....-‘ l LI] I”!!! ”21W" Lilli! Ill [[11 ll! {Hill [Lilli/3! ll 6 This is to certify that the thesis entitled INSTREAM OVERHEAD BANK COVER AND TROUT ABUNDANCE IN TWO MICHIGAN STREAMS presented by Michael David Enk has been accepted towards fulfillment of the requirements for Master nLSciennLdegree in Fisheries & Wildlife ("5—) , A7 ‘ . -) _/ / ‘ /~~{¢ 4,1 / //[/4z{ -( I, \_ \. / Maw/professor Date September 26, 1977 0-7 639 / «NV 0 t. ‘ 1W Aim 2 0 15,37 W313 J‘W‘reéfl ‘ © Copyright by MICHAEL DAVID ENK 1977 INSTREAM OVERHEAD BANK COVER AND TROUT ABUNDANCE IN TWO MICHIGAN STREAMS By Michael David Enk 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 1977 ABSTRACT INSTREAM OVERHEAD BANK COVER AND TROUT ABUNDANCE IN TWO MICHIGAN STREAMS By Michael David Enk The relationship between abundances of instream bank cover and trout was examined in 2.4km of the hardwater Pigeon River, Otsego County, and in 4.5km of the softwater Salmon Trout River, Marquette County. Each study area was divided into lOO-m reference stations. Trout populations were inventoried by mark-and-recapture electrofishing in July and October, 1976. During 1976 summer low flow, the length of stream bank qualifying as overhead concealment for trout was measured in 12 of the stations on the Pigeon River and l8 of the stations on the Salmon Trout River. Overhead bank cover included all submerged undercut banks, overhanging vegetation, and log cover not closer than 15cm to the stream bed and forming over- hangs at least 9cm wide. In the Pigeon River, harboring brook and brown trout, total length of overhead bank cover accounted for 88% of the variation in July number of trout 3150mm long and 72% of the variation in July biomass of trout 150-399mm long. For October brook trout 3150mm, bank cover abundance eXplained 68% of variation in numerical density and 65% of variation in biomass. Brown trout had redistributed Michael David Enk themselves for spawning in October, and at that time, population parameters were poorly correlated with bank cover abundance. In 2.6km of the Salmon Trout River where brook trout were the only salmonid (passage of anadromous fish blocked by a falls), length of overhead bank cover explained up to 83% of the variation in numerical densities and up to 69% of the variation in standing crops of brook trout 3150mm present in July. In October, abundance of cover accounted for 78% of the variation in numerical density and 88% of the variation in biomass of brook trout 3150mm. In combined analyses of Pigeon and Salmon Trout River data, length of bank cover accounted for up to 81% of the variation in July and October trout p0pulations. These strong correlations between indices of cover and trout abundances suggest that availabil- ity of instream bank cover is essential to the production of larger brook and brown trout. Bank cover appeared to be the major factor limiting trout populations in both streams, deSpite differences in fish species composition, water hardness, and hydrologic character- istics. Dedicated to Mrs. Anthony J. Bohte ii ACKNOWLEDGMENTS I would like to extend my sincere thanks and appreciation to Dr. Ray J. White for his guidance and constant assistance throughout this research project, and for his advice during preparation of the manuscript. Dr. White's contagious enthusiasm for the stream environ- ment has served to further motivate my personal efforts, and his ability to generate interest in salmonid research has made this study possible in the first place. I Also, I wish to express my appreciation to Dr. Niles Kevern for his guidance and support of my graduate program from its origin. I am very much indebted to Dr. Graham Larson for the time and effort he has devoted, in the field and at the drafting table, to my instruction in stream mapping techniques. Dr. Larson also provided much-needed equipment and innumerable helpful suggestions during the course of this project. The following people assisted with electrofishing surveys ‘and deserve special thanks for their long, hard hours in the streams: Dr. Ray White, Chuck Bassett, Dick Stanford, Kurt Fausch, Pete Cardinal, Jim Gruber, and Gary Horanburg. I am especially grate- ful to Rosanna Mattingly, who assisted with electrofishing, flow gaging, stream mapping, and cover measurement. This research has been greatly facilitated by the cooperation and hospitality of Dr. Sibley W. Hoobler and the members and staff iii of the Huron Mountain Club. I would especially like to thank the James F. McClelland, Jr., family for their hospitality, and the members of the Huron Mountain Club Fish Committee for their interest in my work. Grants from Dr. S. W. Hoobler and the Huron Mountain Wildlife Foundation have made this study possible, and my appreciation is again extended to them. Finally, I am very grateful to those who have had faith in me, especially my mother and father, and Dr. B. J. Mathis. Their encouragement and support has served to strengthen my educational commitments. iv TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES INTRODUCTION DESCRIPTION OF STUDY AREA Pigeon River . . General Location and Setting Water Quality and Discharge Resident Fishes . . Specific Location and Dimensions . Bed Materials Fishing Pressure Salmon Trout River . General Location and Setting Water Quality and Discharge Fishes Specific Location, Dimensions, and Bed Materials. Fishing Pressure METHODS AND MATERIALS Preparation of Study Area . . Estimation of Trout Populations . Electrofishing Procedure Calculation of Estimates Habitat Studies Streamflow Discharge Measurements Mapping and Bank Cover Measurements . RESULTS AND DISCUSSION . Pigeon River . Population Estimates . Habitat Studies . . . Population- Bank Cover Relationships . Page vii ix Salmon Trout River . . . . . . . . . . . . 51 Population Estimates . . . . . . . . . . . 51 Habitat Studies . . . . . . . . 62 Population- Bank Cover Relationships . . . 65 Combined Analysis: Pigeon and Salmon Trout Rivers . 71 SUMMARY AND CONCLUSIONS . . . . . . . . . . . . 80 LITERATURE CITED . . . . .. . . . . . . . . . . 84 APPENDICES A. Rainbow Trout, Brown Trout, and Coho Salmon in the Salmon Trout River in 1976 . . . . . . . . . 93 8. Habitat Maps for Twelve Stations on the Pigeon River, Michigan . . . . . . . . . . . . . . . 96 C. Habitat Maps for Eighteen Stations on the Salmon Trout River, Michigan . . . . . . . . . . . . . 109 vi Table '10. 11. 12. 13. LIST OF TABLES Description of Salmon Trout River study area Population (number) of brook trout in the Pigeon River in July, 1976.. . Biomass (kg) of brook trout in the Pigeon River in July, 1976.. . . . Population (number) of brown trout in the Pigeon River in July, 1976 . . . . . . . Biomass (kg) of brown trout in the Pigeon River in Ju1y, 1976 . . . . . . . . Population (number) of brook trout in the Pigeon River in October, 1976 . . . . . . . . . Biomass (kg) of brook trout in the Pigeon River in October, 1976 . . . . . Population (number) of brown trout in the Pigeon River in October, 1976 . . Biomass (kg) of brown trout in the Pigeon River in October, 1976.. . . . . . . . Habitat measurements from 12 study stations on the Pigeon River . . . . . . . Correlation coefficients (r) and coefficients of determination (r2) for trout population variables (Y) and total length (m) of overhead bank cover/ station (X) in the Pigeon River . . Population (number) of wild brook trout in the Salmon Trout River below Lower Falls in July, 1976 . Population (number) of wild brook trout in the Salmon Trout River above Lower Falls in July, 1976. vii Page 15 3O 31 32 33 36 37 38 39 42 44 53 54 Table Page 14. Biomass (kg) of wild brook trout in the Salmon Trout River below Lower Falls in July, 1976 . . . . . . 55 15. Biomass (kg) of wild brook trout in the Salmon Trout River above Lower Falls in July, 1976 . . . . . . 56 16. Population (number) of wild brook trout in the Salmon Trout River below Lower Falls in October, 1976 . . . 58 17. Population (number) of wild brook trout in the Salmon Trout River above Lower Falls in October, 1976 . . . 59 18. Biomass (kg) of wild brook trout in the Salmon Trout River below Lower Falls in October, 1976 . . . . . 60 19. Biomass (kg) of wild brook trout in the Salmon Trout River above Lower Falls in October, 1976 . . . . . 61 20. Streamflow discharge (cubic meters per second) at 6 sites along the Salmon Trout River on August 31, 1976 . . . . . . . . . . . . . . . . . 63 21. Habitat measurements from 18 study stations on the Salmon Trout River . . . . . . . . . . . . 64 22. Correlation coefficients (r) and coefficients of determination (r2) for trout population variables - (Y) and total length (m) of overhead bank cover/ station (X) in the Salmon Trout River . . . . . 66 Al. Estimates of rainbow trout populations in the Salmon Trout River in 1976 . . . . . . . . . 94 A2. Estimates of brown trout and coho salmon p0pulations in the Salmon Trout River in October, 1976 . . . . 95 viii Figure 0101-wa 10. 11. 12. 13. 14. 15. LIST OF FIGURES The Pigeon River Pigeon River study area The Salmon Trout River Salmon Trout River study area Biomass of trout in the Pigeon River in July, 1976 Biomass of trout in the Pigeon River in October, 1976 . . . . . . . . Relationship between July brook and brown trout abundances and bank cover in the Pigeon River . Relationship between July total trout abundance and bank cover in the Pigeon River . Relationship between July standing crops of brook and brown trout and bank cover in the Pigeon River Relationship between July total standing crop of trout and bank cover in the Pigeon River . Relationship between October brook and brown trout abundances and bank cover in the Pigeon River . Relationship between October standing crops of brook and brown trout and bank cover in the Pigeon River Biomass of wild brook trout in the Salmon Trout River in July, 1976 . . . . . . Biomass of wild brook trout in the Salmon Trout River in October, 1976 Relationship between July brook trout abundance and bank cover in the Salmon Trout River . . . ix Page 10 12 16 35 35 45 45 47 47 50 50 57 57 67 Figure Page 16. Relationship between July standing crop of brook trout and bank cover in the Salmon Trout River . . 67 17. Relationship between October brook trout abundance and bank cover in the Salmon Trout River . . . . 69 18. Relationship between October standing crop of brook trout and bank cover in the Salmon Trout River . . . . . . . . . . . . . . . . 69 19. Relationship between July total trout abundance and bank cover in the Pigeon River and the Salmon Trout River above Lower Falls . . . . . . . . . . 72 20. Relationship between July total standing crop of trout and bank cover in the Pigeon River and the Salmon Trout River above Lower Falls . . . . . . 72 81. Pigeon River Station 3 97 B2. Pigeon River Station 4 98 B3. Pigeon River Station 5 99 B4. Pigeon River Station 6 100 B5. Pigeon River Station 9 101 B6. Pigeon River Station 10 102 B7. Pigeon River Station 13 103 B8. Pigeon River Station 14 104 B9. Pigeon River Station 19 105 810. Pigeon River Station 20 106 811. Pigeon River Station 21 107 812. Pigeon River Station 22 108 C1. Salmon Trout River Station 3 110 C2. Salmon Trout River Station 5 111 C3. Salmon Trout River Station 6 112 Figure C4. C5. C6. C7. C8. C9. C10. C11. C12. C13. C14. C15. C16. C17. C18. Salmon Salmon Salmon Salmon Salmon Salmon Salmon Salmon Salmon Salmon Salmon Salmon Salmon Salmon Salmon Trout Trout Trout Trout Trout Trout Trout Trout Trout Trout Trout Trout Trout Trout Trout River River River River River River River River River River River River River River River Station Station Station Station Station Station Station Station Station Station Station Station Station Station Station 7 8 1O 13 14 36 38 47 48 50 51 52 53 54 57 xi Page 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 INTRODUCTION Habitat needs of stream salmonids have long been of interest to biologists, fishery managers, and anglers concerned with this resource. Several approaches have been used to investigate possible relationships between physical parameters of streams and trout popu- lations. The most complicated have involved multiple regression of a large number of stream variables onto some measure of trout abun- dance. In a survey of 112 stream segments in Michigan and Wiscon- sin, Hendrickson, Knutilla, and Doonan (1972a) evaluated the rela- tionships between trout p0pu1ations and 29 hydrologic parameters, including channel characteristics, streamflow characteristics, and water quality. They found that trout populations seemed to be limited chiefly by stream temperature, hardness of water, bed mate- rials, instream vegetation, variability of streamflow, and discharge per unit drainage area. However, all correlation coefficients for single hydrologic parameters with trout populations were less than 0.5, suggesting that p0pu1ations in a heterogeneous sampling of streams are not dominated by any single hydrologic characteristic. But when sample size was restricted to stream segments within cer- tain limits of hardness or temperature, for example, higher multiple correlation coefficients resulted. Abundance of fish cover showed poor correlation with trout populations in that study, but the cover values were only visual estimates. Using multiple regression analysis, White, Hansen, and Alexander (1976) further demonstrated the importance of streamflow stability to trout populations in Midwestern U. S. streams. In the Northwestern U. S., Platts (1976) also examined the influence of stream variables in controlling fish populations of 38 streams. Coefficients of determination (r2) for each of his 20 variables with salmonid populations were less than 0.4 in all cases. Although valuable, the multi-variate approach often fails to examine in enough detail each of the variables used. The measure- mentprocedures and criteria are often subjectively established, and the stream variables themselves are hydrologically interrelated and by no means independent. Variation in trout populations along one stream section where water quality and many hydrologic characteris- tics remain fairly constant might be explained largely by one major limiting factor. This factor, in many cases, may be shelter or pro- tective cover (Elser, 1968; Lewis,.1969; O'Connor and Power, 1976). For stream salmonids, habitat differentiation and selection of resting microhabitat are governed by water velocity, turbulence, and cover conditions which include light, water depth, concealment, visual/tactile reference points, and spatial limits (Kalleberg, 1958; Chapman, 1966; Baldes and Vincent, 1969). Baldes and Vincent (1969) also found that the resting microhabitat serves as the hub or focal point for the radius of movement of the fish to other types of microhabitat. Since spatial requirements increase with age and size of fish, the amount of resting microhabitat with adequate cover characteristics may regulate the density of salmonid populations in streams (Kalleberg, 1958; Chapman, 1966; Allen, 1969). According to Onodera (1962), as trout fingerlings grow larger, they begin selecting definite places of residence under objects which cover them from enemies. However, a stream affords only a limited number of shelters for fingerlings, and therefore Onodera suggested that the number of available shelters controls the survival of fingerlings; excess populations are eliminated by flood, competition, and predation. The stream's carrying capacity for trout would be determined in this case primarily by the amount of suitable cover. Much has been learned about the habitat requirements of salmonids through detailed studies of microhabitat preferences. Although microhabitats are almost always associated with some type of shelter, salmonid cover needs are known to vary seasonally, diurnally, by fish species, and by fish size (Saunders and Smith, 1955; Kalleberg, 1958; Hartman 1963, 1965; Chapman, 1966; Gibson and Keenleyside, 1966; McCrimmon and Kwain, 1966; Butler and Hawthorne, 1968; Allen, 1969; Chapman and Bjornn, 1969; Lewis, 1969; Hunt, 1971; Griffith, 1972; Wesche, 1973; Bustard and Narver, 1975a). Undercut banks, overhanging vegetation, submerged objects (logs, stumps, tree roots, boulders), floating debris, water depth and turbulence are all known to provide cover for fish in streams (Giger, 1973; White, 1973). Brook and brown trout show especially strong preference for overhead concealment cover along the stream margin (Hartman, 1963; Gibson and Keenleyside, 1966; Baldes and Vincent, 1969; Lewis, 1969). Furthermore, Kalleberg (1958) and P. W. DeVore and R. J. White (unpublished) observed that larger stream salmonids frequently darted from midstream cover to shoreline cover (rather than vice-versa) when frightened or disturbed. Wesche (1973) conducted a detailed study of trout cover preferences in two Wyoming streams and reported that 85% of all brown trout 3152mm used undercut banks for cover instead of rubble-boulder areas. The value of overhead instream shelter along the bank has been further demonstrated by the success of stream improvement work that included installation of additional bank cover (Tarzwell, 1938; Hale, 1969 in White,l973; Hunt, 1971; Boreman, 1974; White, 1975; Cooper and Wesche, 1976). Trout populations increased in manipu- lated areas with the larger size groups of fish usually responding the most. Also, Boussu (1954) showed that removal of brush cover and overhanging banks from experimental sections in a Montana stream caused decreases in trout populations, with losses being greatest for larger fish. Although the importance of streambank cover to salmonids is widely recognized, criteria for defining and measuring cover have neither been clearly established nor consistently applied in the literature. This is due partly to the diversity of cover needs among various salmonid species and its complexity within mixed salmonid populations. In the West, withdrawal of water from streams for agricul- tural and urban use has damaged fisheries and led to the development of many new methods for assessing the quality and quantity of avail- able fish habitat (Kraft, 1972; Wesche, 1973; Stalnaker and Arnette, 1976; Bartschi, 1976; C00per, 1976; Nickelson, 1976; Tennant, 1976; Waters, 1976). While most of these techniques can adequately describe the loss of fish habitat resulting from flow reduction, they make no attempt to set guidelines for determining the specific attributes of fish cover. Most of the procedures use descriptive variables or subjective rating systems for cover evaluation. One notable exception is the work of Wesche (1973, 1976). His cover rating system utilizes criteria derived from intensive studies of trout cover preferences. The length of overhead bank cover meeting certain criteria is the most important component of his cover rating formula for "catchable" brown trout. These cover ratings were also found to be highly correlated with trout biomass in several streams. Hunt (1971) had reported that the length of bank having permanent overhead cover appeared to limit brook trout populations in 1.6km of a Wisconsin Creek. Although Hunt's criteria for bank cover were slightly different than Wesche's, both research- ers obtained indices of the total length of streambank which afforded overhead shelter for fish. They included only bank cover which was at or below the water surface. Streamside vegetation that is more than a fraction of a meter above the water surface is believed to be of little value as concealment cover for trout (White, 1973). The present study was undertaken to further test the validity of relationships between bank cover and stream trout abundance as proposed by Hunt (1971) and Wesche (1976). Specifically, the objec- tive was to investigate the possible correlation of trout populations with length of streambank cover in two Michigan streams. These streams were widely separated geographically and had different spe- cies compositions, hydrologic characteristics, and dissolved salts concentrations. The Wesche (1973) criteria for determining over- head bank cover were chosen for use in this study because they were based on actual observations of cover usage and had been proven applicable at 11 study sites on four streams (Wesche, 1976). DESCRIPTION OF STUDY AREA Pigeon River General Location and Setting The Pigeon River rises in a cedar swamp at the north edge of a prominent glacial moraine just northeast of the city of Gaylord in the northern part of Michigan's Lower Peninsula (Hendrickson et al., 1973a). The river then flows generally northward for about 70km through coniferous swamp, birch-aspen forests, and hardwood swamp. It draps about 180m from its source to its mouth at Mullet Lake in Cheboygan County (Figure 1). Most of the river flows over glacial till or outwash but downstream sections traverse limestone outcrops (Hendrickson and Doonan, 1970). As much of it lies within State Forest land, the Pigeon River is fairly remote from major roads and other develop- ment. Water Quality and Discharge The water is hard (160-220mg/1 CaCO3) and pH ranges from 7.5-8.5 (Hendrickson et al., 1973a). Due to a large component of groundwater inflow in the head- waters region, discharge is generally uniform in the upper Pigeon River and water temperatures remain relatively cool in summer and warm in winter (Hendrickson et al., 1973a; Benson, 1953a). Records from a U. S. Geological Survey Gaging Station at the DNR Pigeon 7 $4: 1.. SUPERIOR ‘. ‘hvo 1' 7 I l I' 31;: : 34°25‘ I N I +4925! 1 I 1 I 1 ' Afton -—————-—— -o'— .__._ _ ._ I I ‘ " 1 I 1 l : : '1 MICHIGAN 1 T I I 131: I | 1 I ' 1 I. ......... . _____ I I g . I I i ' N 1. :33 l : u ' I I l I I I F ------- 1 “““““““ I 1 1 , I I , IT I .Trout Research [3112 I and Gaging Station I Lansing Club Darn : I l I I I I ,_ _______ \x\__ _R_1_w_____1 I I s‘ruov AREA | I SCALE IT I o 2 4 a wuss 3. m In I I I o z 4 6 KILOMETERS I I 5:5 Luisa. _ - .. ....I Figure 1. The Pigeon River. River Headquarters located about 12km downstream from the study area show a 25-year (1950-1975) average discharge of 2.19m3/sec. Resident Fishes Nearly all of the Pigeon River supports brook and brown trout (Salvelinus fontinalis and Salmo trutta). Certain areas also contain rainbow trout (Salmo gairdneri). Suckers (Catostomidae), sculpins (Cottidae), minnows (Cyprinidae), and darters (Percidae) are present throughout the river. A few other non-trout species are also found in the lower stretches of the stream (Benson, 1953a). Specific Location and Dimensions The study area on the Pigeon River lies in Otsego County in Sections 25 and 36 of Township 32 North, Range 2 West, and Section 1 of Township 31 North, Range 2 West (Figure 1). This portion lies in the upper quarter of the river's length, about 12km upstream from the DNR Trout Research-USGS Gaging Station and about 7km above the Lansing Club Dam. The closest town, Vanderbilt (population: ca. 600), is about 16km west of the study area. . The study section consists of 2.4km of river extending up- stream from the Old Vanderbilt Road bridge (Figure 2). Here, the stream ranges from 7.5-l4m in width, the banks are low, and depth of the channel at its deepest cross-sectional point varies from 40-120cm at the base flow. A few small spring drainages empty into the river along the study area. 10 Figure 2. SCALE 0 .I .2 .BIMLES 0 .I .2 .3 KILOMETERS p—l e N v ' (I K) 9 ,afoofbtidgc H '2 " l2 I3 ‘I _ l4, l5 6 \, 16 m E, 17 c .: 18 IS F U 2C1 21 9 (I 22 C 23 c 24 25 U Pigeon River study area. (Numbers refer to station markers.) 11 Bed Materials Sand and gravel are the major components of the stream bed, but silt and muck occur in areas near the banks. Fishing Pressure Anglers were occasionally encountered by the author in the Pigeon River study area. Indirect evidence of fishing pressure, such as beverage and bait containers, was rarely seen. However, the landowner displays a sign allowing access to fisherman, and the total amount of fishing that occurred in the study area during 1976 is unknown. Salmon Trout River General Location and Setting The Salmon Trout River originates as two small branch streams in the southeastern portion of the Huron Mountains in Mar- quette County in the Upper Peninsula of Michigan (Figure 3). From the junction of these two branches, the river flows generally north- eastward for about 20 stream kilometers and enters Lake Superior. The headwaters and central portion of the stream flow primarily through northern hardwood forest. The lower portion winds through mixed coniferous-hardwood swamp. The river has three major waterfall areas: Upper Falls, Middle Falls, and Lower Falls. A small dam and elongate impound- ment exist on the river about 1km upstream from Lower Falls. Nearly all of the river is remote from public roads, and most of it lies within the boundaries of the Huron Mountain Club. 12 L. SUPERIOR \ L. SUPERIOR Salmon Trout River MICHIGAN o- SCALE ...__STU°Y ABE- I to“ \ o .5 | 2 “11.55 (on LUV" \ ezzzza:Et£an============ r47 Hun g\ o .5 1 2 monarchs ’” / “\\ EEEEEZZJEEEEEEEB umr Falls W, / Clear Cr Falls \ \ \ \ 6' \ ~53 \ as \ N \ 95686“ Eb" $023 r0170” ’I 3293!"’I” Figure 3. The Salmon Trout River. 13 Water Quality and Dischargg_ Chemically, the water is soft-~only 62mg/1 CaCO3--with a pH of around 7.6 (Hendrickson, Knutilla, and Doonan, 1973b). Out- croppings of crystalline bedrock in this part of the Upper Peninsula help to explain the softness of streams in the area. These rocks provide very little soluble material for mineralization of water entering streams such as the Salmon Trout River (Hendrickson et al., 1973b). Mean discharge of the river in Section 12, Township 51 North, Range 28 West (Darby Bend area) was reported by Hendrickson et al. (1973b) to be 1.50m3/sec. Because large areas of Precambrian bedrock in the western part of the Upper Peninsula are exposed or are cov- ered only with a thin layer of glacial deposits, conditions are poor for stable groundwater flow (Hendrickson et al., 1973b). Con- sequently, streams in this area, including the Salmon Trout River, are characterized by wider temperature fluctuations, greater stream- flow variability, and much greater floodflows than coldwater streams in the Lower Peninsula. High, eroded banks and deeply-scoured pools along much of the Salmon Trout River are the result of tremendous flooding that occurs during snowmelt and springtime runoff from the surrounding hills. Fishes Below Lower Falls, the resident fish population of the Salmon Trout River is composed primarily of brook trout, sculpins, ' minnows, darters, and a few rainbOw trout. This part of the river, 14 however, is used heavily by migratory fish from Lake Superior for seasonal spawning activity. In winter and spring, steelhead (rainbow trout), suckers, and burbot (L9ta_lgta) enter the river to Spawn. In late summer and continuing into fall, lake-run (coaster) brook trout move into the lower section of the river up to Lower Falls to spawn. Also in fall, chinook salmon (Oncornynchus tschawytscha), coho salmon (Oncorhychus kisutch), and brown trout make spawning runs into the lower Salmon Trout River. Above the insurmountable Lower Falls-Sheet Rock Falls area, lake migrants are excluded from the stream, and the fish population consists of brook trout, sculpins, minnows, and an occasional brook stickleback (Eucalia inconstans). The Huron Mountain Club conducts an annual stocking of legal- sized brook trout in the Salmon Trout River. The fish are usually stocked in early summer at various locations along the river, includ- ing several sites in the study area. Specific Location, Dimensions, and Bed Materials The study area begins at a point about 100m below a deep pool called Darby Bend and extends upstream about 6km to the base of Middle Falls (Figure 3). This portion of the river lies in Sections 12, 13, and 14 of Township 51 North, Range 28 West. Three small tributaries--Spring Creek, Clear Creek, and Snake Creek--join the river along the study area. The closest town, Big Bay (popula- tion: ca. 250), is about 9km east of the study area. 15 Three basic subdivisions of the study area are described in the following table. Station marker locations are shown in Figure 4. TABLE 1.--Description of Salmon Trout River study area. Station Thalweg Marker c:?32:1 Depth Bed Materials Boundaries (base flow) 1-20 6-17m 21-180cm Generally sand and gravel, larger rubble in some areas. 34-40 7-22m 18-110cm Gravel and rubble with sand intermixed. 40-60 5-12m 24-150cm* Sand and muck in Lower Dam impoundment; gravel, sand, and occasionally rubble upstream from marker 50. *Water depth in Stations 40-50 is regulated by the height of stop- logs in Lower Dam but usually does not exceed 150cm. Fishing_£ressure Fishing in the study area is basically limited to members of the Huron Mountain Club and their guests. Trespassers occasionally enter club prOperty and also fish this part of the river. Most of the study area between station markers 42 and 59 has been designated a "catch-and-release" section by the Huron Mountain Club, meaning that all trout caught are supposed to be returned to the stream immediately. The rate of angler harvest from the river is probably somewhat less than from other streams in the area which are open to the public. 16 6 SCALE : , o .1 .2 .3 MILES 7 " .1 -—-A bi 8 l! 4 .~ “ 2 o EI =2 53 KILOMETERS 9 "' 3 Darby ‘ Io Band 11 ~‘ N '3 \‘ 12 ~ 14 Christy Pool 17 ‘ 15 Ba. 16 .. 18 . Cr srmq p “ 19 20 " «9‘9 k' Lower Falls Sheet Rock Falls 35 ,1 g 36 .. 34 38 ". r. \ 37 42 x 43 43 41 40 ’t c E\\4 .‘n ' QI “ ', 39 \ c : ‘ C ""0 Lower Dom 51 c ~‘ .358 5 52 ' so 4 46 .0 ; - v 30‘ ' s -‘ “ 54 5°6¥ CIGOr (3,- Middle 5 ' 2 Falls 55 Figure 4. Salmon Trout River study area. (Numbers refer to station mark- ers, letters to discharge sites.) METHODS AND MATERIALS Preparation of StudygArea Both the Pigeon River and the Salmon Trout River study areas were initially divided into reference stations by tying brightly- colored flagging to streamside trees at intervals of approximately 100m along the streams. The stations were measured off with a 30-m plastic clothesline marked in meters. Station markers were numbered in the upstream direction and each station was given the number of the marker at its downstream end. Station locations on the two streams are shown in Figures 2 and 4. Estimation of Trout Populations Electrofishing Procedure Mark-and-recapture electrofishing was conducted to inventory trout populations. Two complete sweeps of each study area were made for each inventory. The electrofishing unit consisted of a 2.1—m plastic boat carrying a gasoline-powered 250-VDC, 1.75-kw generator (Pow-R— Gard Model 1736 DCV). Two spring-loaded retracting reels, each con- taining 7.6m of electrical cord, were mounted on opposite sides of the boat's bow and wired to the positive pole of the generator. The two capture electrodes were attached to the free ends of the reel cords. Each electrode consisted of a 135-cm fiberglass handle with 17 18 a head of 4.8mm stainless steel rod bent into a diamond shape about 25cm long. For the grounding electrode,which trailed behind the boat, brass window screening was fastened to the bottom of an 80x30x5cm styrofoam float and connected to the negative pole of the generator. In operation, the electrofishing boat was pulled upstream while two men with capture electrodes swept the channel, jabbed into areas of fish cover, and poked along stream banks. Often, the two electrofishers worked together on one side of the channel, with one moving to the upstream end of log jam, pool, or undercut bank and sweeping downstream with his electrode toward the other man at the lower end. As fish were drawn to the electrodes, they were netted and transferred to a tub of river water carried in the boat. After a lOO-m station had been electrofished, the team either stopped to process the catch or, if few fish had been taken, they placed a dividing net in the collection tub and continued electro- fishing upstream through the next station. During processing, fish captured in each station were anes- thetized with tricaine methane sulfonate (MS-222), measured, weighed, and examined for finclip markings. 0n the first or marking run, the bottom tip of the caudal fin was clipped on every fish. After processing, the fish were held in the stream in a live box until revived from anesthesia, then carried back to the downstream end of the station in which they were captured, and released. This step helps to facilitate the pr0per redistribution of fish in each sta- tion. On the second or recapture run, fish were carefully examined 19 for caudal fin clips during processing, and this time the upper tip of the caudal was clipped to ensure against double-counting of fish that may swim upstream past the team's position overnight. Separate records were kept for each lOO-m station. Trout population inventories were conducted in July and Octo- ber on the Pigeon River. The marking run for the first inventory was made July 18-19, 1976, and the recapture run July 21-22. For the fall inventory, a marking run was made on October 7-8, with the recapture run on October 16-17. The first trout population inventory on the Salmon Trout River began with a marking run June 3-8, 1976. Because of equipment failure, the recapture run could not be made until July 1. 0n the basis of experience gained during the marking run, it was decided that the section of stream from Christy Pool upstream to Lower Falls (Figure 4) would not be electrofished again, owing to treach- erous bed materials (large, slippery rubble), the difficulty of portaging equipment around Lower Falls, and the sparseness of trout populations in that area. Therefore, valid population estimates were made only for Stations 1-17 and 36-59 during the first inven- tory. Both runs of the fall electrofishing inventory for the Salmon Trout River were conducted during October 2-7 in Stations 1-19 and 34-59. For purposes of future growth analyses, special finclip markings were made on fish of the youngest age group (as judged by length-frequency distribution and appearance) during the electro- fishing in both streams. In the Pigeon River, Age-O trout were 20 given a left ventral (LV) fin clip during the October inventory. In the Salmon Trout River, Age-l fish captured below Lower Falls in the June-July electrofishing were given an adipose (A) fin clip. Those caught above Lower Falls were given an adipose and right ventral (ARV) fin clip. During the fall inventory on the Salmon Trout River, all Age-0 trout were given a left ventral (LV) fin clip, and any hatchery trout (identified by color markings and size) caught were given a left pectoral (LP) fin clip. Calculation of Estimates Population estimates were calculated according to the Schaefer modification of the Petersen formula (Regier and Robson, 1967): $=[m(r+u+l) _] (I‘m) estimated population, where P m = number of fish marked during first run, r = number of marked fish recaptured during second run, and u = number of unmarked fish captured during second run. Efficiency of capture by electrofishing tends to increase with total length of fish up to about 250mm, at which point it levels off (Schuck, 1945; Cooper and Lagler, 1956; McFadden, 1961). In order to minimize the error caused by this size-selectivity of electrofishing gear, separate population estimates were made for the fish of each 50-mm length interval from 100mm to 250mm. Fish longer 21 than 250mm were grouped for p0pu1ation estimates. No estimates were made for fish smaller than 100mm because of the extremely low recap- ture rate for these fish. Estimates for each size class were made for the entire Pigeon River study area and for three separate sections within the Salmon Trout River study area: the sections between Darby Bend and Lower Falls, Lower Falls and Lower Dam, and Lower Dam and Middle Falls. These total population estimates were then partitioned into the individual stations within the study areas on the basis of the relative proportions of the sums of m + u for each station. The same procedure was used in combining the data for brook and brown trout in the Pigeon River, then segregating the two species after calculation of a total population estimate for each size class. Cooper (1952) found no difference in the catchability and rate of recapture between brook and brown trout. Similarly, the Salmon Trout River data for stocked and wild brook trout were combined for the calculation of a total population estimate, and the two groups were subsequently separated out during the partitioning of the total into (individual stations. This method of combining data and then sub- dividing total estimates is more accurate because it allows the use of larger individual units in the estimations, especially the number of recaptured fish, upon which the method is based, and probable errors decrease accordingly (C00per, 1952). Estimates of trout biomass in each station were computed as the product of the number and average weight of fish in each size 22 class, the biomasses of the various size classes then totaled. Aver- age weights for each size class were calculated separately for brook and brown trout. In the Pigeon River, several very large brown trout (3500mm) were captured, and these fish were simply added to the final popu- lation and biomass estimates for the station in which they were taken. A single rainbow trout, about 250mm long, was caught in Station 20 but was not entered in the estimate tables because brook and brown trout were the primary focus of this study. In the Salmon Trout River, a few large migratory steelhead (rainbow trout) were caught below Lower Falls during the first inventory. These fish were not included in the population estimates but are listed in Appendix A, Table A1. Separate calculations were made for the sparse population of smaller rainbow trout present below Lower Falls and the results are also given in Table A1. During fall electrofishing, many young-of-the-year coho salmon, several adult coho, and two adult brown trout were caught below Lower Falls (see Appendix A, Table A2); because these are not considered resident stream fish, they are not included in the popu- lation estimates. Also encountered were several large, obviously lake-run (coaster) brook trout, but since spawning migrants and I resident brook trout could not always be positively distinguished, all were included in the population and biomass estimates. Brook trout stocked by the Huron Mountain Club were mistakenly not fin- clipped as arranged, and therefore had to be identified by their 23 hatchery coloration and size (230-350mm). These fish were omitted from the final population estimates. Confidence intervals were calculated for population esti- mates according to the procedure outlined by Davis (1964). In this method, the proportion of recaptures to total catch in the second run [r/(r + u)] determines the appropriate distribution, Poisson or binomial, to be used in deriving 95% confidence limits for the true proportion of marked fish in the population. Then, by reference to the proper table or graph given in Ricker (1975) or Adams (1951), upper and lower limits for r can be obtained and these values used to calculate upper and lower 95% limits for the population estimates of each size class. The new estimates are then partitioned into individual stations in the same manner as the original or best estimate. In any case, the lower population limit for a particular size class of fish in a station can never be less than the total number of fish (m + u) of that size caught in that station. If a calculated lower limit was less than the sum of m + u, it was simply disregarded and the lower limit was given as the value m + u. Finally, confidence intervals for trout biomass were computed by multiplying the upper and lower population limits in each station by the average weight of fish in the corresponding size group. Habitat Studies In order to study conditions in the two rivers at a time when fish cover would most likely be at a minimum, habitat measure- ments were made during summer low flow. Cooper and Wesche (1976) 24 implied that trout populations were limited by stream carrying capac- ity at the lowest flow which occurs in the stream for an extended period of time, in keeping with Liebig's law of the minimum (Odum, 1971). Habitat measurements were made on August 7-21 at the Pigeon River and from August 30-September 13 at the Salmon Trout River. Streamflow Discharge Measurements Initially, discharge measurement sites were established so that stream flow could be monitored for the duration of the summer study period. Discharge measurements were made by stretching a tape measure across the river perpendicular to the direction of flow, and measuring the average water velocity at 30-cm increments along the tape with a Gurley pigmy current meter. For each 30-cm width interval, the product of average velocity x average depth x width was computed to derive an individual discharge value. All of the interval products were then summed to obtain total discharge through the cross section being considered. A single discharge measurement was made on August 10 at Old Vanderbilt Road (station marker 1) on the Pigeon River. The height of the water was also recorded at that time. Because good records were available from the USGS gaging station further downstream, no additional discharge measurements were necessary, but height of the river at the road bridge was noted daily to check stability of flow during the habitat studies. Discharge was measured on August 31 at five strategic loca- tions on the Salmon Trout River and at one place on Clear Creek 25 near its confluence (Figure 4). Clear Creek is the largest tribu- tary in the study area. The flow of water in nearby Snake Creek was too slight to be accurately gauged with our equipment. Specific cross-sectional sites for discharge measurements were selected on the basis of uniformity of flow and substrate. Also recorded at each site was the height of the river at the time of discharge measurement and periodically thereafter during the summer study, again, to detect any changes in stream flow. Mapping2and Bank Cover Measure- Beets. Since it was impossible to make detailed habitat measure- ments of the entire study area on either river in the time avail- able, stations to be studied were chosen on the basis of the first population inventory results. The objective was to obtain a set of stations on each river--12 on the Pigeon and 18 on the Salmon Trout-- with the greatest possible variety of trout abundance. However, certain stations on the Salmon Trout River were intentionally excluded from the selection because they contained large, deep, unwadable pools where habitat measurement would not have been possi- ble and population estimates were not reliable. Also, the first six stations above Lower 0am were removed from consideration due to the pond-like nature of the stream in that area. In order to obtain accurate measurements of channel and thalweg lengths, and to graphically represent the path of the thal- weg and the locations of various kinds of cover, detailed maps of the 26 selected study stations were constructed by a standard compass traverse procedure (Compton, 1962). In the traverse, a series of points were surveyed by measuring the direction and distance from one point in the stream to a second point upstream, and from the second to a third and so on. A compass was used to determine bearing of the stream channel and a 30-m calibrated vinyl clothesline was used to measure distances from point to point as well as stream widths along the traverse. After a suitable scale had been chosen, the maps themselves were drawn in the field on graph paper using the ruled lines as a north-south/east-west grid for plotting bearings. Once a station had been mapped, the path of the thalweg was sketched in while walking along the deepest part of the channel. The actual water depth along the thalweg was also recorded on the map. The most important part of the habitat studies involved the measurement of overhead bank cover, which was defined in this study as solid or nearly-solid overhead cover not closer to the bed than 15cm and extending at least 9cm from the bank in water that is at least 15cm deep. These criteria for usable bank cover were developed (by Wesche (1973, 1976) who reported that trout were never found in overhead bank cover less than 9cm wide, and that 92% of all trout sampled were found in water depths of at least 15cm. To determine if an area of bank cover met these requirements, a special gauge constructed of wooden doweling and plexiglas was fitted along the outer edge of the cover. This gauge had a measur- ing arm which was 9cm wide and 15cm high. If the gauge could be 27 fitted beneath an area of potential bank cover, the length or streamside perimeter of that cover was measured. The gauge was continuously inserted along the edge of the cover as the researcher moved upstream, so that only that portion of bank cover which satis- fied the above criteria would be included in the length measurement. However, individual cover lengths less than 15cm were deemed insig- nificant and disregarded. No attempt was made to measure the actual width of bank cover greater than 9cm wide. In both rivers, there were basically three main types of overhead bank cover: undercut banks, overhanging vegetation, and log cover. These were measured and recorded separately. The length of suitable undercut bank was easily determined by contouring a tape measure along that portion of bank under which the special gauge would fit. Overhanging vegetation consisted of tree branches and roots (primarily speckled alder, Alnus rugosa) extending into the water from the bank. If this mat of vegetation was solid or nearly- solid and continuous from the bank to its outer edge in the stream, the gauge was fitted along it and the length which qualified as overhead bank cover was determined with a tape measure. Log cover included only those single logs, deadfalls, or log jams which were firmly lodged against the bank. If any of these fulfilled the requirements for overhead bank cover, the length of their outer edge or streamside perimeter was measured. In many cases, only a portion of the log cover met the criteria and was included in the length measurement. As a final step, the three types of overhead bank 28 cover were sketched in at their approximate locations on the indi- vidual station maps. Other conspicuous instream objects were also added to the maps to help indicate the relative location of bank cover. RESULTS AND DISCUSSION Pigeon River ngulation Estimates The Pigeon River was generally favorable for electrofishing. Its hard water was conductive of electricity and there were few places where water was too deep to be waded. However, several very large log jams were present in the study area, and these were diffi- cult to electrofish effectively. As a result, populations may have been slightly underestimated in certain stations, but in general, I feel the data provide an accurate description of the trout popula- tion. July standing cr0ps of brook and brown trout 3400mm in length in the Pigeon River study area are shown by station in Tables 2-5. At that time, no Age-I trout were smaller than 100mm, and no Age-O trout were as large as 100mm. Confidence intervals for the total number and biomass of trout in each station and in each size class have been included in the tables. These confidence inter- vals represent the totals of individual confidence limits derived for each size class of fish within each station. For brook trout, fish in the 150-199mm size class comprise the majority of total biomass, with few fish reaching lengths greater than 250mm. Biomass of brown trout is spread over a much greater range of size groups. Whether measured in numbers or 29 30 TABLE 2.--Population (number) of brook trout in the Pigeon River in July, 1976. Total Length (mm) Sta Total 95% CI 100-149 150-199 200-249 250-299 1 25 22 47 35-67 2 29 24 l 54 41-78 3 2 9 l 12 10-16 4 32 31 l 2 66 51-94 5 20 7 l 28 21-43 6 18 18 36 27-51 7 5 9 14 11-19 8 7 15 22 18-30 9 5 l 15 12-21 10 5 3 13 9-19 11,12 5 44 l 50 41-64 13 18 40 58 45-78 14 ll 16 27 21-38 15 5 22 3 30 23-40 16 5 15 20 16-27 17 5 29 l 35 28-46 18 14 37 3 54 43-74 19 ll 33 l l 46 37-62 20 5 27 32 25-41 21 7 16 23 18-31 22 7 46 3 56 45-73 23 5 18 6 29 22-40 24 ll 18 l 1 31 25-43 Total 257 510 26 5 798 624-1095 95% CI 174-414 425-630 20-46 5-5 624-1095 31 TABLE 3.--Biomass (kg) of brook trout in the Pigeon River in July, 1976. Total Length (mm) Sta Total 95% CI 100-149 150-199 200-249 250-299 1 0.64 1.07 1.71 1.31-2.33 2 0.74 1.16 0.21 2.11 1.69-2.86 3 0.05 0.44 0.11 0.60 0.52-0.83 4 0.82 1.50 0.14 0.34 2.80 2.31-3.80 5 0.51 0.34 0.08 0.93 0.72-1.40 6 0.46 0.87 1.33 1.03-1.81 7 0.13 0.44 0.57 0.46-0.74 8 0.18 0.73 0.91 0.76-1.20 9 0.13 0.44 0.14 0.71 0.60-1.01 10 0.13 0.24 0.45 0.82 0.57-1.24 11,12 0.13 2.13 0.11 2.37 1.98-3.05 13 0.46 1.94 2.40 1.91-3.12 14 0.28 0.78 1.06 0.83-1.43 15 0.13 1.07 0.36 1.56 1.19-2.11 16 0.13 0.73 0.86 0.71-1.12 17 0.13 1.41 0.07 1.61 1.31-2.10 18 0.36 1.80 0.41 2.57 2.03-3.51 19 0.28 1.60 0.07 0.17 2.12 1.76-2.77 20 0.13 1.31 1.44 1.14-1.80 21 0.18 0.78 0.96 0.76-1.25 22 0.18 2.23 0.33 2.74 2.19-3.59 23 0.13 0.87 0.60 1.60 1.21-2.27 24 0.28 0.87 0.12 0.17 1.44 1.23-1.94 Total 6.59 24.75 2.99 0.89 35.22 28.22-47.28 95% CI 4.44- 20.61- 2.28- 0.89- 28.22- 10.56 30.56 5.27 0.89 47.28 nomnmpm mum one 813 «~13 818 gram mmuom mmmuwmp owns 8 xmm 32 NON-NNN NON N N O. ON ON NO NO ON, ON .OOOP ON-NN NN N O N N N, ON NN-NN NN N N O N O N FF ON NN-ON ON N N e N ON N NN ON-N_ ON _ O N N F_ _N NN-NP PN N O N O F. ON PO-ON _N N F O N N N. N O_ ON-ON NN N N N N N O N. _N-N. NN N N N N N_ ON-_F N, P N N N N_ O.-O NP _ N N N. NN-NP O_ N N P O N N ON NN-O_ NN P N N e N PP N ON NN-ON NN NOON.NNNVN _ N N N N. NN.P_ O -O O NONNOP N N O. O_-N __ O N N O O NN-N O, F N N N N N O -N N _ N O N N_-OP NP _ P _ N N N N_-O N, N P O O N ON-_F N_ N P N N N e O O_-NN NF p N N N N N ON-NN N_ O N N N N N ON-N N _ N O P NOOOemFO ONO OOO NON NON NON NON ON, OOP A 0 I 0 I. I I. I. 0 NO NNO Papa» OON ONO OOO ONN OON ONN OON ON, OOF NON Nagy OOOONO Peace .NNON .NNOO ON LNONN OONONO NON Ow NOOLO OzOeO CO NLOOEOOV OONOOPOOON--.O NONNO OOOONO OON ON NOONN ONOLO NO NOON NNOEONN--.O NONON 40 25% of the legal-sized brown trout stock is taken by anglers annu- ally. Nyman (1970) also reported that brook trout in some Newfound- land streams had a much higher catchability than brown trout. In Figure 6, October biomass totals are shown for all sta- tions in the study area. As in July, brown trout strongly dominate the biomass figures, but in fall, trout biomass appears to vary even more radically from station to station than in summer. During the October inventory, it was observed that most of the trout were in Spawning condition and that many were near areas which appeared to be spawning habitat. Concentration of fish in scattered areas of suitable spawning substrate may help to explain the more clumped fall distribution. Also, a greater number of very large brown trout (3500mm) were captured in the study area during the October electro- fishing. It is believed that many of these large fish were Spawn- ing migrants, perhaps from the Lansing Club Pond. Older trout in some streams are thought to be more abundant in downstream areas during most of the year and then to move upstream temporarily in the fall to spawning grounds (McFadden and Cooper, 1962). In October, large brown trout occurred erratically in the research area, adding considerable biomass to the stations in which they were captured and further contributing to the irregularity of the fall biomass distribution. Habitat Studies On August 10, 1976, discharge of the Pigeon River at the Old Vanderbilt Road bridge was 1.42m3/sec, and depth of the water 41 at mid-channel was 58cm. Mean discharge further downstream at the USGS gaging station was reported to be 1.61m3/sec for the same date. Moderate precipitation on the night of August 11 caused a slight rise in water level; mid-channel river depth at the road bridge on August 12 was 61cm. No significant precipitation occurred for the remainder of the summer study period, and fluctuations in river height at the bridge were only 1-2cm. USGS records indicated that the Pigeon River remained at summer low flow for the duration of the cover measurement studies. Maps of the 12 stations selected (see p. 25 for selection procedure) for cover measurement are given in Appendix B. Log jams and fallen trees provided the majority of trout cover in the Pigeon River study area. Alders, hardwoods, and cedars occurred along much of the stream bank. Frequently, alder branches and roots protruded into the water and provided measurable bank cover. In several stations, the river flowed through intermittent meadow where undercut banks and grassy overhangs afforded excellent trout cover. Channel and thalweg lengths, as well as total measurements of the three types of cover, are given for the selected stations in Table 10. Population-Bank Cover Relationships A series of simple correlation analyses was made to discern possible relationships between trout abundance and amount of over- head bank cover in the 12 selected stations. Various trout popula- tion parameters were considered separately as dependent variables. 42 TABLE 10.--Habitat measurements from 12 study stations on the Pigeon River. Channel Thalweg Length (m) 0f Overhead Bank Cover Sta “Iii" “Iii“ “"321? 1:333:22; .332. 3 100.6 105.0 4.0 0.0 43.3 47.3 4 105.0 108.8 3.8 0.8 50.9 55.5 5 101.2 103.8 3.9 5.1 31.9 40.9 6 101.2 104.4 30.9 7.3 7.8 46.0 9 103.8 106.2 20.6 6.1 5.8 32.5 10 100.6 103.8 7.4 8.3 15.5 31.2 13 110.0 115.0 10.0 9.2 32.0 51.2 14 105.0 108.1 15.1 9.4 24.4 48.9 19 108.8 113.8 2.3 8.9 68.0 79.2 20 97.5 99.4 4 7 9 6 59.2 73.5 21 101.2 106.9 9,2 13.6 33.6 56.4 22 106.2 110.0 4.6 9.2 47.3 61.1 43 Total length of bank having overhead cover was the independent vari- 2 values and their significance able in each case. Resulting r and r levels are given in Table 11. The non-random selection of stations for habitat study did not introduce positive bias into the analysis because the objective of selection was to obtain a sample with the greatest possible variety of trout abundance. By forcing population parameters to have high variability, this procedure actually tends to produce conservative correlation coefficients. For July populations, all parameters involving numbers of trout 3150mm long showed significant correlation with length of bank cover (Table 11). Abundance of trout less than 150mm in length, however, was not significantly related to bank cover. Furthermore, the number of brown trout 3450mm long was found to be more strongly correlated (r = 0.812) with the amount of bank cover than was the number of brook trout :150mm long (r = 0.665). Relationships between overhead bank cover and the two spe- cies of trout are shown in Figure 7. The greater association of brown trout with bank cover may be due in part to competitive advan- tages enjoyed by the species. Nyman (1970) reported that although brook and brown trout have similar ecological demands, brown trout in some Newfoundland streams were larger and dominated niches having the most favorable cover conditions. The highest correlation between trout abundance/station and bank cover/station is obtained when the total number of brook and brown trout 3150mm is considered (Table 11). On the basis of a 44 TABLE ll.--Correlation coefficients (r) and coefficients of determination (r2) for trout population variables (Y) and total length (m) of overhead bank cover/ station (X) in the Pigeon River. July 1976 October 1976 Dependent Variables (Y) r r2 r r2 No. all trout lOO-l49mm/sta 0.128 0.016 0.340 0.116 No. brook trout 3150mm/sta 0.665* 0.442 0.822** 0.676 No. brown trout 3150mm/sta O.812** 0.659 0.449 0.202 No. all trout 3150mm/sta 0.756** 0.572 0.792** 0.627 No. all trout >150mm/sta-- 0.938** 0.880 -- -- Stations 13 and 22 excluded Biomass all trout 100-149mm/sta 0.114 0.013 0.360 ’ 0.130 Biomass brook trout 3150mm/sta 0.617* 0.381 0.804** 0.646 Biomass brown trout 3150mm/sta 0.448 0.201 -0.164 0.027 Biomass brown trout 150-399mm/sta 0.734** 0.539 0.127 0.016 Biomass all trout.3150mm/sta 0.627* 0.393 0.021 0.001 Biomass all trout l50-399mm/sta o.347** 0.717 0.375 0.141 *Indicates significance at the 5% level. **Indicates significance at the 1% level. 45 50 ' V v v f I Brook Trout Z r! 0.685 40 - :8 0.641: - 10.734 0 . 0’ Brown “out 30 T I, f3 0.812 ’° 1:0.39711- 3.224 20*' No. Trout I‘-|501111'n/Station oo 20 4a 6b ‘ e‘o Length (In) of Overhead Bank Cover/Station Figure 7.. Relationship between July brook and brown trout abundances and bank cover in the Pigeon River. 80 T I I 1 v 6 v I I 4 C O E 60 . ® ‘ Q r: 0.933 E " "Oe,.5‘-lz.77' «I E O 2 4O " «I M ‘5 E ' a 5 20 - . g n . .1 o L 1 1 a I I | ‘ o 23 40 so 80 Length (ml of Overhead Bank Cover/Station Figure 8. Relationship between July total trout abundance and bank cover in the Pigeon River. (Circled points have been excluded from the correlation analysis.) 46 statistical test for significance (« = 0.05) of outlying observa- tions (Grubbs and Beck, 1972), the points for Stations 13 and 22 can be excluded from the correlation analysis. As shown in Figure 8, the remaining ten points form a strong linear relationship with 2 value indicates that 88% of the varia- r = 0.938. The resulting r tion in abundance of trout 3150mm in the Pigeon River can be accounted for by the length of overhead bank cover. Hunt (1971) reported that the number of brook trout greater than 152mm in a Wisconsin creek was highly correlated (r = 0.815) with the length of permanent bank cover, which he defined as stream- bank providing at least 15.2cm of overhang in 30.5cm of water. This correlation remained high (r = 0.809) even after extensive arti- ficial creation of cover had increased the average amount of bank cover by more than 400%/station. Furthermore, Lewis (1969) demon- strated that cover was the single most important factor accounting for variation in numbers of brown trout 178mm or larger in pools of a Montana stream. According to Lewis's description, cover included brush, overhanging vegetation, undercut banks, and other areas pro- viding shelter for larger fish. Correlations between July trout biomass/station and bank cover/station are given in Table 11 and shown in Figures 9 and 10. Biomass of trout less than 150mm was not related to length of over- head bank cover. However, biomass of brook trout 3150mm was signifi- cantly correlated with cover. The total biomass/station value will be influenced heavily by the presence of any very large trout. 47 .5 4°- . I I 1 U V 1— I T a; ,’ Brown Trout \ ’6’ r 3 0.734 I I 5 4.0 .. y = 0.058 x - 0.365 ° 1 .. E o I,’o o 3 Brook Trout , I ’ '10 1- ] ' 3 0.6.7 0 ’ I O o 9 y = 0.030 x " 0.379 0 I o 2 ’ o '3" 2.0- .. .= 15 5 In «11 8 O 5 .— I l I J J l L 1 In °6 20 40 so so Length (m) at Overhead Bank Cover/Station Figure 9. Relationship between July standing crops of brook and brown trout and bank cover in the Pigeon River. Figure‘lO. g so .‘ . . E . (I) \ - . E r=0.847 '8'; 4.0 _ 180.08811'0343 . 1 .53 § ' ‘ .: i 2.0- - ‘3' .- ‘5 .. . O .5 m 0 I l 1 1 n 1 4 A 0 20 4O 60 80 Length (111) of Overhead Bank Cover/Station Relationship between July total standing crop of trout and bank cover in the Pigeon River. 48 Since larger browns may reside near study station boundaries and occupy sizeable home ranges which can extend into an adjacent sta- tion, the stations in which they are captured may not accurately reflect their total cover requirements. Large trout also seem to travel greater distances when frightened by electricity, hence their true stations of residence may in some cases be less certain. Per- haps if stations of greater length had been used, the biomass variability due to large brown trout might not have been so great. These interferences are believed in part to explain the poor correlation between biomass of brown trout 3150mm and abundance of bank cover. If, however, all brown trout 400mm or longer are omitted from the biomass totals, the correlation between brown trout biomass/station and bank cover/station becomes highly significant (r = 0.734). Also, by excluding brown trout 3400mm from the totals for biomass of all trout 3150mm (there were no brook trout 3400mm), a very high correlation with bank cover is obtained (Figure 10). Thus, about 72% of the variation in biomass of all trout 150-399mm in the Pigeon River can be accounted for by the length of overhead bank cover. Stewart (1970 in Hunt, 1971) analyzed physical characteris- tics of 41 study sections in a small Colorado trout stream and found that a combination of hiding and protective shelter was highly correlated with the distribution and density (biomass/station) of brook trout. According to Wesche (1976), a linear relationship existed between the mean cover rating for a stream section and the 49 standing crop of trout present. His cover rating formula involved one component for the length of overhead bank cover, another com- ponent for the area of stream having a water depth of at least 15cm with a substrate diameter of 7.6cm of more (rubble-boulder), and a preference factor for ”catchable" and "subcatchable" trout. In the present study, "rubble-boulder" areas were not measured because of the low preference of larger (3152mm) brook and brown trout for this type of cover. Relationships between October trout p0pu1ations and bank cover are also summarized in Table 11 and depicted in Figures 11 and 12. As previously mentioned, brown trout appear to redistribute themselves extensively in the fall for spawning. This may explain the lack of significant correlation between any of the October brown trout population parameters and the amount of bank cover. Brook trout, on the other hand, displayed an even stronger rela- tionship with bank cover in fall than in summer. Both number and biomass of brook trout 3150mm were highly correlated with the length of overhead bank cover (Figures 11 and 12). One possible explana- tion for increased brook trout association with bank cover during this period was the apparent decrease in brown trout occupation of the preferred sites in their search for more appropriate spawning habitat. In an attempt to test for possible interactions between brook and brown trout in their use of cover, a series of tests for corre- lation was run, using various population density parameters as X 50 f I T I 1 I I I r: 0’! Brown Trout 2.9 4O _ I r=0.449 o . an? ID y= 0.23011 + 4.693 2 . Brook Trout o E ' r = 0.822 '1 8 o y= 0.59411-15.145 2', 20 ~ I ‘ 8 L. I'. . 1: ° . O z: 1 k 1 1 1 J 1 1 . 00 20 40 so 30"— LengthIIn) ot Overhead Bank Cover /Station Figure 11. Relationship between October brook and brown trout abun- dances and bank cover in the Pigeon River. (31 Brown Trout ° 1: I r 8 0.127 5.: 6'0 ' IO y=0.0I4 x +1.ser " (‘13 O Brook Tt’out E r / r = 0.e04 1 E y = 0.033 x - 0.008 8: o I? 1413- - c: 52 0- 1- III a o o ...... P ---,,-—o”’- . g 2.0 I’ Do a .1 8 D g - . ii 3 1+ fiL 1 1 J 1 co 20 40 50 3'6"— Length (m1 at Overhead Bank Cover/Station Figure 12. Relationship between October standing crops of brook and brown trout and bank cover in the Pigeon River. 51 and Y variables. No significant relationships could be demonstrated between brook trout population parameters (abundance, biomass, or trout/cover density, i.e., number or biomass of trout/length of overhead bank cover) and brown trout/cover density. However, because both brook and brown trout showed significant correlation in June with length of bank cover, there was also a positive correlation (r = 0.764) between number of brook trout :150mm/station and number of brown trout 3150mm/station. Such a relationship did not exist for the October trout p0pu1ations. Multiple regression was also used to test the importance of brown trout/cover density as a second variable in accounting for variation in brook trout abundance when length of overhead bank cover served as the first variable. The analysis revealed that brown trout/cover density was not a significant variable in the regression equation (a = 0.05). It should be acknowledged, however, that the above tests were only crude evaluations based on station totals and not on individual cases of cover usage by the two species. There- fore, the possibility of meaningful interaction between brook and brown trout in regard to utilization of cover has certainly not been eliminated. Salmon Trout River Population Estimates Although the softness of the water made electrofishing less effective in the Salmon Trout River, I feel that the majority of study stations were sufficiently sampled for the computation of 52 reliable population estimates. However, deep water prevented elec- trofishing in certain areas of Stations 2, 15, 19, 39, and 55, and estimates may therefore be incomplete or deceptively low for these stations. A series of five long, unwadable pools made electro- fishing nearly impossible in Station 11, which probably explains why no brook trout were captured there. Results and confidence intervals for the July brook trout inventory of the Salmon Trout River are given in Tables 12-15. The sparseness of the brook trout p0pu1ation below Lower Falls (Table 12) is immediately apparent. There are fewer than ten trout 3100mm in nearly all stations, and few of these fish exceed 250mm in length. Above Lower Falls (Table 13), the population of wild brook trout increases, but again, there are very few fish larger than 250mm. Trout in the lOO-l49mm size group comprise the vast majority of the p0pulation. As shown in Figure 13, July biomass totals for brook trout vary considerably from station to station throughout the study area. Above Lower Falls, the lowest biomass (0.21kg) was found in Station 51, while Station 47 contained the greatest biomass of brook trout (3.44kg). I October brook trout abundances are given in Tables l6-l9. Data for the study area below Lower Falls (Tables 16 and 18) include any Lake Superior spawning immigrants (coaster brook trout) present at the time of the fall inventory. Because of these fish, brook trout populations were noticeably increased over summer levels in several stations. Above Lower Falls, the total population of brook trout 3100mm increased from July to October, primarily because of 53 TABLE 12.--Population (number) of wild brook trout in the Salmon Trout River below Lower Falls in July, 1976. Total Length (mm) Sta 100- 150- 200- 250- 300- T°taI 95% CI 149 199 249 299 349 1 1 2 2 5 3-10 2 1 2 2 5 3-10 3 3 3 2-7 4 4 3 7 4-13 5 1 1 1-2 s 1 1 2 2-3 7 1 2 3 2-6 8 3 3 6 4-13 9 4 4 1 9 s-19 10 2 2 1-4 11 12 1 2 2 1 1 7 5-13 13 1 3 1 5 4-10 14 1 2 3 2-6 15 4 2 6 3-11 16,17 10 11 3 3 27 15-57 Total 27 31 21 10 2 91 58-184 95% CI 21-52 ‘15-53 12-42 8-20 2-2 58-184 54 TABLE 13.--Population (number) of wild brook trout in the Salmon Trout River above Lower Falls in July, 1976. Total Length (mm) Sta Total 95% CI 100-149 150-199 ZOO-249 250-299 35 42 I 43 23-94 37 11 3 14 9-31 38 90 5 1 96 52-209 39 45 5 50 28-109 40,41 35 10 4 49 35-75 42,43 26 12 1 39 29-56 44 4 10 8 1 23 15-41 45,46 13 15 8 36 25-60 47 22 16 18 56 37-97 48 9 3 2 14 10-22 49 13 1 2 16 12-25 50 9 9 2 20 15-32 51 11 11 8-16 52 35 10 45 33-65 53 9 1 10 8-15 54 37 7 44 32-63 55 28 10 2 40 28-60 56 20 31 23 17-33 57 61 15 76 56-110 58 46 46 34-66 59 52 9 61 45-88 Total 618 145 47 2 812 551-1367 95% CI 415-1025 110-229 24-111 2-2 551-1367 55 TABLE 14.--Biomass (kg) of wild brook trout in the Salmon Trout River below Lower Falls in July, 1976. Total Length (mm) Sta 100_ 150_ 200_ 250_ 300_ Total 95% CI 149 199 249 299 349 1 0.02 0.12 0.20 0.34 0.18-0.68 2 0.02 0.12 0.18 0.32 0.17-0.65 3 0.07 0.07 0.05-0.16 4 0.24 0.38 0.62 0.37-1.17 5 0.02 0.02 0.02-0.05 6 0.21 0.30 0.51 0.50-0.71 7 0.02 0.33 0.35 0.19-0.70 8 0.07 0.64 0.71 0.47-1.44 9 0.09 0.24 0.17 0.50 0.36-0.98 10 0.12 0.12 0.06-0.24 11 12 0.02 0.12 0.30 0.24 0.42 1.10 0.89-1.77 13 0.02 0.35 0.23 0.60 0.49-1.20 14 0.02 0.30 0.32 0.17-0.64 15 0.24 0.30 0.54 0.27-1.01 16,17 0.23 0.65 0.50 0.74 2.12 1.28-4.26 Total 0.60 1.85 2.84 2.23 0.72 8.24 5.47-15.66 95% CI 0.48- 0.89- 1.62- 1.76- 0.72- 5.47- 1.42 3.43 5.65 4.44 0.72 15.66 56 TABLE 15.--Biomass (kg) of wild brook trout in the Salmon Trout River above Lower Falls in July, 1976. Total Length (mm) Sta Total 95% Cl 100-149 150-199 200-249 250-299 36 0.98 0.06 1.04 0 57-2.27 37 0.26 0.18 0.44 0.32-0.98 38 2.10 0.30 0.10 2.50 1.47-5.35 39 1.05 0.30 1.35 0.84-2.94 40.41 0.66 0.52 0.36 1.54 1.03-2.62 42.43 0.49 0.62 0.18 1.29 1.00-1.81 44 0.08 0.52 0.75 0.33 1.68 1.13-3.00 45,46 0.24 0.78 0.87 1.89 1.20-3.61 47 0.41 0.83 2.20 3.44 2.02-6.96 48 0.17 0.16 0.17 0.50 0.32-0.87 49 0.24 0.05 0.34 0.63 0.41-1.29 50 0.17 0.47 0.23 0.87 0.61-1.55 51 0.21 0.21 0.15-0.30 52 0.66 0.52 1.18 0.85-1.72 53 0.17 0.05 0.22 0.18-0.35 54 0.70 0.36 1.06 0.77-1.52 55 0.53 0.52 0.20 1.25 0.84-2.03 56 0.38 0.16 0.54 0.39-0.75 57 1.15 0.78 1.93 1.42-2.84 58 0.86 0.86 0.64-1.24 59 0.98 0.47 1.45 1.08-2.12 Total 12.49 7.65 5.22 0.51 25.87 17.24-46.12 95% CI 21212 12:18 12 51 3:51 46:12 57 so - Biomass (kg) of Wild Brook ‘Itout If." .13' :I.‘ . -' . . 1 '2-1 4.: ,5 _ .. .' 7.. .. :2; 3:. .;. .: '3): '._". {:7 :1: -.1 V x .1: 4.; _.j. _. . . - '2-? 3'3: .- .-; S. -:- 1'5 3' 2-2 2'. 'I- '-.‘r .-.- .1.‘ .5 .'. '.- g.- , . . . . . . . . :.'. "q. ..: .: .j. .32. I ‘.~1 3‘ . T :1. ~2- . . . . . .. . . . . . ~. 2 .‘ .«3 .‘ 1. , h. '.' .3. '- L. ,-.' ~. . , >:‘ r .- .‘j'. ‘4 ~11. :':, :'I . :2 "1."; j. _-~- _ ‘. |:-- .'_.: .f.‘ 3: ._ . :5. n .'. -.-. n ‘ -. . .- . A .j ._.: .1- :.: _‘ ‘ .-. _.:. ._ .‘.- ;.; _.; .;.; . 1 .. ~. 1 U‘I‘ I .1 .'. '. . .7. 5'. ' 21:? '-: 1:? 1?: -.'-: .‘ L-zi :13 £54 ‘31' .13 ~:' :-' , :~'. 5“ L1: ' Lj.; _I:'. \I-1 .l'- .'_. i. . 'j.’ ';'.'1 "a 1' 'r -.. . 1'1‘ v” :4 5.1 S-f l" -;» -:- -:. ;-: 9. ~:-.' 5; . :.'_, j'. .;.. ;_ :.;, .;.; ;.l I: .21 p: .0; . . . .' .~ _. . .g. .- _. ._a 12;; 1-7- -'.. :17- -:- .' 5:~ '7: :'~" L .. I 5 IO 15 20 35 4O 45 Station Markers Figure 13. Biomass of wild brook trout in the Salmon Trout River in July, 1976. 3.01- .. ‘ 2&3r Biomass (kg) of Wild Brook Trout 20'35 40 Station Markers Figure 14. Biomass of wild brook trout in the Salmon Trout River in October, 1976. ' 58 TABLE 16.--Population (number) of wild brook trout in the Salmon Trout River below Lower Falls in October, 1976. Total Length (mm) Sta 100- 150- 200- 250- 300- >350 T°ta‘ 95% CI 149 199 249 299 349 (length) 1 2 2 1-5 2 2 2 1-5 3 4 4 2-10 4 2 4 4 2 12 4-23 5 6 6 4-14 6 6 6 4-14 7 2 4 6 2-12 8 2 4 6 3-12 9 2 4 4 10 4-19 10 9 7 2 18 8-37 11 12 4 4 2 2 2(350,449) 14 7-27 13 4 2 6 2-11 14 2 2 1-5 15 4 4 2 10 4-21 16 9 4 14 4 31 12-58 17 4 7 4 2 17 6-33 18 6 7 13 6-26 19 4 4 14 22 7-40 Total 70 37 52 18 8 2 187 78-372 95% CI 39-169 10-64 14-88 9-34 4-15 2-2 78-372 59 TABLE l7.--Population (number) of wild brook trout in the Salmon Trout River above Lower Falls in October, l976. Total Length (mm) Sta 100-149 160-199 200-249 250-299 T°ta1 95% CI 34 52 34 2 88 59-148 35 44 28 72 49-120 36 33 9 42 29-67 37 62 12 74 52-117 38 89 32 2 123 84-200 39 35 23 5 63 42-107 40 16 6 22 16-32 41,42 9 15 6 3 33 21-51 43 23 3 26 18-38 44 3 12 9-17 45 9 3 12 9-17 46 25 3 28 21-39 47 21 5 26 20-37 48 14 5 19 14-27 49 18 5 3 26 18-38 50 41 6 3 50 36-72 51 44 3 47 35-67 52 32 3 3 38 27-54 53 25 2 27 20-38 54 35 3 6 44 30-65 55 58 9 6 73 52-106 56 14 3 17 12-24 57 58 18 3 79 57-114 58 30 20 50 37-73 59 41 9 50 38-71 Total 837 259 39 6 1141 805-1739 95% CI ‘518-1219 170-439 15-71 2-10 805-1739 60 TABLE lB.--Biomass (kg) of wild brook trout in the Salmon Trout River below Lower Falls in October, l976. Total Length (mm) Sta 100- 150- 200- 250- 300- >350 T°ta‘ 95% CI 149 199 249 299 349 -— 1 0.02 0.02 0.01-0.05 2 0.02 0.02 0.01-0.05 3 0.04 0.04 0 02-0 11 4 0.02 O.l8 0.46 0.07 1.36 0.52-2.56 5 0.06 . 0.06 0 04-0 15 6 0.06 0.06 0.04-0.15 7 0.02 0.48 0.50 0.l3-0.89 8 0.02 1.07 1.09 0.54-1.92 9 0.02 0.46 1.07 1.55 0.66-2.73 10 0.10 0.31 0.33 0.74 0.31-1.42 11 12 0.04 0.18 0.53 0.62 1.46 2.83 2 10-4.18 13 0.40 0.40 0.80 0 30-1 50 14 0.02 0.02 0.01-0.05 15 0.04 0.31 0.37 0.72 0.28-l.39 16 0.10 0.18 1.46 1.14 2.88 1.08-4 92 17 0.04 0.31 0.59 0.33 1.27 0.43-2.34 18 0.06 0.31 0.37 0.13-0.68 19 0.04 0.18 1.20 1.42 0 41-2.39 Total 0.72 1.65 5.36 4.10 2.46 1.46 15.75 7 02-27 48 0 39- 0 45- 1.44- 2.05- 1.23- 1.46- 7.02- 95% CI 1.81 2:83 9.07 7.68 4.63 1.46 27.48 61 TABLE l9.--Biomass (kg) of wild brook trout in the Salmon Trout River above Lower Falls in October, 1976. Total Length (mm) Sta Total 95% CI 100-149 150-199 200-249 250-299 34 0.71 1.71 0.l4 2.56 1.59-4.62 35 0.60 1.41 2.01 1.29-3.62 36 0.45 0.45 0.90 0.58-1.54 37 0.85 0.60 1.45 0.97-2.44 38 1.22 1.61 0.17 3.00 1.91-5.25 39 0.48 1.15 0.51 2.14 1.35-3.96 40 0.25 0.27 0.52 0.37-0.76 41.42 0.14 0.66 0.74 0.46 2.00 1.00-3.30 43 0.37 0.23 0.60 0.35-0.90 44 0.14 0.13 0.27 0.20-0.38 45 0.14 0.13 0.27 0.20-0.38 46 0.40 0.13 0.53 0.39-0.73 47 0.33 0.22 0.55 0.43-0.79 48 0.22 0.22 0.44 0.34-0.63 49 0.29 0.22 0.50 1.01 0.55-1.55 50 0.65 0.27 0.21 1.13 0.74-l.67 51 0.70 0.l3 0.83 0.61-1.18 52 0.51 0.13 0.31 0.95 0.57-1.40 53 0.40 0.09 0.49 0.35-0.69 54 0.56 0.13 0.56 1.25 0.69-l.99 55 0.92 0.40 0.46 1.78 1.15-2.73 56 0.22 0.13 0.35 0.25-0.49 57 0.92 0.80 0.23 1.95 1.34-2.89 58 0.48 0.88 1.36 1.01-2.01 59 0.65 0.40 1.05 0.80-1.50 Total 12.60 12.27 3.56 0.96 29.39 19.03-47.40 9.33- 8 01- 1 37- 0.32- 19.03- 95% CI 18.33 20.99 6.49 1.59 47.40 62 recruitment of young fish into the 100-149mm size class. However, the October total bi0mass of brook trout above Lower Falls was remarkably similar to the July total. Addition of Stations 34 and 35 to the fall inventory accounted for the greater total biomass in October. As in summer, fish in the 100—149mm size group com- prised the bulk of the fall p0pulation. October biomass totals for all stations sampled are shown in Figure 14. The presence of coaster brook trout obviously inflated the fall biomass in Stations l-l9. Above Lower Falls, the patterns of biomass distribution in October were roughly similar to those observed in July, with the notable exception of Station 47, which had far less biomass than in July. In order to facilitate electro- fishing during the fall inventory, the water level at Lower Dam was drawn down considerably, leaving no water beneath a large log jam in Station 47. This temporary loss of a major cover area is believed to have accounted for the lower autumn biomass in the station, as most of the brook trout captured there during the July inventory were taken from beneath this log jam. Habitat Studies The results of discharge measurements made on August 31, 1976, at six sites (Figure 4) along the Salmon Trout River are given in Table 20. Note that the incoming flow of Clear Creek provides most of the increase in discharge from site E to site C on the river. No significant changes in discharge were detected from site C downstream to site A. Water levels recorded periodically at each discharge 63 TABLE 20.--Streamflow discharge (cubic meters per second) at 6 sites along the Salmon Trout River on August 31, 1976 (Site locations are shown in Figure 4), Site of Nearest Discharge Measurement Station Marker (cms) A 1 0.77 B 18 0.79 C 37 0.79 0* 40 0.16 E 50 0.54 F 60 0.53 *Clear Creek site fluctuated less than l.5cm during the two weeks of summer study. No precipitation occurred during this time. Therefore, the river remained at summer baseflow for the duration of the habitat measure- ments. As previously described, 18 stations for mapping and cover analysis on the Salmon Trout River were selected on the basis of diversity of trout abundance. Finished maps are given in Appendix C. Total lengths of each of the three major types of bank cover are given by station in Table 21. Channel and thalweg lengths for each station have also been included in the table. Below Lower Falls, where the stream flowed primarily through mature hardwood forest, most stations had steep banks with little cover in the form of overhanging vegetation. Substantially undercut 64 TABLE 21.--Habitat measurements from 18 study stations on the Salmon Trout River. Channel Thalweg Length (m) of Overhead Bank Cover 36? Leggth Length Undercut Overhanging Log Total Bank Vegetat1on Cover 3 99.4 106.2 3.0 0.0 13.6 16.6 5 96.2 99.4 13.0 1.1 8.1 22.2 6 101.2 104.4 8.2 4.0 7.6 19.8 7 101.9 103.8 3.0 2.9 1.7 7.6 8 105.0 108.8 3.6 2.7 13.1 19.4 10 101.2 106.9 9.6 2.0 8.8 20.4 13 96.2 100.0 5.8 1.7 6.9 14.4 14 105.0 113.8 9.4 0.0 1.0 10.4 36 92.5 95.6 8.4 4.1 3.5 16.0 38 94.4 96.2 13.8 3.1 17.1 34.0 47 98.8 101.2 1.1 7.7 43.7 52.5 48 102.5 109.4 3.5 9.7 1.5 14.7 50 108.1 115.0 0.9 5.2 9.6 15.7. 51 99.4 103.8 0.0 0.9 6.6 7.5 52 120.0 132.5 5.8 0.8 20.3 26.9 53 110.6 116.9 1.1 0.5 2.9 4.5 54 111.9 123.8 10.4 3.8 11.6 25.8 57 106.2 110.6 17.8 10.6 13.7 42.1 65 banks occurred in some areas. Logs artificially fastened against the bank provided additional shelter in several stations. Between Lower Falls and Lower Dam, undercut banks and log cover occurred erratically in small amounts, and overhanging vegetation was more abundant. Above Lower Dam, the stream meandered for a considerable distance through alder meadow where overhanging roots and branches, undercut banks, and infrequent log jams constituted the majority of fish cover. Hardwoods again began to encroach upon the stream around Station 52, but all three major types of bank cover continued to occur throughout the study area above that point. Egpulation-Bank Cover Relationships Correlation analyses similar to those performed on Pigeon River data were also done on data from the Salmon Trout River. Brook trout population parameters were used as dependent variables while abundance of overhead bank cover served as the independent variable. Correlation coefficients (r) and coefficients of deter- mination (r2) for these analyses are presented in Table 22. Relationships between July number of brook trout 3150mm and bank cover are. significant at the 1% confidence level when all stations are considered, as well as when only stations above Lower Falls or Lower Dam are considered. However, the eight stations below Lower Falls contain sparse populations of brook trout and generally detract from the strength of the correlation (Figure 15). If only the eight selected stations above Lower Dam are considered, 66 TABLE 22.--Correlation coefficients (r) and coefficients of determination (r2) for trout population variables (Y) and total length (m) of overhead bank cover/ station (X) in the Salmon Trout River. July 1976 October 1976 Dependent Variables (Y) r [‘2 r r2 No. brook trout 3]50mm/sta-- 0.822** 0.676 0.495* 0.245 all stations included No. brook trout 3150mm/sta-- O.851** 0.724 0.452 0.204 stations above Lower Falls only No. brook trout >150mm/sta-- 0.909** 0.826 0.518 0.268 stations above—Lower Dam only No. brook trout >150mm/sta-- -- -- 0.882** 0.778 stations above-Lower Falls only, excluding #38 and #47 Biomass brook trout 3100mm/sta-- O.870** 0.757 0.442 0.195 all stations included Biomass brook trout 3100mm/sta-- 0.944** 0.891 0.390 0.152 stations above Lower Falls only Biomass brook trout 3100mm/sta-- 0.963** 0.927 0.367 0.135 stations above Lower Dam only Biomass brook trout 3100mm/sta-- -- -- 0.863** 0.745 stations above Lower Falls only, excluding #38 and #47 Biomass brook trout 3150mm/sta-- O.738** 0.545 0.383 0.147 all stations included Biomass brook trout 3150mm/sta-- 0.787** 0.619 0.447 0.200 . stations above Lower Falls only ' Biomass brook trout ZlSOmm/sta-- 0.829* 0.687 0.491 0.241 stations above Lower Dam only Biomass brook trout >150mm/sta-- -- -- 0.939** 0.882 stations above LowEr Falls only, excluding #38 and #47 *Indicates significance at the 5% level. **Indicates significance at the 1% level. N0. Trout zI501'nn1/ Station Figure 15. Biomass (kg) Trout z150mm /St0tion Figure 16. 4 V f I T I 4.0 2.0 67 I) r = 0.822 (000) y: 0.548 x — 5.445 . 2h 3 0.851 (00) J )1: 0.555 x - 4.315 03 31 r = 0.909 (0) y: 0.556 x - 5.551\ Length(tn) of Overhead Bank Cover/ Station Relationship between July brook trout abundance and bank cover in the Salmon Trout River. (:1 - stations below Lower Falls; 0 - stations between Lower Falls and Lower Dam; a - stations above Lower Dam) I) r: 0.738 (c1001 1: 0.041: - 0.361 2) rs 0.787 (00) e .. 3'0 y: 0.045 x - 0.466 3) r8 0.829 ('1 ”0.040 x - 0.422 \ ' 4'0 ‘ 60 Length (m) of Overhead Bank Cover/ Station Relationship between July standing crop of brook trout and bank cover in the Salmon Trout River. ( 1:1 - stations below Lower Falls; 0 - stations between Lower Falls and Lower Dam; o - stations above Lower Dam) 68 the highest correlation (r = 0.909) between bank cover and abundance of brook trout 3150mm is obtained, This is an even stronger rela- tionship than that reported by Hunt (1971) between permanent bank cover and abundance of brook trout greater than 152mm, for which r was 0.815. I July biomass totals for brook trout 3100mm and_:150mm were also highly correlated with abundance of bank cover (Table 22, Figure 16). Above Lower Falls, brook trout in the 100-149mm size group comprise nearly half of the total July biomass, which may explain why a greater correlation coefficient is obtained when bio- mass of trout 3100mm (instead of 3150mm) is plotted against length of overhead bank cover. Correlations between brook trout biomass and bank cover also tend to improve as stations below Lower Dam are excluded from the analysis. Results of bank cover correlation tests for October brook trout population parameters are also given in Table 22. Correlation coefficients are very low when all stations are included or when all stations above Lower Falls or Lower Dam are included. The presence .of coaster brook trout in stations below Lower Falls confounds the October population data and probably explains the poor results obtained in analyses inVolving these stations. Correlations between cover and trout populations in stations above Lower Falls are depressed by the effects of Stations 38 and 47. The points for these two stations, circled in Figures 17 and 18, deviate most substan- tially from expected points and may be excluded from the analysis 69 40 V I 1 I r g G) 3.5 3° '1 Q r . 0.882 E y=OA32:-0382 CE: 20 ° 4 52 Al ‘5 2 P: '0. 00 e d 3 ° <9 1 L J l 1 00 20 4‘0 10 Length (m) of Overhead Bonk Cover/ Station Figure 17. Relationship between October brook trout abundance and bank cover in the Salmon Trout River. (0 - stations between Lower Falls and Lower Dam; o - stations above Lower Dam. Circled points have been excluded from the correlation analysis.) po 5? Biomass (kg) Trout 2|50mm/Station Figure 18. o? r 8 0.939 y 8 0.02411 - 0.022 l l 20 40 Length (111) of Overhead Bonk Cover / Station 60 Relationship between October standing crop of brook trout and bank cover in the Salmon Trout River. (C) - stations between Lower Falls and Lower Dam; e - stations above Lower Dam. Circled points have been excluded from the correlation analysis.) 70 according to the "outlier" significance test (cc = 0.05) of Grubbs and Beck (1972). In addition, there are experimental reasons for doubting the validity of cover measurements for Stations 38 and 47. A huge crib of logs from a former logging dam is lodged against the bank in Station 38, forming a vast area of fish cover which was probably underestimated by the method of cover measurement used in this study. Because a large number of trout were obtained from this device during the July inventory, an especially intense electro- fishing effort was made around these logs during the October inven- tory. The result was a high population estimate for a station having a good deal of unmeasured cover. 0n the other hand, most of the measured cover in Station 47 was rendered unuseable to fish during the October inventory because of the lowered water level, as described previously (p. 62). Points for the remaining eight stations above Lower Falls describe a close relationship between overhead bank cover and both measures of brook trout density (Figures 17 and 18). Abundance of overhead bank cover/station accounts for about 78% of the variation in number of trout 3150mm/station and more than 88% of the variation in biomass of trout 3150mm/station. By October, more brook trout have been recruited from the 100-149mm size group into the 150-199mm group, and as a result, stronger correlation with bank cover is obtained when biomass of trout 3150mm (instead of 3100mm) is used as the dependent variable. 71 Combined Analysis: Pigeon and Salmon Trout Rivers Although these two rivers differ markedly in many respects, the possibility of a continuous relationship between cover and trout abundance in both streams demanded investigation. In comparison, the Pigeon is the larger of the two rivers, having an 80% greater baseflow discharge in the study area than the Salmon Trout River. Furthermore, the Pigeon River, which contains brook and brown trout, generally has more overhead bank cover/station as well as greater numbers and biomass of trout/station than the Salmon Trout River 'above Lower Falls, where brook trout are the only salmonid present. In Figure 19, the relationship between July trout abundance and bank cover in both the Pigeon and Salmon Trout Rivers is depicted. The resulting r value--0.869--suggests a strong correla- tion, and the coefficient of determination indicates that about 76% of the variation in abundance of trout 3150mm in both rivers can be explained by abundance of overhead bank cover. Stations below Lower Falls on the Salmon Trout River were not included in the analysis because of the inconsequential numbers of brook trout and (the possibly confounding effects of anadromous salmonids there. Biomass of trout in July also proved to be highly correlated (r = 0.901) with length of overhead bank cover in a combined analy- sis of the two study streams (Figure 20). In this test, only trout 150-399mm in length were entered into the biomass totals in order to avoid the distortion caused by very large brown trout in the 2 Pigeon River, as explained previously. According to the r -value for this relationship, length of overhead bank cover accounts for 72 8G I l T I I U U j . 60" r a 0.869 1 y a 0.954 x - 11.546 No. All Trout 2150mm / Station 06‘9"“‘9" 20 1 4o 60 L 80 Length (m) of Overhead Bonk Cover/ Station Figure 19. Relationship between July total trout abundance and bank cover in the Pigeon River (a) and the Salmon Trout River above Lower Falls (0). . 6.0V *1 l 1 j I 1 1 I r= 0.901 y: 0.0911: ‘ 1.191 '1 Biomasdkg) All Trout Ibo-399mm/Station 8C1 Length (m) of Overhead Bank Cover/ Station Figure 20. Relationship between July total standing crop of trout and bank cover in the Pigeon River (0) and the Salmon Trout River above Lower Falls (0). 73 81% of the variation in biomass of trout 150-399mm long in both rivers. Several complications arise when attempts are made to corre- late October population parameters with length of bank cover in the two streams. Biomass totals for the Pigeon River are severely dis- rupted by the redistribution of spawning brown trout, while rela- tionships in the Salmon Trout River are clouded by the effects of Stations 38 and 47. The only strong correlation derived for October trout data from both streams was between length of overhead bank cover/station and total number of trout 3150mm/station, in all Pigeon River stations and all Salmon Trout River stations above Lower Falls except #38 and #47. The r value for this analysis was 0.897. Considering only October brook trout populations in the Pigeon River stations and Salmon Trout River stations above Lower Falls (excluding #38 and #47), an r of 0.800 was obtained for the correlation of biomass of trout 3150mm with length of bank cover. The existence of a bank cover-trout abundance relationship that is continuous throughout study stations from two very different Michigan streams is remarkable. It suggests that despite major dif- ferences in water chemistry and flow regimes, quantity of streambank shelter is still the principal limiting factor of trout density in many areas of both these rivers. However, hasty speculation concern- ing means of boosting trout populations in these streams should be avoided. It may not be possible, for example, to raise the trout carrying capacity of the Salmon Trout River to Pigeon River levels 74 by increasing the amount of bank cover beyond the highest observed cover abundances. Other environmental factors,such as severe winters, lack of spawning habitat, or floods, might limit the population before it could fill the increased cover capacity. Nevertheless, these results suggest that under conditions existing in the study areas during 1976, the length of overhead bank cover played a major role in regulating trout abundance. It would seem that some enlarge- ment of trout p0pu1ations may be possible by augmenting bank cover in stations having little shelter of this kind. According to Hunt (1976), the mean number of legal-sized brook trout in one section of Lawrence Creek, Wisconsin, increased by over 190% and the mean biomass of legal-sized brook trout increased nearly 180% during the six years following addition of bank cover/wing deflectors to that stream section. Other factors may also have affected stream carrying capac- ity and trout p0pu1ations in the study streams. According to Platts (1976), it is the proper combination of many conditions that is significant in producing a fishery resource. Benson (1953b) felt that ground water seepage, by determining the location and number of suitable spawning areas, was an important condition limiting trout production in many parts of the Pigeon River. Certainly, availabil- ity of spawning sites and variation in reproductive success may influence trout abundance in the Pigeon and Salmon Trout Rivers. McFadden and Cooper (1962) found a positive relationship between standing crops of fish and water conductivity in six Pennsylvania 75 streams, such that hardness and specific conductance might be inter- preted as measures of water fertility and biological productivity of trout streams. According to these criteria, the hard water of the Pigeon River, with its high specific conductance (380 micromhos), would be presumed more "fertile" than the soft, poorly-conductive (130 micromhos) water of the Salmon Trout River. One important fac- tor that can have a crucial effect on trout populations is angler- caused mortality (Schuck, 1945; Cooper, 1952, 1953; McFadden, 1961; McFadden and Cooper, 1962, 1964; Gard and Seegrist, 1972). Stations on the two streams used in this study are undoubtedly subjected to differential fishing pressures. Although spring flooding occurs in both rivers, the Salmon Trout River is known to experience severe highwater during Spring snowmelt. These floods may have devastating effects on fish populations as well as fish cover. Late winter or spring floods can destroy eggs and alevins incubating in spawning redds, subject adult fishes to harmful if not lethal stresses, and seriously reduce the macrofauna food supply (Needham and Jones, 1959; Elwood and Waters, 1969; Mundie, 1969; Seegrist and Gard, 1972). Interspecific competition may also have an important influ- ence on stream trout populations. The sympatric occurrence of brook and brown trout in the Pigeon River study area raises questions concerning the significance of ecological interactions in population regulation. In the Salmon Trout River below Lower Falls, seasonal invasion by spawning migrants from Lake Superior, as well as tempor- ary occupation of the river by their offspring, may seriously limit 76 resident trout p0pulations. Furthermore, the introduction of legal- sized hatchery trout into the Salmon Trout River could have impor- tant effects on the growth and reproduction of wild trout (Allen, 1962). In this study, the significance of specific habitat features other than total length of bank cover was not investigated. Hater velocities beneath cover were not measured, although it has been clearly demonstrated that the selection and use of microhabitats by salmonids is governed heavily by current velocity (Chapman, 1966; Hickham, 1967 in Stalnaker and Arnette, 1976; Baldes and Vincent, 1969; Giger, 1973; Wesche, 1973; Banks, Mullan, Wiley, and Dufek, 1974 in Stalnaker and Arnette, 1976). Proximity of cover to the thalweg or to principal lines of drift, which may have important implications for feeding efficiency (Everest and Chapman, 1972; Jenkins, 1969), was also not measured. Another unmeasured habitat variable was the number, size, and depth of pools, which several researchers have reported to be highly related to the production of larger trout (Tarzwell, 1937; Shetter, Clark, and Hazzard, 1949; Hunt, 1971). Because all these factors and others can influence the abundance and distribution of trout in streams, it is not difficult to understand why abundance of overhead bank cover fails to account for all the observed variability in trout density in the Pigeon and Salmon Trout Rivers. In view of all other influences, it is remark- able that bank cover abundance alone can account for a major portion (SS-88%) of the variability in numbers and biomass of trout 3150mm in both separate and combined analyses of the two streams. 77 Now that the major significance of overhead bank cover to brook and brown trout has been demonstrated, the mechanisms by which this type of fish cover might act upon stream carrying capacity should be examined. First and foremost, overhead bank cover provides concealment and protection from predators (Giger, 1973; White, 1973). Saunders and Smith (1962) felt that availability of suitable hiding places was a dominant factor in delimiting the carrying capacity of a Prince Edward Island stream for older trout. They found a signifi- cant increase in the percent survival of fingerling brook trout and a marked increase in the number of Age-II trout following altera- tions which increased the number of hiding places. The observations of Gibson and Keenleyside (1966), McCrimmon and Kwain (1966), and Butler and Hawthorne (1968) indicated that the strong preferences of trout for overhanging cover was due to their photonegative responses which caused them to seek security in shaded areas. Bank cover can also provide shelter from swift currents. Hartman (1963) reasoned that summer association of brown trout with cover, shade, and regions of moderate water velocity may serve primarily for the (development of an efficient feeding strategy. Brook trout are also known to rest in covered or shaded areas from which they may emerge to snatch food, but where they can also quickly retreat into dark corners when alarmed (Gibson and Keenleyside, 1966). According to Baldes (1968 in Giger, 1973), the number and diversity of microhabitats is directly proportional to the potential carrying capacity of the stream environment. Since salmonid 78 microhabitats are almost always associated with cover (Kalleberg, 1958; Hartman, 1963, 1965; Nickham, 1967 in Stalnaker and Arnette, 1976; Baldes and Vincent, 1969), it is certainly possible that the amount of overhead bank cover may limit to some degree the number of suitable microhabitats, and thus greatly influence the carrying capacity of the stream for brook and brown trout. In other words, because of territoriality and competition for a limited number of favorable positions, the amount of territory with sufficient cover may regulate maximum densities of salmonid populations in streams (Allen, 1969; Lewis, 1969). Furthermore, log jams, overhanging vegetation, and other items of instream overhead bank cover often provide greater visual isolation, which reduces territory size and may allow the density of fish to increase (Kalleberg, 1958; Chapman, 1966; Allen, 1969). I Finally, Hunt (1969) proposed another means by which stream- bank cover might affect trout abundance. He felt that increases in a trout population after habitat improvement which involved installa- tion of permanent bank cover were largely the result of increased overwinter survival. Maciolek and Needham (1952) noted that severe winter conditions cause extremely high mortalities of trout, and other researchers have observed that salmonids in general display a strong preference in winter for stream banks with overhanging or submerged cover (Needham and Jones, 1959; Bustard and Narver, 1975b). This close association with cover is believed to be a mechanism for gaining shelter from currents and remaining in suitable 79 reaches of stream over winter (Hartman, 1963). Bustard and Narver (1975a) found evidence that low overwinter survival of coho salmon and rainbow trout in one stream reflected low availability of suit- able winter cover along the stream bank. Clearly, winter is a critical time for stream salmonids--a period when harsh environmental conditions and severe physiological stresses can greatly reduce fish p0pu1ations. The availability of stable overhead bank cover may indeed become crucial to trout survival during this time. SUMMARY AND CONCLUSIONS Possible relationships between instream overhead bank cover and abundance of brook and/or brown trout were investigated on 2.4km of the Pigeon River and on 4.5km of the Salmon Trout River in ’ Michigan's Lower and Upper Peninsulas, respectively. These streams differed in chemical content, hydrologic charactersitics, and fish species composition. Numerous study stations, each about 100m long, were estab- lished on both rivers. Mark-and-recapture electrofishing was used to determine trout populations in July and October of 1976. During August and September, when both rivers were at summer low flow, measurements were made of discharge and habitat character- istics. Maps showing channel width, thalweg location, thalweg depth, and bank cover were drawn for 12 Pigeon River stations and 18 Salmon Trout River stations. The length of overhead bank cover as defined by Wesche (1973, 1976) was measured in these stations. Bank cover was classified into three types: undercut banks, over- hanging vegetation, and log cover. In nearly all Pigeon River stations, brown trout were less numerous than brook trout, yet comprised the majority of the biomass. Total trout biomass/station in July ranged from 1.26-8.73kg. During October, trout were distributed more erratically, apparently for spawning. While number and biomass of legal-sized brook trout 8O 81 declined from July to October, total trout biomass/station in October ranged from 1.90-12.12kg. Log jams and fallen trees comprised the majority of overhead bank cover in the Pigeon River. In many stations, however, undercut banks and overhanging vegetation provided substantial bank cover. In Pigeon River stations, total length of overhead bank cover accounted for 44% of the variation in July number of brook trout 3150mm, 66% of the variation in July number of brown trout 3150111111, and 88% of the variation in July number of all trout 315011111. Length of bank cover also explained 38% of the variation in July biomass of brook trout 3150mm. A significant relationship between bank cover and July biomass of brown trout 3150mm could only be obtained by excluding fish 3400mm in length from the analysis, thereby yielding an r2 of 0.54. The length of overhead bank cover/ station explained 72% of the variation in July biomass of all trout 150-399mm/station. For October brook trout 3150mm in the Pigeon River, bank cover abundance accounted for 68% of the variation in numerical density and 65% of the variation in biomass. Interferences result- ing from upstream spawning migration are believed to explain the lack of significant correlation between length of bank cover and number or biomass of brown trout 3150mm in October. In the Salmon Trout River, brook trout populations were very sparse below Lower Falls. The greatest July brook trout biomass/station in this area was l.lOkg. A few small rainbow trout and several very 82 large lake-run rainbows were also present. Above Lower Falls, brook trout were more abundant, but there were few fish exceeding 250mm in length. Total biomass/station in July ranged from 0.22-3.44kg. In October, the presence of spawning immigrants from Lake Superior increased the number and biomass of brook trout below Lower Falls. Brook trout biomass/station ranged from O—2.88kg. Several large coho salmon, two large brown trout, and numerous fingerling cohos and browns were also present. Above Lower Falls, October brook trout biomass/station ranged from 0.27-3.00kg. Log cover was the predominant type of overhead bank cover in the Salmon Trout River, but was much less abundant than in the Pigeon River. Undercut banks and overhanging vegetation were also present in much of the study area. Correlations between brook trout abundance and length of bank cover in the Salmon Trout River were generally stronger when stations below Lower Falls were omitted from the analysis. For July populations of brook trout 3150mm, total length of overhead bank cover accounted for 72-83% of variation in number and 62-69% of variation in biomass in stations above Lower Falls. For October brook trout 3150mm, bank cover abundance explained 78% of the varia- tion in number and 88% of the variation in biomass, again in stations above Lower Falls. Several strong correlations between amount of bank cover and trout abundance were demonstrated in combined analyses of Pigeon River and Salmon Trout River data. Length of bank cover accounted 83 for 76% of the variation in July number of all trout 3150mm, 81% of the variation in July biomass of all trout 150-399mm, and 80% of the variation in October number of all trout 3150mm in the Pigeon River and Salmon Trout River above Lower Falls. These results indicate that overhead bank cover provides an important element in the habitat of brook and brown trout larger than 150mm. The strong correlations between abundances of cover and trout suggest that bank cover is the major factor limiting trout populations in both streams, despite differences in chemical, bio- logical, and hydrological characteristics. LITERATURE CITED 84 LITERATURE CITED Adams, L. 1951. Confidence limits for the Petersen or Lincoln index used in animal population studies. J. Wildl. Mgt. 15: 13-19. Allen, K. R. 1962. The natural regulation of population in the Salmonidae. New Zeal. Sci. Rev. 20: 58-62. Allen, K. R. 1969. Limitations on production in salmonid populations in streams. Pages 3-18 jn_T. G. Northcote, ed. Symposium on Salmon and Trout in Streams. H. R. MacMillan Lectures in Fisheries, Univ. Brit. Col., Vancouver. Baldes, R. J. 1968. Microhabitat velocity occupied by trout. M.S. Thesis, 0010. State Univ., Fort Collins. 33 pp. Cited jn_ R. D. Giger. 1973. Streamflow requirements of salmonids. Oregon Wildl. Comm., Fed. Aid Project AFS-62-l Job Final Rep. 117 pp. Baldes, R. J. and R. E. Vincent. 1969. Physical parameters of microhabitats occupied by brown trout in an experimental flume. Trans. Am. Fish. Soc. 98: 230-238. Banks, R. L., J. W. Mullan, R. W. Wiley and D. J. Dufek. 1974. The Fontenelle Green River trout fisheries: considerations in its enhancement and perpetuation, including test flow studies of 1973. U. S. Fish Wild. Serv., Salt Lake City. 74 pp. Cited jg_C. B. Stalnaker and J. L. Arnette. l976. Methodologies for determining instream flows for fish and other aquatic life. Pages 89-138 ifl_C. B. Stalnaker and J. L. Arnette, eds. Methodologies for the Determination of Stream Resource Flow Requirements: an Assessment. U. S. Fish Wildl. Serv., Office of Biol. Services, Western Water Allocation. Bartschi, D. K. 1976. A habitat-discharge method of determining instream flows for aquatic habitat. Pages 285-294 jg_J. F. Osborn and C. H. Allman, eds. Proceedings of the Symposium and Specialty Conference on Instream Flow Needs: Volume II. Am. Fish. Soc., Bethesda, Md. Benson, N. G. 1953a. The relationship among certain ecological conditions and trout populations in the Pigeon River. Ph.D. Dissertation, Univ. Mich., Ann Arbor. 85 Benson, 86 N. G. 1953b. The importance of ground water to trout popu- lations in the Pigeon River, Michigan. Trans. N. Am. Wildl. Conf. 18: 269-281. Boreman, J. 1974. Effects of stream improvement on juvenile rainbow Boussu, trout in Cayuca Inlet, New York. Trans. Am. Fish. Soc. 103: 637-641. M. F. 1954. Relationship between trout populations and cover on a small stream. J. Wildl. Mgt. 18: 229-239. Bustard, D. R. and D. W. Narver. 1975a. Aspects of the winter ecol- ogy of juvenile coho salmon (Oncorhynchus kisutch) and steel- head trout (Salmo gairdneri). J. Fish. Res. Bd. Can. 32: 667-680. Bustard, D. R. and D. W. Narver. 1975b. Preferences of juvenile coho salmon (Oncorhynchus kisutch) and cutthroat trout (Salmo clarki) relative to simulated alterations of winter habitat. . J. Fish. Res. Bd. Can. 32: 681-687. Butler, R. L. and V. M. Hawthorne. 1968. The reactions of dominant trout to changes in overhead artificial cover. Trans. Am. Fish. Soc. 97: 37-41. Chapman, 0. W. 1966. Food and space as regulators of salmonid popu- lations in streams. Am. Nat. 100: 345-357. Chapman, 0. W. and T. C. Bjornn. 1969. Distribution of salmonids in streams with Special reference to food and feeding. Pages 153-176 jg_T. G. Northcote, ed. Symposium on Salmon and Trout in Streams. H. R. MacMillan Lectures in Fisheries, Univ. Brit. Col., Vancouver. Compton, R. R. 1962. Manual of field geology. John Wiley and Cooper, Cooper, Cooper, Sons, New York. 378 pp. C. O. and T. A. Wesche. l976. Stream channel modification to enhance trout habitat under low flow conditions. Water Resources Ser. No. 58, Univ. Wyom. Water Resources Res. Inst., Laramie. 107 pp. E. L. 1952. Rate of exploitation of wild eastern brook trout and brown trout populations in the Pigeon River, Otsego County, Michigan. Trans. Am. Fish. Soc. 81: 224-234. E. L. 1953. Growth of brook trout (Salvelinus fontinalis) and brown trout (Salmon trutta) in the Pigeon River, Otsego County, Michigan. Pap. Mich. Acad. Sci., Arts, and Lett. 38: 151-162. 87 Cooper, G. P. and K. F. Lagler. 1956. Appraisal of methods of fish population study: Part III, the measurement of fish popula- tion size. Trans. N. Am. Wildl. Conf. 21: 281-297. Cooper, J. L. 1976. United States Forest Service stream survey procedure: Northern Region. Pages 300-313 jg_J. F. Osborn and C. H. Allman, eds. Proceedings of the Symposium and Specialty Conference on Instream Flow Needs: Volume 11. Am. Fish. Soc., Bethesda, Md. Davis, W. S. 1964. Graphic representation of confidence intervals for Petersen population estimates. Trans. Am. Fish. Soc. 93: 227-232. Elser, A. A. 1968. Fish populations of a trout stream in relation to major habitat zones and channel alterations. Trans. Am. Fish. Soc. 97: 389-397. Elwood, J. W. and T. F. Waters. 1969. Effects of floods on food consumption and production rates of a stream brook trout population. Trans. Am. Fish. Soc. 98: 253-262. Everest, F. H. and D. W. Chapman. 1972. Habitat selection and spatial interaction by juvenile chinook salmon and rainbow trout in two Idaho streams. J. Fish. Res. Bd. Can. 29: 91-100. Gard, R. and D. W. Seegrist. 1972. Abundance and harvest of trout in Sagehen Creek, California. Trans. Am. Fish. Soc. 101: 463-477. Gibson, R. J. and M. H. A. Keenleyside. 1966. Responses to light of young Atlantic salmon (Salmo salar) and brook trout (Salvelinus fontinalis). J. Fish. Res. Bd. Can. 23: 1007- 1024. Giger, R. D. 1973. Streamflow requirements of salmonids. Oregon Wildl. Comm., Fed. Aid Project AFS-62-1 Job Final Rep. 117 pp. Griffith, J. S., Jr. 1972. Comparative behavior and habitat utili- zation of brook trout (Salvelinus fontinalis) and cutthroat trout (Salmo clarki) in small streams in northern Idaho. J. Fish. Res. Bd. Can. 29: 265-273. Grubbs, F. E. and G. Beck. 1972. Extension of sample sizes and percentage points for significance tests of outlying observa- tions. Technometrics 14: 847-854. 88 Hale, J. G. 1969. An evaluation of trout stream habitat improvement in a north shore tributary of Lake Superior. Minn. Fish. Invest. 5: 37-50. Cited jn_R. J. White. 1973. Stream channel suitability for coldwater fish. Proc. 28th Ann. Mtng. Soil Conserv. Soc. Am.: 61-79. Hartman, G. F. 1963. Observations on behavior of juvenile brown trout in a stream aquarium during winter and spring. J. Fish. Res. Bd. Can. 20: 769-787. Hartman, G. F. 1965. The role of behavior in the ecology and inter- action of underyearling coho salmon (Oncorhynchus kisutch) and steelhead trout (Salmo gairdneri). J. Fish. Res. Bd. Can. 22: 1035-1081. Hendrickson, G. E. and C. J. Doonan. 1970. Reconnaissance of the Pigeon River, a cold-water river in the northcentral part of Michigan's Southern Peninsula. U. S. G. S. Hydrologic Investigations, Atlas HA-333. Hendrickson, G. E., R. L. Knutilla and C. J. Doonan. 1973a. Hydrol- ogy and recreation of selected cold-water rivers of the St. Lawrence River basin in Michigan, New York, and Wisconsin. U. S. G. S. Water Resources Invest. 8-73. 73 pp. Hendrickson, G. E., R. L. Knutilla and C. J. Doonan. 1973b. Hydrol- ogy and recreation on the cold-water rivers of Michigan's Upper Peninsula. U. S. G. S. Water Info. Ser. Rep. 4. 39 pp. Hunt, R. L. 1969. Effects of habitat alteration on production, standing corps and yield of brook trout in Lawrence Creek, Wisconsin. Pages 281-312 in T. G. Northcote, ed. Symposium on Salmon and Trout in Streams. H. R. MacMillan Lectures in Fisheries, Univ. Brit. Col., Vancouver. Hunt, R. L. 1971. Responses of a brook trout p0pu1ation to habitat development in Lawrence Creek. Tech. Bul. No. 48, Wis. Dept. Nat. Resources, Madison. 35 pp. Hunt, R. L. 1976. A long-term evaluation of trout habitat develop- ment and its relation to improving management-related research. Trans. Am. Fish. Soc. 105: 361-364. Jenkins, T. M., Jr. 1969. Social structure, position choice and micro-distribution of two trout species (Salmo trutta and Salmo gairdneri) resident in mountain streams. Anim. Behav. Monogr. 2: 55-123. Kalleberg, H. 1958. Observations in a small stream tank of terri- toriality and competition in juvenile salmon and trout (Salmo salar L. and Salmo trutta L,). Rep. Inst. Freshwater Res. DrottnTngholm, Sweden 39: 55-98. 89 Kraft, M. E. 1972. Effects of controlled flow reduction on a trout stream. J. Fish. Res. Bd. Can. 29: 1045-1411. Lewis, S. L. 1969. Physical factors influencing fish populations in pools of a trout stream. Trans. Am. Fish. Soc. 98: 14-19. Maciolek, J. A. and P. R. Needham. 1952. Ecological effects of win- ter conditions on trout and trout foods in Convict Creek, California. Trans. Am. Fish. Soc. 81: 202-217. McCrimmon, H. R. and W. H. Kwain. 1966. Use of overhead cover by rainbow trout exposed to a series of light intensities. J. Fish. Res. Bd. Can. 23: 983-990. McFadden, J. T. 1961. A population study of the wild brook trout, Salvelinus fontinalis. Wildl. Monog. No. 7. 73 pp. McFadden, J. T. and E. L. Cooper. 1962. An ecological comparison of six populations of brown trout (Salmo trutta). Trans. Am. Fish. Soc. 91: 53-62. McFadden, J. T. and E. L. Cooper. 1964. Population dynamics of brown trout in different environments. Phys. 2001. 37: 355- 363. Mundie, J. H. 1969. Ecological implications of the diet of juvenile coho in streams. Pages 135-152 jg_T. G. Northcote, ed. Symposium on Salmon and Trout in Streams. H. R. MacMillan Lectures in Fisheries, Univ. Brit. Col., Vancouver. Needham, P. R. and A. C. Jones. 1959. Flow, temperature, solar radiation, and ice in relation to activities of fishes in Sagehen Creek, California. Ecology 40: 465-474. Nickelson, T. 1976. Development of methodologies for evaluating instream flow needs for salmonid rearing. Pages 588-596 35 J. F. Osborn and C. H. Allman, eds. Proceedings of the Symposium and Specialty Conference on Instream Flow Needs: Volume II. Am. Fish. Soc., Bethesda, Md. Nyman, O. L. 1970. Ecological interaction of brown trout, Salmo trutta L,, and brook trout, Salvelinus fontinalis (Mitchill), in a stream. Can. Field-Nat. 84: 343-350. O'Connor, J. F. and G. Power. 1976. Production by brook trout in four streams in the Matamek watershed, Quebec. J. Fish. Res. Bd. Can. 33: 6-18. Odum, E. P. 1971. Fundamentals of ecology. third edition. W. B. Saunders Co., Philadelphia. 574 pp. 9O Onodera, K. 1962. Carrying capacity in a trout stream. Bul. Fresh- water Fish. Res. Lab. Tokyo 12(1): 1-41. Platts, W. S. 1976. Validity of methodologies to document stream environments for evaluating fishery conditions. Pages 267- 284 jg_J. F. Osborn and C. H. Allman, eds. Proceedings of the Symposium and Specialty Conference on Instream Flow Needs: Volume 11. Am. Fish. Soc., Bethesda, Md. Regier, H. A. and D. S. Robson. 1967. Estimating population number and mortality rates. Pages 31-66 ig_S. D. Gerking, ed. The Biological Basis of Freshwater Fish Production. John Wiley and Sons, New York. Ricker, W. E. 1975. Computation and interpretation of biological statistics of fish populations. Bul. Fish. Res. Bd. Can. 191. 382 pp. Saunders, J. W. and M. W. Smith. 1955. Standing crops of trout in a small Prince Edward Island stream. Can. Fish. Cult. 17: 32-39. Saunders, J. W. and M. W. Smith. 1962. Physical alteration of stream habitat to improve brook trout production. Trans. Am. Fish. Soc. 91: 185-188. Schuck, H. A. 1945. Survival, population density, and movement of the wild brown trout in Crystal Creek. Trans. Am. Fish. Soc. 73: 209-230. Seegrist, D. W. and R. Gard. 1972. Effects of floods on trout in Sagehen Creek, California. Trans. Am. Fish. Soc. 101: 478- 482. Shetter, D. S., O. H. Clark and A. S. Hazzard. 1949. The effects of deflectors in a section of a Michigan trout stream. Trans. Am. Fish. Soc. 76: 248-278. Stalnaker, C. B. and J. L. Arnette. l976. Methodologies for deter- mining instream flows for fish and other aquatic life. Pages 89-138 jfl_C. B. Stalnaker and J. L. Arnette, eds. Methodolo- gies for the Determination of Stream Resource Flow Require- ments: an Assessment. U. S. Fish Wild. Serv., Office of Biol. Services, Western Water Allocation. Stewart, P. A. 1970. Physical factors influencing trout density in a small stream. Ph.D. Dissertation, Colo. State Univ., Fort Collins. 78 pp. Cited ig_R. L. Hunt. 1971. Responses of a brook trout population to habitat development in Lawrence Creek. Tech. Bul. No. 48, Wis. Dept. Nat. Resources, Madison. 35 pp. 91 Tarzwell, C. M. 1937. Experimental evidence on the value of trout stream improvement in Michigan. Trans. Am. Fish. Soc. 66: 177-187. Tarzwell, C. M. 1938. An evaluation of the methods and results of stream improvement in the Southwest. Trans. N. Am. Wildl. Conf. 3: 339-364. Tennant, D. L. 1976. Instream flow regimes for fish, wildlife, recreation and related environmental resources. Pages 359- 373 jg_J. F. Osborn and C. H. Allman, eds. Proceedings of the Symposium and Specialty Conference on Instream Flow Needs: Volume II. Am. Fish. Soc., Bethesda, Md. Waters, 8. F. 1976. A methodology for evaluating the effects of different streamflows on salmonid habitat. Pages 254-266 in J. F. Osborn and C. H. Allman, eds. Proceedings of the Symposium and Specialty Conference on Instream Flow Needs: Volume II. Am. Fish. Soc., Bethesda, Md. Wesche, T. A. 1973. Parametric determination of minimum stream flow for trout. Water Resources Ser. No. 37, Univ. Wyom. Water Resources Res. Inst., Laramie. 102 pp. Wesche, T. A. 1976. Development and application of a trout cover rating system for IFN determinations. Pages 224-234 in J. F. Osborn and C. H. Allman, eds. Proceedings of the Symposium and Specialty Conference on Instream Flow Needs: Volume II. Am. Fish. Soc., Bethesda, Md. White, R. J. 1973. Stream channel suitability for coldwater fish. Proc. 28th Ann. Mtng. Soil Conserv. Soc. Am.: 61-79. White, R. J. 1975. Trout population responses to streamflow fluc- tuations and habitat management in Big Roche-a-Cri Creek, Wisconsin. Verh. Internat. Verein. Limnol. 19: 2469-2477. ‘White, R. J., E. A. Hansen and G. R. Alexander. 1976. Relationship of trout abundance to stream flow in Midwestern streams. Pages 597-615 jg_J. F. Osborn and C. H. Allman, eds. Pro- ceedings of the Symposium and Specialty Conference on Instream Flow Needs: Volume II. Am. Fish. Soc., Bethesda, Md. Wickham, G. M. 1967. Physical microhabitat of trout. M.S. Thesis, Colo. State Univ., Fort Collins. 42 pp. Cited jg_C. B. Stalnaker and J. L. Arnette. l976. Methodologies for determining instream flows for fish and other aquatic life. Pages 89-138 jn_C. B. Stalnaker and J. L. Arnette, eds. Methodologies for the Determination of Stream Resource Flow Requirements: an Assessment. U. S. Fish Wildl. Serv., Office of Biol. Services, Western Water Allocation. APPENDICES 92 APPENDIX A RAINBOW TROUT, BROWN TROUT, AND COHO SALMON IN THE SALMON TROUT RIVER IN 1976 93 TABLE Al.--Estimates of rainbow trout populations in the Salmon Trout River in 1976 {2:316 lOO-l49mm 150-199mm 200-249mm .3250mm (length) Month Jun Oct Jun Oct Jun Oct Jun Sta_ 1 1 2 l 3 1 l 4 l 1 5 1 1(587) 6 l D 7 8 l 3 l 3(270,405,618) 9 l 10 1 ll 12 3 l3 1 14 15 l 1 16,17 1 5 2 l 18 19 3 1 Total 6* 10 12 3* O 3* 4* *Actual number of fish captured 94 95 TABLE A2.--Estimates of brown trout and coho salmon populations in the Salmon Trout River in October, 1976 Brown Trout Coho Salmon--Total Length Sta (Total Length) too-149mm 150-199mm 3200mm (length) 1 1(103) ‘5 2 29 3 24 4 73 5 1(540) 90 1(605) 6 24 7 69 1 8 24 9, 102 1(53o) ,0 24 11 106 1(577) 12 8 ,3 33 14 1(645) l5 ,5 53 16 53 2(350,570) 17' 8‘ 18 29 19 16 Total 3* ' 870 1* 5* *Actual number of fish captured APPENDIX B HABITAT MAPS FOR TWELVE STATIONS ON THE PIGEON RIVER, MICHIGAN 96 9995.119 ‘0 s cation marker ------- thalweg (depth-cm) JJ-LLLL undercut bank M overhanging vegetation /fl§$. log cover S CALE (I i 10“"- ZOmeters Figure 81. Pigeon River Station 3. 97 98 Figure 82. 5129202 —0 station marker ....... thalweg (depth-cm) JJ-LLU’ undercut bank 7E § overhanging j \ vegetation /§§$. log cover 5 CALE (I 3 IO“ ZOmeters Pigeon River Station 4. 99 5599.112 .0 station marker ------- thalweg (depth-cm) ,U-LLLL undercut bank N overhanging vegetation /${#. log cover \‘ - SCALE (557“ 5 LO meeters \ ~ \ ( ‘\ \ o "‘ \\ 9:57 \ \ x, \ \ 1, \(461 \ \ \ \ \x \(70) x \x '3’- \ ,o,’4 \ / 1\ - ‘NESE:) 1 100‘.‘ x~ ‘s {0 “ t N 0 (43T‘\‘ 7 0 (52) Figure 83. Pigeon River Station 5. 100 \ ~99 \ \ \ \ .\ \ ‘\ \ i431 - \ , \ \ \ \ \ \ \ \6& ~.'\\ ‘\\\ Figure 84. 9.25.599 —0 station marker ------- thalweg (depth-cm) .U-U-U’ undercut bank m overhanging vegetation ¢¢‘% log cover SCALE 0 5 IO ZOmeters BEE-1 HI-U—————i ‘-~ .“ Pigeon River Station 6. lOl LEGEND -(::> station marker ------- thalweg (depth-cm) llllll undercut bank N overhanging vegetation /#r log cover Figure BS. Pigeon River Station 9. 102 Figure 86. $9.859. —-0 station marker ------- thalweg (depth-cm) M undercut bank N overhanging vegetation /fi%. log cover 5 CALE (I z 10__ ZOmeters Pigeon River Station 10. 103 9.5.9.912 N ---0 station marker ....... thalweg (depth-cm) .U-LUJ’ undercut bank 9 overhanging 6|) (kt §§: vegetation /1 1 />#€- log cover 11071 /. SCALE / ._____.... , Q 5 1o zometers '(‘651 ‘1'. \ Figure 87. Pigeon River Station 13. 104 56.95119. —0 station marker I ------- thalweg (depth-cm) J-LLLLL undercut bank - 78 § overhanging )‘Q vegetation \\ \ /%&. log cover \ (w\\\ ‘\ 1 . \\\ g) 0071‘\\==, . X .\ SCALE \ (i 2 IO ZOmeters (701‘, 1 \ l 1 1 1 BEN 1 1 1 l E: 1 1 N g (6‘); ’ 7 I I I . ’ I x / I 1’ 1”, / - "57' 1921 7:" I I I 80 Figure B8. Pigeon River Station 14. 105 m —--0 station marker ....... thalweg (depth-cm) Hllll undercut bank M overhanging vegetation /% log cover S CALE (i 2 IO ZOmeters 2 Figure B9. Pigeon River Station 19. 106 m N —-O station marker ....... thalweg (depth-cm) @ II I | I I undercut bank /’(92) overhanging ,mt :15: gsvegetation I , . :I \ /% log cover \ . l.. (E 2 IO ZOmeters u"- ;7 . f/Z \ 9'. \ \ \ \ \\ \/° ‘ \x. VN\GR¢~ ‘==1 (1073‘~\ "2.. I, no] (fl§\\ \ \\\ \ \\ b F (64‘\ ’ {0‘0 ‘s‘ \ ‘s - \ \\\ \ \ \\ a I .51? Figure 310. Pigeon River Station 20. 107 £92132 —-0 station marker - ----'-- thalweg (depth-cm) JJ-LLLL undercut bank m overhanging vegetation / 35%. log cover 8 CALE (i ; IO ‘ ZOmeters Figure 811. Pigeon River Station 21. 108 M —0 station marker ------- thalweg (depth—cm) W undercut bank m overhanging vegetation ffi% log cover SCALE Worm ters 2 Figure BIZ. Pigeon River Station 22. APPENDIX C HABITAT MAPS FOR EIGHTEEN STATIONS ON THE SALMON TROUT RIVER, MICHIGAN 109 LEGEND --0 station marker ------- thalweg (depth-cm) ,LLLLLL undercut bank Moverhanging vegetation /% log cover 5 CALE (i g '0' ZOmeters MOXJ g '-- H O ~-- -”’ Figure C1. Saimon Trout River Station 3. 110 111 LEGEND N ---0 station marker ------- thalweg (depth—cm) I | l I I |,. undercut bank m overhanging o vegetation />}§é§- log cover M o 5 IO ZOmeters 51:14 RH I—-—-—q‘ Figure C2. Salmon Trout River Station 5. 112 25.122319. N ---0 station marker ....... thalweg (depth-cm) JJ-LUJ’ undercut bank W overhanging / vegetation 'flfifié log cover SCALE 0 5 no zpmeters ' midi—{H Figure C3. SaTmon Trout River Station 6. 113 229m —0 station marker ...... - thalweg (d epth-cm) W undercut bank W overhanging vegetation I/fiffi- log cover SCALE Q 5 L0 zpmeters WWI-*4 - ---- --' 3 \ ‘—.. Figure C4. SaTmon Trout River Station 7. 114 LEGEND --0 station marker ------- thalweg (depth-cm) JJ-l-UJ' undercut bank ’I overhanging . \ 7W S; : vegetation (8,5; ’23? log cover / SCALE A’s?) % 3 IO ZOme ters , \ I I (64) I I 1' / I I ll (92» I I I I . I I I [(46) I S I /’ Q Q” /‘ (37) I I I I I, ((2!) N K 1 \ \ \ \ 1 1 (271‘. Figure C5. Salmon Trout River Station 8. 115 LEGEND _ —0 station marker N ....... thalweg (depth-cm) W undercut bank m overhanging vegetation I9" 21% log cover SCALE 2 3 IO ZOme ters Figure C6. SaTmon Trout River Station 10. 116 21.522122 —0 station marker - ------ thalweg (depth-cm) m undercut bank N overhanging vegetation /fl%. log cover gems. (E s :9 meeters Figure C7. Salmon Trout River Station 13. ll7 LEGEND -—O station marker ....... thalweg (depth-cm) ||Il||v undercut bank m overhanging vegetation ’% log cover SCALE L .35 I9 meeters mill-1H Figure C8. (34) l I 1. 1 l I I (37» I I I I I (432/, ‘ I I I I ’I (58)! I” I” I’ (37) I I I I' (46 I I I I” (58),“ s a’ \9 I/ Q ’I 640/ I Salmon Trout River Station l4. 118 LEGEND —0 station marker ------- thalweg (depth-cm) JJJ-LU' undercut bank m overhanging vegetation yfiyé log cover SCALE 2 3 IO ZOmeters Figure C9. Salmon Trout River Station 36. 119 LEGEND —0 station marker ...... - thalweg (depth-cm) W undercut bank m overhanging vegetation é"%~ log cover SCALE 2 3 IO ZOmeters Figure ClO. Salmon Trout River Station 38. 120 ‘CID (m»' 'l -- Figure Cll. ' LEGEND —O station marker ....... thalweg (depth-cm) l||I|| undercut bank M overhanging vegetation fffi. log cover 8 CALE 5 '0 20 me te rs Salmon Trout River Station 47. 12] LEGEND N -—O station marker ....... thalweg (depth-cm) |lll||— undercut bank m overhanging vegetation yfifié log cover SCALE 0 5 I9 29meters 3141-1 H H l-—-——:i Figure C12. Salmon Trout River Station 48. 122 LEG END —-O station marker ------- thalweg (depth-cu)H J—U-LLL undercut bank M overhanging vegetation /% log cover SCALE 3 i '0' 20meters Figure C13. Salmon Trout River Station 50. 123 item. —-O station marker N .. - - - - ‘~- thalwe g (depth-cm) JJ-LLU' undercut bank m overhanging vegetation I’fifi. log cover S CALE (i ; IO ZOmeters Figure Cl4. Salmon Trout River Station 51. 124 Figure C15. LEGEND N --0 station marker ------- thalwe g (depth-cm) llllll undercut bank M overhanging vegetation /% log cover SCALE (1 2 I0“ ZOmeters Salmon Trout River Station 52. 125 m -0 station marker ------- thalweg (depth-cm) ’JJ-LLlJr undercut bank M overhanging vegetation /%. log cover 8 CALE (Z ; IO ZOmeters Figure C16. Salmon Trout River Station 53. 126 LEGEND -—0 station marker ....... thalwe g (depth-cm) J—U-J-U' undercut bank moverhanging .. @ vegetation / ”I yfifié log cover 1' I SCALE o a 19 zometers ;\\\\ HER HHI————-—1 ’1 1‘\ ‘ .. \ G2“ \ \ 1 I I \ 1851; \‘ \ ~ ’I , 5~ 62" ’ {7" ' (37) I I I I New 1 1 I I l 1' (49): Q? L,‘ ,l ...... -9192/ ’ Figure Cl7. Salmon Trout River Station 54. 127 49) \ ‘\ \ ‘ \ 3|) / \ \ H‘PK‘ a£r~. \\ 1 (88I1\ \ 3 ‘1 i3 1 an»: I I 1 I I I I / I 1 IN «‘84) ' ‘\ \(95) K “ ‘ ‘ s ‘ LEGEND —O station marker ....... thalweg (depth-cm) ||llllr undercut bank I m overhanging vegetation /% log cover SCALE Q 5 19 zpmeters ElingHEP---3 Figure Cl8. Salmon Trout River Station 57.