AN ASSESSMENT OF THE INTRODUCTION OF SUMMER STEELHEAD INTO MICHI GAN by David G. Fielder A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science School of Natural Resources The University of Michigan 1987 Committee members Associate Professor James S. Diana, Chairman Adjunct Professor W. Carl Latta Ex-off ic io Examiner Paul W. Seelbach ACKNOWLEDGMENTS This research was completed with the help of a great many individuals. I am extremely thankful to all who were involved with this project and I have tried to acknowledge these below. I apologize to any who may have been omitted. This research was initiated and funded largely by a series of generous grants from the Flint River Valley Chapter of the Michigan Salmon and Steelhead Fishermen's Association. I extend to the members and officers of that organization my sincerest gratitude for their commitment to this work and to the sport fisheries of Michigan. My special thanks to H. Shinabarger of the Flint Chapter for his dedication to the project. I thank D. Calhoun, also of the Steelheader's Association, for his interest, support, and role in initiating this project. My wholehearted thanks to the hundreds of volunteer fishermen who participated as data collectors and as friends, so many are special, yet too numerous to list. My thanks to the Manistee Chapter of the Steelheader's Association for their help. This research was additionally funded and supported through the Michigan Department of Natural Resources, specifically the Institute for Fisheries Research. Only through the assistance and support of the department and Institute was a project of this magnitude possible. W. C. Latta was invaluable as a source of administrative support, i i encouragement and technical guidance. P. W. Seelbach was extremely helpful through advice and encouragement. I thank him f Or the review of this manuscript and for his friendship. Numerous other Institute biologists and personnel assisted in various ways including: R. D. Clark, J. B. Gapczynski, B. A. Gould, R. N. Lockwood, B. A. Lowell, J. W. Merna, J. R. Ryckman, J. C. Schneider, A. D. Sutton, and G. M. Zurek. Many other individuals of the Michigan Department of Natural Resources cooperated and assisted in this project. MOS t notable were District 6 personnel, who provided extensive assistance with field operations. My sincere thanks to J. A. Allen, L. Frankenberger, R. L. Hay, and the various creel census clerks. Also, I thank G. Rakoczy of the Charlevoix Great Lakes Station for his help. Many of the faculty and students in the School of Natural Resources, The University of Michigan contributed to this project. J. S. Diana provided encouragement and technical assistance throughout the project and invaluable editorial help. My sincerest thanks to him for his patience and support. Much of the field work for this research was conducted in and around Manistee, Michigan. I am grateful to the many residents and merchants for their assistance and interest. Specifically, I thank the Donald Johnson family of Cadillac, Michigan, for their friendship and many fishing trips. ii i I thank my parents, the late Gordon Fielder, Jr., and Anita Fielder for their emotional and financial support and for showing me the value of an education. Their patience, love, and unwavering support contributed more than they know to my development as a biologist. Finally, I thank my wife Candi for her unmeasurable patience, love, and support, without which this endeavor would not have been possible. iv TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . i i LIST OF TABLES . . . . . . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . . . . . . v i i LIST OF APPEND ICES . . . . . . . . . . . . . . . . . vi i i ABSTRACT . . . . . . . . . . . . . . . . . . . . . . ix INTRODUCTION . . . . . . . . . . . . . . . . . . . . l METHODS . . . . . . . . . . . . . . . . . . . . . . 8 Volunteer Research Anglers . . . . . . . . . . . 8 Creel Census . . . . . . . . . . . . . . . . . . l 4 RESULTS . . . . . . . . . . . . . . . . . . . . . . l6 DISCUSSION . . . . . . . . . . . . . . . . . . . . . 33 LITERATURE CITED . . . . . . . . . . . . . . . . . . 48 APPENDICES . . . . . . . . . . . . . . . . . . . . . * 52 Table LIST OF TABLES Locations and numbers of each summer steelhead strain stocked in Michigan rivers in l984. . . Number of summer steelhead and other species caught by volunteer anglers in lakes and rivers, July 1984 to November 1986. Percent of total number in parentheses. . . . . . . . . . Chi-square contingency table for salmonids compared between locations. * = 2,593. 584, P<0.000l. Comparison done also between steelhead lake catch, X* =l 6.429, P<0.000l. Data were collected from 1984 through l886. . Chi-square test Of monthly steelhead proportions for a seasonal expansion in steelhead catch . . . . . . . . . . . . . . . . Estimated number of Great Lakes steelhead, summer steelhead, and salmon caught in the Big Manistee River (excluding Manistee Lake), from April to November 1985 and from May to October l986 (95% confidence limits in parentheses) . . Estimated total catch for Skamania and Rogue steelhead in the Big Manistee and Boyne rivers in 1986. The estimates for the Big Manistee River were determined by the l886 creel census. The estimate for the Boyne River was determined by extrapolation from effort ratios in the 1986 C e Il SllS (95% confidence limits a re in parentheses) . . . . . . . . . . . . . . . . . Mean total lengths (inches) at age 3+ for each steelhead type considered in this study. . . . Page l 7 19 27 29 30 31 vi LIST OF FIGURES Figure l. Locations in Michigan where summer steelhead were stocked in 1984 . . . . . . . . . . . . . Percent monthly catch in rivers from July 1984 to November 1986 for the four strains of summer steelhead: (a) Rogue, (b) Skamania, (c) Umpqua, and (d) Siletz . . . . . . . . . . . . . . . . . Percent monthly catch in rivers for (a) summer steelhead (strains combined), (b) Great Lakes steelhead, and (c) salmon, from July 1984 to November 1986. . . . . . . . . . . . . . . . . Per Cent composition by strain of summer steelhead catches from rivers, July 1984 to November 1986. & G © © & © © o © © © º G ſº O e Monthly seasonal expansion of river steelhead fishing created by summer steelhead (+ indicates summer steelhead percent, x indicates f all-run Great Lakes steelhead percent) . . . . Page 2l 22 24 26 vi i LIST OF APPEND ICES Appendix l. Fin-clip patterns for the four summer steelhead strains. Rogue steelhead-– left pectoral and right ventral. Skamania steelhead--adipose and right pectoral. Siletz steelhead--adipose and both ventrals. Umpqua steelhead--adipose and left Ventral . . . . . . . . . . . . . . . . . . Volunteer research angler brochure. . Research anglers' diary given to volunteer fishermen (actual size). A copy of the cover and all pages are shown. Pages 3 and 4 were filled out for each fishing trip, 24 copies were included in each diary. . . . . . . . . . A copy of summer steelhead research angler button given to volunteer fishermen. Button was l l /2 inches in diameter, color was blue on maize. . . . . . . . . . . . . . . . . . . . Volunteer research angler effort during the project for all rivers and ports combined. The original effort goal was approximately 36 hours per month. . . . . . . . . . . . . . . . . . . Percent monthly catch by volunteer anglers in lakes f Or (a) summer steelhead (strains combined), (b) Great Lakes steelhead, and (c) Salmon . . . . . . . . . . . . . . . . . . . . Estimated number of Great Lakes steelhead, summer steelhead, and salmon caught in Manistee River from Tippy Dam to Bear Creek, Bear Creek to Manistee Lake, and Manistee Lake, l 985 and 1986 (95% confidence limits a re in parentheses) . . . . . . . . . . . . . . . . . vi i i Page 53 54 6l 63 64 65 66 ABSTRACT Steelhead (Salmo gairdneri) are not native to the Great Lakes but have been introduced since the late l800s. Steelhead in the Great Lakes make their spawning migrations in the fall or spring. In l975 the State of Indiana introduced the Skamania strain of steelhead which migrate during the summer months. The Skamania strain has been very successful in Indiana. In l984 the State of Michigan introduced four stains of summer steelhead (Rogue, Skamania, Siletz, and Umpqua) into several Great Lakes tributaries to expand the existing steelhead river fishing season to the SUllſtrºle I e This thesis tested the hypothesis that the introduction of summer steelhead expanded river steelhead fishing into the summer months. Volunteer research anglers were used to report fishing activities as a means of documenting summer steelhead returns. A creel census was also employed on one of the stocked rivers to document return.S. The summer steelhead did significantly expand the river steelhead fishing season. Volunteer angler data and creel census results showed first date of river catch was July with a peak in August. This provided for about 2 months of new angling for river steelhead. The Rogue strain returned first (lº 84) and the other three strains returned by the third summer (1986). Steelhead and salmon sport catches were statistically different between lake and river ix locations with salmon generating more of a lake fishery and steelhead generating more of a river fishery. Summer steelhead contributed a significantly greater proportion of lake catch than Great Lakes steelhead. The lengths of fish at a known age were similar between summer and Great Lakes steelhead, indicating similar growth rates. The level of catch per unit effort generated by the summer steelhead introductions was very low. Future stocking efforts will require larger stocking numbers and annual releases in consistent locations. Some additional potential benefits and problems of summer steelhead introductions exist. INTRODUCTION Many forms of steelhead (Salmo gairdneri) (anadromous rainbow trout) have been introduced into the Great Lakes since the late l800s (MacCrimmon l971; MacCrimmon and Gots 1972). Presently some self-sustaining populations of steelhead exist in the Great Lakes (Biette et al. l981; Seelbach l886), and sport catch is a combination of both hatchery and wild fish. Steelhead found in the Great Lakes exhibit tWO primary river migration patterns. SOme individuals make their river migration in the late fall and winter months while others enter only in the spring months (Biette et al. 1981; Seelbach l886). These two groups will be referred to here as fall-run and spring-run Great Lakes steelhead, respectively. These populations of steelhead have provided an excellent sport fishery in the lakes and rivers especially during their spawning migration. On the Pacific Coast of the United States, steelhead are divided into two types. Winter steelhead make a spawning migration back to their natal rivers during the winter months, usually November through April. Summer steelhead make their spawning migration during the summer months, usually May through September (Withler lº 66). The two types differ in other characteristics as well. Winter steelhead are sexually mature when entering the river, while summer steelhead are sexually immature during entrance (Smith l860, 1969; Withler lº 66; Leider et al. 1984). Spawning is frequently earlier for summer steelhead (Leider et al. 1984) yet this characteristic seems variable and needs further documentation. There may also be slight morphological differences between the two types (Smith 1969). These life history characteristics are different enough that the two types have been frequently identified as separate races of the same species (Smith l860; Withler lS66; Behnke l872; McKern et al. l S74; Leider et al. l.984). However, Allendorf and Utter (1979), based on electrophoric data, divided steelhead into "coastal and inland" groups and not by anadromy or season of return to fresh water. For the purpose of this thesis summer and winter steelhead will be considered separate races of the same species. In a particular river system, specific strains of each race may have evolved (i.e., the Rogue River strain) or strains may have been developed artificially in a hatchery. The Skamania strain of summer steelhead was developed at the Skamania Hatchery in Washington where biologists selectively bred individuals for summer return, winter spawning, and large size (older age at return) (Millenbach l873). The life history pattern of spring-run steelhead in the Great Lakes is most similar to that of winter-run steelhead (Biette et al. 1981), and it was probably winter strains which were most commonly introduced. Introduced steelhead seem to have retained the major characteristics of their life history from their native range (Behnke l872; Biette et al. 1981; Seelbach l986). Michigan's fall-run steelhead differ from spring-run fish mainly in their time of river return but also in the degree of sexual maturity at river entry. similar to summer Steelhead, fall-run Great Lakes steelhead are generally immature sexually at the time of their migration and ripen sometime prior to spawning (Seelbach l886). Behnke (lº 72) believed that the two populations are genetically distinct and reproductively isolated. It has been suggested that Great Lakes f all-run steelhead are analogous to coastal summer steelhead with timing differences related to climatic conditions in each locale (Biette et al. 1981). However, there are fall-run steelhead on the Pacific Coast as well as winter- and summer-run steelhead (Neave 1949; Shapovalov et al. 1954; Royal lS72). In 1975 the State of Indiana introduced S Ulſtliſtler steelhead into Lake Michigan from eggs imported from the Skamania Hatchery in Washington. Despite only limited stream habitat, Indiana successfully established a summer steelhead fishery. Some of the creeks that receive the hatchery stocks are marginal trout streams at best and are very warm for trout occupation. Nevertheless, large runs of Skamania summer steelhead congregate in Lake Michigan near the river mouths, creating a very popular lake fishery. After the fish run the creeks, some stream fishing takes place. Indiana chose the Skamania summer steelhead for stocking because of its reputation for tolerating warmer water temperatures and from only limited success with other steelhead (W. D. James, Indiana Department of Natural Resources, Indianapolis, personal communication with J. A. Scott, Michigan Department of Natural Resources, l082; Armstrong lSS5). In 1983 the Michigan Department of Natural Resources (MDNR) imported four strains of summer steelhead. The state's purpose for introducing summer steelhead was to improve steelhead fishing by expanding the existing river fishery to the summer months. This hypothesis of seasonal expansion forms the basis of this thesis. The Skamania strain was imported from Indiana, while the Rogue, Siletz, and Umpqua strains were imported from those rivers in Oregon. Each of these summer steelhead strains will henceforth be termed Skamania steelhead, Rogue steelhead, Siletz steelhead, and Umpqua steelhead for simplicity. By March 1984, the state had l38,000 summer steelhead smolts for release. The fish were approximately 7 to 8 inches in length. Nine Michigan rivers were chosen by MDNR to receive smolts (Figure l, Table l). Smolts were f in clipped with distinctive patterns to allow strain identification by biologists and fishermen (Appendix l). The objectives of this study were: l) To statistically compare the numbers returning by month for Great Lakes steelhead and Sll ſtuſſler steelhead so as to establish a basis for evaluating Cherry Creek Boyne River N Betsie River Big Monistee River Au Soble River Little Monistee River Pere Marquette River White River \–% Muskegon River T Figure l. Locations in Michigan where summer steelhead were stocked in 1984 Table l. Locations and numbers of each summer steelhead strain stocked in Michigan rivers in l984. Location Number Strain Muskegon River 20,000 Skamania l 7, 000 Umpqua Big Manistee River 20,000 Skamania l6,000 Rogue White River 20,000 Siletz Betsie River 8,000 Rogue Boyne River 8,000 Rogue Au Sable River l9,000 Skamania 17,000 Rogue Pere Marquette River 10,000 Skamania l(), 000 Umpqua Little Man is tee River 5,000 Rogue 5,000 Umpqua 5,000 Siletz Cherry Creek 18,000 Siletz 2) whether summer steelhead did expand river fishing into the summer months. To estimate total sport catch of Skamania and Rogue summer steelhead and Great Lakes steelhead in the Big Manistee River from a creel census conducted in l985 and 1986. METHODS Sport angler catch was the primary data collected in order to monitor migration runs and the contribution of summer steelhead to the state's steelhead fishery. Sport catch was a convenient source of data because the data were readily available through volunteer research anglers, and sport catch was the goal of these introductions. The use of volunteer anglers also allowed the monitoring of a large geographical area. Catch data were also obtained through a MDNR Creel census on the Big Manistee River. SOme additional summer steelhead reports were supplied from MDNR biologists through weir or field operations and from the general public. Volunteer Research Anqlers Volunteer fishermen were usually recruited by attending meetings of clubs such as the Michigan Salmon and Steelhead Fishermen's Association. A slide presentation was given explaining the project and requesting help. Volunteers' names, addresses, and telephone numbers were collected and they were supplied with an assessment kit. This kit consisted of an explanatory brochure, a project diary to document their catch, and a button designating them as a Summer Steelhead Research Angler (see Appendices 2, 3, and 4). Some additional volunteers were recruited through the mail. Volunteers were allowed to fish as they normally would, but were asked to check for f in clips on any steelhead Caught. In order to monitor effort, fishermen were told to make an entry on one diary page for each fishing trip even if no fish were caught. Data recorded were: date of trip, hours fished, location, number of fish caught, and species of fish caught. If a fin-clipped steelhead was caught, additional data recorded included clip pattern, total length, and specific location. The goal of this analysis was to have each stocked river and port (except Cherry Creek) covered with enough research anglers to provide reports for at least two weekdays and one weekend day every week, continuously from July l 984 through November 1986. Approximately l80 volunteers were registered from mid-l984 through late 1985 and about 300 from late l885 through late 1986. Active participation was defined as the percent of volunteers who actually returned diaries. Anglers were requested to return diaries as they became filled. Blank replacement diaries and thank you letters were mailed in response. Twice during the project, approximately halfway through and In ear completion, all diaries (including partially filled diaries) were recalled. Anglers who did not return diaries were mailed a postcard, again requesting the return of their diaries. Anglers who still did not return diaries were contacted by telephone. If diaries were then not returned, l 0 it was assumed that the volunteer angler had not participated. Volunteer research angler catch (both lake and river) for salmon and steelhead was used to calculate percent monthly catch. The percent for each month was determined by dividing the monthly catch for each species or strain by the project total for that species or strain. Percent monthly catch of summer steelhead utilized all reports of summer steelhead, not just those of the diaries. Percent monthly catch is partly dependent on changes in effort but allows for a comparison of seasonal return for steelhead types despite large differences in total numbers returning. Volunteer angler catch for all types was also subdivided into total frequencies between open water (lake) and river. These are termed volunteer angler catch frequencies and utilize all summer steelhead reported. Catch data for summer steelhead strains were used to calculate percent of each strain in the total summer steelhead catch. This portrays the contribution of each strain to the monthly river catch of summer steelhead throughout the project and utilizes all summer steelhead reported. This percent was calculated by dividing the monthly catch of each strain by the total monthly catch of all summer steelhead strains combined and was termed percent Composition. This should not be confused with the previously mentioned percent monthly catch of each strain. Catch per unit effort (CPUE) of volunteer anglers for summer ll steelhead was extremely low and missing for many months (except on the Boyne River for part of the project) because many of the reported summer steelhead came from sources other than the volunteer diaries. Some reasons for this are discussed later and volunteer angler CPUE is not presented. Boyne River volunteer CPUE is used in SOIſle later calculations and explained then. For the purpose of the hypothesis test, seasonal expansion in the state river steelhead fishery was defined as the establishment of summer steelhead catch in a monthly proportion significantly greater than that for Great Lakes steelhead. Because Michigan already benefits from steelhead returns in fall through spring, summer (late May through early September) is the only season available for expansion. If the summer steelhead strains were collectively caught in greater proportions during a month other than the Great Lakes steelhead, then there would be an expansion. Two time periods were chosen for comparison Of steelhead catch. They were July through December 1985 and July through November (end of sampling) lº&6. The two time periods effectively allow for the comparison of summer steelhead returns to that of f all-run Great Lakes steelhead. The fall run of Great Lakes steelhead is closest in timing to summer runs and represents the best comparison for summer steelhead. Summer steelhead may continue to be caught in the river during the winter and spring months but this represents no seasonal expansion. There was insufficient l2 summer steelhead returns to allow a comparison during the first summer and fall of the project in l984. The statistical procedure used for the comparison of steelhead types in each month of each time period was the Chi-square test using a 2x2 contingency table for comparison of proportions in two independent samples (Snedecor and Cochran l97l). The data used in the test were volunteer angler catch frequencies for summer and Great Lakes steelhead caught in the rivers. Summer steelhead strains were combined for the test. The procedure was applied to each month in each time period. The procedure effectively compares two proportions, in this case the proportion of each type of steelhead caught in a specific month relative to the total steelhead caught of each type for the entire time period. The table was structured with steelhead type (summer or Great Lakes) versus number caught for that month and number not caught (the remainder caught for the rest of that time period). The observed frequencies are compared to expected frequencies calculated under the null hypothesis of equal proportions. The expected frequencies are determined by multiplying the corresponding row and column totals and dividing by the sample size (Snedecor and Cochran l97l). This and all statistical tests were run using a significance level of 0.05. Other comparisons performed between steelhead types included catch frequencies between open water and rivers and an examination of length at age. Volunteer angler catch l3 frequencies were compared in order to determine if summer steelhead and Great Lakes steelhead or salmon (salmon were included as a point of reference) differ in their tendency to generate a lake or river fishery. The Chi-square test (Snedecor and Cochran l971; Remington and Schork lº& 5) was again used. The expected values were calculated with the hypothesis that the row and column classifications were independent. The expected frequencies were calculated the same as described above for the comparison of proportions. Patterns of dependence were identified by the difference between the observed frequencies and those expected if both locations were equally represented in all three groups. Mean length for a known age was also compared between the summer steelhead strains and fall-run Great Lakes steelhead using a test of equality of the means of two samples whose variances are assumed to be unequal (Sokal and Rohlf lS81) referred to hereon as t tests. The procedure compared total length in inches for the summer steelhead strains reported by volunteer anglers from August through November 1986, to Great Lakes fall-run steelhead of the same age collected at the Little Manistee River weir during fall l383 and l384. The age of comparison for steelhead was 2.5 years after Smolting. This procedure allowed some comparison of growth rates. l 4 Creel CenSuS A stratified creel census was conducted on the Big Manistee River in l985 from April to mid-November, and in l986 from May to mid-October. The survey was conducted by the MDNR and followed methods and calculations described by Ryckman (1981). The study area was from Tippy Dam (the first upstream barrier) down to and including Manistee Lake at Stronach, Michigan. The census provided estimates of seasonal catch per unit effort (CPUE) and total catch for each species and strain by month. For all estimated catches, 95% confidence limits were also calculated (Ryckman 1981). The l886 census included separate catch estimates for summer steelhead strains but the l885 census did not. Summer steelhead catch estimates for the l885 census were derived monthly from the ratio of steelhead with summer steelhead fin-clip patterns to steelhead without f in clips. Total steelhead catch, estimated by the creel census, was multiplied by this ratio to determine summer steelhead catch. Some estimates of monthly and seasonal summer steelhead catch were also possible for the Boyne River because volunteer effort was relatively large and consistent enough to provide meaningful CPUE data. These estimates were calculated from a ratio of volunteer to creel census effort from the Big Manistee River. That ratio was multiplied by the volunteer effort on the Boyne River for an estimate of l 5 total effort on the Boyne. This was then multiplied by the CPUE of volunteers on the Boyne River, which yielded an estimate of total summer steelhead caught. A correction factor, based on the different efficiencies of volunteers and average anglers, was calculated from the ratio of volunteer angler steelhead CPUE to the estimated steelhead CPUE for the l885 Big Manistee River creel census. This was multiplied by the total estimate above to determine the actual number of summer steelhead caught on the Boyne River. The estimates generated by the creel censuses and the extrapolations are useful in two ways. The monthly estimates could confirm timing trends identified in the volunteer angler data. Secondly, the total estimates and CPUE allow for a measure in magnitude of the contribution made to the fishery by the summer steelhead plants. This information may be useful in determining future stocking effort needed to generate adequate returns. RESULTS Active participation of the volunteers was 30% (54 anglers) from July 1984 to September 1985. There was loº (47 anglers) active participation of the volunteers from September 1985 to November 1986. A total of l C, 879 hours of effort was logged in project diaries by volunteer anglers. This effort included 51% on open water and 49% in rivers. Effort was not evenly distributed throughout the project duration (Appendix 5). Rivers receiving the most consistent annual effort were the Big Manistee and Boyne. POrtS receiving the most consistent effort throughout the open- water seasons were Manistee and Muskegon. The effort objective of two weekdays and one weekend day of angling trips per weeks was not achieved consistently on any of the study rivers although the Big Manistee River came very close. This inconsistency in effort for most rivers makes seasonal return data for that specific location tenuous. Sport catch and effort combined for all rivers or ports should allow inferences about individual strains. Total catch of all species reported during the project was 8,490 fish (Table 2). Of this, 59% were caught in open water and 41% in rivers. Out of the total, 280 (3%) were summer steelhead. This total includes lS 8 reported by sources other than volunteer diaries. Salmon caught by volunteers consisted primarily of Chinook (Oncorhynchus l6 Table 2. Number of summer steel head and other species caught by volunteer anglers Pen Cent Of t O tal number in parentheses. in lakes and rivers , July 1984 to November 1986. Summer steel head Great Lakes Other Other Location RO Sk Um Si TO tal steel head Sal mon tr OU t species TO tal Lake 2 34 1 O 14 6O 239 3, 274 1, 339 68 4, 98 O (21) (12) (8O) (69) (27) (59) River 83 79 26 32 22O 1, 694 8 1.9 59C) 187 3, 5 1 O (78) (88) (2O) (31) (73) (41) Combined 85 1 13 36 46 28O 1, 933 4, O93 1, 929 255 8, 49C) (3) (23) (48) (23) (3) ( 1 OO) | Ro: Rogue, Sk: Skamania, Um : Umpqua , Si : Si letz . s 18 tshawytscha) and coho (Oncorhynchus kisutch). Other trout caught in the open water were usually lake trout (Salvelinus namay cush) and brown trout (Salmo trutta). Other trout caught in the rivers were usually brown and rainbow trout. The majority of the volunteer angler catch for both summer steelhead and Great Lakes steelhead occurred in the rivers (78% and 88%, respectively). Contrarily, the majority of salmon were caught in the lake (80%). Lake catch was defined as any catch in a Great Lake or shoreline lake such as Manistee Lake. In an effort to determine if the proportion of lake or river catch varied with species, the Chi-square contingency test was applied to these data (Table 3). The species caught was strongly dependent on location (P<0.0001). Salmon were more readily caught in lakes, while Great Lakes and summer steelhead were more readily caught in rivers. Summer steelhead had 21% of their total catch occurring in the lakes while only l2% of Great Lakes steelhead were caught there (Table 2). The proportion of summer steelhead caught in lakes (Table 3) WaS significantly greater than Great Lakes steelhead (P<0.000 l). Lake catch in the Great Lakes for salmon and steelhead is largely restricted to the open-water SeaSOIn (approximately May through September). Peak lake catch by volunteers of both summer Steelhead and salmon occurred in August . The peak year for summer steelhead catch in the lakes was lº&6 (Appendix 6). Volunteers' catch of Great 19 Table 3. Chi-square contingency table for salmonids compared between locations. X* = 2, 593. 584, P-0.000l. Comparison done also between steelhead lake catch, X* =l 6. 429, P-0.000l. Data were collected from 1984 through l886. Observed frequencies Expected frequencies Species Lake River Lake River Salmon 3, 274 819 2, 319 l, 774 Great Lakes steelhead 23.9 l, 694 l, 0.95 838 Summer steelhead 60 220 159 l2l 20 Lakes steelhead in lakes was slightly higher in September than August for lS86, but the reverse was the case in 1985. The Rogue steelhead dominated the volunteer angler catch of summer steelhead for the first 22 months of the project (Figure 2). The Rogue strain was caught in the rivers from fall lS 84 through early lº& 5. No river catch was observed in June 1985. Peak river catch of Rogue steelhead occurred in July 1985 and tapered off through the fall months. River catch of Rogue steelhead increased again in early lº& 6 but dropped off by June. Rogue steelhead were again caught in the river starting in August l SS6 and continuing through the fall (Figure 2). The Skamania, Umpqua, and Siletz steelhead were largely absent from the river catch until July 1986 (Figure 2). Peak river catch of Skamania steelhead occurred in August l986, 2.5 years after stocking. Peak river catch of the Umpqua and Siletz steelhead also occurred in August 1986 (Figure 2). The river catch for all summer steelhead strains combined peaked in August 1986 (Figure 3). Most of these were the Skamania steelhead, but some of each strain were caught at this time. River catch of Great Lakes steelhead by volunteers started in fall 1984 and continued through the early summer, peaking in November 1985 (Figure 3). In 1985, the lowest percent of river catch for Great Lakes steelhead occurred in August. This was almost exactly opposite the summer steelhead catch for lS85 which peaked in July. River catch 50 - 40 - 30 - 20 - 10 - 21 (a) m||1, J A S O N D J F M A M J J A S O N D J F M A M J J A S O N 1984 1985 50 - (b) 40 - 30 - 20 - 10 - J A S O N D J F M A M J J A S O N D J F M A M J J A S O N 1984 1985 50 - (c) 40 - 30 - 20 - 10 - —"—""— J A S O N D J F M A M J J A S O N D J F M A M J J A S O N 1984 1985 50 - (d) 40 - 30 - 20 - - 10 - J A S O N D J F M A M J J A S O N D J F M A M J J A S O N 1984 1985 Figure 2. Percent monthly catch in rivers from July 1984 to November 1986 for the four strains of summer (c) Umpqua, steelhead: (a) Rogue, (b) Skamania, and (d) Siletz. 22 i 35 - 30 - 25 - 20 - 15 - 10 - J A S O N D J F M A M J J A S O N D J F M A M J J A S O N 1984 1985 | 1986 (b) J A S O N D J F M A M J J A S O N D J F M A M J J A S O N 1984 1985 | 1986 (c) J A S O N D J F M A M J J A S O N D J F M A M J J A S O N 1984 1985 | 1986 Figure 3. Percent monthly, catch in rivers for (a) summer steelhead (strains combined), (b) Great Lakes steelhead, and (c) salmon, from July lº&4 to November l 986. - 23 of Great Lakes steelhead increased again in fall and continued through winter and spring. The lowest river catch of Great Lakes steelhead in l986 occurred June through August, opposite again of peak summer steelhead river catch at this time. River catch of salmon peaked in September l985 and l 986 with river catch of salmon beginning at least by August each year (Figure 3). The test for seasonal expansion was applied to the last two summers and falls during the project span. With in each of these two time periods the proportion of summer steelhead caught were tested by month with the proportion of Great Lakes steelhead caught. The stain composition of summer steelhead returns, however, varied throughout the project (Figure 4). The first summer and fall of l S84 river catch of summer steelhead was comprised entirely Of Rogue steelhead. During time period one (July to December 1985) river catch was dominated by Rogue steelhead but included some of the other three strains. Time period two (July to November 1986) showed a dramatic shift in river catch to a mixture of all four strains. Skamania steelhead comprised the largest proportion of fish caught in time period two. Besides strain contribution, the summer steelhead catch in the three seasons also differed in relative magnitude (Figure 3). The first summer and fall consisted of only five Rogue steelhead, time period one-–31 summer steelhead, and time period two--lal summer steelhead. The increase in 24 SN|UNNNNNNNNNNNNNNNNN （ ） |NINNNNNNNNNNNNNN LNNNNNNNNNNNNNNNNNNNN ， . . . . . . . RŅſ[NNNNNNNNNNNNNNNNNS LĪKNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN| [NNNNNNNNNNNNNNNNNNNNNNNNNNNN.， ， , ， ، ، ، Ņ ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، ، SÑŅ（ ） ĶN Ñ |- （ - （ ） · · - Ñ I-ILI| 1I|TLI===ſ±#-I-I CDOOO COCO Boyne River - Au Sable River -> Pere Marquette River — Cherry Creek - Little Man is tee River — The purpose of this project is to evaluate the returns of summer steelhead exactly ocked in order to accurately ident ify each strain. into nine strains 20, 000 l 7, 000 20, 000 l6, 000 20, 000 8,000 8,000 l9, 000 l 7, 000 l 0, 000 l 0, 000 l6, 000 5,000 5,000 5,000 Skamania Umpqua Skamania Rogue Siletz Rogue Rogue Skamania Rogue Skamania Umpqua Siletz Umpqua Rogue Siletz g § to as many of these rivers as possible. The future stocking of these strains is dependent on evaluation of returns from these plants. This project is the only O In e planned t O evaluate returns; the refore, the future of the summer steelhead program in Michigan could be highly dependent on data generated by it and the eff Orts put out by each participant. Evaluation Methods (l) River returns. --River I et U r In S of summer steelhead will be evaluated by three operations: l catch of research fishing teams on each river, 2) an intensive C reel census on the Big Man is tee River, and 3) fish returns to the weir on the Little Man is tee River. Research angling teams may be set up to intensively fish each of the * * / rivers indicated in the background section. The number of anglers per team can vary. The teams will be expected to regularly fish each river and keep records of the i r catch in diaries provided by the project. Each team will be responsible for insuring that at least one member fishes l weekend day and 2 weekdays every week. An entry in the diary should be made even if no summer steelhead are caught that day. In addition to date, biological data on length and clip pattern will be recorded for each steelhead collected (see page ll). It is very important that each river be fished systematically. Since we do not know how soon (in years and in months) to expect returns from each stra in , we must regularly fish each river throughout the open-water months. 8 Pe ITS Ons catch ing Clipped Steel head should contact Dave Fielder soon after, So We Can keep track of the status of each run. Returns from the research angling team On the Big Man is tee River will be Compared to total catch determined by a C re el Census on the river. TWO Census clerks will estimate total catch by Counting anglers and evaluating total Catch of each angler interviewed. This Comparison of team catch will be used to eXtrapo late results from the Big Man is tee to the other river systems. We hope to initiate the Creel census by Summer l 985. The weir on the Little Man is tee River is normally operated in the fall to Collect Chinook salmon eggs for the hatchery system. The Michigan DNR plans 9 to extend the operation of the we i r from July through September t O evaluate returns of summer steelhead. (2) Open water returns. --Data Of steelhead caught in the open Water fishery will be obtained by utilizing charter captains from each source river area as collectors of data, and by utilizing the DNR lake analyses done for Lakes Michigan, Huron, and Superior. Charter captains and other of f shore anglers will be recruited to form local teams again, much like those for the river areas. These people will also be requested to fill in diaries and collect the basic information described for the river analysis. The diagram on page ll indicates the measurement and f in-clip patterns to be evaluated. 10 § SUMMER STEELHEAD Summer steelhead have been planted in several Michigan rivers in 1984. They can be identified by distinctive f in-clip patterns. If you catch 3. f in-clipped steelhead, please collect the following data: Date Location F in-clip pattern Total length Hours fished l 2 3 4 5 Return these data to : Dave Fielder Institute for Fisheries Research 212 Museums Annex Building Ann Arbor, Michigan 48109 (3 l 3) 663 – 3554 Adipose (A) | Total Length Clip patterns Right Left Right Left A + RP Pectorals Ventral s A + LV + RV (RP, LP) (LV, RV) LP + RV A * LV Study conducted by: The University of Michigan Michigan Steelheaders Michigan Department of Natural Resources ll 35 APPenci X 5. Research anglers ' diary given to volunteer fishermen (actual size ). A copy of the cover and all pages are shown. Paqes 3 and 4 were filled out for each Fishing trip, 24 copies were included in each diary Complete a page of this diary for each day you fish, whether or not you catch steelhead . Also, collect the data for all steelhead caught, clipped, or unclipped. \; hen finished , return to : ! Institute for Fisheries Research Summer Steelhead 212 Museums Annex Building Ann Arbor, Michigan 48109. David Fielder for Michigan (313) 663-3554 For each Steelhead, please record : (1) date of catch (2) location (as specific as possible) (3) clip pattern (4) total length (5) hours fished 2 Adipose (A) oa t t e r n s p : ch: Le f t A + RP l Pect c : a l S A - LV → RV *º-º-º: | (RP, LP R.P - RV" A • LV § Appendix 3. Clips Length Continued: Location Date Start time Finish time - Hours fished Number of fish caught Clipped steelhead Unclipped steelhead SalmOn T rout OtherS Please record steelhead data on the back. 63 Appendix 4. A copy of summer steelhead research angler button given to volunteer fishermen. Button was l l /2 inches in diameter, color was blue on maize. 64 Volunteer research angler effort during the project for all rivers and ports combined. S SN Ñ ~ Ñ N SNS S . The J F M A M J J A S O N 1986 S Ñ· SN . SNS Ñ Ñ Ñ S original effort goal was approximately 36 hours J F M A M J J A S O N D 1985 aeº o + }Q) + C|× 5 8CC D-|-）*）*）*） 3-4= =C Cl)<[ <ſ2 C.*（O Ğ � INc/o CD | ()！= ×<， ÆII—yr–– †I——— I|| – 5© c > c > O O O CO O §© c -> C > O O C) C <Ç