern Michigan (Figure l). Originating in Duck and Green Lakes southwest of Traverse City, the Betsie flows in a westerly direction to its mouth in Lake Michigan at Frankfort. The river is approximately 82 km long and drains an area of 67,149 hectares in Grand Traverse, Manistee, and Benzie counties. Over fourteen thousand hectares within the watershed are state—owned as part of the Pere Marquette State Forest and the Betsie River State Game Area. In 1973 the Betsie River was listed in the state’s Natural Rivers Program. 11 .38“ 2: E 3:83 Ea 833$ 35:3 .33 8.35:3 832 Eouwoanto: 05 E @832 @2225.» 83m 380m 2:. A 8sz 3 m o — h _ Ex 95 Ema 2:283:85. tense: EEO Loam Egom 2:3 xooco :mo , Ema Ema 9.3 $90 ammumoEoI . ;m_:om moxmn xozo w 590 5.9522 Z toicm. u uefilumw axe-I 12 The river flows through the Highland, Newaygo, and Manistee regional landscape ecosystems which is reflected in a patchwork of geologic features (Albert et al. 1986). The watershed is characterized by a topography of gently rolling hills, several glacial moraines and outwash plains. Soils are primarily classified within the Rubicon- Grayling series. This association is dominated by sandy soils interspersed with loamy soils. Therefore the soil is very permeable and has low water holding capacity. Approximately 60% of the watershed is forested by northern hardwoods. Coniferous forests are restricted to poorly drained areas of outwash channels (Albert et al. 1986). The Betsie River discharges an estimated 8.5 m3 /s at Thompsonville Road (Anonymous 1970) and between 4.0 m3 /s and 12.8 m3 Is at Homestead Darn throughout the year (Newcomb 1998). Water quality is considered good and is comparable to the Pere Marquette and Boardman Rivers (Hartig and Stifler 1978). Nutrient levels are low. Nitrate and phosphate concentrations are typically 0.10mg/L or less. River water sampled at Thompsonville had a pH of 8.2 and an alkalinity (CaCO 3 ) of 135mg/L (Anonymous 1970). The river has been modified by the existence of three dams. The first upstream from Lake Michigan is the Homestead Dam, lying 20 km above the mouth. Prior to 1972, the dam prevented the passage of anadromous fish. Salmonids were trapped and physically carried over the barrier and allowed to continue their spawning migration (W icklund and Dean 1958). In 1972, the dam was modified into a low head lamprey weir. Strong swimming fish like steelhead are now able to negotiate the weir on their own. l3 Thirty-six kilometers upstream from Homestead is the site of the former Thompsonville Dam. The dam maintained a head of 3 meters and prevented upstream fish passage. Since its failure in March, 1989, migratory fish are no longer excluded from the upstream portion of the watershed. The mouth of the Little Betsie River, the largest tributary in the system, is 0.5 km above this site. Six kilometers below the outlet at Green Lake is the Grass Lake Dam. Originally established for logging operations, the Grass Lake Wildlife Flooding is now managed for waterfowl. The dam maintains a head of l- 2 meters. The Betsie is classified as a marginal trout stream by the Michigan Department of Natural Resources (Wicklund and Dean 1958). However, typical of a varied temperature regime, the watershed supports a diverse fish community including many game species. In addition to steelhead, coho salmon, Oncorhynchus kisutch, and chinook salmon make spawning runs and produce smolts annually. Resident salmonids, such as brook trout, Salvelinusfontinalis, and brown trout, Salmo trutta, are common in coldwater Dair Creek and the Little Betsie River. Walleye, Stizostedion vitreum, northern pike, Esox lucius, and white sucker, Catostomus commersoni, predominate in the lower river. Northern pike and centrarchids also inhabit Grass Lake (Carbine 1945). The Hatchery - the Orsini Hatchery is dedicated solely to the production of steelhead smolts. The 2100 sq. ft. building is located on the river between Homestead Dam and the Thompsonville Dam site, 52 km upstream from Lake Michigan. An artesian well releases 600 liters per minute of 9 ° C water through the hatchery. This privately owned facility is sponsored and maintained by the Manistee County Sport Fishing 14 Association with contributing funds donated by the Michigan Department of Natural Resources and other sport fishing groups. Steelhead smolts are released directly into the river from the hatchery near highway M-1 15. Numbers released have increased from 29,000 in 1991 to 55,000 in 1996 and account for the dramatic rise in stocking levels (Figure 2). The Orsini Hatchery fish are first generation offspring of wild parents from the Little Manistee River. Spawners are obtained by MDN R personnel at the Little Manistee River weir facility. The fertilized eggs are hatched at the state owned Wolf Lake Hatchery. Fingerlings are then transported to the Orsini Hatchery where they are reared to smolt size yearlings and released into the Betsie River. Numbers of Steelhead Planted 60000 50000 40000 30000 20000 10000 15 Orsini Hatchery Production ‘T‘To'r'r. 82 83 84 85 86 87 88 89 90 9192 93 94 95 96 Year of Stocking Figure 2. Stocking record for steelhead stocked in the Betsie River. Orsini Hatchery began operations in 1991. METHODS Data Collection I assessed the adult steelhead population primarily through scale analysis along with other data collected from individual fish. Adult scale samples were obtained through a collection program by river guides and volunteer anglers fishing the Betsie River from 1994 - 1996. Guides and anglers were supplied with envelopes for the return of the scale samples. Scales were removed from the preferred area between the posterior edge of the dorsal fin and the anterior edge of the anal fin above the lateral line (Scamecchia 1979-, Knudsen and Davis 1985). Anglers were asked to record total length, sex, date, location of capture, and the presence of any fin clips (Appendix A). In order to describe the distribution of sport caught steelhead, I divided the segment of the river open to fishing during the extented season into three sections. The section from Highway 31 down to the mouth at Frankfort was termed the lower section, the river from the Homestead weir to Highway 3 lwas labeled as the middle section, and the area above the Homestead weir was considered the upper section. In 1996 the number of scales collected by guides and volunteer anglers was supplemented by sampling emigrating kelts caught in a block net above Homestead Dam. As part of a concurrent smolt study (Newcomb 1998), the blocknet was designed to capture smolts but also captured adults. The constricted area of the block net was 16 l 7 electrofished for three nights in May during the emigration of spent adults. In 1997 guides collected scales fi'orn marked hatchery fish only. A target sample size for each year was set at 200. For scale pattern analysis on an “unknown” mixed stock, a sample size of 100 is recommended by Conrad (1985) when a sample of 200 “knowns” is obtained for each stock. The criteria for distinguishing between wild and hatchery stock of Lake Michigan steelhead was developed by Seelbach and Whelan (1988) with “known” wild and hatchery samples of 622 and 346 respectively. Therefore, a target of 200 “unknowns” per year assured an adequate number of readable scales for the analysis. A percentage of yearling hatchery smolts released from the Orsini Hatchery during 1993-1995 were marked with a right pelvic fin clip (Ralph Hay, MDNR Personal Communication). I contacted management agencies around Lake Michigan to request their port creel surveys and lake catch records in order to document the lake distribution of right pelvic fin clipped steelhead. I also solicited river creel surveys and weir and ladder reports to assess straying by Betsie River marked fish. Scale Analysis Prior to examination I prepared the scales by soaking them in a mild detergent and manually cleaning them to remove epidermal tissue and dirt. The scales were then rinsed with distilled water and mounted between two glass microscope slides. Up to 10 scales were mounted per fish. 18 I acquired scale pattern data with a computerized imaging system. Images of each scale were captured through an Olympus dissecting microscope using a Cohu RS-17O video camera and displayed on a Sony Trinitron HR monitor. The monitor is interfaced with a microcomputer and Optimas image-processing software (Bioscan 1989). Measurements were obtained along a 360° axis line from the scale focus towards the anterior edge using a mouse and the menu driven Optimas program. Origins - I determined origins of sampled fish fi'om the Betsie River with the scale pattern criteria developed by Seelbach and Whelan (1988). After identifying the first stream/hatchery annulus, the width of the five circuli just inside the first annulus is compared to the width of the five circuli just outside of the annulus resulting in a numeral determination called “ratio 23” (Figure 3). Using ratio 23, I assigned fish to either hatchery or wild origin for each year’s spawning run. Validation was possible by the presence of a fin clip on a hatchery fish. Calculation of relative abundance required hatchery and wild assignment frequencies and classification error rates. I determined frequencies for my analysis from blind readings of known origin adult scales archived at the Institute of Fisheries Research, University of Michigan. The frequencies of origin assignment were as follows: P“ (0.06) = hatchery classified as wild PW,I (0.102) = wild classified as hatchery P“ (0.898) 2 wild classified as wild Pllll (0.94) = hatchery classified as hatchery “— 5th circulus distal 0.1338 mm 1 l 1'_ Annulus J 0.7337 mm r — 5th circulus proximal Enlarged Measurement Area Figure 3. Scale from a wild steelhead showing measurement area for Ratio 23 determination. 20 I calculated the proportion of wild steelhead adjusted for classification error according to the equation in Worlund and Fredin (1962): 1)1: (waxpw) + Phw(l—Pw) where PW is the observed proportion of wild steelhead from ratio 23 determinations and PI is the adjusted proportion predicted by the above equation. An estimate of the true proportion of wild steelhead was then obtained through a maximum likelihood solution (Millar 1987, 1990). I used the following equation in a maximum likelihood scenario to determine the true relative contributions of wild steelhead: Log(L) = n. x 1080’.) + II: x logo-P.) where Log(L) is the objective function to be maximized, n, is the of observed number of wild steelhead in a spawning run, and n2 is the number of observed hatchery steelhead in the same run. I calculated the variance of each year’s wild contribution estimate by the equation: -1 .. n2 "' “1 1+ Phw(-1+Pw)—(Pw Pw)2 (Phw -Pthw +Pwpww)2 Var (Phw - P" )2 I: l 21 The variance equation was developed by using the maximum likelihood parameter estimate in the second derivative of the likelihood function. The inversion generates the the variance. Life history characteristics - Scale reading is also a widely used method of aging fish and analyzing grth (Jearld 1983). The long history of using scales in age and growth studies was summarized by Carlander (1986). I used steelhead scale aging methods described by Davis and Light (1985), Seelbach and Beyerle (1984), and Jones (unpublished). Previous spawning checks were identified according to descriptions by Seelbach and Beyerle (1984) and Hartman (1959). Stream and lake grth are distinguishable by circuli spacing and a smolting check. Circuli laid down during lake residency are much more widely spaced than during stream growth. Based on these criteria I estimated age structure and spawning history for both hatchery and wild fish. I used the European nomenclature for fish age description as described by Schwartzberg and Fryer (1989). The number of stream annuli (number of winters a fish spent in the stream) is designated by an Arabic numeral followed by a period and the number of lake annuli. Thus, a steelhead that spent two years in the Betsie River watershed as a juvenile and two years in Lake Michigan would have a recorded age of 2.2 (Figure 4). A repeat spawner is indicated by an “S” after the lake age followed by the next lake year. If the above mentioned fish had spawned after its first lake year it would be recordedasZ.1S1. 22 2nd Lake Annulus ___ aggra‘ lst Lake Annulus \\‘ . . >§\;Z‘i\\§\ ,. \QVV‘ \ *\ . -. V , \ .\. .‘ $5: ‘ \ ‘» . ‘. ~,-% }«§\\:§K \Q“_;V‘f‘l\\\ \- "7: .1 :..' ‘Wx ‘\ .w \\ 17% 2nd Stream Annulus lst Stream Annulus Scale Focus \ ’1 .4. \\ in“ \\ . , 1‘ \\ "1 V711,". ”1),, l u , x .2 cl Figure 4. Image of a steelhead scale from a 4 year old fish (2.2) collected in the spring of 1995. The wild steelhead spent 2 years in a stream environment and 2 years in Lake Michigan. 23 Beamish and McFarlane (1983) stressed the need for age validation when using scales. Validation techniques include mark and recapture of known age fish, and length fiequency analysis. Numerous studies by the MDNR have validated the accuracy of steelhead scale aging methodology in the Great Lakes (Paul Seelbach, MDNR Personal Communication). The precision or repeatability of my age determinations was estimated by an average percent error (APE) index (Bearnish and Foumier 1981) using the equation: N APE (Average Percent Error) = $172 Fl 1 R|X9_x1i [-13% x, JHOO, where N is the number of fish aged, R is the number of scale readers, X H is the ith age of the jth fish, and X J is the average age of the jth fish. When multiplied by 100 the equation becomes the index of average percent error for a set of age determinations. The index ranges from 0 to 100, and indicates higher precision with smaller index values. Three readers independently aged a 50 scale subset from the Betsie River steelhead scale collection to obtain APE estimates of stream, lake, and total age. The coeflicient of variation (CV) is also a strong estimator of reproducibility according to Chang (1982). I calculated CV for each fish and averaged over all fish. The percent error contributed by each observation can be estimated by the following index of precision (Chang 1982): where D is the index of precision value, CV is the coefficient of variation for each fish, and R is the number of scale readers. I calculated D for each fish and averaged over all fish aged. 1 determined proportions of fish maturing at lake age for wild and hatchery populations. Mean age-at-maturity was calculated by sex and origin. I computed mean lengths at lake age and age-at-maturity by year, sex, and origin. Analysis of variance (ANOVA) and student’s t-tests were used to compare group means. Mean lengths at lake age were used to develop growth trajectories with respect to sex and origin. For each year, sex ratios were determined as percent malezfemale for an overall ratio and a ratio at each lake age by origin. I tested for differences between observed sex ratios and a 50:50 ratio with a chi-square test. I compared spring migration timing based on the weekly catch at one location in the watershed. Although fishing effort may have been variable throughout the season due to weather and other factors, I assumed equal catchability between wild and hatchery fish. Median weekly dates of capture for wild and hatchery fish were compared each year. Distributions of migration timing were compared with chi-square tests. For all statistical tests, I used an or = 0.05. Smolt size influence - I tested for smolt size influence on eventual adult return by comparing mean scale radius fi'om migrating smolts captured at the Homestead Dam 25 during 1993-1994 , with mean smolt check radius from returning adults of the same cohort. I used a student’s t-test to statistically compare mean smolt radius by cohort and origin. To graph the relationship, I converted smolt check radius to smolt length using simple linear regression for both hatchery and wild smolts. The traditional Fraser-Lee equation was not used because of an apparent saltatory growth pattern in the scatter plot of scale radius and fish length. The residual plot also indicated a systematic error pattern which suggested the Fraser—Lee model would over predict length for smaller individuals and under predict length for larger fish. Therefore, wild and hatchery smolt length estimates were derived from back-calculations based on the following predictive regression equations relating smolt scale radius with smolt total length: TL: 9.0697 x SR + 11.491 , where TL is fish total length and SR is scale radius. The regression was highly significant with residuals normally distributed indicating no heterogeneity of variance (P < 0.001 , r2 = 0.39, N = 283). The regression for wild smolts also was highly significant (P < 0.001, r2 = 0.54, N = 165). Residual variance was homogenous. Wild smolt lengths were estimated with the equation: TL = 10.109 x SR + 9.7239. 26 Observed, measured smolt lengths from the 1993 and 1994 Newcomb (1998) data set were then plotted against smolt lengths back-calculated fiom returning adults of the same cohort. RESULTS River guides and anglers collected scale samples and data from 191, 131, and 157 steelhead during the years 1994, 1995, and 1996 respectively (479 total). Fish sampled in the fall were considered as part of the following spring spawning run. Sixty six kelts were sampled during three nights in May of 1996. An abbreviated and targeted collection program in 1997 produced 10 RV clipped fish. Steelhead were obtained throughout the three study sections of the watershed, but primarily from the middle study section of the river below the Homestead weir (Table 2). Of the 555 steelhead sampled for the study, 48 had scales that were regenerated or otherwise unusable for analysis. Therefore, 507 samples were used to determine origin and life histories. Origins Relative proportions of wild steelhead in the spawning runs of 1994, 1995, and 1996 were estimated to be 0.457 (0.044 SE), 0.404 (0.054 SE), and 0.299 (0.046 SE) respectively. Wild fish made up 0.505 ( 0.077 SE) of the kelt population in 1996 (Figure 5). These estimates of wild steelhead are all much smaller than the estimate made by Seelbach and Whelan (1988). They represent a significant change in the relative contributions of wild and hatchery steelhead to the Betsie River fishery since 1984 ( x: = 58, df=l, P< 0.001). 27 28 Table 2. Distribution of sampled sport catch by river section, 1994 - 1996. The kelt sample is not included. STUDY SECTION ' Return Year Lower Middle Upper n 1994 9% 78% 13% 191 1995 22% 58% 20% 131 1996 33% 54% 13% I57 ' Sections defined in Methods 29 —L O O I III Hatchery l Wild A O) m 0 C O l l 1 Proportion of Run (%) N O 0 l 1 984 1 994 1 995 1 996 1996 Kelt Seelbach and Whelan [988 Figure 5. Relative contribution of hatchery and wild steelhead to the Betsie River fishery, 1994-1996. The 1996 kelt population was sampled with blocking nets and electrofishing gear. The 1984 sample is included for comparison. 30 Hatchery strays from other rivers identified by unique fin clips contributed up to 11% of the sampled fish (Appendix B). Life HistoryCharacteristics Aging Precision - The precision of my age classifications was determined from a 50 scale subsample of the Betsie River collection. There were 6 aging differences among the three independent scale readers when assigning stream age. The resulting average percent error (APE), coefficient of variation (CV), and the index of precision (D) were all under 5% (Table 3). The scale readers produced only four discrepancies in estimating lake age. The index of APE, CV, and D were under 2% for lake age assignments (Table 4). The scale readers disagreed 10 times in total age determination. As a result, the APE, the CV, and D were all calculated to be under 3% (Table 5). Age Structure and Composition - The overall age structure of wild steelhead varied between years but was dominated by either the 1.3, 2.2, or 2.3 age category during the spawning years 1994-1996. In contrast, the age structure of hatchery steelhead remained consistent throughout the study. The 1.2 and 1.3 age categories comprised over 70% of the returning hatchery steelhead in all three years. Eighteen age categories were identified among the sampled wild fish and 11 different age categories were identified in the hatchery sample (Figures 6, 7, and 8). The majority of wild and hatchery steelhead returned at a lake age of 3 but ranged in age from 1 to 5 lake years. Wild and hatchery fish differed slightly in lake age patterns 31 Table 3. Estimated stream ages and associated APE' ,CV 2 , and D 3 from three independent readers. The fifty scale subsample taken from the 1994 -l996 Betsie River collection. Estimated Stream Age Reader n 1 2 3 APE CV D 28 1 l 1 0 0 0 16 2 2 2 0 0 0 I 1 1 2 0.3333 0.4330 0.250 I 1 2 2 0.2667 0.3464 0.200 1 2 1 2 0.2667 0.3464 0.200 1 2 1 1 0.3333 0.4330 0.250 I l 2 2 0.2667 0.3464 0.200 1 1 1 2 0.3333 0.4330 0.250 Average 0.0360 0.0468 0.0270 I APE = Average Percent Error 2 CV = Coefficient of Variation 3 D = Index of Precision 32 Table 4. Estimated lake ages and associated APE' , CV 2 , and D 3 from three independent readers. The fifty scale subsample taken from the 1994-1996 Betsie River collection. Estimated Lake Age Reader n 1 2 3 APE CV D 7 1 1 1 0 0 0 14 2 2 2 0 0 0 21 3 3 3 0 0 0 4 4 4 4 0 0 0 1 3 2 3 0.1667 0.2165 0.1250 1 4 4 3 0.1212 0.1575 0.0909 1 4 2 3 0.2222 0.3333 0.1925 1 4 3 4 0.1212 0.1575 0.0909 Average 0.0126 0.0173 0.010 I APE = Average Percent Error 2 CV = Coefficient of Variation 3 D = Index of Precision 33 Table 5. Estimated total ages and associated APE' , CV 3 , and D 3 from three independent readers. The fifty scale subsample taken from the 1994-1996 Betsie River collection. Estimated Total Age Reader n 1 2 3 APE CV D 5 2 2 2 0 0 0 7 3 3 3 0 0 0 18 4 4 4 0 0 0 10 5 5 5 0 0 0 1 5 4 5 0.0952 0.1237 0.0714 1 5 5 4 0.0952 0.1237 0.0714 1 5 3 4 0.1667 0.2500 0.1443 1 3 3 4 0.1333 0.1732 0.1000 1 5 6 6 0.0784 0.1019 0.0588 1 4 3 4 0.1212 0.1575 0.0909 1 3 3 4 0.1333 0.1732 0.1000 1 5 4 5 0.0952 0.1237 0.0714 1 5 4 4 0.1026 0.1332 0.0769 1 4 5 5 0.0952 0.1237 0.0714 Average 0.0223 0.0297 0.0171 I APE = Average Percent Error 3 cv = Coefficient of Variation '3 D = Index of Precision 34 2.2s1s1 Wild 1994 n=77 2.131 (n ‘3 1 231 E5 . o O) < 2.3 2.1 1 '2 r 1 r r r r 0 5 10 15 20 25 30 35 40 Percent 138‘ Hatchery 1994 1.231 n=94 1.1s1s1 m 2.2 U) a o 1.5 (D 2’ 1.4 1.3 1.2 1.1 0 5 1015 20 25 30 35 40 45 50 55 Percent Figure 6. Age class structure of wild and hatchery steelhead from the Betsie River, 1994. 2.331 2.23131 2.231 3.3 2.4 2.3 2.2 2.1 1.4 1.3 1.2 1.1 Age Class 1.331 1.2313131 1 .231 1.5 1.4 Age Class 1.3 1.2 1.1 Figure 7. Age class structure of wild and hatchery steelhead from 35 the Betsie River, 1995. Wild 1995 =45 5 10 15 20 25 30 Percent Hatchery 1995 n=69 5 1O 15 20 25 30 35 40 45 50 Percent 36 2.351 2.2313131 Wild 1996 2.23131 "=75 2.281 “‘7‘ 5322633 Age Class ." r‘ 7‘ r" 1“ 1° 1° to on b 01 N 0) 48 cut d l 1 I 0 5 1 0 1 5 20 25 Percent 1 '381 Hatchery 1996 1.23131 n=128 1.231 AgeClass: (I) Pee-1N7; dNOD-hN-i l l l l l l l 0 5 10 15 20 25 30 35 40 Percent Figure 8. Age class structure of wild and hatchery steelhead from the Betsie River, 1996. 20- 1O-1 Proportion of sample (%) Proportion of sample (%) u o 37 Repeat Spawners I Hatchery I Wild 60 1 2 3 4 5 Lake Age Maiden Spawners Hatchery I Wild Figure 9. Proportions of hatchery and wild maiden and repeat spawners according to lake age, 1994-1996. 38 (Figure 9). Wild fish matured earlier than hatchery fish, with mean lake ages of 2.6 and 2.8 years respectively (t=2.66, df=164, P=0.009). Age at maturity also differed between the sexes. Both wild and hatchery males matured earlier than females (wild: t=2.40, df=81, p=0.0186; hatchery: t=2.74, df: 134, P=0.007). Steelhead returning after 1 lake year were always precocious males in both the wild and hatchery samples (Table 6). The wild adult population displayed an average composition of 38.5% stream age- 1 fish, 61% stream age-2 fish, and 0.5% stream age-3 fish. Although most wild adults smolted at stream age-2, the age at smolting as inferred from adult scales varied between years. (Table 7). For example, in 1995 the 55% of returning wild adults entered Lake Michigan as smolts after 1 year in the watershed. Hatchery adults were virtually all age-1 smolts (99%). Repeat Spawning Frequency - The majority of steelhead sampled, regardless of their origin or year of return, were on their maiden spawning run. Repeat spawning frequency was higher among wild steelhead than among hatchery steelhead. Eighteen percent of wild fish had spawned previously and 10% of hatchery fish were repeat spawners. A contingency table test indicated the difference between wild and hatchery repeat spawning frequency to be significant ( 13 = 6.76, df = l, P < 0.01). A variety of repeat spawner age categories were observed in the 3 year sample for both wild and hatchery populations (Figures 6, 7, and 8). As expected, the predominant repeat spawner groups were those that had spawned only once previously. The percentage of wild and hatchery repeat spawners progressively diminished with each successive 39 Table 6. Lake age structure of maiden adult steelhead returning to the Betsie River, 1994 - 1996. Percentage of maiden adults according to lake age Mean lake age Origin Sex n l 2 3 4 5 at maturity Wild Male 81 18 28 45 6 3 2.5 Female 82 0 37 56 6 l 2.7 Combined 163 9 32 51 6 1 2.6 Hatchery Male 129 10 29 46 12 2 2.6 Female 135 0 24 60 15 1 2.9 Combined 264 5 27 53 14 l 2.8 40 Table 7. Distribution of age at time of smolting for wild adult steelhead according to year of return. STREAM AGE Return Year % Age-l % Age-2 % Age-3 1994 30 70 0 1995 55 43 2 1996 37 63 0 Weighted Mean 38.5 61 0.5 41 Table 8. Frequency of repeat spawning incidence in wild and hatchery steelhead returning to the Betsie River, 1994-1996. Spawning m W history n Percent SE n Percent SE Maiden 163 82 2.7 269 90 1.7 l spawn 26 13 2.4 23 7.7 1.5 2 spawn 8 4 1.4 6 2 0.8 3 spawn 2 l 0.7 1 0.3 0.3 Total repeat 36 18 30 10 42 spawning episode. No steelhead had spawned more than 3 times prior to its current spawning run (Table 8). Incidence of repeat spawning was more prevalent among females than in males. Numbers of female repeat spawners outnumbered male repeat spawners in all three years for both wild and hatchery fish. Females represented 72% of wild repeat spawners and 63% of hatchery repeat spawners. Sex Ratios - The overall sex ratio for wild steelhead was 1: 1.2 (malezfemale). Although the ratio slightly favors females, it was not significantly different from a 1:1 ratio ( 12 = 2.020, df=], P: 0.155). The overall sex ratio for hatchery steelhead (1:1.1) also did not significantly depart from a uniform sex ratio ( 13 = 1.215, df= 1, P: 0.270). Male to female ratios varied according to return year. Number of males exceeded the number of females in 1994 for both wild and hatchery steelhead. Females predominated in 1995, 1996 and the wild and hatchery kelts sampled in 1996 (Table 9). These yearly differences where not significant except in 1996 for wild kelts ( x3" = 4.172, df= 1, P=0.041). Seventy percent of male kelts (wild and hatchery) were infected with Saprolegnia, a fungus infection common in stressed or injured salmonids. Sex ratios of returning steelhead also varied with lake age. Higher male proportions were related to earlier lake ages. Ratios for lake age-1 through age-5 were 1:0, 1:1.2, 1:1.2, 1:2, and 1:3 respectively for wild fish. Ratios for hatchery fish at each lake age were 1:0, 1:.89, 1:1.4, 121.9, and 1:1.5 (Figure 10). 43 Table 9. Sex ratios for adult wild and hatchery steelhead according to return year. Asterisk indicates a significant difference (P < 0.05) from a 1M:1F sex ratio WILD M:F HATCHERY M:F Return Year n % Ratio n % Ratio 1994 39:37 51:49 12.95 48:46 51:49 1:.96 1995 21:25 47:53 1:1.2 36:38 49:51 1:1.1 1996 20:27 43:57 1:1.4 55:74 43:57 1:1.3 1996 Kelt 9:20 * 31:69 1:2.2 12:18 40:60 1:1.5 Total 89: 109 45:55 1:1.2 139:158 47:53 1:1.1 [:1 Wild Female I Wild Male 100 80 ‘E 60 0 2 8 4o 20 0 . 1 2 3 4 5 Lake Age L'J Hatchery Female I Hatchery Male 100 1- —~ —» ~—» ‘-—1‘ 80 E 60 0 0 2 4o 20 o . 1 2 3 4 5 Lake Age Figure 10. Gender composition by lake age in samples of wild and hatchery steelhead returning to the Betsie River during 1994-1996. Frequency % Frequency "/5 N O .5 0| .3 O 0| 0 N O .A 0" .5 O U! 0 Hatchery 1994 X = 67.4 mammawwmmumuww Length (cm) Length (cm) 38 42 46 50 54 58 62 66 70 74 78 82 86 90 .’ ‘.~"A’.‘-. r . .. ’, I- z‘?.,‘..-.. .‘Xx, ”59’4” mmmwmwwwmnmwwm Length (cm) 45 38 42 46 50 54 58 62 66 7O 74 78 82 86 90 Length (cm) 20 Wild x = 65.6 15 . 10 b 5 . 0 wuwwawwmmnmamm Length (cm) 20 mumwawmmmunwwm Length (cm) Figure 11. Length frequency distributions of wild and hatchery steelhead by return year. Lengths are grouped in two centimeter intervals. 46 Length and Growth - Yearly length frequency distributions, as depicted in Figure 11, illustrate the size structure of the spawning populations. Lengths ranged from 39cm to 89cm for wild steelhead and 38cm to 99 cm for hatchery steelhead. Annual overall mean lengths remained consistent throughout 1994-1996. Analysis of variance indicated no length differences among years or origin (Table 10). Lengths at lake age also remained consistent during the three years of study (Figure 12). Wild lengths at lake age did not differ statistically between years, nor did hatchery lengths (Table 10). Adult steelhead lengths did vary according to age, sex, and spawning history (Table 11). Lengths at maturity of male hatchery spawners increased with each year of lake residence. Wild lengths at maturity of male spawners reached an apparent asymptotic level at lake age—4 whereas hatchery males did not. Comparative length trajectories depicted in Figure 13 portray these growth patterns. The mean lengths of hatchery males at lake age were not significantly different than the respective lengths of wild males (t- tests, P > 0.05). Maiden wild and hatchery females were not recruited to the river fishery until lake age-2. Their growth trajectories, however, project a similar pattern to that of the corresponding male steelhead (Figure 13). Mean lengths of hatchery females were not significantly different from their wild counterparts except at lake age-3 (t=2.058, df=46, P = 0.045). The stream age of wild steelhead appeared to significantly affect adult length only at lake age-l. Two year old smolts were longer than one year old smolts only at lake age- 1(t=2.614, df=8, P = 0.031). After the first year in Lake Michigan, wild adult lengths 47 Table 10. Two way analysis of variance tests for year and origin effects on steelhead length, 1994-1996. Lake Age Effects n Degrees of F P Value Freedom Overall year 495 2, 493 0.76 P = 0.4690 origin 1, 493 3.42 P = 0.0649 Lake agel year 27 2, 25 1.60 P = 0.2262 origin 1, 25 0.65 P = 0.4297 Lake age2 year 118 2, 1 16 1.29 P = 0.2799 origin 1, 1 16 0.01 P = 0.9248 Lake age 3 year 240 2, 238 1.64 P = 0.1953 origin 1, 238 3.53 P = 0.0616 Lake age 4 year 71 2, 69 2.05 P = 0.1366 origin 1, 69 0.57 P = 0.4538 48 80 — g a El 0 70 r o O O A “ROM. E 3, P I a 60 . . 91 5 '2 ‘33 50 1 °3 .4 19 o 3 O :15 4o _ Wild 30 . . 1994 1995 1996 90 — D so »- D 9 o A Lake Age E 70 » o 0 o 0 E3 50 r I I I '2 O :I 03 .4 r so. , o D5 e 40 r Hatchery 30 I I 1994 1995 1996 Figure 12. Mean total length at lake age for wild and hatchery steelhead by return year, 1994 - 1996. 49 Table 11. Mean total length (cm) by age class and sex for wild and hatchery adult steelhead from the Betsie River, 1994 - 1996. WILD HATCHERY Male Female Male Female Age Class Mean SE n Mean SE n Mean SE n Mean SE n Maiden Spawners 1.1 42 0.6 6 47 1.8 13 1,2 59 0.9 8 60 1.2 9 53 0.9 37 61 0.8 33 1.3 72 1.2 15 68 1.0 22 70 0.6 60 69 0.4 81 1,4 31 6.3 2 74 2.5 2 77 1.1 15 75 0.9 20 1.5 79 0 2 85 2-3 3 81 - l 2.1 44 1.1 7 2,2 61 1.7 15 60 0.7 21 62 - 1 63 - l 2.3 70 1.0 21 67 0,9 24 2,4 77 3.3 3 71 2.2 3 2.5 77 - l 72 — 1 3.3 69 - 1 Repeat Spawners 1.18181 60 1.3 2 1,251 61 - I 69 1.0 7 68 1.0 6 1.28181 73 2.2 4 1.2818181 76 - 1 1.381 77 - l 74 3-8 3 75 2-5 2 75 2-0 8 2.181 53 2.5 3 2.281 58 0.2 3 66 0.8 13 2.28181 78 3.8 5 2.2818181 77 3.8 2 2.381 76 3.6 2 74 3.9 3 50 90 r' 80 p Male . r ' 70 - 60- Total Length (cm) 30 . I i . 30 _ Female ' .- 70L 60- Total Length (cm) 5° * —o—wnd - l- Hatchery 30 . . u . Lake Age Figure 13. Mean length at maturity (cm) of maiden wild and hatchery steelhead according to sex returning to the river during 1994 - 1996. Stream ages pooled for wild steelhead at each lake age. 51 90 r 80 - E 3' 70 - 5 u C 3 E 60 ~ ,2 '-0—Stream Age 1 50 _ - I- Stream Age2 40 I I I T I 1 2 3 4 5 Lake Age Figure 14. Mean length at maturity (cm) according to stream age of wild adult steelhead returning to the Betsie River during 1994 -1996. 52 were virtually the same, regardless of stream age history (t-tests, P > 0.05). Wild steelhead that smolted at age-1 had achieved the same length as age-2 smolts of equal lake residence (Figure 14). Sex influenced adult length for wild and hatchery fish (Table 12). Wild males were longer than wild females in the lake age-3 category (t=3.559, df=37, P = 0.001) and at lake age-4 (t=2.576, df=5, P = 0.049). Lake age-2 males were not significantly longer than the corresponding females (t=0.215, df=21, P: 0.832). Hatchery males were also longer than hatchery females at lake age-3 (t=2.015, df=58, P = 0.048). Hatchery females were longer than males at lake age-2 (t=2.652, df=29, P = 0.01). Spawning history also influenced adult length. Mean lengths of maiden spawners were generally greater than lengths of repeat spawners of equal lake age (Table 13). Among wild steelhead significant length differences occurred between lake age-2 and -3 maiden and repeat spawners (t-tests, P < 0.01). A similar difference occurred between hatchery lake age—3 maiden and repeat spawning steelhead (t=4.728, df= 14, P = 0.0003). Annual increments of length diminished with each year of lake residence with the exception of lake age-5 hatchery fish . There was no difference in growth rates between wild and hatchery steelhead as measured by walford plots. Comparative walford plots indicated similar growth (Figure 15). Growth coefficients (K) were essentially the same for wild and hatchery fish. Migration Timing - Wild and hatchery migrants were caught in the Betsie River during the fall and winter months. The majority of these fish, however, were captured in the lower river. Respective spring run timing, as measured by weekly catch at the 53 Table 12. Comparison of mean total length (cm) by sex for maiden wild and hatchery steelhead. Significant length differences denoted by asterisk (* P< 0.05; **P< 0.01; ***P< 0.001; NS = no significant difference). LENGTH Male Female Origin Lake Age n mean n mean P Value Wild 2 23 60.2 30 60.0 NS 3 37 70.8 47 67.5 *** 4 5 78.6 5 72.1 * 5 3 78.3 1 71.8 NS Hatchery 2 38 58.1 33 61.2 * 3 60 70.5 81 69.1 * 4 15 77.2 20 75.4 NS 5 3 85.1 1 81.3 NS 54 Table 13. Comparison of mean total length (cm) by spawning history for wild and hatchery steelhead. Significant length differences denoted by asterisk (* P< 0.05; **P< 0.01; ***P< 0.001; NS = no significant difference). LENGTH Maiden Repeat spawners spawners Origin Lake Age n mean 11 mean P Value Wild 2 53 60.2 3 53.3 ** 3 84 69.0 17 65.1 ** 4 10 75.5 14 74.6 NS 5 4 78.3 2 77.5 NS Hatchery 2 72 59.6 0 - ' - 3 141 69.7 15 67.9 *** 4 35 76.2 14 75.3 NS 5 4 83.8 1 76.2 NS I no hatchery lake age 2 repeat spawners in sample 55 30° 1 Wild ,2 80 _ K=0.4697 L, = 84.8 E 60 E ‘9 <0 40 _l / / 20 " ,r ’ I y = 0.6252x + 31.779 , , ’ R2 = 0.9968 0 1' 1 j 1 l I 0 20 40 60 80 100 Length(t) 300 ' Hatchery , , K=0.4452 80 L, = 89.5 E 60 .‘c" I?" a, 40 -J I 20 . I,” y=0.6407x+32.176 , z ’ R2 = 0.9836 0 1" n 1 l n I 0 20 40 60 80 100 Length(t) Figure 15. Walfort Plots for wild and hatchery steelhead and their calculated growth coefficients (K) and asymptotic lengths (L,) 56 Homestead weir, was congruous between wild and hatchery steelhead (Figure 16). The distribution of wild and hatchery migration timing was not significantly different in any year (Chi-square tests, P > 0.05) Peak immigration of wild fish coincided with the peak immigration of hatchery fish in all three years. Median date of capture for both wild and hatchery immigrants occurred in the first week of April in 1994 and 1995 and the second week of April in 1996. Migration timing appeared to differ between sexes. Based on weekly catch, males (wild and hatchery) returned earlier than females (Figure 17). The distribution of male migration dates was significantly earlier than that of females for wild and hatchery steelhead (Chi-square tests, P < 0.001). Male steelhead made up 78% of the fall and winter catch. Smolt Size Influence Influence on Return - I first measured size-dependent selection for return directly from scale data without the potential biasing effects of back-calculation procedures. The form and relative magnitude of selection was determined with two independent samples of the 1993 and 1994 cohorts taken before and after lake residence. Because scale radius is proportional to fish length, it can therefore be used as an index of size. Smolt scale radius when compared with adult smolt check radius suggested that the probability of return is size dependent for steelhead smolts. Mean smolt scale radii of surviving hatchery adults from the 1993 and 1994 cohorts were significantly greater than the corresponding scale radii from the same cohort prior to lake residence (Table 14). The greatest difference in mean scale radius 57 35 , 1994 Wild 3° ' — - Hatchery 25 - 20 ~ °/o 15 - 10 - 5 _ 0 r a 1 2 3 4 1 2 3 4 1 2 3 4 March April May Figure 16. Migration timing based on the weekly catch at Homestead weir from March through May. 58 251 , Male - - -Female March April May 25 - Hatchery Mme v ' \ 20 . ' ‘ - - -Female Figure 17. Migration timing by sex of wild and hatchery steelhead based on average weekly catch at the Homestead weir from March through May, 1994-1996 59 between hatchery adults and smolts occurred in 1993. Adult frequency distributions of scale radii in both years were highly skewed and shifted toward the larger scale sizes. Wild steelhead comparisons showed a similar size dependent relationship where larger smolts have a greater probability of returning as adults. Adults from the 1994 cohort had significantly larger smolt scale radii than did the 1994 smolts measured prior to lake residence. Mean adult smolt scale radii from the 1993 cohort were larger than the scale radii of the corresponding 1993 smolts. However, the difference was not significant (Table 14). The greatest difference in mean scale radii between wild adults and smolts occurred in 1994. As with hatchery adults, wild adult distributions of scale radii were shifted toward large scale intervals. Graphical comparison of smolt lengths derived from back-calculation of adult scales and observed measured smolt lengths also suggests that survival is higher for fish that are large size at smolting. Differences in length distributions were apparent mainly in the smaller length intervals (Figures 18 & l9). Surviving adult hatchery fish had significantly different smolt length distributions than did migrant smolts measured at Homestead in both cohort years (Chi-square tests, P < 0.0001; Figure 18). Smolt length distributions of wild adults from the 1994 cohort were significantly ( x3 = 31.01, df= 12, P = 0.002) shifted towards larger sizes when compared to wild smolts measured in 1994 (Figure 19). Adult smolt lengths from the 1993 cohort were shifted toward the larger sizes as well. The distribution, however, was not 60 Table 14. Comparison between adult smolt check radius (mm) and smolt scale radius (mm) of wild and hatchery steelhead according to smolt year. Significant scale radius differences denoted by an asterisk (* P< 0.05; **P< 0.01; ***P< 0.001; NS = no significant difference). SMOLT SCALE RADIUS (mm) Adult Smolt Origin Smolt Year n mean n mean P Value Wild 1993 45 .994 10 .890 NS 1994 32 1.04 101 .905 * * * Hatchery 1993 122 .950 10 .854 * 1994 39 .956 272 .905 * 61 50 I Observed Lengths From Homestead 1993 HatCherV Cohort 40 3 liafittécalculated Lengths From P < 0.0001 3? Z: = 77.3 I o c o :1 U 9.3 IL 15 16 17 18 19 20 21 22 23 24 Smolt Length (cm) 35 I Observed Lengths From Homestead 1994 Hatchery 30 ‘ E] Observed Lengths From Orsini c°h°n 25 l I Back-calculated Lengths From Adults . P < 0_0001 a? Z: = 77.5 I 20 — o r: o a. 15 - 2 u. 10 ~ 5 . 0 Smolt Length (cm) Figure 18. Frequency distributions of observed lengths from migrating hatchery smolts measured at the Homestead weir and back- calculated smolt lengths of surviving adult steelhead from the same cohort. Smolts were also measured at the hatchery prior to release in 1994. 62 Frequency (%) -* N to 01 O 01 ..L O 45 8 Frequency (%) I Observed Lengths From 1923:11ch Homestead o I Back-calculated Lengths From Adults P = 0.248 13 =12.5 Smolt Length (cm) I Observed Lengths From Homestead 1994 Wild 3 I Back-calculated Lengths From Adults Cohort l P = 0.002 12 = 31.01 12 14 16 18 20 22 24 Smolt Length (cm) Figure 19. Frequency distributions of observed lengths from migrating wild smolts measured at the Homestead weir and back- calculated smolt lengths of surviving wild adult steelhead from the same cohort. 63 significantly different than the distribution from the measured smolt lengths of the same cohort (12 = 12.5, df= 13, P = 0.248). In both smolt years, wild and hatchery mean back-calculated lengths were greater than observed, measured lengths (Table 15). Mean differences between back-calculated and observed lengths ranged from 0.5cm to 1.3cm. Length differences between male and female smolts determined from scales were not significant for either wild or hatchery steelhead (t-tests, P > 0.05). As expected, smolt size increased with stream age according to observed and back-calculated lengths of wild smolts (Table 15). Differences between back-calculated and observed lengths decreased with stream age. Differences were significant at stream age-l but not significant at stream age-2. This result implies that size selection is more pervasive among the smaller stream age-1 smolts than among stream age-2 smolts. Influence an Age at Maturity - The length of smolts had some apparent influence on the number of years spent in Lake Michigan prior to spawning. Generally, as years of lake residence increased, smolt length decreased, meaning that large smolts returned earlier than small smolts (Table 15). Smolt length differences between lake age-2 and -3 fish were not significant in either wild or hatchery steelhead (t—tests, P > 0.05). But the difference between lake age-1 hatchery fish and older lake age hatchery fish was significant (t=2.584, dfi17, P = 0.02). Wild lake age-l smolt lengths, although larger on average, were not significantly different than smolt lengths of other lake age fish ((t=1.422, df=l3, P = 0.178). 64 Table 15. Observed smolt lengths (OL) and back-calculated smolt lengths (BL) from adults in relation to the smolt cohort year, the stream and lake age, the sex, and the year of return with all cohorts pooled. WILD HATCHERY Category Mean CL CL Mean BL BL BL Mean OL OL Mean BL BL BL (cm) n (cm) n SE (cm) n (cm) n SE Smolt Year 1993 19.3 48 19.8 45 1-6 19.8 65 20.3 116 l -1 1994 19.2 2 2 5 20.3 3 7 1 -5 18.8 740 20.1 38 1-4 Stream Age 1 17.4 164 19,0 37 1.0 2 20.5 93 20.6 46 13 Lake Age 1 20.3 14 1.6 215 18 3.3 2 19.7 36 I .7 20,2 60 1.5 3 19.9 33 l .5 20,0 75 2.0 4 18.6 I - .Sfl Female 19.8 44 I .6 20.2 89 l 3 Male 19.7 39 1-6 20.3 65 1-9 Return Year 1994 20.2 77 l .7 20,3 97 1.5 1995 19.8 47 1-7 20.5 71 1-5 1996 202 75 1.7 205 130 2.0 65 1993 Smolt Cohort Hatchery Wild A 22 r- 22 P -°—Ausmom S - O - Age18molts v _ b - ¢ ' ”ZSMC g 21 21 . ‘b _ - 11 U) 20 " 2° " ‘6 f3, 19 ~ (1) 19 ~ ' .‘3 C D . , 3 ~ -a 18 l l J I 18 l l I J o 1 (94) 2 (95) 3 (96) 4 (97) 0 1(94) 2 (95) 3 (96) 4 (97) Lake Age and Return Year Lake Age and Return Year 1994 Smolt Cohort Hatchery Wild A 22 — 22 .. -°-Ansmom (E, - O - Ago1$mom V - In - 1191129111011: 921 r 21 l- A :0. . . a ”E, 20 ~ 20 . °"""" E D ‘6 ‘ , g 19 r 19 - ‘ . E El o _l 18 1 1 1 18 1 1 1 0 1 (95) 2 (96) 3 (97) 0 1 (95) 2 (96) 3 (97) Lake Age and Return Year Lake Age and Return Year Figure 20. Relationship between mean back-calculated length at smolting and lake age of adult steelhead from the Betsie River, 1994-1997. 66 Very large hatchery male smolts appeared to return earlier than the rest of the hatchery population as lake age-l jacks (Figure 20). The same relationship between smolt length and lake age was not as apparent in the wild cohorts. Lake Distribution and River Straying Lake Distribution - After leaving the harbor at Frankfort as smolts, the geographic dispersal of hatchery steelhead was determined from the open lake fishery. Lake Michigan ports sampled in creel surveys covered the length and breadth of the lake from Burns Harbor, Indiana up to Big Bay de Noc in Michigan and along the east and west shoreline. Data from these surveys provided information about the distribution of uniquely marked Betsie River steelhead (Figure 21). Based on returns, RV clipped fish were just beginning to recruit to the fishery in 1994 and were limited to Michigan waters from Onekama to Ludington. By 1995, these fish were well distributed throughout the southern two thirds of the lake. In 1996, this dispersal pattern repeated itself with fish being caught in the waters of all four states bordering Lake Michigan. During the three years, 47% of the Betsie River hatchery returns were caught in Michigan waters, 28% in Wisconsin, 16% in Indiana, and 9% in Illinois. No RV clipped fish were recorded north of Frankfort. Recoveries of Betsie River steelhead in southern Lake Michigan waters represent a minimum movement of 370 km. Ages of the RV clipped fish (determined by management agency scale readers) ranged from 1.1 to 1.3 and thus represented the 1993-1995 hatchery cohorts. Total 67 lengths ranged from 46-75cm. These fish were harvested during the lake fishery from April through September. According to agency records, RV clipped steelhead were frequently caught along with other uniquely marked steelhead, suggesting intermingling of stocks in the open lake. River Straying - Fin clipping of pre-smolts at the Orsini Hatchery began in 1993. Based on weir records, ladder reports, and stream creel surveys, homing imprecision of Betsie River hatchery steelhead was documented in 1995 and 1996 (Figure 22). Straying was observed in five non-natal streams of Lake Michigan as far south as the St. Joseph River on the east shore and the Root River on the west. Characteristics of the stray fish were similar to fish that had homed accurately back to the Betsie River. Total lengths ranged from 49 - 81cm. Ages of stray steelhead were not documented in most cases. However, length distributions suggest that the majority of strays were lake age-2 and -3 fish. Straying tendency appeared to be independent of sex. The sex ratio was 1:1. This characteristic was similar to the ratios found in the fish sampled in the Betsie River. Most of the stray steelhead (92%) were spring migrants. Eighty eight percent of the recovered strays were observed in weirs, 8% from stream creel surveys, and 4% from ladder reports. Michigan’s four northern Lake Michigan weirs recorded only one Betsie River stray (Little Manistee River Weir). Medusa Creek, Platte River, and Boardman River weir facilities are operated in the fall months only and sample approximately 400 fish. The Little Manistee River weir is 68 operated in the fall and spring. Approximately 400 steelhead are sampled for biological data during each run. The other strays were found in streams far from the vicinity of the Betsie River. Two Wisconsin rivers shared the largest proportions of recorded strays. Located on the Kewaunee River, the Besadny Facility recorded Betsie River strays in 1995 and 1996. The Root River Steelhead Facility also counted Betsie River strays in both years. These two weir facilities are operated during spring and fall salmonid runs. Over 2000 steelhead are examined at both weirs each year. Wisconsin creel clerks on the Sheboygan River sampled one Betsie River steelhead in 1995 and one in 1996. The St. Joseph River attracted two known Betsie River fish. One stray was captured in the Berrien Springs Dam ladder during sampling operations in 1995. Indiana creel clerks sampled the other stray steelhead upstream of the French Paper Company Dam in 1996 69 Z Betsie River Sheboygan Ludington Wisconsin Michigan Milwaukee . 0 South Haven Waukegan Benton Harbor Illinois Michigan City Indiana Figure 22. Distribution in Lake Michigan of steelhead stocked as smolts in the Betsie River during 1993 4995. RV clipped steelhead recovered in 1994 - *; 1995-Q;1996—.. 70 . ‘ ad / Kewaunee H. -' I i I I 0 Little Manistee H. Sheboygan R. I I O ', . Michigan Wisconsin { I I \ RootH. 2). . '. 000 ...... s 1 Illinois ,' O S" ' H’ r . I. . _ Indiana Figure 23. River straying by steelhead stocked as smolts in the Betsie River during 1993 - 1995. RV clipped adults recovered in I995 - O; 1996 -. . DISCUSSION Origins Accuracy - Classification accuracy and error rates, as assessed from the subsample of Seelbach and Whelan’s (1988) archived scale collection, compare favorably with similar studies. Based on the one parameter of ratio 23, I determined assignment frequencies of .898 and .940 for wild and hatchery steelhead respectively for an overall accuracy of 92%. Unwin and Lucas (1993) separated wild and hatchery chinook salmon with accuracy rates of 82-90% using a single scale parameter. Using circuli spacing, Barlow and Gregg (1991) achieved an accuracy rate of 83% in discriminating between wild and hatchery barramundi, Lates calcarifer. Bernard and Myers (1997) used a six parameter technique to separate wild and hatchery steelhead from North Pacific populations and obtained a 94% accuracy rate in their test samples. Quantifiable differences in the scales of Betsie River wild and hatchery steelhead presumably allowed for the same level of accuracy in my samples as in the archived sample. Although not all hatchery steelhead were marked, 96% of the RV clipped fish were correctly identified as having a hatchery origin. Moreover, ratio 23 may be a better discriminator of Betsie River fish than for fish from a more benign system. Scale characteristics such as circuli spacing are strongly influenced by environmental factors like temperature and food availability (Bhatia 1931; Willett 1994). In response to cold 71 72 temperatures and reduced feeding, circuli spacing narrows. Conversely, spacing widens with increased temperatures and feeding. Such patterns are accentuated with temperature extremes and are less so with more uniform temperatures. The Betsie River is a thermally diverse watershed with relatively wide ranging temperatures (Newcomb 1998). Scales from fish that grew in the Betsie environment would be expected to exhibit patterns reflecting this type of temperature regime. Indeed, stream annuli on wild Betsie River steelhead were generally distinct and easily defined. Also the difference between the mean ratio 23 values of Betsie River wild and hatchery steelhead was greater than the difference between the archived wild and hatchery sample. As the difference between ratio 23 values increases, separation of wild and hatchery fish becomes increasingly accurate. Composition of Spawning Runs - Assuming equal catchability of hatchery and wild fish, wild steelhead made up 30-46% of the returning adult population in the three study years. This relative contribution is much lower than that measured just 10 years previously. The downward trend in relative abundance of wild fish can be interpreted in two ways. First, the decrease may be only relative to the increase in hatchery numbers. Stocking levels of Betsie River hatchery steelhead have increased four-fold since the early 1980’s. Fish are also released at a larger size and higher up in the watershed, both of which should increase returns. Therefore, the same number of wild fish would contribute a smaller proportion due to an increase of hatchery fish returning to the Betsie River. The other possible explanation for the lower contribution is that fewer wild fish are being produced in the Betsie River. When the Thompsonville Dam failed, sediments held behind the dam were released downstream. Much of the watershed’s best spawning 73 gravels were compromised. If wild production is less than what it was in the early 1980’s, then the lower proportions reflect an actual decline in wild steelhead numbers. Regardless of the possible explanations for the reduction in relative abundance, wild steelhead are still making important contributions to the fishery. Proportions of wild steelhead in other marginal tributaries of Lake Michigan that receive hatchery fish have ranged from 55% in the Muskegon River (Seelbach and Whelan 1988) to 3-11% in the St. Joseph and Grand Rivers (Seelbach et al. 1994). The contribution of wild steelhead is much greater in the high quality streams of northwestern lower Michigan where little stocking is required to maintain the fisheries (Seelbach 1987). Stray hatchery steelhead from rivers other than the Betsie comprised an unknown percentage of the spawning runs. Guides and anglers recorded the clip type of marked steelhead from their catch. Anglers, however, were not aware of the maxillary clips used by the state of Wisconsin during the 1994 and 1995 seasons. In addition, not all hatchery steelhead stocked into Lake Michigan are marked (Anonymous 1975-1995). Therefore, the relative abundance of strays could not be estimated. The origins of the uniquely marked fish appear to be widespread throughout Lake Michigan (Appendix B). Although fish did stray from as far as the St. Joseph River, Indiana, the majority of recorded strays had a Wisconsin origin. These fish, which potentially made significant contributions to the spawning effort, were not only from allopatric sources, but also of differing strains. The Ganaraska, Skamania, and Chambers Creek strains were all represented by the stray steelhead as well as the Little Manistee strain used by Michigan. 74 An interesting and statistically significant difference existed between the 1996 estimate of wild and hatchery contributions based on the fishery (30% wild) and the 1996 kelt estimate sampled with block nets and electrofishing gear (51% wild). Kelts, by definition, are survivors of spawning. Because the majority of steelhead caught in the fishery upstream from Homestead were hatchery fish (Appendix C), it does not seem likely that the difference is related to greater numbers of wild fish utilizing that section of the watershed. The Orsini Hatchery, from which they were reared and released, would naturally exert a strong influence on the area of return chosen by hatchery fish (Slaney et al. 1993). The hatchery is located upstream of Homestead. Furthermore, based on data from scale analysis, wild steelhead survived maiden spawning episodes and returned to spawn again more frequently than hatchery steelhead. The lower relative abundance of hatchery kelts is probably associated with poorer spawning survival and not fewer hatchery fish returning to the upper watershed. Numerous other studies concur with this finding. Leider et al. (1986) reported a lower incidence of repeat spawning among hatchery steelhead in Washington. Scale data from 16 Vancouver Island streams confirmed a higher repeat spawning frequency among wild steelhead than among hatchery steelhead (Hooton et al. 1987). Life History Characteristics Life history characteristics of fish represent a combination of genetic constraints and adaptive responses to environmental pressures (Ricker 1972; Schaffer and Elson 1975). The variability of these characteristics are what enabled the steelhead to colonize 75 Table 16. Summary of similarities and distinctions in life history traits of Betsie River wild and hatchery steelhead. SIMILARITIES Origin Migration Timing Length Growth Rate Coefficient Sex Ratio Wild April median date 66cm K = 0.47 1:1 Hatchery April median date 67cm K = 0.45 1:1 DISTINCTIONS Origin Age at Maturity Age Structure Repeat Spawning Freq. Wild 2.6 lake years 18 age categories 18% Hatchery 2.8 lake years 11 age categories 10% 76 and adapt to the localized conditions of Great Lakes tributaries. In the Betsie River I found both life history distinctions and similarities between wild and hatchery steelhead which are summarized in Table 16. Four of the seven parameters examined were similar (sex ratios, lengths, growth, and migration timing) and three were distinct (age structure, age at maturity, and repeat spawning frequency). Age Structure and Composition - The majority of returning wild steelhead were lake age-2 and —3 fish as were returning hatchery fish. Lake age-3 steelhead are the norm for Little Manistee fish which serve as broodstock for the Orsini Hatchery (Seelbach 1993). However, a greater breadth in the age structure of wild steelhead was evident in the number of age categories. A varied age structure may be an adaptive response to a marginal stream such as the Betsie. According to Schaffer and Elson (1975), when environmental conditions are harsh and unpredictable, selection favors individuals that are capable of spawning at different ages. Saunders and Schom (1985) suggested that the variability in age structures of Atlantic salmon is a safeguard against reproductive failure of any one year class. Individuals from one year class return over multiple years, thereby ensuring some contribution from that cohort. Therefore, a population confronted with the Betsie River, might be expected to send its spawners at a variety of different ages. Males of wild and hatchery origin had an earlier maturation schedule than females. This maturation pattern is typical of Pacific and Great Lakes steelhead populations (Tipping 1991; Biette et al. 1981). The difference in age at maturity between male and female fish is thought to be related to the difference in gonadal investment 77 between the sexes (Moyle and Cech 1988). Males require less growth before reaching sexual maturity than females. Wild steelhead as a whole matured earlier than hatchery steelhead. I also found a higher proportion of lake age-I males (jacks) in the wild Betsie River sample than in the Betsie River hatchery sample. Age at maturity is influenced by genetic as well as environmental factors (Gall et al. 1988). Therefore, steelhead maturity can be manipulated by selective breeding practices as demonstrated by Tipping (1984, 1991). Age at maturity was delayed in the progeny of hatchery stock at the Cowliz Trout Hatchery when only older adults were used as spawners. In Michigan, lake age-l males are not used as spawners at the Little Manistee River weir facility (Peter Makoweski, MDNR, Personal Communication). In many Great Lakes tributaries, greater than 70% of the returning adults spent 2 years growing in the stream before smolting (Biette et al. 1981; Seelbach 1993). On average, 61% of returning Betsie River wild adults were stream age-2 smolts. Newcomb ( 1998), however, found the majority of Betsie River wild juveniles to be stream age-1 smolts. The apparent discrepancy is presumably related to higher mortality of the stream age-1 smolts. This hypothesis is supported by comparing smolt length frequencies of adults and smolts from the same cohort and will be discussed further in the section concerning smolt size influence. Similar observations in Great Lake populations, where the majority of emigrating smolts were stream age-l, but the majority of returning adults were stream age-2 smolts, were made by Kwain (1981), Stauffer (1972) and Karges (1987). 78 The percentage of returning stream age-1 and -2 adults was not constant between years. Stream age classes of emigrating smolts in tributaries to the Finger Lakes were also found to vary. The annual variation was attributed to stream flow and temperature. Low flow and high temperatures caused the early descent of stream age-l fish (Northcote 1969). A similar environmental mechanism may be operating in the Betsie River watershed. Repeat Spawning Frequency - Much of the difference between wild and hatchery age structures can be explained by their frequency of repeat spawning. As discussed above, repeat spawning was more prevalent among wild steelhead than in hatchery steelhead. Life history theory predicts how fish vary reproductive effort in response to their environment (Schaffer 1974; Mitton and Lewis 1989). Species in unpredictable habitats place a premium on multiple spawnings. Hutchings (1993) anticipated a high degree of iteroparity in brook trout populations when associated with unstable, harsh streams. Conversely, Seelbach (1993) suggested that the stable flows of the Little Manistee River would select for fewer spawnings by larger steelhead. The optimal reproductive strategy for a Betsie River steelhead, if the watershed is viewed as harsh and unpredictable, would be to retain an iteroparous life history. Sex Ratios - Proportions of males to females were close to a 50:50 ratio in both wild and hatchery samples. An even malezfemale ratio is normally optimal in vertebrate populations with open, polygamous mating systems (Karlin and Lessard 1986). Although both wild and hatchery fish were evenly divided between male and female, deviation 79 from a 50:50 ratio occurred in the kelt sample, where females were favored. Differential mortality between the sexes is attributed to the longer duration on the spawning grounds by the males and their territorial defense of the redd (Shapovalov and Taft 1954). Saprolegnia infections were also more prevalent in males than in females. As a result, the majority of repeat spawners are female. Relatively equal sex ratios in the spawning populations are then maintained in part by the yearly contributions of precocious males. Withler (1966), Kwain (1971), and Seelbach et al. (1994) all reported equal proportions of male and female spawners in steelhead populations. In other systems, salmonid sex ratios have been inadvertently altered by hatchery practices. The Kalninka River chum salmon, Oncorhynchus keta, population changed from a 50:50 ratio to a ratio favoring males as a result of selecting only early returning fish for breeding purposes (Altukhov and Salmenkova 1990). Length and Growth - Lengths of wild and hatchery fish did not differ between calendar years. Similarly, lengths at age also did not change significantly over the three study years. Annual differences, as indexed by the dominant lake age-3 group, were 1.5cm or less in the wild sample and 1.4cm or less in the hatchery sample. Because adult length is primarily a function of lake age (Seelbach and Beyerle 1984), conditions for growth were apparently stable and equal for hatchery and wild steelhead in Lake Michigan during the years of study. Seelbach (1989; 1994) also found little variation in steelhead lengths between years. Considering the stable size structure in Lake Michigan, Seelbach (1994) suggested that population levels are at a point of equilibrium out in the lake. 80 Examination of length frequencies revealed distinct modes only at the smaller sizes. Therefore, length frequency analysis of Betsie River steelhead may not be appropriate for accurate assignment of age groups. Size distributions better describe the status or balance of a population (Ney 1993). For example, the 1995 length frequency histogram reflects the near absence of age-l hatchery fish recruited to the fishery in that year. I found no difference in male lengths between wild and hatchery fish of the same lake age. Nor did I find differences in female length except at lake age-3, for which hatchery females were significantly longer than wild females. Other comparative studies involving lengths of wild and hatchery steelhead offer mixed results. Hooten et al. (1987) and Peterson (1979) found Pacific populations of wild and hatchery steelhead to be of equal length at equal age. Seelbach and Miller (1993) also found similar lengths in a Great Lakes population of wild and hatchery fish. However, wild Kalama River steelhead were longer than their hatchery counterparts (Leider et al. 1986). Finally, hatchery steelhead from the Cowlitz River were smaller than wild steelhead until the source of hatchery broodstock changed from a domesticated strain to a stream-specific broodstock (Tipping 1984). Lengths of wild male and female steelhead appeared to reach an asymptotic level whereas hatchery lengths did not. Fish growth normally decreases gradually with size and age. The normal approach to an asymptotic length may have been masked by the small sample size at the older age groups (age-5) and the possible presence of Skamania strain steelhead in the hatchery sample. Skamania steelhead were stocked into the Betsie River in the late 19805 and early 19905 and may have recruited to the fishery during the 81 study. Skamania steelhead frequently reach greater length and mature at a later age than the Little Manistee strain (Fielder 1987). I did find length differences related to sex and spawning history in both wild and hatchery steelhead. The overall mean length of females was greater than the overall length of males in each year. Because of a deferred maturity schedule, females sampled each year were older and thus longer than males. Males, however, were generally longer than females at specific lake age. Males grew significantly longer than females during the third and fourth year of lake residence. Females, as opposed to males, channel more energy into sexual maturity and less energy into somatic growth. Hence, males were both the smallest (precocious jacks) and largest fish in the adult population. Maiden spawners observed during the study were longer than repeat spawners of lake age-2 and -3. In preparation for spawning, steelhead divert energy into gonadal growth and then deplete reserves during migration and spawning activity. If spawners survive, additional reserves are used to recover from the loss of condition. Maiden steelhead from the same cohorts remain in a lake environment devoting energy to somatic growth. Interestingly, significant differences between maiden and repeat spawners were not observed in the older age categories of either wild or hatchery fish. If maturity is, in part, related to length (Griffith, 1993), then the similarity between maiden and repeat spawner lengths at older lake ages can be attributed to slower growing individuals that have yet to mature. This older group of maiden spawners would then obscure any length differences associated with repeat spawning. 82 Smolt age of wild steelhead influenced adult size only at lake age-1 where 2 year old smolts were longer as adults. After the first season in Lake Michigan, Betsie River adults exhibited equivalent lengths at lake age regardless of time spent in the watershed prior to smolting. Hooten et al. (1987), Kwain (1981), and Karges (1987) all published similar results where smolts of different age and size ultimately achieved the same adult length. Incremental growth rates of wild and hatchery steelhead were not distinguishable. Both wild and hatchery fish obtained considerable length during their first 2 years of lake residence gaining 40.2cm and 39.3cm respectively. Growth coefficients were also nearly identical indicating similar growth. Growth (and growth estimates) can be influenced by food availability, competitive interactions, weather conditions, size selective sampling and mortality (Van Den Avyle 1993). It can thus be inferred that wild and hatchery steelhead are experiencing the same environmental pressures while in Lake Michigan. Length and growth data from the river sport fishery are not necessarily representative of the entire Betsie River steelhead population. By sampling only the spawning population, may have introduced bias because non-maturing members of the same cohorts remain in the lake and are not sampled. Consequently, lengths determined from the river fishery are actually a function of growth an_d maturity. However, for the purposes of comparing relative length and growth, these data are beneficial descriptors of wild and hatchery fish at a critical point in their life history. The data may also be useful in comparing Betsie River fish with steelhead from other systems. Lengths of Betsie River steelhead are within the range reported in other 83 Lake Michigan populations (Biette et al. 1981; Hansen and Stauffer 1971; Seelbach et al. 1994). Migration Timing - Timing is an important trait for the long term survival of an anadrOmous population. Streams may not be in suitable condition if returning adults are not adapted to the watershed. Spawning too early or too late adversely affects embryo development and fry survival (Gharrett and Smoker 1993). Spawning out of synchrony with optimal conditions can have a negative effect on spawner survival (Leider et al. 1984). Although timing is mediated somewhat by temperature and flow, it is primarily under genetic control (Gharrett and Smoker 1993). Numerous examples of altered run timing have been documented when wild populations were supplemented with hatchery fish (Tipping 1984; Leider et al. 1986; Steward and Bjomn 1990). Given the long history of steelhead runs reported in the Betsie River (Wicklund and Dean 1958) migration timing does not appear to be maladapted to the watershed. During the study, spring migration of wild and hatchery fish closely paralleled each other. Median migration dates coincided and occurred in the first or second week in April. The duration of the spring runs also matched, ranging from 6-11 weeks. Wild and hatchery males consistently entered the fishery earlier than females indicating earlier onset of migration. Shapovalov and Taft (1954) and Withler (1966) documented similar behavior in Pacific steelhead populations. The migration profile of Betsie River steelhead also resembles the migrations of most Great Lakes populations summarized by Biette et a1. (1981). 84 Aging Precision - Scale aging based on the comparative results of three readers exhibited considerably consistency (Tables 3 - 5). The APE and CV values were uniformly low across stream, lake and total age determinations indicating a high level of precision. The index of precision (D) value assigned the percent error contributed by each observation (Chang 1982). Steelhead life histories are quite varied which can potentially lead to erroneous aging. Although low, stream age APE had a higher value relative to the lake age APE value. Betsie River scale readers, therefore, where consistently more in agreement when assigning lake ages than stream ages. Total age was determined with an intermediate level of precision. These indices (APE, CV, and D) can all be compared to the precision levels obtained in other scale aging evaluations. For example, the APE values determined in this study for stream, lake, and total age were 3.60%, 1.26%, and 2.33% respectively. Karges (1987) calculated an APE for stream, lake, and total age of 4.56%, 4.65%, and 4.03% respectively during an Ontario steelhead study. In Lake Michigan, the APE for total age averaged 2.17% for chinook salmon otolith aging (Hesse 1994) and 3.63% total age APE for chinook salmon scale aging (Wesley 1996). Smolt Size Influence Influence on Return - One of the most important factors influencing the adult return of anadromous salmonids is smolt size. A long history of research has shown a positive relationship between steelhead smolt length and rate of adult return (Larson and 85 Ward 1954; Wagner 1967; Parkinson and Slaney 1975; Ward and Slaney 1988; Seelbach et a1. 1994). Likewise, my research suggests that the probability of return is size dependent for Betsie River smolts. Mean smolt lengths of returning adults were larger than mean lengths of smolts from the same cohort measured during emigration. This observed inverse of Lee’s phenomenon could have other explanations beside the apparent size based mortality of the smaller fish. Back-calculation error could potentially bias results. Therefore, I used data directly measured from scales as a surrogate for smolt size as well as data based on back-calculation techniques. Both approaches showed that larger smolts were more likely to return as adults than their smaller cohort members. Several plausible mechanisms may be associated with smolt size. Size selective mortality has been demonstrated for sockeye salmon (West and Larkin 1987), chinook salmon (Neilson and Geen 1986) and steelhead (Hume and Parkinson 1987). In each of these studies, the smallest members of the cohort suffered the greatest risk of predation. Predation on emigrating Betsie River smolts would presumably also select against the smallest fish. Predator avoidance would be enhanced by larger size (Ward and Slaney 1990). Hence, the surviving adults would exhibit a greater mean smolt length than the entire cohort would at smolting. Predation by piscivores reduced the numbers of hatchery smolts reaching the Baltic Sea by 26% in a Norwegian stream (Larsson 1985). Smolt loss from avian predators can also be substantial. Wood (1987) found common mergansers, Mergus merganser, to be a major source of mortality for salmon and steelhead smolts in British Columbia. Predators found in the Betsie River system include walleye, northern pike, 86 and bowfin, Amia calva, all of which potentially prey on small steelhead smolts. Even the merganser may prey on Betsie smolts given the French name for the river, “Bec Scies”, meaning saw-toothed duck. Size based failure to completely migrate may be another mechanism that explains smolt length differences. Residualized smolts which do not leave the river revert to parr. Because they do not mature in a lake or ocean, neither do they recruit to the fishery as large adult steelhead. In addition, their in-stream survival is thought to be low (Seelbach 1987). It is the smaller sized smolts that consistently show a high rate of residualism (Ewing et al. 1984; Ward and Slaney 1990). In the Betsie River, for example, the mean smolt length of returning adults from the 1994 hatchery cohort was 20.1cm. The smolts from the same cohort measured at the Homestead weir averaged 18.8cm, while smolts measured at the hatchery prior to release averaged only 13.7cm. Differences in smolt length between the hatchery and the weir may have resulted from a high percentage of small smolts remaining in the stream as residuals. Clearly the smaller smolts were less likely to return as adults than the larger smolts. They either failed to migrate, succumbed to predation and other forms of mortality, or both. Generally, wild steelhead data suggests a similar size dependent relationship between smolt length and adult return. However, the magnitude of difference between smolt length and adult smolt length was greater in the hatchery sample than in the wild sample. Predation mortality may be higher for hatchery steelhead than that experienced by wild fish (Berejikian 1995). Seelbach and Miller (1993) did not detect any size 87 dependent survival in wild steelhead in a Lake Superior tributary. But they did find evidence of higher survival of large hatchery fish stocked in the same stream. Length frequencies of Betsie River wild fish were significantly different between smolts and surviving adults in the 1994 cohort but not in the 1993 cohort. Lack of significance may be attributed to a real lack of length difference or a small sample size. Nonetheless, as Ricker notes (1969), even a small shift in mean length and size distribution requires a correspondingly large selective mortality exerted on the population. Size selective pressures operating against wild age-l smolts were more apparent than in age-2 smolts. Differences between lengths of emigrating smolts and smolt lengths of returning adults decreased with stream age. Intuitively, the smaller age-l smolts would suffer greater mortality and thus produce greater smolt length differences between emigrating smolts and those that returned as adults. Stauffer (1971) and Kwain (1981) found lower proportions of stream age-l steelhead in their adult samples than in their smolt samples of the same cohort. Both authors attributed higher mortality of age-l smolts for the inconsistency. Influence an Age at Maturity - Although large smolts are advantageous to a population in that greater size results in greater survival and return, large size may also affect age at maturity. Maturity schedules of steelhead are thought to be governed by size as well as genetic factors (Tipping 1991). Wagner (1967) found smolt size inversely related to time spent in the ocean. Ward et a1. ( 1989) also provided evidence that smolt size is related to age at maturity in a Pacific steelhead population. 88 I followed Betsie River wild and hatchery cohorts according to lake age and year of return to substantiate smolt length influence on maturity. Very large hatchery smolts appeared to mature earlier than other cohort members. Smolt lengths of lake age-1 hatchery fish were significantly longer than the smolt lengths of older lake age steelhead. Smolt lengths of lake age-2 and -3 fish, however, were not different, indicating size had no differential influence on maturity at these ages. Smolt length influences on maturity were not evident in wild cohorts. Length differences were not significant between lake ages. Partridge (1985) and Tsumura et al. (1987) documented premature sexual development in extraordinarily large hatchery males (>260mm) leading to precocious behavior. In addition, Neilson and Geen (1986) concluded that faster growing males of a chinook salmon cohort matured at an earlier age and often returned as jacks. Similarly, large wild smolts appeared to return more frequently as ocean age-1 steelhead in Vancouver Island populations (Hooten et al. 1987) But no size relationship was evident between ocean age-2 and -3 fish. Back-calculated smolt sizes of the ocean age-2 and -3 fish were approximately the same, as in the Betsie River sample. Reverse Lee’s phenomenon can sometimes be explained by a population’s maturity schedule. Smaller size fish may not mature and recruit to a fishery as quickly as larger fish giving the impression of larger size at age if the immature fish are never sampled. Following Betsie River cohorts through time, past the large age-1 fish, indicates that the probability of adult return is smolt size dependent and not an artifact of maturity schedule. 89 Lake Distribution and River Straying Lake Distribution - Inferences on the geographical distribution and movements of steelhead in lentic environments are frequently based on marked fish recovered by anglers. In Lake Michigan, port creel surveys suggested wide dispersal of Betsie River hatchery steelhead after leaving the harbor at Frankfort. By 1996, Betsie River fish were recorded in waters of all four states bordering the lake. Widespread dispersal of marked steelhead in Lake Michigan was also noted by Seelbach et al. (1994), substantiating the high mobility of these animals. Habitats located in the southern two thirds of the lake appeared to be more attractive to the steelhead. Hatchery fish from the study were not reported in surveys north of Frankfort. Fishing effort in northern Lake Michigan may have influenced the likelihood of recovering RV clipped steelhead. However, Hansen and Stauffer (1971) reported a prevailing southerly movement in Lake Michigan according to recoveries from their steelhead study. Miller et al. (1983) found Columbia River juvenile steelhead to also disperse in one direction (north) upon entering the Pacific coast. Curiously, RV clipped steelhead were absent from the creel along Michigan’s coast from south of Ludington to Holland. This is surprising in light of the heavy fishing activity out of the ports in this area (Rakoczy and Svoboda 1994). In Lake Ontario, Haynes et al. (1986) explained steelhead distribution by temperature and thermal fronts. Generally, steelhead location was associated with water temperatures averaging 9°C and the edge of thermal fronts. These thermal fronts, known on the surface as “scum lines” concentrate terrestrial insects, a preferred food item of steelhead trout (Jude et al. 1987). 90 Lake Michigan chinook and coho salmon are restricted to the southern basin of the lake by temperature constraints during the winter and early spring (Sommers et al. 1981). Non-maturing steelhead may likewise be influenced by temperature which may, in turn, explain the catch of Betsie River fish in southern Lake Michigan. River Straying - Steelhead are renowned for their abilities to “home” back to the natal stream. Early this century, Taft and Shapovalov (1938) first documented the high degree of precision by which steelhead return to their stream of origin. Numerous other studies have verified this fidelity (Lister et al. 1981). Yet straying into non-natal streams has also been documented in steelhead populations. Straying is not entirely detrimental nor benevolent. It can be the mechanism whereby underseeded streams are colonized and can protect populations from localized environmental catastrophes (Moring 1993). Conversely, extensive straying could potentially impact the genetic integrity of discrete stocks and may reduce individual fitness (Lister 1981). I documented stray Betsie River hatchery steelhead in 5 non-natal streams scattered around Lake Michigan. The life history characteristics of the stray fish appeared to be similar to those fish which homed accurately. Because streams are monitored differently and some not at all, only the occurrence and not the magnitude of straying could be identified. The extent of straying by Betsie River wild steelhead could not be documented and is unknown. One stray hatchery fish was observed in a stream of the same region as the Betsie River. The others strayed to distant streams. Biette et al. (1981) found straying in Great Lakes populations primarily in adjacent or nearby streams 91 to the natal stream. Seelbach and Miller (1993), however, reported extensive straying of hatchery steelhead to distant streams. Evidence indicates that homing salmonids return to the same spawning area from which they emerged as fry. This finding led to the “sequential imprint hypothesis” (Lister et al. 1981). Olfactory cues stored during smolt emigration and later recalled in reverse order allow migrating adults to return to the site where they were spawned. Imprecise homing is thought to be related to inaccurate olfactory senses or a disruption of olfactory cues (Leider 1989). After the eruption of Mount St. Helens, Leider (1989) found substantial straying of steelhead from impacted streams. He attributed the straying to increased turbidity and wide ranging temperatures which disrupted sensory acuity. In the case of hatchery supplementation, off site releases away from the rearing station and low in the watershed frequently increase straying (Lister et al. 1981; Chapman et al. 1997). Hatchery steelhead stocked in the Betsie River from the Orsini Hatchery are reared within the watershed and released on site relatively high in the system - a practice shown to minimize straying of cultured fish. Indeed, steelhead from the Orsini Hatchery have been known to home all the way back into the hatchery building itself. CONCLUSIONS AND RECOMMENDATIONS The results of this study indicate that the Betsie River steelhead fishery is supported by both hatchery and wild fish. Although still a substantial contributor (3 0- 46%), wild steelhead do not match the relative proportion (93%) reported by Seelbach and Whelan (1988). Management should give careful consideration regarding the purpose of the hatchery program on the Betsie River. If the only goal for the fishery and hatchery is to provide an adequate local harvest to meet angler demand, then wild contribution is not a concern. Large hatchery releases will likely fill any void in catch rates. If the abundance and sustainability of wild populations in Lake Michigan is a management goal, as put forth by the MDNR Fisheries Division (MDNR 1997) and the Great Lakes Fishery Commission (GLFC 1992), then the Betsie River population warrants further scrutiny. Estimating harvest or actual numbers of wild adult spawners would be an appropriate management objective. Quantifying wild steelhead numbers answers the question of whether a real decline of wild fish exists or merely a decline relative to hatchery stocking levels. The three year study gives only a brief overview into the Betsie River steelhead population. Long term data sets are much more valuable in deciphering trends than data sets limited to a few years. Other than continuing the study, one method for extending the Betsie River data would be to examine past scale collections. An annual creel survey was conducted on the Betsie River during the mid to late 1980’s. Fishing effort and harvest estimates were calculated along with the collection of scale samples. Origins 92 93 have never been determined from these scales. Currently, the scales are being archived at the MDNR research station at Charlevoix (Jory Jonas, MDNR, Personal Communications). An analysis of these scales would help in establishing long term trends in the Betsie River without any further scale collection efforts. Ratio 23 was an accurate discriminator of Betsie River steelhead origins. Until all hatchery steelhead stocked in Lake Michigan are marked, ratio 23 should remain a valuable tool for estimating wild and hatchery composition in a mixed population. Unfortunately, the origins of 9% of the sampled Betsie River fish could not be determined because of poor scale quality. Many steelhead have scales that are regenerated or that have reabsorbed edges. I found that removing at least 10 scales from each fish in the preferred area was necessary to ensure a usable scale sample. I found life history characteristics of wild and hatchery steelhead to be both similar and distinct. Length and growth, migration timing, and sex ratios were nearly identical. Conversely, age at maturity, repeat spawning frequency, and age structure, were not alike. The Betsie River is a marginal trout stream at best (Newcomb 1998) and will likely always need supplementation to satisfy current angling pressure. Ideally, the life history patterns of the stocked hatchery fish should parallel those of the wild fish. Having phenotypes similar to locally adapted steelhead increases the odds of success for hatchery fish and will not compromise the wild population (Kapuscinski and Jacobson 1987) The characteristics of the Little Manistee River strain may be similar enough to wild steelhead in the Betsie River to justify its continued use as a donor stock. Genetic 94 differences between the two may, in fact, be small. In addition, the apparent differences in traits may be due to energetic or environmental factors related to hatchery rearing. However, after over 100 years and multiple generations, discrete stocks of steelhead have been identified in Lakes Ontario and Superior (Ferguson et al. 1993; Krueger et al. 1994). Phenotypic traits as well as underlying genetics differed between populations in the two lakes. Therefore, it would be prudent to continue the preliminary genetic study of Lake Michigan steelhead begun by Epifanio (1996). Hatchery production now contributes the majority of fish in the river fishery. Large smolts stocked into the Betsie River provided the greatest return of adult steelhead. Efforts to raise large smolts of 19cm or more should be encouraged in order to ensure consistent returns. Small smolts may not only be lost to the fishery, but may also adversely affect wild parr (Steward and Bjomn 1990). Further research directed at the effects of smolt size on maturity would also be beneficial in defining an appropriate size range for stocked steelhead hatchery smolts. Straying by Betsie River hatchery fish was evident in several Lake Michigan streams. Because hatchery fish are currently raised and released relatively high in the watershed, additional options to reduce straying are few. In light of the strays from the Betsie River as well as steelhead straying into the Betsie River, managers should not view Lake Michigan tributaries as entirely isolated reproductive units. Finally, the angling public has long held the Betsie River and its steelhead fishery in high esteem. It remains a valuable economic and ecological resource that merits continuing evaluation and protective vigilance. 95 APPENDICES APPENDIX A ANGLERS £3- YOURHELPISNEEDED! mmmmwmmmmmmmrmmwm 112de 199San‘lsu1'lgl%nn HOW'IOSAMPIE l) SameStolOsuisun'rgapodetlaifeJmflns’dedthfishhelowarflhfmt d‘thedorsdfin 2) lrsrirt‘L ' l ‘L Flau‘L-T;. ‘L ‘ ".— Smierenmfldosmt Inwmbeafatflpmeduealdfishflntmlnnfledgaflywimmym(mddn) rumvedamhereunmdtotlnmter ifthired. 3) lugflisluldbenmredfnmtlrtipofthmnoflnenldthtail 4) Duck for fin clip arxl write dip locatim at scale envelope. Fish may have mltifle dim. Scrape Scales Adipose Fin (AD) Right Maxillary (RM) h— 7 '4 V - — _ Right Ventral (RV) ‘ Right ectoral(RP) Total Length mmamwmmnmémQ—Wemmrymdnyammh and I‘Inll-IKLLBI 1 r 1, 1:}, a ' v r 1, yuwfiLl 11mm Myrnftryurprfidpfiafl Timmy J. Mm Plan (517) 3362760 Jiml-larbedt (517)3551&l Wilma-chm Fax(517)m1699 MdingtateUnva-sty 96 APPENDIX B Table 17. Marked hatchery strays and probable origin from the Betsie River fishery 1994-1996 and the 1996 kelt sample. YEAR DATE LENGTH FIN PROBABLE PROBABLE (mm) CLIP STRAIN ORIGIN 1994 4/5 649 LP Skamania St. Joseph R., IN 1994 4/17 813 AD L. Manistee various rivers, MI 1994 4/19 787 AD L. Manistee various rivers, MI 1994 4/20 622 LP Skamania St. Joseph R., IN 1995 3/23 584 ADRP Skamania St. Joseph R., IN 1995 4/13 578 RP L. Manistee various rivers, MI 1995 4/14 572 RP L. Manistee various rivers, MI 1995 4/14 578 ADLV Ganaraska Kewaunee or Root R, WI 1995 4/26 559 LP Skamania St. Joseph R., IN 1995 4/29 610 RP L. Manistee various rivers, MI 1996 12/3 419 RP L. Manistee various rivers, MI 1996 4/13 775 RM Skamania Kewaunee or Root R, WI 1996 4/13 699 RM Ganaraska Kewaunee or Root R, WI 1996 4/24 648 LV Skamania Kewaunee or Root R, WI 1996 5/2 737 LP Skamania St. Joseph R., IN 1996 5/5 711 LP Skamania St. Joseph R., IN 1996 5/5 648 RP L. Manistee various rivers, MI 1996 5/15 597 RP L. Manistee various rivers, MI 1996 5/15 508 RP L. Manistee various rivers, MI 96 kelt 5/14 737 LMRV Chambers Cr Sheboygan R., WI 96 kelt 5/14 686 LM Skamania St. Joseph R., IN 96 kelt 5/ 14 787 LM Chambers Cr Kewaunee or Root R, WI 96 kelt 5/ 14 706 LM Chambers Cr Kewaunee or Root R, WI 96 kelt 5/19 732 LM Chambers Cr Kewaunee or Root R, WI 96 kelt 5/19 673 RMRP Ganaraska Sheboygan R., WI 96 kelt 5/23 749 RM Skamania Kewaunee or Root R, WI Marked hatchery strays as percentage of sample: 1994 - 2.1% (n=191) 1995 - 4.6% (n=131) 1996 - 5.7% (n=157) 96 Kelt - 11% (n=66) 97 APPENDIX C Table 18. Relative composition in 1996 of wild and hatchery steelhead by river section and gear. PERCENTAGE Origin Entire Lower Middle Upper Kelt River 1 Section ' Section ' Section "3 Samle 2‘3 Wild 30% 40% 32% 14% 51% Hatchery 70% 60% 68% 86% 49% ' Hook and line sample 2 Temporary block net sample 3 Sampled from steelhead caught above Homestead weir 98 APPENDIX D Table 19a. Preliminary data of Betsie River steelhead collected from the 1994 sampled fish. Sample ID# Date Length (cm) SEX Site Coll Clip I-IIW AGE 7 1 94 25-Mar 2 M HSD JW H 1.3 2 94 25-Mar 58 M HSD JW W 2.2 3 94 25-Mar 64 F HSD JW H 1.2 4 94 25-Mar 64 F HSD HS H 1.3 5 94 26-Mar 53 M HSD EH H 1.2 6 94 26-Mar 61 M HSD EH W 1.281 7 94 26-Mar 71 F HSD EH ? H.281 8 94 26Mar 83 M FREDS DT W 1.3 9 94 26-Mar 82 M FHEDS DT H 1.1818181 10 94 26-Mar 76 F RR HS H 1.4 11 94 26-Mar 46 M HSD HS H 1.2 12 94 26-Mar 69 F HSD HS W 2.3 13 94 26-Mar 64 M MDWS HS W 2.2 14 94 26-Mar 74 M HSD H8 W 2.3 15 94 26Mat 66 M HSD HS H 1.2 16 94 26-Mar 76 M HSD HS W 2.3 17 94 26-Mar 71 M HSD HS W 2.3 18 94 26-Mar 76 M HSD H8 W 2.3 19 94 27-Mar 64 M HSD HS H 1.3 20 94 27-Mar 74 F HSD HS W 2.3 21 94 28-Mar 69 F HSD HS W 1.3 22 94 28-Mar 79 M HSD HS H 1.4 23 94 28-Mar 76 F HSD HS W 1.3 24 94 28-Mar 61 M HSD HS ? F12 25 94 28-Mar 69 F HSD HS W 2.381 26 94 28-Mar 62 F HSD EH H 1.2 27 94 28-Mar 69 M HSD EH W 2.3 28 94 28-Mar 64 M HSD HS W 2.2 29 94 29—Mar 58 M HSD HS W 1.2 30 94 29-Mar 69 F HSD HS ? H.2 31 94 29-Mar 66 F HSD HS W 2.281 32 94 29-Mar 66 M HSD HS W 2.3 33 94 29-Mar 71 F HSD HS H 1.3 34 94 29-Mar 61 M HSD HS W 1.2 35 94 29-Mar 66 M HSD HS W 2.2 36 94 29-Mar 69 M HSD HS H 1.281 99 100 Table 19a (cont’d) Sample ID# Date Length Em) SEX Site Coll Clip H/W AGE 74 37 94 29-Mar M HSD HS H 1.3 38 94 29-Mar 74 M HSD HS H 1.3 39 94 29-Mar 69 F HSD HS H 1.3 40 94 30-Mar 71 F HSD HS H 1.381 41 94 30-Mar 69 F HSD HS H 1.3 42 94 31-Mar 74 M HSD HS H 1.3 43 94 31-Mar 69 F HSD HS W 2.2S1 44 94 31-Mar 71 M HSD HS W 1.3 45 94 31-Mar 58 F HSD HS H 1.2 46 94 31-Mar 77 M HSD HS H 1.3 47 94 31-Mar 53 F HSD HS W 2.2 48 94 31-Mar 66 F HSD HS H 1.3 49 94 31-Mar 58 M HSD HS H 1.2 50 94 31-Mar 72 M HSD HS W 1.3 51 94 31-Mar 84 M HSD HS W 2.4 52 94 1-Apr 71 F HSD EH W 2.28181 53 94 2-Apr 61 F HSD EH ? R2 54 94 2-Apr 62 M HSD EH H 1.1 55 94 2-Apr 53 M HSD EH H 1.2 56 94 2-Apr 74 M HSD EH H 1.3 57 94 2-Apr 79 M HSD EH H 1.3 58 94 2-Apr 66 NONE TN W 2.3 59 94 2-Apr 41 NONE RR TN H 1.1 60 94 2-Apr 66 F HSD HS W 2281 61 94 2-Apr 84 F HSD PS H 1.381 62 94 3-Apr 61 F HSD EH W 1.2 63 94 3-Apr 79 M HSD EH ? R4 64 94 3-Apr 71 F HSD EH W 1.3 65 94 3-Apr 71 F HSD EH H 1.3 66 94 3-Apr 67 F HSD EH H 1.2 67 94 3-Apr 72 M HSD JW H 1.3 68 94 3-Apr 58 F HSD JW H 1.2 69 94 4-Apr 71 F RR PS H 1.3 70 94 4-Apr 69 M HSD PS W 2.3 71 94 4-Apr 56 F HSD PS ? R2 72 94 4-Apr 71 M HSD PS W 1.3 73 94 4-Apr 61 M HSD PS H 1.2 74 94 4oApr 74 F HSD PS H 1.3 75 94 4-Apr 71 F HSD PS W 2.4 76 94 4-Apr 58 M HSD PS W 2.181 77 94 4-Apr 64 F HSD PS ? R2 78 94 4-Apr 50 M HSD PS H 1.1 79 94 4-Apr 62 F HSD PS H 2.2 80 94 4-Apr 72 F FREDS DT H 1.3 81 94 4-Apr 81 F FREDS DT W 1.331 101 Table 196 (cont’d) Sample ID it Date Length (cm) SEX Site Coll Clip I-I/W AGE 92 94 4-Apr 69 M FREDS DT W 2.3 93 94 4-Apr 70 F HSD EH H 1.3 94 94 4-Apr 77 M HSD EH w 1.331 95 94 4-Apr 67 F HSD EH w 1.3 86 94 441;» 70 F HSD JW w 1.3 97 94 4-Apr 69 F HSD JW w 2.3 99 94 5-Apr 56 M HSD JW w 2.2 99 94 5-Apr 70 M HSD JW H 1.3 90 94 5-Apr 70 F JW H 1.3 91 94 s-Apr 67 F HSD JW LP H 1.3 92 94 5-Apr 64 F HSD JW ? R3 93 94 5-Apr 57 F LT w 2.2 94 94 5-Apr 67 F LT W 2281 95 94 5-Apr o M LT H 1.3 96 94 5-Apr 70 F HSD PS w 2.3 97 94 6-Apr 74 F HSD PS H 1.4 99 94 6~Apr 64 F HSD PS ? R2 99 94 7-Apr 61 F RR PS H 1.2 100 94 7-Apr 67 F RR or H 1.3 101 94 7-Apr 59 M RR DT H 1.2 102 94 8-Apr 69 M RR PS w 2.3 103 94 8oApr 75 M PS H 1.3 104 94 8-Apr 74 M PS H 1.281 105 94 9-Apr 47 M RR or w 2.1 106 94 9-Apr 76 M RR DT H 1.331 107 94 9oApr 69 M LOWER JW H 1.3 109 94 9-Apr 74 M RR or w 1.3 109 94 9-Apr 66 F HSD EH 2 R2 110 94 9-Apr 74 M HSD EH w 2.3 111 94 9-Apr 61 F HSD EH W 2.2 112 94 9Apr 59 M LOWER JW H 1.2 113 94 9-Apr 69 F Hso EH H 1.3 114 94 10-Apr 75 F LOWER LT H 1.4 115 94 11-Apr 71 F HSD EH H 1.3 116 94 11-Apr 69 F HSD EH H 1.3 117 94 11-Apr 79 F HSD EH w 138181 119 94 11-Apr 91 M HSD EH ? R4 119 94 11-Apr 56 F HSD EH w 1.2 120 94 11-Apr 66 F HSD EH w 2.3 121 94 11-Apr 61 F HSD EH 7 R2 122 94 11-Apr 66 F HSD EH w 2.3 123 94 12-Apr 43 M HSD PS w 2.1 124 94 12-Apr 71 M PSUTKA JW H 1.3 125 94 13—Apr 46 M PSUTKA JW ? R2 126 94 1:3-Apr 59 M HSD PS W 1.2 Table 193 (cont’d) 102 Sample ID 11 Date Length (cm) SEX Site Coll Clip HIW AGE 127 94 14-Apr 66 F Hso PS H 1.3 129 94 14-Apr 55 M PSUTKA JW w 2.2 129 94 15-Apr 57 F HSD LT w 2.2 130 94 15-Apr 76 F HSD PS w 2.3 131 94 16—Apr 76 F Hso PS H 1.4 132 94 16-Apr 64 F HSD PS H 1.3 133 94 16-Apr 56 M H5O PS H 1.2 134 94 16-Apr 72 M FREDS OT H 1.3 135 94 17-Apr 74 M FREDS OT H 1.3 136 94 17-Apr 90 M H5O LT w 2.351 137 94 17-Apr 91 M HSD PS AD H 1.4/1.5 139 94 17-Apr 59 M HSD LT H 1.2 139 94 17-Apr 69 F HSD LT 2 R281 140 94 17-Apr 99 F LT H 1.5 141 94 17-Apr 66 M HSD PS w 2.3 142 94 17-Apr 66 F H5O PS H 1.3 143 94 17-Apr 59 F PIER PS w 2.2 144 94 17-Apr 56 M HSD PS H 1.2 145 94 18—Apr 67 F HSD LT w 2.3 146 94 18-Apr 66 F HSD LT H 1.3 147 94 18-Apr 66 F HSD LT w 1.3 149 94 18-Apr 66 F H5O PS H 1.3 149 94 19-Apr 66 M HSD PS H 1.3 150 94 19-Apr 67 F HSD LT w 2.3 151 94 19-Apr 79 F HSD LT AD H R.25151 152 94 19-Apr 67 M PSUTKA JW H 1.2 153 94 19-Apr 67 F PSUTKA JW w 1.3 154 94 19-Apr 44 M PSUTKA JW w 2.1 155 94 19-Apr 69 F HSD EH H 1.3 156 94 19-Apr 64 F HSD EH H 1.3 157 94 20-Apr 71 F HSD EH H 1.3 159 94 20-Apr 71 M Hso EH H 1.3 159 94 20~Apr 71 F HSD EH H 1.351 160 94 20-Apr 79 M HSD EH H 1.4 161 94 20-Apr 62 M PSUTKA JW LP H R2 162 94 20-Apr 94 F HSD PS w 2.25151 163 94 21-Apr 53 M H5O PS w 2.2 164 94 22-Apr 74 M HSD PS w 2.3 165 94 22-Apr 69 F FREDS OT H 1.3 166 94 22-Apr 66 F H5O LT w 2.3 167 94 23Apr 91 F HSD PS H 1.5 169 94 23-Apr 62 M FREDS OT H 1.2 169 94 23-Apr 74 M JW w 2.3 170 94 24-Apr 60 F FREDS OT W 2.2 171 94 24-Apr 60 M HSD PS H 1.2 103 Table 19a (cont’d) Sample ID# Date Length (cm) SEX Site Coll Clip HIW AGE 172 94 24-Apr 66 F HSD PS H 1.3 173 94 26-Apr 74 F HSD PS H 1.4 174 94 27-Apr 54 F FREDS OT 7 R2 175 94 27-Apr 64 F HSD PS w 2.3 176 94 28-Apr 46 M HSD PS H 1.1 177 94 29-Apr 60 M HSD PS 2 R2 179 94 29-Apr 74 F PSUTKA JW w 1.3 179 94 3671;» 74 NONE HSD PS H 1.3 190 94 30-Apr 57 M RR OT H 1.2 191 94 5-May 69 NONE LT H 1.3 192 94 6-May 71 F FREDS OT H 1.3 193 94 6-May 44 M FREDS OT H 1.1 194 94 6-May 60 F FREDS OT H 1.2 195 94 6May 67 M FREDS OT H 1.3 196 94 26-Mar 74 M Hso HS w 1.3 197 94 26-Mar 69 M H5O HS w 2.3 199 94 31-Mar 64 M HSD HS H 1.3 199 94 2-Apr 66 M Hso HS H 1.2 190 94 4-Apr 74 M HSD H5 2 R3 191 94 31-Mar 59 M HSD HS w 2.3 104 Table 19b. Preliminary data of Betsie River steelhead collected from 1995 sampled fish. Sample ID 11 Date Leflgth (cm) Sex Site Coll Clip HIW AGE F1 94 29-Oct 70 F GRACE JW H 1.3+ F2 94 29-Oct 44 M RR JW W 2.0+ F3 94 25-Nov 77 M RR JW W 2.3+ F4 94 25-Nov 55 M RR JW H 1.1+ F5 94 25-Nov 72 M RR JW W 1.2+ F6 94 27-Nov 69 M RR JW W 2.2+ F7 94 29-Nov 62 M M20 JW W 1.1+ F8 94 29-Nov 72 M M20 JW H 1.2+ F9 94 1-Dec 74 M LOWER JW H 1.2+ F10 94 9-Dec 64 M RR JW H 1.2+ 1 95 4-Jan 71 M HSD H 1.4 2 95 4-Jan 64 F HSD W 1.3 3 95 22-Mar 69 M RR JW W 3.3 4 95 24-Mar 69 M HSD H 1.281 5 95 25-Mar 69 M HSD H 1.3 6 95 3-Apr 81 F HSD H 1.4 7 95 3-Apr 61 M HSD H 1.2 8 95 3-Apr 89 F HSD W 2.28181 9 95 3-Apr 76 F HSD W 1.2813181 10 95 4-Apr 58 M HSD PS W 1.2 11 95 4-Apr 69 M HSD H 1.281 12 95 4-Apr 69 F HSD W 1.3 13 95 4-Apr 42 M HSD W 2.1 14 95 5-Apr 69 M PTSUKA JW W 1.3 15 95 7-Apr 81 M HSD H 1.3 16 95 6~Apr 67 F HSD JW W 2.4 17 95 7-Apr 79 M HSD H 1.4 18 95 8-Apr 69 F HSD TN H 1.3 19 95 8-Apr 71 F HSD H 1.3 20 95 8-Apr 69 F HSD H 1.3 21 95 8-Apr 64 M HSD H 1.3 22 95 8-Apr 53 F PETE H 1.2 23 95 8-Apr 71 F W 1.3 24 95 9-Apr 48 M PETE H 1.1 25 95 10-Apr 89 M PETE ? R5? 26 95 10-Apr 89 M PETE H 1.5 27 95 10-Apr 61 F HSD W 1.2 28 95 10-Apr 81 F HSD H 1.381 29 95 10-Apr 69 M PTSUKA JW H 1.3 30 95 12-Apr 84 F HSD H 1.4 31 95 13-Apr 61 F HSD H 1.2 32 95 13-Apr 71 F HSD ? R.3 105 Table 19b (cont’d) Sample ID it Date Length(cm) Sex Site Coll Clip HIW AGE 33 95 13-Apr 59 F RR RP H R2 34 95 13~Apr 65 F RR w 1.3 35 95 13-Apr 72 M RR JW w 1.3 36 95 13-Apr 56 M RR w 1.2 37 95 14-Apr 57 M RR ET RP H 1.2 39 95 14-Apr 59 M RR LV H 1.2 39 95 14-Apr 57 M RR H 1.2 40 95 14-Apr 70 F RR JW H 1.3 41 95 15-Apr 76 F HSD H 1.3 42 95 15-Apr 79 F HSD H 1.4 43 95 15-Apr 57 F RR JW w 2.2 44 95 16-Apr 72 F PTSUKA JW H 1.3 45 95 17-Apr 61 F RR OT H 1.2 46 95 17-Apr 70 F RR OT w 1.3 47 95 18-Apr 65 F FREDS DT H 1.3 49 95 18-Apr 56 F FREDS OT H 1.2 49 95 19-Apr 71 F HSD H 1.3 50 95 19-Apr 69 F HSD H 1.3 51 95 19-Apr 74 M HSD H 1.3 52 95 19-Apr 59 F HSD JW H 1.2 53 95 20-Apr 64 M HSD H 1.3 54 95 20-Apr 53 F H5O H 1.2 55 95 20-Apr 64 M HSD w 2.2 56 95 20-Apr 69 F HSD H 1.251 57 95 20—Apr 66 F HSD w 1.2 59 95 20-Apr 61 M HSD ? R2 59 95 21-Apr 69 F PTSUKA JW 7 R8 60 95 21-Apr 55 M PTSUKA JW w 1.1 61 95 22-Apr 67 F KURICK JW H 1.351 62 95 22-Apr 57 F KURICK JW w 1.3 63 95 22-Apr 77 F KURICK JW H 1.3 64 95 22-Apr 76 F KURICK JW w 1.4 65 95 23-Apr 71 F FREDS OT w 1.4 66 95 26-Apr 56 M KURICK JW LP H 1.2 67 95 26-Apr 65 F KURICK JW H 1.3 69 95 26-Apr 67 F KURICK JW w 1.3 69 95 28-Apr 55 F FREDS OT w 2.2 70 95 29-Apr 71 M KURICK JW H 1.3 71 95 29-Apr 65 F FREDS OT H 1.2 72 95 29~Apr 64 F FREDS OT w 1.3 73 95 29-Apr 61 M KURICK JW 7 R2 74 95 29-Apr 61 F KURICK JW RP H H R2 75 95 29-Apr 65 M KURICK JW w 2.3 76 95 30-Apr 59 M KURICK JW H 1.2 106 Table 19b (cont’d) Sample ID it Date LM (cm) Sex Site Coll Clip l-IlW AGE 77 95 30—Apr 64 F KURICK JW H 1.2 79 95 3-May 73 M HSD OT H 1.3 79 95 3-May 40 M HSD OT w 2.1 90 95 3-May 64 F HSD OT w 2.3 91 95 3-May 70 F HSD OT H 1.3 92 95 3-May 63 F HSD OT H 1.2 93 95 5-May 69 F FREDS OT w 2.25151 94 95 9-May 64 F Hso JH w 2.251 95 95 10-Mar 65 F LOWER JW w 2.3 86 95 10-Mar 73 NONE LOWER JW w 2.351 97 95 10-Mar 64 M LOWER JW H 1.3 99 95 12-Mar 65 M HSD H 1.3 99 95 15-Mar 61 M HSD w 1.2 90 95 15.Mar 50 M H5O 7 R1 91 95 17-Mar 76 M HSD H 1.3 92 95 17-Mar 74 M HSD w 2.3 93 95 18-Mar 71 F HSD H 1.3 94 95 18-Mar 76 M HSD JK 7 DET 95 95 18-Mar 61 F CARL ? R2 96 95 18-Mar 66 M HSD H 1.251 97 95 18-Mar 76 F PETE H 1.3 99 95 18-Mar 94 M PETE H 1.4 99 95 18-Mar 59 F HSD PETE w 1.2 100 95 18-Mar 62 M HSD H 1.2 101 95 19.Mar 56 F :7 R181 102 95 19-Mar 76 F PETE 2 R3515 103 95 19-Mar 74 M HSD w 1.3 104 95 19-Mar 74 M HSD H 1.3 105 95 20-Mar 56 F HSD PETE w 1.2 106 95 20-Mar 90 M MAX ? DET 107 95 20-Mar 99 M HSD w 1.4 109 95 21-Mar 74 F HSD LB w 2.251 109 95 21-Mar 74 F MAX H 1.3 110 95 23-Mar 59 F HSD RPAD H 1.2 111 95 23-Mar 76 M HSD H 1.4 112 95 29-Mar 71 M ? R3 113 95 28-Mar 56 F w 1.2 114 95 29-Mar 61 M H 1.2 115 95 3-Apr 64 M HSD w 1.3 116 95 3-Apr 76 F HSD TN H 1.3 117 95 3-Apr 69 F HSD TN H 1.2 119 95 3-Apr 59 F HSD TN 7 R2 119 95 4-Apr 76 F HSD JW w 3.451 120 95 19-May 39 M H5O TN w 1.1 121 95 14-May 70 F HSD DW H 1.4 107 Table 19c. Preliminary data from Betsie River steelhead collected from 1996 sampled fish. Sample ID it Date Length (cm) Sex Site Coll Clip HIW Age 11 95 29-Oct 41 M RR JW w 1.0+ 21 95 29-Oct 65 M RR JW H 1.2+ 31 95 13-Nov 69 M M-22 HE H 1.2+ 41 95 30-Nov 59 M RR JW w 2.1+ 51 95 3-Dec 42 M RR JW RP H 1.0+ 1 96 9-Feb 53 M HSD EH RP H 1.2 2 96 9-Feb 94 F Hso EH H 1.4 3 96 9-Feb 91 F HSD EH w 2.451 4 96 9-Feb 77 M Hso EH w 2.5 5 96 22-Feb 66 F HSD EH w 2.251 6 96 22-Feb 64 F HSD EH w 2.251 7 96 22-Feb 69 F HSD EH 2 R251 9 96 22-Feb 76 M HSD EH H 1.4 9 96 22-Feb 94 M HSD EH H 1.4 10 96 22-Feb 79 M HSD EH w 1.5 11 96 23-Feb 71 M HSD EH H 1.251 12 96 23-Feb 69 M H5O EH ? R3 13 96 23-Feb 74 M HSD EH H 1.3 14 96 23-Feb o M HSD EH H 1.3 15 96 23~Feb o M HSD EH 7 R281 16 96 25-Feb 71 F ws H 1.3 17 96 26-Feb 42 M RR JW W 1.1 19 96 26—Feb 43 M RR JW w 1.1 19 96 27-Feb 42 M RR w 1.1 20 96 27-Feb 66 M RR JW RV H 1.3 21 96 2—Mar 50 M RR JW 2 R2 22 96 2-Mar 69 M RR JW H 1.4 23 96 13-Mar 66 F HSD SA w 2.3 24 96 13-Mar 61 F HSD SA w 2.3 25 96 13-Mar 64 F HSD SA w 1.25151 26 96 16-Mar 79 F HSD H 1.3 27 96 16-Mar 69 F RR JW H 1.3 29 96 16Mar 72 M RR JW H 1.351 29 96 17-Mar 71 M HSD H 1.3 30 96 17411161 64 F HSD H 1.3 31 96 17-Mar 43 M HSD H 1.2 32 96 22-Mar 61 F MOUTH w 2.3 33 96 22-Mar 65 F MOUTH H 1.3 34 96 22-Mar 41 M MOUTH H 1.1 35 96 22-Mar 67 F MOUTH H 1.2 36 96 28-Mar 76 F RR JW ? R.4 108 Table 19c (cont’d) Sample ID it Date Length (cm) Sex Site Coll Clip HIW Age 37 96 29-Mar 51 M RR JW w 1.151 39 96 30-Mar 49 M MOUTH H 1.1 39 96 1-Apr 67 F RR JW H 1.3 40 96 2-Apr 69 F RR JW w 1.351 41 96 2-Apr 71 M RR JW w 1.3 42 96 5-Apr 62 F RR JW w 1.3 43 96 6-Apr 75 F RR JW W 2.4 44 96 6-Apr 64 F RR JW H 1.2 45 96 7-Apr 70 M RR JW w 2.3 46 96 7-Apr 65 F RR JW RV H 1.3 47 96 7-Apr 72 M RR JW Rv H 1.3 49 96 7-Apr 65 M RR JW H 1.3 49 96 8-Apr 43 M RR JH w 1.1 50 96 8-Apr 64 M RR JH w 2.2 51 96 8-Apr 62 F MOUTH W H 1.2 52 96 8-Apr 64 F MOUTH Tw H 1.3 53 96 9-Apr 73 M HSD JH Rv H 1.3 54 96 9-Apr 74 F HSD JH H 1.3 55 96 9-Apr 75 M H50 JH w 1.4 56 96 9-Apr 66 M H50 JH H 1.3 57 96 9-Apr 69 F MOUTH 9.1 H 1.251 59 96 10—Apr 39 M RR JW H 1.1 59 96 10-Apr 60 F RR JW w 2.3 60 96 10-Apr 64 F HSD EH H 1.25151 61 96 10—Apr 61 F HSD EH w 2.2 62 96 10-Apr 66 M H50 w 1.3 63 96 1o-Apr 76 F H50 H 1.351 64 96 11-Apr 67 M RR JW H 1.3 65 96 11-Apr 60 F RR JW H 1.2 66 96 11-Apr 71 M HSD H 1.3 67 96 11-Apr 61 M PETE ? R2 69 96 11-Apr 66 F PETE H 1.2 69 96 11-Apr 69 F PETE H 1.3 70 96 12-Apr 64 F PETE H 1.2 71 96 12-Apr 64 F H50 H 2.3 72 96 12-Apr 71 M HSD H 1.2 73 96 12-Apr 79 M HSD H 1.4 74 96 12-Apt 76 M HSD H 1.4 75 96 12-Apr 64 F H50 w 1.251 76 96 12-Apr 67 F H50 H 1.3 77 96 12-Apr 59 M H50 w 2.2 79 96 12—Apr 69 F HSD H 1.3 79 96 12-Apr 79 F W 1.451 90 96 12-Apr 74 F HSD H 1.351 91 96 12-Apr 76 M JONES MENO 7 REG l 09 Table 19c (cont’d) Sample ID it Date Lerlgth (cm) Sex Site Coll Clip HIW Age 92 96 12-Apr 64 M HSD W 2.2 93 96 12-Apr 66 F JONES MENO H 1.251 94 96 12-Apr 69 F HSD PETE w 1.3 95 96 12-Apr 74 F PETE H 1.4 96 96 12-Apr 76 F PETE H 1.4 87 96 12-Apr 61 F PETE H 1.3 99 96 12-Apr 64 M JONES MERCIER H 1.3 99 96 13-Apr 70 F RR JW RM w 2.3 90 96 13-Apr 65 M RR M H 1.3 91 96 13-Apr 77 M RR JW RM H 1.351 92 96 13-Apr 61 F MOUTH w 2.2 93 96 13-Apr 69 F MOUTH H 1.3 94 96 13-Apr 79 F MOUTH TN RV H 1.3 95 96 13-Apr 60 F MOUTH TN W 2.2 96 96 13-Apr 66 M H50 TN 7 R2 97 96 13Apr 76 F MOUTH w 2.3 99 96 13-Apr 51 M MOUTH H 1.2 99 96 17-Apr 75 M 7 R3 100 96 17-Apr 76 F HSD M w 2.25151 101 96 17-Apr 72 M H 1.3 102 96 19-Apr 56 F HSD H 1.2 103 96 19-Apr 66 F H50 H 1.3 104 96 21-Apr 70 F PSUTKA JW Rv H 1.3 105 96 22-Apr 71 F PSUTKA JW H 1.3 106 96 23-Apr 69 F HSD EH ? DET 107 96 23-Apr 69 F H50 EH ? OET 109 96 23-Apr 66 F HSD EH ? DET 109 96 23-Apr 70 F PSUTKA Jw H 1.4 110 96 24-Apr 69 M PSUTKA M H 1.3 111 96 24-Apr 60 M PSUTKA JW H 1.2 112 96 24-Apr 67 M PSUTKA M H 1.251 113 96 24-Apr 67 M PSUTKA JW H 1.3 114 96 24-Apr 65 F PSUTKA JW LV H 1.251 115 96 24-Apr 71 F HSD H 1.251 116 96 24-Apr 69 F HSD H 1.3 117 96 25-Apr 46 M H50 H 1.1 119 96 25-Apr 76 F HSD H 1.4 119 96 25-Apr 71 F HSO H 1.3 120 96 26-Apr 60 F HSD JW Rv H 1.3 121 96 27-Apr 77 M PSUTKA Jw Rv H 1.3 122 96 27-Apr 69 M PSUTKA M W 2.3 123 96 27-Apr 66 F PSUTKA JW H 1.3 124 96 28-Apr 67 F PSUTKA JW H 1.3 125 96 28-Apr 90 M PSUTKA M H 1.4 126 96 28-Apr 77 M PSUTKA JW H 1.4 127 96 30~Apr 74 M PSUTKA Jw RV H 1.3 Table 190 (cont’d) 110 Sample ID # Date Length (cm) Sex Site Coll Clip HIW Age 129 96 30-Apr 66 M KURICK JW W 1.3 129 96 30-Apr 70 F PSUTKA JW RV H 1.3 130 96 1-May 59 M H50 JH H 1.2 131 96 2-May 61 F HSD JH w 2.2 132 96 2-May 69 F HSO JH w 2.3 133 96 2-May 79 F H50 JH H 1.351 134 96 2-May 74 M HSO JH LP H 1.4 135 96 2-May 66 F HSO JH w 2.3 136 96 2-May 74 F HSD JH H 1.3 137 96 2-May 59 F HSD JH H 1.2 139 96 2-May 61 F HSD JH w 2.2 139 96 2-May 74 F H50 JH H 1.4 140 96 3-May 71 F HSD JH 0 H 1.251 141 96 5-May 65 F HSO TN RP H 1.3 142 96 5-May 71 F H50 TN LP H 1.351 143 96 9-May 76 F H50 MW H 1.4 144 96 13-May 71 F HSD MW H 1.3 145 96 13-May 66 F H50 MW H 1.3 146 96 13-May 66 F HSD MW H 1.3 147 96 14-May 47 M H50 TN w 2.1 149 96 15-May 51 M H50 RP H 1.1 149 96 15-May 60 F HSD RP H R2 150 96 15-May 62 F H50 w 2.2 151 96 15-May 69 F H50 H 1.25151 152 96 15—May 91 F HSD w 2351 Kelts '1k 96 14-May 79 M HSD JH '2k 96 14-May 79 M HSD JH LM w 1.5 '3k 96 14-May 70 M HSD JH LM H 1.3 4k 96 14-May 55 M H50 JH H 1.2 '5k 96 14—May 72 M HSD JH 6k 96 14-May 67 M HSO JH H 1.2 7k 96 14-May 74 F H50 JH Rv H 1.3 8k 96 14-May 74 F HSD JH w 2.2s1s1s1 9k 96 14-May 66 F HSD JH w 2.2 10k 96 14-May 64 F HSD JH w 2.2 11k 96 14-May 74 F HSD JH RVLM H 1.3 12k 96 14-May 72 F HSO JH w 1.351 13k 96 14-May 77 M H50 JH w 1.3 14k 96 14-May 51 M HSD JH RV H 1.2 15k 96 14-May 79 F HSO JH H '23 16k 96 14-May 65 F HSD JH W 2.2 17k 96 14-May 62 F H50 JH w 2.3 18k 96 14-May 69 F HSD JH H 1.3 19k 96 14-May 60 F H50 JH ? REGEN 20k 96 14-May 91 M HSD JH LM H 1.3 Table 19c (cont’d) lll Sample ID # Date Length (cm) Sex Site Coll Clip HIW Age 21k 96 14-May 73 M HSD JH W 2.4 22k 96 14-May 66 M HSD JH H 1.3 23k 96 14-May 90 F H50 JH w 1.3 24k 96 14-May 63 F HSD JH w 2.3 25k 96 14-May 72 F H50 JH w 2.5 26k 96 19-May 69 F H50 JH LM H 1.4 27k 96 19-May 66 F HSD JH H 1.2 28k 96 19-May 74 F HSD JH H 1.3 29k 96 19-May 64 F HSD JH w 2.3 30k 96 19-May 71 F HSD JH w 1.3 31k 96 19-May 59 F H50 JH H 1.15151 32k 96 19-May 39 M HSD JH Rv H '2.1 33k 96 19-May 67 F HSO JH RPRM H 1.3 34k 96 19-May 60 F H50 JH w 2.2 35k 96 19-May 62 M HSD JH w 2.2 36k 96 19-May 62 F HSD JH w 2.2 37k 96 19-May 56 F HSD JH w 2.2 38k 96 19-May 71 M H50 JH H 1.3 39k 96 19-May 64 F HSD JH w 1.2 40k 96 19-May 66 M HSD JH H 1.3 41k 96 19-May 65 F HSD JH H 1.2 42k 96 19—May 74 M HSD JH w 2.3 43k 96 19-May 62 F HSD JH H 1.2 44k 96 19-May 60 M HSD JH w ? 2.2 45k 96 19-May 65 F HSD JH w 1.3 46k 96 19-May 72 F HSD JH RV H 1.3 47k 96 19-May 75 M HSD JH Rv H 1.3 48k 96 19-May 72 M H50 JH H 1.3 49k 96 19-May 71 F H50 JH w 1.3 50k 96 19-May 67 F HSD JH H 1.3 51k 96 19-May 66 F H50 JH H 1.251 5211 96 19-May 39 H50 JH Rv H ' R1 5311 96 23-May 59 M H50 JH H 1.2 54k 96 23-May 71 M HSO JH 7 R3 5511 96 23-May 75 M HSD JH w 2.4 56k 96 23-May 69 M HSD JH RM H 1.3 57k 96 23-May 65 M HSD JH w 2.2 58k 96 23-May 59 F HSD JH H 1.2 59k 96 23—May 64 F HSD JH H 1.3 60k 96 23-May 60 F H50 JH w 1.2 61k 96 23-May 65 F HSD JH w 2.2 62k 96 23-May 75 F HSD JH H 1.25151 63k 96 23-May 71 F HSD JH :7 R251 64k 96 23May 71 F HSD JH H 1.3 65k 96 23-May 67 F HSD JH w 1.3 66k 96 23-May 75 M H50 JH w 2.3 LITERATURE CITED Albert, D.A., S.R. Denton, and B.V. Barnes. 1986. Regional landscape ecosystems of Michigan. The University of Michigan, Ann Arbor. Allendorf, F.W., D. Bayles, D.L. Bottom, K.P. Currens, C.A. Frissell, D. Hankin, J .A. Lichatowiich, W. Nehlsen, P.C. Trotter, and T.H. Williams. 1997. Prioritizing Pacific salmon stock for conservation. Conservation Biology 11:140-152. Altukhov,Y.P. and EA. Salmenkova. 1990. Introductions of distinct stocks of chum salmon, Oncorhynchus keta (Walbaum), into natural populations of the species. Journal of Fish Biology 37:25-33. Annonymous. 1970. A survey of background water quality in Michigan streams. Michigan Department of Natural Resources, Water Resources Commission Report, Lansing. Anonymous. 1975-1995. Fish stocking records. Michigan Department of Natural Resources, Lansing. Barlow, CG. and BA. Gregg. 1991. Use of circuli spacing on scales to discriminate hatchery and wild barramundi, Lates calcarifer (Bloch). Aquaculture and Fisheries Management 22:491-498. Beamish, R.J. and GA. McFarlane. 1983. Validation of age determination estimates: the forgotten requirement. Transactions of the American Fisheries Society 75:237-256. Berejikian, BA. 1995. The effects of hatchery and wild ancestry and experience on the relative ability of steelhead trout fry (Onchorhynchus mykiss) to avoid a benthic predator. Canadian Journal of Fisheries and Aquatic Sciences 52:2476-2482. Bernard, KL and K.W. Myers. 1996. The performance of quantitative scale pattern analysis in the identification of hatchery and wild steelhead (Oncorhynchus mykiss). Canadian Journal of Fisheries and Aquatic Sciences 53:1727-1735. Bethe, J. And P. Krasnowski. 1979. Stock separation studies of Cook Inlet sockeye salmon based on scale pattern analysis, 1977. Alaska Department of Fish and Game, Informational Leaflet No. 180, 31p. 112 113 Bhatia, D. 1931. On the production of annual zones in the scales of rainbow trout (Salmo irideus)). Journal of Experimental Zoology 59:45-49. Biette, R.M., D.P. Dodge, R.L. Hassinger, and Stauffer. 1981. Life history and timing of migrations and spawning behavior of rainbow trout (Salmo gairdneri) populations of the Great Lakes. Canadian Journal of Fisheries and Aquatic Sciences 38: 1759-1771. Bioscan. 1989. Optimas User’s Guide and Reference, Washington, DC, Edmonds, 482pp. Carbine, W.F. 1945. An investigation of Grass Lake, Benzie County Report No. 1025, Fisheries Division, Michigan Department of Natural Resources, Institute for Fisheries Research, Ann Arbor. Carlander, K.D. 1986. A history of scale and growth studies of North American freshwater fishes.Pages 3-14 in R.C. Summerfelt and GE. Hall [ed.] Age and growth of fish. Iowa State University Press, Ames, Iowa. Chang, W.Y.B. 1982. A statistical method for evaluating the reproducibility of age determination. Canadian Journal of Fisheries and Aquatic Sciences 39: 1208-1210. Chapman, D., C. Carlson, D. Weitkamp, G. Matthews, J. Stevenson, and M. Miller. 1997. Homing in sockeye and chinook salmon transported around part of their smolt migration route in the Columbia River. North American Journal of Fisheries Management 17:101-113. Conrad, R. 1985. Sample sizes of standards and unknowns for a scale pattern analysis. (Unpublished memorandum). Alaska Department of Fish and Game, Sports Fishery Division, 333 Raspberry Road, Anchorage, Alaska. Damsgard, B. 1991. Smolting characters in anadromous and resident Arctic charr, Salvelinus alpinus. Journal of Fish Biology 39:765-774. Davis, ND. 1987. Variable selection and performance of variable subsets in scale pattern analysis (Document submitted to annual meeting of the INPFC 1987). 47pp. FRI-UW-8713. Fisheries Research Institute, University of Washington, Seattle. Davis, ND. and J .T. Light. 1985. Steelhead age determination techniques (Document submitted to annual meeting of the INPFC, Tokyo, Japan, November 1985), 41pp. FRI-UW-8506 Fisheries Research Institute, University of Washington, Seattle. Epifanio, J. 1996. An evaluation of genetic diversity in Lake Michigan steelhead (Oncorhynchus mykiss) populations. Progress Report:F-53-R Study. Michigan Department of Natural Resources, Ann Arbor. 114 Ewing, R.D., M.D. Evenson, E.K. Birks, and AR. Hemmingsen. 1984. Indices of parr- smolt transformation in juvenile steelhead trout (Salmo gairdneri) undergoing volitional release at Cole Rivers Hatchery, Oregon. Ferguson, M.M., R.G. Danzmann, and S.K.A. Amdt. 1993. Mitochondrial DNA and allozyme variation in Ontario cultured rainbow trout spawning in different seasons. Aquaculture 117:237-259. Fielder, D.G. 1987. An assessment of the introduction of summer steelhead into Michigan. Michigan Department of Natural Resources Fisheries Report 1948, Ann Arbor. Friedland, K.D., C. Esteves, L.P. Hansen, and RA. Lund. 1994. Discrimination of Norwegian farmed, ranched and wild-origin Atlantic salmon, Salmo salar L., by image processing. Fisheries Management and Ecology. 1:117-128. Gall, G.A., J. Baltodano, and N. Huang. 1988. Heritability of age at spawning for rainbow trout. Aquaculture 68293-102. Garling, D. And S. Dann. 1995. Fisheries Special Report No. 74. In Status and Potential of Michigan Natural Resources. Michigan Agricultural Experiment Station, Michigan State University. Gharrett, A.J., and W.W. Smoker. 1993. Genetic components in life history traits contribute to population structure. Pages 197-202 in J .G. Cloud and G.H. Thorgaard (ed). Genetic conservation of salmonid fishes. Plenum Press, New York. GLFC (Great Lakes Fishery Commission). 1992. Strategic vision of the Great Lakes Fishery Commission for the decade of the 19905. GLFC, Ann Arbor, Michigan. Griffith, J .S. 1993. Coldwater streams. Pages 405—425 in CC. Kohler and W.A. Hubert (ed). Inland Fisheries Management in North America. American Fisheries Society, Bethesda, Maryland. Hansen, M.J. and T.M. Stauffer. 1971. Comparative recovery to the creel, movement, and growth of rainbow trout stock in the Great Lakes. Transactions of the American fisheries society 100:336-349. Hartig, J .H. and ME. Stifler, 1979. Water quality and pollution control in Michigan. Michigan Department of Natural Resources, Environmental Protection Bureau, Lansing. 115 Hartman, W.L. 1959. Biology and vital statistics of rainbow trout in the Finger Lakes Region, New York. New York Fish and Game Journal 6: 121-177. Haynes, J .M., D.C. Nettles, K.M. Parnell, M.P. Voiland, R.A. Olson, and J .D. Winter. 1986. Movements of rainbow steelhead trout (Salmo gairdneri) in Lake Ontario and a hypothesis for the influence of spring thermal structure. Journal of Great Lakes Research 12:304-313. Helle, J .H. 1981. Significance of the stock concept in artificial propagation of salmonids in Alaska. Canadian Journal of Fisheries and Aquatic Sciences 38: 1665-1671. Hesse, J .A. 1994. Contribution of hatchery and natural chinook salmon to the eastern Lake Michigan sport fishery, 1992-1993. Master’s thesis. Michigan State University, East Lansing. 82p. Hoar, W.S. 1976. Smolt transformation: evolution, behavior and physiology. Journal of Fisheries Research Board of Canada 33: 1234-1252. Hooton, R.S., B.R. Ward, V.A. Lewynsky, M.G. Lirette, and AR. Faccin. 1987. Age and growth of steelhead in Vancouver Island populations. Province of British Columbia Fisheries Technical Circular No. 77. Hume, J .M.B. and EA. Parkinson. 1988. Effects of size at and time of release on the survival and growth of steelhead fry stocked in streams. North American Journal of Fisheries Management 8:50—57. Hutchings, J .A. 1993. Adaptive life histories affected by age-specific survival and growth rate. Ecology 673-684. Ihssen, P.E., H.E. Booke, J .M. Casselman, J .M. McGlade, N.R. Payne, and FM. Utter. 1981. Stock identification: materials and methods. Canadian Journal of Fisheries and Aquatic Sciences 38: 1838-1855. Jearld, A., Jr. 1983. Age determination. Pages 301-324 in Nielsen and Johnson. Fisheries Techniques. American Fisheries Society, Bethesda, Maryland. Jones D.E. unpublished. Handbook for interpretation of steelhead trout scales in southeast Alaska. Alaska Department of Fish and Game, Sport Fish Division, Petersburg. Jude, DJ ., F.J. Tesar, S.F. Deboe, and T.J. Miller. 1987. Diet and Selection of Major Prey Species by Lake Michigan Salmonines, 1973-1982. Transactions of the American Fisheries Society 1 16:677-691. 116 Kapuscinski, AR. and L.D. Jacobson. 1987. Genetic guide for fisheries management. Minnesota Sea Grant Research Report 17, Department of Fisheries and Wildlife, University of Minnesota, St. Paul. Karges, R.G. 1987. Life history, reproductive success, and abundance of rainbow trout (Salmo gairdneri) in the Ganaraska River, Ontario. Master’s thesis. University of Waterloo, Ontario. Karlin, S. And S. Lessard. 1986. Sex Ratio Evolution. Princeton University, Princeton, New Jersey. Knudsen, CM. and ND. Davis. 1985. Variation in salmon scale characters due to body area sampled. (Document submitted to annual meeting of the International North Pacific Fisheries Commission, Tokyo, Japan, November 1985) 59pp. FRI-UW-8504, Fisheries Research Institute, University of Washington, Seattle. Krueger, C.C., D.L. Perkins, R.J. Everett, D.R. Schreiner, B. May. 1994. Genetic Variation in Naturalized Rainbow Trout, (Oncorhynchus mykiss) from Minnesota Tributaries to Lake Superior. Journal of Great Lakes Research 20(1):29-316. Kwain, W.H. 1971. Life history of rainbow trout (Salmo gairdneri) In Batchawana Bay, eastern Lake Superior, Journal of the Fisheries Research Board of Canada 28:771- 775. Kwain, W.H. 1981. Population dynamics and exploitation of rainbow trout in Stokely Creek, Eastern Lake Superior. Transaction of the American Fisheries Society 1102210-215. Larson, R. And J. Ward. 1954. Management of steelhead trout in the state of Washington. Transactions of American Fisheries Society 84:261-274. Larsson, PO. 1985. Predation on migrating smolts as a regulating factor in Baltic salmon, Salmo salar L., populations. Journal of Fish Biology 26:391-397. Latta, WC. 1958. Age and growth of fish in Michigan. Fish Division Pamphlet No. 26. Michigan Department of Conservation. Lansing, MI. Latta, WC. 1974. A history of the introduction of fishes into Michigan. Pages 83-96 in Michigan Fisheries Centennial Report 1873-1973. Michigan Department of Natural Resources, Fisheries Division. Lansing, MI. Leider, SA. 1985. Precise timing of upstream migrations by repeat steelhead spawners. Transaction of American Fisheries Society 114:906-908. 117 Leider, SA. 1989. Increased straying by adult steelhead trout, Salmo gairdneri, following the 1980 eruption of Mount St. Helens. Environmental Biology of Fishes 24:219- 229. Ieider, S.A., M.W. Chilcote, and J .J . Loch. 1986. Comparative life history characteristics of hatchery and wild steelhead trout (Salmo gairdneri) of summer and winter races in the Kalama River, Washington. Canadian Journal of Fisheries and Aquatic Sciences 43: 1398-1409. MacCrimmon, HR. 1971. World distribution of rainbow trout (Salmo gairdneri). Journal of Fisheries Research Board of Canada 28:663-704. D MacLean, J .A. and DO. Evans. 1981. The stock concept, discreteness of fish stocks, and fisheries management. Canadian Journal of Fisheries and Aquatic Sciences 38:1889- 1898. McKeown, PB. 1984. Fish Migration. Croon Helm, London and Sydney. ' Mahoney, E.M. D.B. Jester and GO J amsen. 1991. “Recreational fishing in Michigan” in D.M. Spotts (Ed.) Travel and Tourism in Michigan: a statistical profile. East Lansing, Mi. Travel, Tourism, and Recreational Center, Michigan State University. MDNR. 1994. Hatchery Program Review Document. Michigan Department of Natural Resources, Fisheries Division, Lansing. MDNR. Fisheries Division Management Plan: fiscal year 1996-97. [Online] Available http://www.dnr.state.mi.us/dept/mgmtplan/mgmtfish.htm, September 18, 1997. Millar, RB. 1987. Maximum likelihood estimation of mixed stock fishery composition. Canadian Journal of Fisheries and Aquatic Sciences 44:583-590. Millar, RB. 1990. Comparison of Methods for Estimating Mixed Stock Fishery Composition. Canadian Journal of Fisheries and Aquatic Sciences 47:2235-2241. Miller, D.R., J .G. Williams, and CW. Sims. 1983. Distribution abundance and growth of juvenile salmonids off the coast of Oregon and Washington, summer 1980. Fisheries Research 2: 1-17. Mitton, J .B. and W.M. Lewis. 1989. Relationships between genetic variability and life history features of boney fishes. Evolution 43: 1712-1723. Moring, J .R. 1993. Anadromous stocks. Pages 553-580 in CC. Kohler and W.A. Hubert (ed.). Inland Fisheries Management in North America. American Fisheries Society, Bethesda, Maryland. 118 Moyle, P. B., and J .J . Cech. 1988. Fishes: an introduction to ichthyology. Prentice-Hall, Englewood, New Jersey. Neilson, J .D. and G.H. Geen. 1986. First-Year growth rate of Sixes River chinook salmon an inferred from otoliths: effects on mortality and age of maturity. Transactions of the American Fisheries Society 115128-33. Newcomb, T. J. 1998. Productive capacity of the Betsie River watershed for steelhead smolts. Ph.D. dissertation. Michigan State University, East Lansing, Michigan, pp166. Ney, J .J . 1993. Practical use of biological statistics. Pages 137-158 in CC. Kohler and W.A. Hubert (ed). Inland Fisheries Management in North America. American Fisheries, Bethesda, Maryland. Northcote, TC. 1969. Patterns and mechanisms in the lakeward migratory behavior of juvenile trout. Salmon and trout in streams. H.R. MacMillan lectures in Fisheries, University of British Columbia, Vancouver, BC. pp. 181-204. Parkinson, EA. and RA. Slaney. 1975. A review of enhancement techniques applicable to anadromous gamefishes. Province of British Columbia Fisheries Management Report 66, Victoria, BC. Patridge, RE. 1985. Effect of steelhead trout smolt size on residualism and adult return rates. Idaho Department of Fish and Game. Lower Snake River and Wildlife Compensation Plan, contract no. 14-16-001-83605, Boise, Idaho. Pauley, G.B., B.M. Bortz and M.F. Shepard. 1986. Species profiles: life histories and environment requirements of coastal fishes and invertebrates: steelhead trout. US. Fish and Wildlife Service Biological Report 82 (11.62). Peterson, N .M. 1979. Biological characteristics of wild and hatchery steelhead, Salmo gairdneri, in two Oregon rivers. Master’s thesis. Oregon State University, Oregon. Rakoczy, GP. and RF. Svoboda, 1994. Sportfishing catch and effort from the Michigan waters of Lake Michigan, Huron, Erie, and Superior, April 1, 1992 - March 31,1993. Michigan Department of Natural Resources, Fisheries Division, Technical Report 94-6, Ann Arbor. Rand, P.S., DJ. Stewart, P.W. Seelbach, M.L. Jones, and LR. Wedge. 1993. Modeling Steelhead Population Energetics in Lakes Michigan and Ontario. Transactions of the American Fisheries Society 122:977-1001. 119 Ricker, W.E. 1969. Effects of size-selective mortality and sampling bias on estimates of growth, mortality, production, and yield. Journal of the Fisheries Research Board of Canada 26:479-541. Ricker, W.E. 1972. Hereditary and environmental factors affecting certain salmonid populations, p. 19-60. In R.C. Simon and RA. Larkin (ed.). The stock concept in Pacific salmon. H.R. MacMillan Lectures in Fisheries, University of British Columbia, Vancouver, BC. Saunders, R.L. and CB. Schom. 1985. Importance of the variation in life history parameters of Atlantic salmon (Salmo solar). Canadian Journal of Fisheries and Aquatic Sciences 42:615-618. Scamecchia, D.L. 1979. Variation of scale characters of coho salmon with sampling location on the body. Progressive Fish-Culturist 41 :132-135. Schaffer, W.M. 1974. Selection for optimal life histories: the effects of age structure. Ecology 55:291-303. Schaffer, W.M. and PF Elson. 1975. The adaptive significance of variations in life history among local populations of Atlantic salmon in North America. Ecology 56:577-590. Schwartzeberg, M., and J. Fryer. 1989. Experiments in identifying hatchery and naturally spawning stocks of Columbia Basin spring chinook salmon using scale pattern analysis. Columbia River Inter-Tribal Fish Commission Technical Report 89-3. Seelbach, PW. 1987. Smolting success of hatchery-raised steelhead planted in a Michigan tributary of Lake Michigan. North American Journal of Fisheries Management 7:223-23 1 . Seelbach, PW. 1993. Population biology of steelhead in a stable-flow, low-gradient tributary of Lake Michigan. Transaction of the American Fisheries Society 122:179- 198. Seelbach, P.W. and GB. Beyerle. 1984. Interpretation of the age and growth of anadromous salmonids using scale analysis. Michigan Department of Natural Resources. Fisheries Technical Report 84-5, Ann Arbor. Seelbach, P.W. and GE. Whelan. 1988. Identification and contribution of wild and hatchery steelhead stocks in Lake Michigan tributaries. Transactions of the American Fisheries Society 1 17:444-451. 120 Seelbach, P.W. and ER. Miller. 1993. Dynamics in Lake Superior of hatchery and wild steelhead emigrating from the Huron River, Michigan. Michigan Department of Natural Resources. Fisheries Research Report 1993, Ann Arbor. Seelbach, P.W., J .L. Dexter, and ND. Ledet. 1994. Performance of steelhead smolts stocked in southern Michigan warmwater rivers. Michigan Department of Natural Resources. Fisheries Research Report 2003, Ann Arbor. Shapovalov, L. and AC. Taft. 1954. The life histories of the steelhead rainbow trout (Salmo gairdneri) and silver salmon (Onchorhynchus kisutch) with special reference to Waddell Creek, California, and recommendations regarding their management. California Department of Fish and Game, Fish Bulletin 98. Sacramento. Slaney, P.A., L.Berg and AF. Tautz. 1993. Returns of hatchery steelhead relative to site of release below an upper-river hatchery. North American Journal of Fisheries Management 13:558-566. Sommers, L.M., C. Thompson, S. Tainter, L. Lin, T.W. Colucci, and J .M. Lipsey. 1981. Fish in Lake Michigan: distribution of selected species. Publication No. MICHU- SG-81-600. East Lansing, Mich. Michigan Sea Grant Program, Michigan State University. Stauffer, T.M. 1972. Age, growth, and downstream migration of juvenile rainbow trout in a Lake Michigan tributary. Transaction of the American Fisheries Society, 101:18- 28. Steward, C.R. and T.C. Bjomn. 1990. Supplementation of salmon and steelhead stocks with hatchery fish: a synthesis of published literature. US. Department of Energy, Bonneville Power Administration, Division of Fish and Wildlife, Technical Report 90-1. Portland, Oregon. Taft, AC. and L. Shapovalov. 1938. Homing instinct and straying among steelhead (Salmo goirdnerii) and silver salmon (Onchorhynchus kisutch). California Department of Fish and Game 24:118-125. Tipping, J. 1984. A profile of Cowlitz River winter steelhead before and after hatchery propagation. Washington Game Department, Report 84-11, Olympia, Washington. Tipping, J. 1991. Heritability of Age at Maturity in Steelhead. North American Journal of Fisheries Management 1 1:105-108. Tsumura, K., J .M.B. Hume, and BM. Chan. 1987. Effects of size at release in rainbow trout (Salmo gairdneri) stocked in a winterkill lake. Province of British Columbia Fisheries Management Report No. 88. 121 Unwin, M.J. and DH. Lucas. 1993. Scale characteristics of wild and hatchery chinook salmon (Onchorhynchus tshowytscho) in the Rakaia River, New Zealand, and their use in stock identification. Canadian Journal of Fisheries and Aquatic Sciences, 50: 2475-2484. VanDenAvyle, M.J. 1993. Dynamics of exploited fish populations. Pages 105-135 in CC. Kohler and W.A. Hubert (ed). Inland Fisheries Management in North America. American Fisheries Society, Bethesda, Maryland. Wagner, H.H. 1967. A summary of investigations of the use of hatchery-reared steelhead in the management of a sport fishery. Oregon State Game Commission, Fisheries Report 5. Ward, BR. and P.A. Slaney. 1988. Life history and smolt-to-adult survival of Keogh River steelhead trout (Salmo gairdneri) and the relationship to smolt size. Canadian Journal of Fisheries and Aquatic Sciences, 45:1110—1122. Ward, BR. and P.A. Slaney. 1990. Returns of Pen-Reared Steelhead from Riverine, Estuarine, and Marine Releases. Transactions of the American Fisheries Society 119:492-499. Ward, B.R., P.A. Slaney, AR. Facchin, and R.W. Land. 1989. Size-biased survival in steelhead (Oncorhynchus mykiss): back-calculated lengths from adults’ scales compared to migrating smolts at the Keogh River, British Columbia. Canadian Journal of Fisheries and Aquatic Sciences, 46: 1853-1858. Wesley, J .K. 1996. Age and growth of chinook salmon in Lake Michigan: verification, current analysis, and past trends. Master’s thesis. University of Michigan, Ann Arbor. 93p. West, C.J., and P.A. Larkin. 1987. Evidence for size selective mortality of juvenile sockeye salmon (Oncorhynchus nerka) in Babine Lake, British Columbia. Canadian Journal of Fisheries and Aquatic Sciences 44:712-721. Wicklund, R.G. and BC. Dean. 1958. Betsie River watershed: survey and plans report. Michigan Department of Conservation, Fish Division, Lake and Stream Improvement Section, Lansing. Willett, DJ. 1994. Use of temperature to manipulate circuli patterns on scales of silver perch, Bidyonus bidyanus (Mitchell), for the purpose of stock discrimination. Fisheries Management and Ecology 1:157-163. 122 Withler, LL. 1966. Variability in life history characteristics of steelhead trout (Salmo gairdneri) along the Pacific Coast of North America. Journal of Fisheries Research Board of Canada 23:365-392. Wood, CC. 1987. Predation of juvenile Pacific salmon by the common merganser (Mergus merganser) on eastern Vancouver Island. I: predation during the seaward migration. Canadian Journal of Fisheries and Aquatic Sciences 44:941-949. Worlund, DD, and RA. Fredin. 1962. Differentiation of stocks. Pages 143-153 in N .J . Wilimovski, (ed). Symposium on pink salmon. H.R. MacMillan Lectures in Fisheries, University of British Columbia, Vancouver, BC. MIC 111117111111llllilllllllllll5’