R I t H I‘ o m; ({‘1 ' 1.? ‘I I I s fin "In f \t- N?! ’ m v1 I 1;: I‘ S‘ b “51%;? If“ V If‘ v r w L" ‘ v 4 "I I ,s I. WWW"?! . . tummwmm“ 43.1an ,I III. ,I. ‘l ‘I l Wu! 3 ImLI .7 ‘Ir‘ 4.3 .. I I . I . I... . .».I, I...“ IIHIMW I. ., ,. {.mItIIIIww: wwm, wfiw. I "INHMW t. .me u. LI (’1‘? {3:}! , ' I "I I}! I I! I.“ I -. I IIIIIIIWWW I .4. In !I, \I. I I,» I IIIIIImIrI .II I I IIIIIIIFIJIIbuNmHIIIUVIfloflIKI! ! II IIJaI. B! IN. IIIIIISII n.) 4.! I. VWMWNMIMEMWAMRIIH IanmbIIIWIHIIImw IIIIW I. III. I). .II @939 I I III IIIIuHWMIIIIIImIIImquvmnIIII I, L NI, . , , :3!!! ”mtg.“ I‘ 1‘ _! .51“: l «IIIJIII $811. .. I fiIIQIIMWIMIIW morn”. .I..II I IE I. I. “I, , IIIIIIIII I IIIIIIIIIII. a? ..... r.. ,II «IMMHI IIIIAI. .IIII III \IJIIMuI . .thmmMWMWMmIIMIWmuI IIRIIIINIII IfiIWuIIfifiufiwmfluulIWdfigwmfi. Ina“... I III IIrtlqucc. IIIHIIPIiundILnIIIIIIIIXIzAh{.IIIIVIIIHRWII LXI «@AUDBIIIUIII III. .. . I . IIIIIIIIII . IIIIMIIIIIK ._ IIIIIIIIIIII IIIMWPIIIIIII . . I I I! IIIIAIII!!lle.nIII§1L!tIUUuR “Input”! II! I \IiII! .. .III I “mull: «WI! . my? H‘wdflwmmfl! I I “I I IIIKIF‘A II IIIIIN .I I. . I $3...” my IAI. I , I ) IIIII...I&.. . II IIWIWMIHIwMI—MH .II.III . . I.I I :Itw! km.“ I «#IIMME .l . .I. I IIILI: alvfit .I I I. . IrvIIIwiILIIImmLIIIINu. In! \! huffing In?” WWII a... m. . II . I I \hKInIIIII'l‘ 1!: If"; I II. Inn! . IKIIIImunmv I {fill III .IMWIIII! I IN tanIII :3th flu! “HIII III!!! I I: Kant! . ,I! I III. II III! Iuwlor. I+IIIIIIII IIQIWIIHIIIEIIIIWI! I. (III “N. I I l umurxnu. I tWIHIIdIIJIIIIIIIY ! WIxcganIIIFUIHMuII I I .!I I I .I I I II .I l I I. Iv .I'I‘rII. FIVMIIII.» I Jul. . I I I I I I WHIIHIIRILI III, L! I. . ..rI!!4!!JIIInNIrII!.r II I . Immmwwnf I §+nMImIIuI . lo, I I. I I . I IIIIulILIII . Ivlclvk II a? I . flmuIWIILIUIMIWL! I) IIImmauquIII. I HI .I. IIIrtmeLItI Iv...) IIIAIM 3 Iotwfif, JQIIUIIII Him! . ,I I. .hImIIIIIMII! .I .L I I tiny! ‘ v I HeuIII IIIIII ImhIVIHIIImh WWIII7IILI!II IIII L? I IWIIJIIManMIma .. r III II III I ! I I .c . I I I I IIIIIJIIIIIIJHHII. . I I I . I . 3"th .InndNuII .311“? I. 5!; I . . I . I I . I I I ”($3111 (3.! !. U 74d... “Irma!“ $.st I am?“ I. I. I I. . II , .I I. u. II I Ivfimwnhwdi gawk II Winn In. I I !.I I I I. .. I II . ,I I. I. I . uvfloflr, I! I IIIKIILIAIII 3! LerI JIuWIImIIIImn. I. . I I LI 0! I I .I 1' oiughu. . Ia! I I, I III. .I II I, ., , kvlhuviflflxIMmIII1fiNII1IannIiI .IIIxI .IIHnmeAIU fiNxflMfluWWLI I .IuInrwIImIIIIIIIIIII. ,. . IMIHIfiunflUIIIII InIIIwIINImIquIISIIIIIIIImS Arman» IIIWIIwInmanI. IL . I I... III! IWnImvhfivxumnlImemflthIiiw PIIIMIMW. I .«ufiMqu III"! mmwthlgtbnflyI If; I I sham: I “$1. vflnuhflynvmmgunvx .71. .I. Ale t: h... I .! IIIJI ! II\u.9I I e. at Ip.flul\ I! I Dla. II ! I.. I. I I . .flII v..I‘II!! ‘0 ’UWF III!- . I.-.w,IIII.II..II.I..II IIIIIIIIIIIIIIIIIIIWIIIIIIIIIIWIHRI mIIIIIIII ITIIMIIIIIIIIIIIII. «I %I.I I I HIE. I II . ,I 4113 Ivfimvwfimmflft. . .. . ,I..I. I IlhuI!!llIIIII.v .I I! IRHIHIHWIVI .HmIIIVIotu. I IN I I I [JIIIInLInuIlI !I_m~u!IlI!I. .IIIII .IIItIIII. nuPIlIchquII II . I M! I. IIIIanIIIIIlIIIIMWImnIMMInNMIIIMuMIIIthUJ 1 Ian.» inn..." n I I’SIfiuiflItfitIfuluumumy IHMrMde. I . Lkmvwnzdmtflw! I I .I _ .I II I . IIIJIIwJuAInIIDIIINVIIuJdeI§ I IIIIFNIIIIWIQW. I 14. IIHIRWIIIII IhI II 3“? IcfiIlJIhv l I M»..L :III III .1 A IQ. Ive ls! IIIMNIIII.IIn1 II . II I . .u I, IIEIIIIIIIIIIIII ISIII IIIIIIIII -IwaflIwunIIIiIm . III .. . III I I lirWWunvmflmfififlt (undamuxvhu I mHHIEWNII!nI:LIII!I“uXIRRmumuuMI I IIWIII #1.. .1 IIIIHIMIII} IIIIMmImI,‘ ,IHI. -! .1 I II I IV “hi 1 Cal I 1‘!“ I III» III.» 1.3; in“. IWWFIIIIII I I "I... I I III . III) U4 ,IAfiI. haul}! ”.1 I. I .!.. I ’15. ‘. ‘I.I.. low IN“? . I. 9‘! 17»? III!!! I...!.!!.. . I ”mm 4 flnnzIwnfl‘MnnwfileIIV 111“an I I .3 yo... RIIHwWIIHMUfiInflWdHIFUNIWQIwMHIuHIP II I I I I I "I l . I 1 .I I , IWIIIII i .I I. I II . .I... I thIWrIII‘IQIIJINW HJI . I . .IIIIMIIwIdIII In! IIIIIIIIIIIJIIIIII III I A!’ IIII III I IIII I!L.(I I . IIIHIQIII I II I . I I I I! In»! I III . I I III I .I I I! I! III!» II I. II I I I :ng In? I I. "I“: III} II!!! I . I II, I” . I I.1IMIII.WWWWI&II¥IW!nwmIAI I . I IIIIIIMIRI. IIJInmIImImquIIIfi I I I I I I. IIIU.I lflanunIMWuInll . . I“. III In! LII I. I!!! I nIIiI I II" I, II “0:. I . . I. . I I. II... I I II I I H .I . . .II I I fiImIIIIIIIHIIH I .I III-III"! Ifiwvlh\_tq!fu.mmn!.u|om.!|.. , I I . I. I .uvn. I... ,n I IIIIIIII; III I IIIIIIIIIIIIIIIIIIIIIIIIII . I . . .I .IIImmem I. . 1, Nu I. I III .I 3! . I, . I .I I .IIIIIIIHIJI III: I . I I I I I . I. Ann» I I II I I. I II I. I I .II I I I .5343 (3.51,! I .,II I I I .I . . .IIIIIIHIIII II: I! Is. Ilfilix|w Lil II. u .I . I . Iain a! I II». 135% mm fin I I III! I IIIIIIEqunnIrIIII I III. , .IIIIIIIIIIIIII . IInIImIImuII. Inflow. ImIIIIIIIIIIIfiIEI . I I. . I I IIIIIIIIbfluI IIHJEMm-IMHIU‘IOI! III”. . . III . . I III I. I I "thunk.“ . III..I. 11‘ IIBIIIMMMRWIHIWIE , I. I I . I II II. , . I . Idmuflwxnmv: I I , Rd I In“ I .I II...!...I!I1.u .IwIsmmquIoImvnIflIMnUmmquwmmhh I I II I. : I. .II I . .. . . IIumGInInuI.J.hI..1NdInu3IIIWm«mMuMWIwhflrrlxhhmn JIIIuwnMmeI—MIIih» . I I III!.,1IIIIIII I! If“... II. I , I d . I; 0 NW“. Jail“! I .I . KEIWI. . I I IINIIIII w IRIMI. . I .I ..I.. - -.ra.-'-.- ' ---.' ,. .-_ This is to certify that the thesis entitled Feeding Habits of Salmonids in Michigan Waters of Eastern Lake Michigan and Southern Lake Superior presented by Stuart Nelson Kogge has been accepted towards fulfillment of the requirements for M.S. degreein Fisheries & Wildlife m 17&gwmz/ Major professor Date 10/17/86 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date ,stamped below. MSU LlBRARlES SP ,, 1,. 3 135‘? . , a 9?. :31: FEEDING HABITS OF SALMONIDS IN MICHIGAN WATERS OF EASTERN LAKE MICHIGAN AND SOUTHERN LAKE SUPERIOR BY STUART NELSON KOGGE 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 1985 (/“, I .‘ ABSTRACT FEEDING HABITS OF SALMONIDS IN MICHIGAN WATERS OF EASTERN LAKE MICHIGAN AND SOUTHERN LAKE SUPERIOR BY STUART N. KOGGE Stomach contents from 4,390 salmonids from eastern Lake Michigan and 158 from southern Lake Superior were analyzed to determine feeding habits. In 1983, 974 (64.38%) salmonids from eastern Lake Michigan contained prey items while 1,256 (46.51%) contained prey in 1984. The percent of salmonids feeding in 1984 was significantly lower (p<.001) than in 1983. The decline in feeding fish is presumed to be indicative of a decline in forage availability. .Alewife (Alosa pseudoharengus (Wilson)) once dominated the diets of Michigan's sport fish (chinook salmon Oncorynchus tshawytscha (Walbaum); coho salmon, Oncorynchus kisutch (Walbaum); lake trout, Salvelinus namaycush Walbaum; rainbow trout, Salmo gairdneri Richardson; brown trout, Salmo trutta Linnaeus, and atlantic salmon, _S__a_l_mg gala; Linnaeus; (Wright,1968; Chiotti, 1973; McComish and Miller, 1976). Decline in alewife populations and opportunistic feeding behavior by salmonids led to the preponderance of smelt (Osmerus mordax (Mitchill)) and bloaters (Coregonus hoyi (Gi11)), the most abundant species, in 1984. -Differences in feeding habits and intensity occurred across regional boundaries of the lake and throughout the year with respect to various salmonid species. In Grand Traverse Bay lake trout predominated and fed extensively on smelt. Likewise, smelt was the predominant prey item of Lake Superior salmonids. They showed a larger diversity and paucity of prey items consumed reflecting the smaller forage base of Lake Superior, in comparison to Lake Michigan. ACKNOWLEDGEMENTS I would like to thank my major professor, Dr. Niles Kevern, for providing this opportunity and for guidance and advice throughout the study. Dr. Richard Snider provided valuable advice and instruction throughout my stay at Michigan State University and 19m deeply grateful for his time given in reviewing the manuscript. Dr. William Taylor administered help and guidance throughout my years at M.S.U. for which I'm very thankful. Special thanks are in order to Chuck Pistis, District Extension Sea Grant Agent, whose help and knowledge paved the way for the success of the project. I wish to thank John McKinney and Ron Kinnunen, District Extension Sea Grant Agents, Gerald Draheim, County Extension Director, Bob Koch, fisheries biologist-Indiana DNR, and Bill Carlson, Carlson Fisheries, for their extended help in both the field and the fishing community. I would also like to thank my comrade Mike Nurse whose help in the field and the lab were greatly appreciated. Sha Miao, Lisa Harris, Eric Alexander, David Tushman, Andy Loftus, Kim Decker, and Russ Hanshue for their help in the field, lab, and compilation of the data. Most importantly, I would like to thank all the charterboat captains, sportsmanls clubs (idL Steelheaders), and anglers for their support and help, without whom this project would never have left the dock. Last of all, I would like to thank Ron Kimball, Bill Martin, Walt Olmstead, Ken Whitney, Bolhouse and sons, and the "guys" in Ludington's port marina .(who always knew what I was there for) for their knowledge and cooperation. Financial support for this study was provided by the following: the Michigan Sea Grant College Program (grant number R/GLF/18, funds from the Office of Sea Grant, National Oceanic and Atmospheric Administration, ELS. Dept. of Commerce and the State of Michigan); the Michigan Agricultural Experiment Station; the Sport Fishery Research Foundation; and the Coastal Management Program, MDNR. TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . LIST OF FIGURES . . . . . . . . INTRODUCTION . . . . . . . . . . DESCRIPTION OF THE STUDY AREA . METHODS . . . . . . . . . . . . 1983 SAMPLING APPROACH 1984 SAMPLING SCHEME . . . . . SAMPLE ANALYSIS . . . . . . . STATISTICS . . . . . . . . . . RESULTS AND DISCUSSION . . . . . LAKE MICHIGAN GENERAL FEEDING HABITS, 1983 . FEEDING HABITS BY REGION AND FISH SIZE AND PREY SELECTION GENERAL FEEDING HABITS, 1984 . FEEDING HABITS BY REGION AND FISH SIZE AND PREY SELECTION SEASON SEASON FEEDING HABITS AND WATER DEPTH . . . GRAND TRAVERSE BAY GENERAL FEEDING HABITS, 1983-84 LAKE SUPERIOR GENERAL FEEDING HABITS, 1983-84 SUMMARY AND CONCLUSIONS . . . . LITERATURE CITED . . . . . . . . APPENDICES A-E . . . . . . . . . iv 10 11 11 12 13 17 17 24 34 55 64 70 90 99 101 104 108 112 10. 11. LIST OF TABLES Page Seasonal delineation of salmonid diet research . 12 Percent (number feeding/sample size) of feeding salmonid species caught in eastern Lake Michigan, 1983. .. . .. . .. . 17 Frequency of occurrence of prey items in 1983 salmonid diets (percent of diet in parentheses) . . . . . . . . . . . . . 22 Frequency of occurrence of organisms making up the pooled categories -- insects, microcrustaceans, and other fish species, as well as miscellaneous items consumed in eastern Lake Michigan, 1983 . . . . . . . . . . . . . . . . . . . . . 23 Relative percent of feeding salmonid species by region (sample sizes in parentheses), 1983 . . . . . . . . . . . . . . 25 Relative percent of feeding salmonid species by season (sample sizes in parentheses), 1983 . . . . . . . . . . . . . . 26 Total prey items consumed, by region, by the three major salmonids in eastern Lake Michigan, 1983. . . . . . . . . . . . . . 28 Total prey items consumed, by season, by the three major salmonids in eastern Lake Michigan, 1983. . . . . . . . . . . . . . 29 Percent frequency of occurrence, by region, of various prey items consumed by the three major salmonids in eastern Lake Michigan, 1983 . . . . . . . . . . . . . . . . 31 Percent frequency of occurrence, by season, of various prey items consumed by the three major salmonids in eastern Lake Michigan, 1983 . . . . . . . . . . . . . . . . 32 Percent (number feeding/sample size) of feeding salmonid species caught in eastern Lake Michigan, 1984. . . . . . . . . . 55 24. Percent frequency of occurrence of various prey items consumed by Lake Superior salmonids, 1983-84 . . . . . . . . . . . . . . 102 LIST OF FIGURES Figure Page 1. Study sections, ports, and regions within eastern Lake Michigan. . . . . . . . . . . . . 7 2. Southern shoreline of Lake Superior representing study section III . . . . . . . . 9 3. Percentage of each major prey item contributing to the diet of salmonids in eastern Lake Michigan, 1983 . . . . . . . . 20 4. Percentage of each major prey item contributing to the diet of Chinook salmon in eastern Lake Michigan, 1983 . . . . . . . . 36 5. Percentage of each major prey item contributing to the diet of coho salmon in eastern Lake Michigan, 1983 . . . . . . . . 38 6. Percentage of each major prey item contributing to the diet of lake trout in eastern Lake Michigan, 1983 . . . . . . . . 40 7. Percentage of each major prey item contributing to the diet of steelhead in eastern Lake Michigan, 1983 . . . . . . . . 42 8. Salmonid size class predation on alewife and other prey items in eastern Lake MiChigan’ 1983 O O O O O O O O O O O I O O O O 45 9. Chinook salmon size class predation on alewife and other prey items in eastern Lake MiChigan, 1983 O O O O O O O 0 O O O O O O O O 47 10. Lake trout size class predation on alewife and other prey items in eastern Lake MiChigan, 1983 O O O O O O O O O O O O O O O O 49 11. Mean length of alewife consumed by salmonids of various size classes in eastern Lake MiChigan’ 1983 O O O O O O O O O O O O O O O O 51 12. Salmonid size class predation on organisms within the zooplankton category in eastern Lake Michigan, 1983. . . . . . . . . . . . . . 54 INTRODUCTION The balance between predator and prey populations is important in the management of a fishery. Forage base relationships have been greatly restructured, both positively and negatively, by the introduction of a species and species extinction. Invasion of the sea lamprey (Petromyzon marinus Linnaeus) and alewife (Algsa pseudoharengus (Wilson)) led to the collapse of various fish stocks and support of an economically productive sport fishery in the Great Lakes (Smith, 1964; Christie, 1974; Stewart et al., 1981).]311ake Michigan the alewife has become well established and presently constitutes a signif- icant portion of the forage base (Stewart et al., 1981). Alewife were first reported in Lake Michigan in 1949 (Miller, 1957), although earlier establishment could have gone undetected due to low abundance levels (Smith, 1970). Alewife proliferated in the lake during a period of low piscivore abundance, brought about by invasion of sea lamprey and its predation on lake trout (Salvelinus namaycush Walbaum) (Smith, 1964; Smith, 1970; Brown, 1972). Further alewife population increases occurred after the collapse of lake trout populations by 1950 (Christie, 1974). In the 1960's sea lamprey control had been staged (Baldwin, 1968). Increasing alewife populations, along with rainbow smelt (Osmerus mordax (Mitchill)), a species introduced in 1912, created yet a few more imbalances in the system. Namely, 1) extreme reduction and extinction of several native species in Lake Michigan through competi- tion for food and predation by both alewife and smelt and 2) occurrence of massive alewife dieoffs and large spawning runs (Wells and McClain, 1973; Stewart et al., 1981). Reasons for large alewife dieoffs have not been fully investigated but are believed to be due to a combination of climatic conditions, their low tolerance for cooler water temperatures, and high population densities (Smith, 1970; Hatch and Brown, 1978). These previous events led to the introduction of pacific salmon and attempted rehabilitation of lake trout into the Great Lakes (Tody and Tanner, 1966). Both have provided noteable control over alewife populations as well as better resource’ utilization. Since addition of pacific salmon into the Great Lakes, most of the bordering states and provinces have seen alewife populations reduced to more aesthetic levels, res- toration of native fish stocks, and development of economically productive sport fisheries. Michigan, for example, realized an annual valueaof about $350 million from its Great Lakes salmonid fishery in 1983 (Jester, personal communication). The popularity of salmon fishing has grownmultifold since the first introductions of coho salmon (Oncorynchus kisutch (Walbaum)) and Chinook salmOn (Oncorynchus tshawytscha (Walbaum)) in 1966 and 1967. other species comprising Michigan's salmonid fishery include: lake trout, rainbow trout (Salmo gairdneri Richardson), brown trout (Salmo trutta Linnaeus), and atlantic salmon (Salmo salar Linnaeus). Lake run rainbow trout are frequently referred to as steelhead and will be so designated in the remainder of this thesis. Increasing popularity of these gamefish over the past decade has led to the organization of several interest groups (eug. Michigan Steelheaders and Salmon Unlimited), economic expansion of numerous ports, increased annual stocking, and growing concern for the fisheries subsistence (Pistis, personal communication). Salmonid fishery recruitment has been accomplished primarily by stocking. This artificial recruitment alters predator-prey density feedback mechanisms associated with natural reproduction, and may lead to a flooding of the system with predators. Ideally natural responses would keep predator populations in line with their prey (Ricker 1975; Stewart et al., 1981). With the increasing number of salmonids being planted into Lake Michigan each year, numerous questions arise regarding natural balance of predator-prey populations. For instance, is the forage base capable of sustaining continual predator salmonid influx‘? Most importantly, will salmonids switch to other forage species given a decline in the alewife population? Alewife constituted the bulk of salmonid diets for the past 15 years (Smith, 1970; McComish and Miller, 1976; Stewart et al., 1981). Also, would switching to other forage species allow for self-sustainment and further stability of the fishery? Smelt, bloater'(Coregonus hoyi.(Gill)), and yellow perch (Perca flavescens (Mitchill)) are viewed as being potential prey items in reconstituting the forage base. Collaboration among the Great Lakes states in setting stocking rates, with both a qualitative and quantitative understanding of salmonid diets and responses to a changing forage base will help in managing the Great Lakes' salmonid fishery (Michigan Department of Natural Resources). Finally, it is on the premise of Stewart et a1. (1981), where salmonid stomach contents revealed trends of the forage base in western Lake Michigan, that this study was undertaken for eastern Lake Michigan and southern Lake Superior salmonids. The objectives of this study were to: 1) determine by year, region, season, size, and species the feeding habits (diet composition) of salmonids constituting Michigans' sport fishery and; 2) relate diet changes with various biotic factors ( ine. forage base and prey distributions). STUDY AREA The eastern portion of Lake Michigan from Michigan City, Indiana, northward, to Leland, Michigan, constituted study section I. This was subdivided geographically into three regions, south, middle, and north, corresponding to Lake Michigan's statistical districts MM-8, MM-7, and MM-6 and MM-5, respectively (Smith et a1. 1961). Ports of the three regions are shown in Figure 1. Both east and west bays of Grand.Traverse Bay were sampled independently of section I. Their seclusion and lack of continuity, both physically'and.biologicallyy with that of Lake Michigan led to this decision. Grand Traverse Bay was section II (MM- 4). The southern shoreline of Lake Superior represented study section III (Fig. 2). Twelve major fishing ports along the coastline from Michigan City to Leland were sampled. Within a given port, from 1 to 10 individual cleaning stations and charterboat docks were frequented to obtain salmonid data. Figure 1.--Study sections, ports, and regions within eastern Lake Michigan. Io ESSOUTHERN REGION 9 Q EJMIDDLE REGION §;7 + ++ @NORTHERN REGION + .STUDY SECTION II ++ ELAND + TRAVERSE CITY LUDINGTON PENTWATER MONTAGUE-WHITEHALL LIJ :‘: GRAND HAVEN <3: .1 WISCONSIN ._ _ _ _.,._-.. SAUGATUCK ILLINOIS SOUTH HAVEN BENTON HARBOR-ST. JOSEPH NEW BUFFALO MICHIGAN INDIANA .HHH cofluoow mosum mcflucomoumou uofluomom oxcq mo ocflfiouocm combusomll.m ousoflm a ouuonqumz Hobnmm Hoodoo \QWQ IVA) \OJMV .AWHNNUHWWMUV av u A" Aflzhafimflmw METHODS Collection of salmonid data followed different routes in 1983 and 1984, although sampling efforts each year were concentrated on weekends when larger Sample sizes and diversity among fish could be obtained. Reasons for differences between the two sampling schemes will be discussed in following sections. Charterboat docks, fish cleaning stations and fishing tournaments allowed for large numbers of fish to be sampled. Anglers provided information on salmonid depth at capture and water depth; while length, weight, lamprey scars and sex were recorded whenever possible. Fish stomachs were removed and placed in numbered whirl-pac (plastic bags) matching recorded data. The "stomach" was defined as beginning at the esophagus and ending at the hind guts The intestine was retained only where measurable material was found; this was the case for only a few insect feeding salmonids. Whirl-pacs were then filled with 15% formalin, sealed and returned to the lab for analysis. Fish utilized in this study were caught from near the surface to a depth of 140 meters, and within 25 kilometers of shore. The majority of salmonids were brought into port between 10:30 a.m. and 1:30 p.m., 3:00 p.m. and 5:30 p.m., and 8:00 p.m. and dusk. Although no records were made at exact capture time, larger numbers of fish appeared to be caught between.6:30 aJm and 9:00 aJm.and.at dusk. 10 11 1983 SAMPLING APPROACH Weekly sampling commenced in early spring on April 15, and ran through October; Spring sampling focused primarily on the southern region where the majority of fishing was concentrated. Spring fishing tournaments in the southern region were numerous and provided an opportunity for collecting large sample sizes. As spring progressed into summer, many anglers moved to the middle region of Lake Michigan. Fall fishing concentrated more heavily in the northern region. tuna» sampling followed major catches and movements of salmonids northward up the coastline (Sommers et al., 1981). This approach allowed for the collection of large sample sizes deemed necessary due to fluctuations occurring within salmonid diets (Peterson et al., 1982). Major diet changes were observed to occur in the course of one day among hundreds of fish (personal observation). 1984 SAMPLING SCHEME Weekly sampling commenced on April 1 and ran through September. Early spring sampling focused on the southern region of Lake Michigan because of the: 1) higher relative number of anglers and; 2) an attempt to sample salmonids feeding extensively’on zooplankton. Numerous salmonids were found feeding on microcrustacea in spring (mostly southern region) of 1983. The 1984 sampling scheme continued with collections from each of the twelve major ports throughout spring, summer and fall. This allowed for 12 detection and comparison of feeding habit changes on the basis of region and season. Salmonid data collected were seasonally labeled as either spring , summer, or fall samples (Table 1). Table.l,--Seasonal delineation of salmonid diet research. SPRING SUMMER FALL 1983 April 15-May 31 June 1-July 31 Augustl-Sept.:u) 1984 Aprill-Junelo June 11-August 10 Augustll-Sept. 30 SAMPLE ANALYS I S Analysis of stomach contents consisted of identi- fication, enumeration, length measurement (mm), and state of decomposition of each prey item. Food items ranged from freshly ingested organisms to unidentifiable masses. Identification, where possible, was generally taken to species for macroinvertebrates, microcrustacea and fish and to family for terrestrial and aquatic insects. Regression equationS‘utilized.forldetermining'total fish lengths from caudal peduncle lengths are given in Appendix A. In 1983 a sedgewick rafter cell was used to estimate the large number of microcrustacea collected, While in 1984, fewer numbers appeared, allowing for total counts to be made. 13 STATISTICS Descriptive statistics were utilized to rank the importance of various salmonid prey items by species, region, season, size, depth, and year. The two main descriptive statistics used were frequency of occurrence and percent of diet. The non-normal distribution of the data did not allow for parametric tests to be made. Thus, tests for independence and significant differences between parameters and descriptors were done using non-parametric statistics, primarily chi-square (Siegel, 1956). Tests for normality followed Sokal and Rohlf (1969) on the basis of skewness. Frequency of occurrence was calculated by taking the number of individual stomachs containing a given prey item and expressing it as a percent of the total number of stomachs analyzed (Lagler, 1956). This value expressed the apparent occurrence of a prey item in a salmonid represen- ting its utilization by the predator. This statistic, however, did not represent the actual occurrence and total mass relationship of a prey item to the diet. The proceeding weighting system was developed to provide a better descriptor (percent of diet) for presenting bulk relationship and actual occurrence of prey items in salmonid diets. The total value of the percent of diet statistic for each salmonid stomach equaled 1.0. Prey items within a 14 stomach were represented as a fraction of the total. The percent that a prey item contributed to salmonid diets was calculated by modifying Paloheimo's (1979) indice for the proportion of a prey item in a given predator. This was done by substituting Paloheimo's Ri and Pi values, volume searched and probability of capture for prey item i, for WV. percent of dieti = ni(WV)i/ nj(WV)j WV was an arbitrarily weighted value incorporating the selection of various prey items by a predator on the basis of prey size, nutritive value, and ease of capture (Werner; 1974; Werner'and Hall, 1974; Rottier'and.Tucker, 1982). Subscript j represented all prey items. All prey fish species were weighted equally regardless of size and species. There are several reasons for this judgement. Smaller pelagic species were easier for salmonids to capture compared to larger prey species on the basis of swimming speed and ease of capture (Wardle, 1977). More smaller sized forage fish are needed to comprise a feeding, equal in energy expenditure, to one of fewer, larger individuals (Werner, 1974). Rottiers and Tucker (1982) reported caloric contents of forage fish.(alewife, smelt and sculpins) to be similar, thus a consumption shift from one species to another would not be expected to significantly alter energy gain. Furthermore, the following were evident after data analysis: 1) there was a rare occurrence of vastly different size classes of fish 15 being consumed within a single feeding and; 2) rare occurrence of a single feeding inclusive of both benthic and pelagic species. The weighting procedures were applicable regardless of the state of decomposition of the various prey items. For example, four intact smelt and one decomposed alewife in a stomach would represent 80% and 20% of the diet, respectively. This continuity is based on the equal probability of catching a salmonid prior to or following a feeding on any given prey item. In 1983, fish demonstrated the ability to pack themselves full of fish and continue to behave in a feeding manner. Reasons for the weighting values for insects and microcrustaceans are as follows. The nutritional value of insects, on a total weight basis, was assumed to be much lower than that of fish. However, less energy was expended by the predator in consuming insects than fish. Werner (1974) noted that when prey were sessile, or much less active than the predator, the former would be conceived as a potential meal. Insects consumed were also smaller in size than fish species and thus given a 1:9 (.111) energetic value based on size, nutritional value, and energy expenditure (ease of capture) differences. Microcrustaceans consumed were very small ranging from 2 mm for Qaphnia spp. to 20 mm for Mysis oculata relicta (Loven). A value of .055 (1/18) was given on the basis of 16 size, signifying that the presence of 18 organisms within a salmonid depicted selection of that prey item and not its consumption by chance. In the case of salmonids feeding heavily on Qaphnia spp., occurrence of a fish species seemed insignificant and was presented as such. In calculating the contribution of various fish species, insects, and microcrustaceans to salmonid diets the following weighting values (WV) were given. fish (all species) = 1.0 insects = 1/9 (9 insects to 1 fish) =.111 microcrustaceans - 1/18 (18 micro#rustaceans to 1 fish) . 055 For example, if a stomach contained 5 alewives, 3 insects and 2 Daphnia spp. then the percent of diet values for that individual fish was: alewife = (5*l.0)/(5*l.0)+(3*.lll)+(2*.055) = .919 insects = (3*.111)/(5*1.0)+(3*.111)+(2*.055) = .061 microc. = (2*.055)/(5*1.0)+(3*.111)+(2*.055) = .020 Values were then summed for each prey item per stomach to obtain percent of diet values for the various regions, seasons, years, and species. Great precision can not be claimed with these microcrustacean and insect estimates. IPotential errors are assumed to be small because of these organism's small contribution to the overall salmonid diet. RESULTS AND DISCUSSION GENERAL FEEDING HABITS,1983 The stomachs of 1,513 salmonids were examined for food items in 1983. The percent of feeding fish, by species, was similar with the biased exception of steelhead, brown trout, and atlantic salmon. Their sample sizes were relatively small (Table 2). Food items were found in 974 (64.38 %) stomachs, while 540 (35.62 %) were empty. Table 2.--Percent (number feeding/sample size) of feeding salmonid species caught in eastern Lake Michigan, 1983 CHINOOK S. COHO S. LAKE T. STEELHEAD BROWN T; ATLANTIC S; % 61.1 68.51 63.74 84.62 28.57 50 (468) 372 109 (22) (2) (1) (765) (542) (171) (26) (7) (2) ALL SPECIES: 64.38 (974)/(1513) Wright (1968) reported a much larger percent of feeding fish in 1967 than that of 1983. He found that 93.4% of 1,581 lake trout stomachs examined from Lake Michigan contained food items. Differences may be ex- plained on the basis of collection methods, fish sizes and the forage base» Wright (1968) used gill nets, while in the present study fish were caught by angling. Angling causes a larger bias towards obtaining empty stomachs 17 18 since it samples fish presumably in a feeding mode. Secondly, Wright (1968) primarily sampled immature lake trout (<60 cm) while those examined in 1983 were legal size and larger (30.48 to 85 cm). Most importantly, the biomass of the forage base in the earlier 1970's was larger, predominated by alewife, compared to a smaller forage base in the early 1980's, predominated by smelt and bloaters (Wells and Hatch, 1983) (Appendix B). Decline of forage in 1983 decreased the amount of contact between predator and prey resulting in a smaller percent of feeding fish. This is a principle of fisheries management (Ricker, 1975). The major salmonid food items in Lake Michigan were fish (77.1 %), mainly alewives and smelt (Figure 3). Zooplankton also constituted a large proportion of their diet. This was primarily caused by the large number of smaller salmonids (45 - 70 cm) feeding extensively during spring on cladocerans in southern Lake Michigan. Other microcrustacea, Pontiporeia alfigig Smith and Mysis relicta oculata (Loven), were found throughout the year, and were included as zooplankton. Other investigators also found salmonids utilizing these invertebrates (VanOosten and Deason, 1938; McComish and Miller, 1967; Wright, 1968). Additional prey items were bloaters, perch, sculpins (9933313 spp.), 9-spine sticklebacks (Pungitius pungitius (Linnaeus)), insects, unidentified fish (fish remains), and other fish species. The latter included species rarely occurring in the diets (< 0.40 %), e.g. trout perch 19 mxmq cumummo Ca mpacoE . . . . How mo uoflo map Op mcfiuobauucoo EODH woum Donne commmwwammmmcwouwm .m ousm - " Him 20 xm.zwumz.ozm zucxzcnmoom xN.iumzm.N3 wcummzc xw.mueo.lm omcucczmo_z: xm.onvw.m mumps xz.zuve.ll .nxucpm mzimm - m xe.ou¢ zcansow X¢.oum.e rumba xm.mnmm.sm anemone xv.avuwb.mov mmszmc Nm.~Num.mow Hmmzw pmHo Lo szuzwm ouzozmmw 21 (Percopsis omiscomaycus (Walbaum)),shiners (Notropis spp.), johnny darters (Etheostoma nigrum Rafinesque), and lake Whitefish (Coregonus clupeaformis (Mitchill)). Terrestrial insects blown into the lake were consumed by surface feeding salmonids. The most numerous flying insects consumed were from the family Noctuidae (moths). Presence of various prey items and their contribution to the diet of various salmonid species, are listed in Table 3. Organisms constituting insect, zooplankton, and other categories, as well as miscellaneous items consumed are given in Table 4 with respect to their frequency of occurrence . 22 Table 3.--Frequency of occurrence of prey items inl983 salmonid diets (percent of diet in parentheses). CHINOOK é; gggg E; gggg g; STEELHEAD BROWN 3; ALEWIVES 64.24 20.70 69.72 9.09 50.00 (57.6) (16.95) (64.33) (10.00) (50.00) SMELT 26.98 34.68 25.69 45.45 50.00 (19.52) (24.57) (18.14) (26.37) (50.00) BLOATERS 7.92 12.10 0.92 0.00 0.00 (5.69) (8.24) (0.31) (0.00) (0.00) PERCH 0.86 0.81 0.00 0.00 0.00 (0.64) (0.35) (0.00) (0.00) (0.00) SCULPINS 0.21 0.00 2.75 0.00 0.00 (0.21) (0.00) (2.75) (0.00) (0.00) 9-SPINES 1.50 2.42 0.92 0.00 0.00 (0.99) (1.56) (0.92) (0.00) (0.00) OTHER FISH 1.07 0.81 0.92 0.00 0.00 (0.68) (0.53) (0.05) (0.00) (0.00) UNIDENT- 7.06 8.06 5.50 0.00 0.00 IFIED FISH (5.24) (5.98) (3.94) (0.00) (0.00) INSECTS 0.64 4.84 3.67 45.45 0.00 (9.17) (1.74) (2.28) (12.55) (0.00) 200- 10.06 43.28 9.17 59.09 0.00 PLANKTON (0.24) (40.07) (7.28) (51.10) (0.00) 23 Table 4.--Frequency of occurrence of organisms making up the pooled categories - insects, zooplankton,and other fish species, as well as miscellaneous items consumed in eastern Lake Michigan, 1983. FREQUENCY pg OCCURENCE Invertebrates Crustacea Amphipoda Pontiporeia affinis 5.85 Cladocera Daphniaspp. 19.51 Eucopepoda .10 Mysidacea Mysis relicta oculata 4.00 Insecta Coleoptera Carabidae .21 Diptera .21 Ephemeroptera Hexagenia spp. .21 Hemiptera Corixidae .21 Notonectidae .41 Pentomatidae .51 Hymenoptera .31 Lepidoptera Noctuidae 1.44 Orthoptera Acrididae .10 Trichoptera .51 Fish Clupeidae Alosa cerepedianum .31 Cyprinidae Notropis spp. .21 Percidae Etheostoma nigrum .21 Salmonidae Corigoninae Coregonus clupeaformis .21 Oddities Cigarette butts .10 Driftwood .10 Feathers .21 Macrophytes .10 Mollusca Pisidium sp. .10 Plastic .41 Rocks .21 Rope (yellow) .10 FEEDING HABITS BY REGION AND SEASON Studies of salmonid feeding habits have demonstrated differences between geographical location and time of year (Wright, 1968; Chiotti, 1973; Rybicki and Keller, 1978; Eck and Wells, 1983). The 1983 northward sampling scheme demonstrated a significant correlation between the percent of feeding salmonids by season and region. Spring samples were correlated with those of the southern region. The same occurred for summer and middle region and fall and northern region (Tables 5 and 6). The percent of feeding fish was higher in the spring (mostly southern region), resulting from littoral zones being more heavily fished and containing a larger array of available forage. In spring numerous forage species inhabited littoral zones for both feeding and spawning. Smelt aggregated in Shallower regions prior to spawning. Bloaters, perch, and other smaller'fish species utilized zooplankton rich littoral zones for feeding, spawning and their temperature preference (Wells and Beeton, 1963; Brandt et al., 1980; Sommers et al., 1981). Salmonids moved into the deeper cooler waters in summer, where forage was more dispersed, resulting in a lower percent of feeding fish (mostly the middle region). They moved back into the littoral zones in the fall for spawning and a presumed temperature preference allowing for overlap with forage 24 25 Table EL--Relative percent of feeding salmonid species by region (sample sizes in parentheses), 1983. NORTH ALL REGIONS CHINOOK S. COHO S. LAKE T. STEELHEAD BROWN T. ATLANTIC S. 53.60 (152) 65.40 (25) 55.10 (49) 100.00 (1) 00.00 (0) 00.00 (0) 60.80 (401) 53.20 (299) 45.80 (24) 80.00 (5) 00.00 (4) 50.00 61.10 (765) 68.51 (542) 63.74 (171) 84.62 (26) 28.57 (7) 50.00 ALL SPECIES 77.86 (551) 55.70 (227) 57.01 (735) 64.38 (1513) 26 Table 6.--Relative percent of feeding salmonid species by season (sample sizes in parentheses), 1983. SPECIES SPRING SUMMER FALL ALL SEASONS CHINOOK S. 66.70 57.52 59.14. 61.10 (231) (113) (421) (765) COHO S. 89.70 69.57 52.50 68.51 (223) (22) (297) (542) LAKE T. 76.00 46.30 52.40 63.74 (96) (54) (21) (171) STEELHEAD 88.90 75.00 75.00 84.62 (18) (4) (4) (26) BROWN T. 66.70 00.00 00.00 28.57 (3) (l) (3) (7) ATLANTIC S. 00.00 00.00 50.00 50.00 (0) (0) (2) (2) ALL SPECIES 77.93 56.70 56.15 64.38 (571) (194) (748) (1513) 27 abundant shallows (Sommers et al., 1981). Chinook and coho salmon have optimal temperature ranges of 12 C to 12.5 C and 11 C to 12 C, respectively, thus allowing for their observed seasonal movement (Sommers et al., 1981). Reallocation of energy in the fall for reproduction by mature salmon reflected a decline in the percent of feeding. fish. This behavior was also reported by Wright (1968) and Chiotti (1973) for immature Lake Michigan lake trout and by Frantz and Cordone (1970) for mature Lake Tahoe lake trout. These fish fed most avidly during spring and less so in fall. Of those fish found feeding in the fall, at least 70 % were immature, showing a division of energy expendi- ture between spawning and non-spawning individuals. Among feeding fish, alewives were the most abundant species consumed, followed by smelt and bloaters. Coho salmon, however, consumed more smelt.in the southern and middle regions, as well as, in spring and fall (Tables 7 and 8). Relatively larger numbers of smelt were consumed by all species during spring and fall caused by spring spawning smelt aggregating in littoral zones. These fish dispersed in summer, and semi aggregated again in fall, following water temperature and forage availability (Sommers et al., 1981). Thus, smelt were more readily available for consumption because of their large numbers and distribution overlap with the salmonid predators. Likewise, salmonids became more pelagic during summer months (middle and northern regions), in response to water temperature, creating a distribution overlap with alewife 28 HAHN mmSH ome ock mmHomam HH< HAN moHH HNmH H3 NRA «Hm em 03H mew «AH 6mm cam H9 .UCESmcoo memuH xwua Hmuoenn.m pome 29 NmmN mama «mm mmw mmHummm AA< AAN NSHH ANmH A3 oNA 66A . Nm SNH SAH NAH NON mwm Hmmm mzom9 .commwm >3 .Coesmcoo mEOuH xoua Hmuoeni.w CHOCH 30 and bloater (Wells, 1968; Brown, 1972; Brandt, 1978; Janssen and Brandt, 1980). Predator-prey distributions reflected the large percentage of these forage fishes in predator diets during this time. Perch also made an appearance, correlated with similar habitat overlap. Table 9 presents the frequency of various prey items in salmonid diets by region. Several general observations can be observed: 1) frequency of prey items differ greatly among species, 2) occurrence of prey species shifts between regions, 3) a large percent of chinook and coho salmon feeding on microcrustaceans, 4) presence of smelt in diets correlating with its distribution overlap with salmonids, 5) presence of bloaters in the middle and northern regions of the lake associated with their larger abundance (Hatch and Wells 1983), 6) large numbers of bloaters occurring in coho salmon, 7) a slight rise in frequency of occurrence of perch in the middle region, especially coho salmon, and 8) lake trout maintaining high levels of alewife predation throughout the lake. Similar observations can also be made by viewing the frequency of occurrence of prey items by season, as well as the presence of perch and bloaters in the fall (Table 10). Alewives occurred more frequently in lake trout and chinook salmon than in coho salmon. Coho salmon diets were more evenly diversified. The cohos opportunistic habits are presumed to reflect a quicker response to a changing forage base. This was observed by the higher occurrence of the more abundant forage fish, smelt and bloaters, in coho diets (Hatch and Wells, 1983). 31 .Hmamo AH.0 0N.mq 00.0H 00.0 0H.m Nw.0 Hq.m m0.mH mm.wH 00.0 00.nm 00.0N nomon m0.m «0.0 Q0.0 00.0 mm.0 H<.0 05.m 00.HH 00.0 Nw.m m0.m 0Q.H mHUmmZH 0m.m 00.0 00.A 00.0 N0.mH 50.0 mm.ma m0.mH m¢.m 00.0 em.m 00.¢ .Bzmon: N0.0 Hm.0 ~0.H 00.0 AN.H Nw.0 00.0 mm.m 05.m 00.0 00.0 00.0 mmmeo N0.0 Nq.m 0m.H 00.0 0A.m mm.m 00.0 00.0 00.0 00.0 00.0 00.0 mszmi0 mA.N 00.0 HN.0 00.0 00.0 0030 00.m 00.0 mm.H Hq.H 00.0 00.0 szADUm 00.0 Hm.0 00.0 00.0 00.0 00.0 00.0 Hw.0 mq.m 00.0 00.0 00.H mommm «0.0 0H.NH N0.A 00.0 Nq.mm mq.0 00.m 0N.mm m0.0H 00.0 N0.H 0m.0 mme OLH mo .conOH >2 .oocouusooo 0o xocoscouu accouomiu.0 wamH 32 .9m0m0 >H.0 0N.mq 00.0H 00.0 00.0 Hw.0 00.0 m>.wH 0H.mH 0m.0 0m.>> mq.HN nom0Hz >0.m «0.0 «0.0 00.0 00.0 00.0 00.0 mm.0 mm.H HH.¢ 0m.m 0m.H m90mmzH 0m.m 00.0 00.9 0H.mH Hq.0 m0.q 00.NH mN.0 mm.q q>.m 0m.N 0m.H .9zmoH20 N0.0 H0.0 >0.H 00.0 0N.H Hw.0 00.0 00.0 00.0 00.0 0m. m0.H mm:90 N0.0 N¢.N 00.H 00.0 >>.m N0.N 00.0 00.0 00.0 00w0 00.0 00.0 mszmu0 m>.N 00.0 HN.0 00.0 00.0 00.0 00.5 00.0 00.0 >m.a 00.0 00.0 szqsom 00.0 Hm.0 00.0 00.0 N0.H H0.H 00.0 00.0 00.0 00.0 00.0 00.0 :Ummm N0.0 0H.~H N0.> 00.0 q>.qH mq.0 00.0 mN.Hm 00.0 00.0 00.H 00.0 mm9N.>N Hm.Nq mq.Hm 00.0 00.0 mH.mH >>.0N 0m.Hm 00.0N 94020 N>.00 09.0w qw.qo N0.Hm 00.mm HH.¢0 00.00 m>.00 mm.~> 00.00 00.m 0m.00 mmHzmu< .9 mx2 emesmcoo mEODH >ouc msoHHm> on“ 00 .commcm >b .wocouuzooo.wo >ocoscoum ecooummnu.0fi oabm9 33 Frequency of occurrence for alewives in the northern region for lake trout was 63.4 %, significantly exceeding that of 49.3 % found in 1972 by Chiotti (1973). Differences between the two may indicate a prey preference as well as the inability of salmonids to switch prey items. Further evidence for this condition can be seen from.the occurrence of sculpins in 37.4 % of lake trout stomachs in 1972 (Chiotti, 1973) and only 9.09 % in 1983. Even though the ratio of alewife to sculpins in 1972 greatly surpassed that of recent years (Wells and Hatch, 1983; Wells and McLain, 1972). Relative consumption of sculpins by lake trout in 1972 was greater than in 1983. This could indicate a lag-—time in lake trout feeding response to forage species abundance. During the spring, occurrence of microcrustacea in southern Lake Michigan salmonids was obvious (Table 10). Qaphnia spp. made up over 70% of the spring zooplankton category (by relative weighting procedures). Possible reasons for their dietary presence can be attributed to: 1) occurrence of large pulses with an overlap in predator habitat; 2) forage of another type unavailable and; 3) increased sight recognition of these organisms because the majority were epipphial (Birge and Juday, 1922; Richman, 1958; Mellors, 1975). Regions were used to delineate the salmonid population based on: 1) changes occurring in the forage base and diets across regional boundaries; 2) similarity to delineation by season and; 3) for standardized comparison to other 34 pertinent studies utilizing regions according to Michigans statistical districts. To further illustrate changes occurring among species and regions the percent of diet descriptor was used. Alewife was the major prey item in salmonid diets throughout Lake Michigan, especially the middle and northern regions (Figures 4-6). Zooplankton represented a large percentage of the coho salmon diet in the southern region, but were less evident in chinook salmon and lake trout in that region. Bloaters became a significant portion of the diet in the middle region. Insects remained insignificant throughout the year, not exceeding 5% of any diet. The exception was steelhead, in which insects constituted 19.6 %, by occurrence of the yearly diet (Figure 7). Perch remained insignificant. FISH SIZE AND PREY SELECTION Numerous studies on the relationship of predator length to prey length have been reported (Chiotti, 1973; McComish and Miller, 1976; Eck and Wells, 1983; Hagar, 1985). Chiotti (1973) analyzed.predator—prey length re- lationships using mean alewife prey length where more than one fish of a given species was present in an individuals stomach. The large sample size and number’of prey items found in fish stomachs during the present study permitted the use of Chiotti's method. The mean alewife length was obtained by averaging all the various sized alewives 35 .mNmH .cmmHeon Oxmq cumbmmo :H coEHMm xoochO mo uoHc 0:0 00 0cHuanHucoo EOHH >OHQ Hones some 00 ommHCOOHomnn.v OHsmflm 36 x:.0 2092204000N NN.0 090002: xN.0 00:0:9200:20 N0.0 00:90 xm.0 20¢004x0:90 02:00I0 xN.0 2:04:00 N0.0 .20000 x0.0 0090040 x0.0: 94020 20:00: 0400:: N0.0v 00:3040 020:000 44: N0.00 00:3040 NN.0:H0N.0H 2092204000~ x0.:uv0.: 00:0:9200:20 xv.Nu9v.N 00:90 NN.:M0~.: 2:04:00 V0.:n00.: 20000 N0.N:NN0 N: 0090040 N0.m:n00.m: 940:0 20:000 200:9002 xv0 0L:304¢ 20:000 20029000 Nm.:0 00:304: xvo. 090002: N:.P 00:0:9200:20 N0.— N00420:90 02:00n0 x0.0 0090040 Nv.0N 20922040000 Nb.0 090002: Nv 00:0:9200:20 N: 20:00 NN.0 0090040 N:.NN 940:0 37 cumummo CH .mmmH .cmmHeon mxmq coeHmm 0:00 00 HTHU may on 0cHuobHuucoo EouH >OHQ momma comm 00 ommucooummll.m 0H30H0 38 Nm.0: 00:3040 x0.vN 940:0 XN.0 020:000 440 20:00: 20029002 Nov 2092204000~ .0 o Nbomm mwuwmzH 04.43N5e Am.m oultchmast N0. 00:90 N0.: eupeonm\iiiiiiiiiiiiiiiiiii Nm.o 20000 20:000 0400:: N:.0: 00:3040 N:.NN 940:0 x0.:N 0090040 xLa00420:90 02:00l0 A NHm Anmzw um.m mtlzmne NN.NH ZQAez¢31o0N Nm.o mAummz_ Am.mH Anuzm Am.mH QLHLHHZLQHZS Am.o NLH¢OHSIIIIIIII, NN Am.4 eLon oullleuolza Am.o ”A0082: x0.9 r0000 xu.m 2092204000~ N0.0 090002: N0: 00:0:9200:20 Nh.0 00:90 N0.m . "00420:90 02:0010 x0.0: 0090040 20:00: 20029000 NP.05 2092204000~ 39 .mmmH .cmchon @264 snoummo CH HSOHD oxc: 00 uoHU 0:0 00 mcHusnflHucoo ETDH >OHQ HohmE coco 00 mmmpcoouomnl.0 demHm 020:000 44¢ 40 20:000 0400:: xu.w 090002: N0.m 00:0:9200:z: x00. 000:0 N0.0 2 200420:90 02:0010 Nb.~ 2:04:00 N0.0 0090040 N:.0: 940:0 xv.h . 2092204000~ Nh.m 090002: x0.0: 00:0:9200:z: N0.m 2:04:00 x~.: 0090040 20:000 20029002 x0.v0 00:2040 20:000 20029000 N0.v0 00:2040 Nh.m 090002: N:.0 0090:9200:2: xv.0 00:90 N0.0 "00420:90 02:00:: N0 2:04:00 N0 0000 Nh.h 2092214000~ No.0 090002: Nv.: 2:04:00 N0.0N 940:0 41 .mmm: \cmoflcoflz mxmq cumummm CH 0meHmmum 00 ymHU mg: on mcflusnauucoo EmuH xmum Homme comm 00 mmmucmoummll.9 musmam 42 x0.0* 940:0 N0.m: 090002: N0.b0 20902040000 000: 02:000 I 000240090 43 consumed by the predator. Error occurring from averaging widely separated length measurements was presumed minimal because of adults and juveniles occupying different strata (Janssen and Brandt, 1980; Brandt et al., 1980; Crowder et al., 1981). Independent plots of chinook salmon, coho salmon, and lake trout total lengths versus mean total lengths of ingested alewife resulted in no significant correlations (r): chinook salmon .216, coho salmon .133, and lake trout -.02. A scatterplot, illustrating the lack of correlation for chinook salmon is presented as an example in Appendix C. Graphing salmonid size class predation on alewives and salmonid predation on the various alewife size classes allows few observations: 1) larger salmonids (>75 cm ) fed more often on alewife than did smaller salmonids (<75 cm), while smaller to medium sized salmonids (>50 and <75 cm) fed more frequently on prey items other than alewife (Figure 8); 2) the percent that alewivescontributed to the "percent of diet" of chinook salmon (>65 cm) was fairly constant (60-81 95) (Figure 9); 3) lake trout showed no obvious size predation on alewives (Figure 10). Salmonid species fed primarily upon larger alewives (Figure 11). Although species size distributions illustrate more large chinook salmon were sampled than lake trout (Appendix D), chinook salmon still appeared to show a 44 .mmma .cmmflcoflz mxmq cuwummm CH mewufl xmwm nmcuo 0cm mMflBmHM :0 coflumpmum mmMHU mNHm 0Hc05:mmlu.m demH0 45 FEEDING 0N RNY GIVEN \ FEEDING ON RLENIVES ITEM W/A V//////A WA WA @— ////; ////////; \\ \\ \\\ 8 6 4 .. 20 SGINOW'IVS :IO HSQWON 35.2 47.5 52.5 57.5 62.5 67.5 72.5 77.5 82.5 87.5 92.51000 SALMONID LENGTH CLASS MIDPOINT (CM) 46 .mmm: .cmmflsoflz mxmq Cumummm CH mEmuH >mu0 uwCuo UCm ®0H3mHm Co CoHumpmum meHU oNHm COE:mm xOOCHCUII.m mucmHm //////////// 7/////////// ///////////////////////// W- W \\\r 35.2 47.5 52.5 57.5 62.5 67.5 72.5 77.5 82.5 87.5 92.51000 33% W _ §§ //// :5 ////// *- W /L a 4 NNNNNNN NOW'IVS )IOONIHO :IO HBSWON CHINOOK SALMON LENGTH CLASS MIDPOINT(CM) 48 Cuwbmmm CH .mmma .cmmflnon mxmq memuH wmum umcuo UCm mezmHm Co COHumcmHQ mmm:o mNHm boom: mxmqul.OH mH90H0 E- /_ %- RNY FEEDING 0N X\IA U) N N 1008.]. EMV'I :IO HSSWHN O U) 0 LD 0 c-n ---a 35.2 47.5 52.5 57.5 62.5 67.5 72.5 77.5 82.5 LAKE TROUT LENGTH CLASS MIDPOINT (CM) 50 mwxg Chwummw CH mmmmmgnv wNflm WSOHHM> MO mUHCOEHmm ha cmESmCOU meBmHM .mmm: .cmchon 00 CumCmH cmmz:l.HH mH50H0 51 l \\l\\ W T W l RLL SPECIES CHINOOK SRLMON LHKE TROUT \\V [/ V/////// 1 7% l I 90 >Q\V\§ I 70 140 120 _ SEAIMB'IV SNIWHSNOO HBSWHN O D O 03 O (D O V” O N O 110 130 150 170 190 210 230 45 MEAN ALEWIFE LENGTH CLASS MIDPOINT (cm) 52 higher affinity for larger alewives than lake trout. Fish between 50.1 and 60 cm in length consumed the majority of zooplankton found in the salmonid diets (Figure 12). Coho salmon fed more extensively upon organisms within the zooplankton category followed by steelhead and chinook salmon. In stream, lake, and river habitats rainbow trout at times have been reported to feed extensively'on these zooplankters, especially D_aphnia spp. (Gailbraith, 1967; Taylor and Gerking, 1979). No evidence of Qaphgia spp. consumption by salmonids within Lake Michigan was found in the literature searched. The number of salmonids feeding on insects by size was random, with no evident pattern (Appendix E). 53 .mmm: .cmmHson mxmq Cumummm CH muommumo CouxCM:QooN mCu CHCuHB mEmHCmmuo Co CpommeQ mmmHo mNHm UHCOEHM0II.NH ®H50H0 RLL SPECIES ,/ cone SRLHON [ 5 0\\\\\\\\\\\\\\\\\\\\\\\\\\\\ \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\V OOOOOOOOO GGGGGGGG NOLMNV'IdOOZ NO SNIGSEd HSGINnN 35.2 47.5 52.5 57.5 62.5 87.5 72.5 77.5 82.5 87.5 92.5100.0 SALMONID LENGTH CLASS MIDPOINT (CM) 55 GENERAL FEEDING HABITS, 1984 The stomachs of 2,705 salmonids were examined for food contents in 1984. Food items were found in 1,255 (46.41 %) of the stomachs, while 1,449 (53.59 %) were empty. Species sampled varied with respect to percent feeding (Table 11). Table lJ"-—Percent (number feeding/sample size) of feeding salmonid species caught in eastern Lake Michigan,l984. CHINOOK S. COHO S. LAKE T. STEELHEAD BROWN T. ATLANTIC S. % 35.55 47.90 63.06 73.40 54.67 100.00 (449) (330) (366) (69) (41) (1) (1263) (689) (582) (94) (75) (1) ALL SPECIES: 46.51 (1256)/(2705) These values differed significantly from chinook and coho salmon percentages obtained in 1983. There was an average 38.42 % decrease in salmonids feeding from 1983 to 1984. Chinook salmon decreased 72.1 96, coho salmon, 43.3 %; and lake trout, 1.08 %. The two following observations further typify a decline in feeding fish : 1) increased state of decomposition of prey items in salmonid stomachs and; 2) a decrease in stomach fullness from 1983 to 1984. Reasons for the apparent decline in feeding salmonids could be related to a declining forage base, an increased number of stocked salmonids, or a combination of both. 56 The major food items of salmonids in Lake Michigan were again fish (91.6 %), primarily alewives, smelt, and bloater (Figure 13). Zooplankton did not constitute a significant portion of the diet as it did in 1983. The decline of zooplankton in salmonid diets may be explained by the majority of fishing occurring in deeper water during 1984 than in 1983. This resulted in catches of more pelagic than littoral salmonids. Forage fish-zooplankton interactions occurring within the lake should result in larger zooplankton populations following a decline in jplanktivorous predators (eug. alewife) (Stewart et al., 1981). Leading to the prediction of zooplankton becoming a potential prey item. Sample collections in late fall 1984 have shown the dietary reoccurrence of zooplankton. Their proportion and frequency is still in the analysis stage (Nurse, personal communicationL.Fmture sampling should result in a better understanding of zooplankton in salmonid diets. Occurrence of various prey items in salmonid diets are liSted in Table 12, as well as the percent that each item contributes to the salmonid species diet. Organisms comprising insect, zooplankton, and other categories, including miscellaneous items found in salmonid stomachs, are shown in Table 13. The frequency of bloaters and perch in salmonid diets increased noticeably from 1983 to 1984, while alewives and smelt decreased.Thesenumbersindicateashiftin salmonid feeding habits towards a diet which better reflects the 57 .vmma .cmmflsoflz mxmq Cumummm CH mpHCOEHmm 00 uwHU mCu ou mCHuCCHuuCoo EmuH >mu0 Cohme Comm mo momCCmoummll.mH musmHm 58 xb.:u¢m.:w 2090204000N xm.0uN0.¢h 090002: x0.0u00.0:: 00:0:9200:z: x:.:ummr.v: 00290 x:.b0uvv:.00v x:.Nummh.0N 00::040 ...00400:90 02:00am xv.:umw.0: \\\\\\‘|llll- 2204266 x2.sum42.mm 20000 x0.0:u0:0.vm: 0090040 xv.0:u¢00.:00 94020 11 59 Table 12.--Frequency of occurrence of prey items in the salmonid diets in 1984 parentheses). (percent of diet in ALEWIVES SMELT BLOATERS PERCH SCULPINS 9-SPINES OTHER UNIDENT- IFIED FISH INSECTS ZOO- PLANKTON 37.50 (38.18) 16.21 (15.53) 23.63 (25.94) 6.05 (5.63) 0.00 0.00 1.76 (1.60) 0.78 (0.49) 9.38 (0.92) 0.98 (0.77) 3.71 (2.65) CHINOOK S. COHO S. 19.00 (17.30) 29.45 (33.52) 12.83 (11.41) 10.45 (8.90) 1.66 (1.50) 3.56 (3.86) 2.85 (2.58) 10.21 (10.6) 7.60 (8.04) 2.38 (2.31) 51.30 (58.26) 11.47 (10.04) 10.82 (9.68) 4.98 (4.70) 4.33 (3.55) 1.73 (1.60) 1.95 (1.09) 10.39 (9.17) 2.60 (1.77) 0.43 (0.12) STEELHEAD 21.74 (14.08) 17.40 (12.91) 7.25 (2.21) 17.40 (12.50) 2.90 (0.83) 2.90 (1.41) 1.45 (0.11) 4.35 (3.18) 60.87 (51.32) 1.45 (1.44) 41.46 (39.02) 17.07 (14.84) 4.88 (4.88) 26.83 (21.75) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 9.76 (9.76) 7.32 (7.32) 2.44 (2.44) 60 Table 13.--Frequency of occurrence of organisms making up the pooled categories - insects, zooplankton, and other fish species, as well.as miscellaneous items consumed, 1984. FREQUENCY 9E OCCURRENCE Invertebrates Crustacea Amphipoda Pontiporeia 1.52 Cladocera - Daphnia pulex .72 Q; retrocurva .08 D. longiremis .08 Hyallela azteca .08 Leptodora kindtii .16 Copepoda Cyclops spp. .08 Calanoid .08 Decapoda .08 Mysidacea Mysis oculata relicta .32 Insecta Coleoptera 6.45 Carabidae 2.47 Chrysomelidae .56 Coccinellidae .64 Curculionidae .24 Dytiscidae .64 Elateridae .24 Gyrinidae .08 Haliplidae .08 Hydrophilidae .08 Lampyridae .72 Diptera .96 Muscidae .72 Syrphidae .08 Tabanidae .08 Tipulidae .08 Ephemeroptera Ephemeridae Hexagenia spp. .08 Hemiptera 1.35 Corixidae .32 Pentomatidae 1.19 Homoptera Cicadellidae .24 Lepidoptera Noctuidae 2.23 Orthoptera Tettigoniidae .08 Trichoptera Brachyceridae .08 61 Table 13 (cont'd). FREQUENCY 9g OCCURRENCE Fish Clupeidae Alosa cerepedianum .48 Centrachidae Lepomis macrochirus .08 Cyprinidae Notropis spp. .32 Percidae Etheostoma nigrum .40 Percopsis omiscomaycus .24 Salmonidae Salvelinus namaycus .08 Corigoninae Coregonus clupeaformis .24 Miscellaneous items consumed Cigarette butts .16 Soil (dirt clot) .08 Plastic .16 Sticks .08 Wood .08 62 present forage base. The forage base in Lake Michigan has undergone major species shifts within the:past.20 years. Alewife popula- tions have undergone reductions in number as well as changes in size class distribution (Jester, personal communication) possibly resulting from salmonid predation and inter-specific competition (Stewart et al., 1981; Crowder and Magnuson,l982; Crowder and Binkowski, 1983). These changes in the alewife population should results in a decline in numbers and sizes consumed. The frequency of larger alewife size classes ingested by salmonid predators declined from 1983 to 1984. An increase in perch in salmonid diets was seen from 1983 to 1984. Future predation on perch seems evident with any further decline in alewife populations. However, mis- representation of perch as a major prey item may have occurred in 1984 due to a large year class. Perch have been shown to undergo large population fluctuations (Eshenroder, 1977; Nelson and Walburg, 1977; Smith, 1977). The rate at which salmonid feeding habits have changed in relation to a declining alewife population is not well understood. Evidence of a lag time between salmonid predators in response to a changing forage base may exist. This is based on lake troutls continued predation on alewives compared to coho and chinook Salmon's increased predation on bloaters and perch. Lack of change might also be explained by lake trout's presumed selective preference 63 for alewife. A realistic approach in viewing the consumption of various prey species by salmonids was by viewing species distribution overlap. Lake trout inhabit a deep water benthic strata of the lake as compared to chinook and coho salmon which are more pelagic. The large occurrence of alewives in lake trout was understandable because of adult alewives mainly concentrating on the bottom during the day (Brandt et al., 1980; Janssen and Brandt, 1980; Crowder et al., 1981). The majority of fish sampled were caught between 6:30 a.m. and 9:00 p.m.. Coho salmon fed more extensively'on smeltg bloaters, perch and juvenile alewives which were more pelagic and littoral. These findings and observations of salmonid and forage species juxtaposition in the lake exemplify'the.salmonids opportunistic nature for feeding on the most abundant and available prey items. Numerous studies by Murdoch (1969 and 1975) have shown "switching" to occurr, increased feeding, by a predator, on prey items which are higher in relative abundance. Opportunistic feeding is further substantiated by the occurrence of various forage species ingested at different locations and depths within the lake. Specific cases include occurrence of: l) gizzard shad (Dorosoma cepedianum (Lesseur)) in salmonids caught at the mouths of rivers flowing into Lake Michigan; 2) trout-perch, 9-spine sticklebacks, sculpins, and mysids at deeper depths primarily in lake trout and; 3) insects in steelhead and coho salmon taken at or near the surface. 64 FEEDING HABITS _B_X REGION AND SEASON A.more diversified sampling scheme in 1984 allowed for regions and seasons to be independently analyzed. In the southern and northern regions of Lake Michigan, fewer feeding fish were found (sampled fish containing food) than in the middle region (Table 14). These values contrasted those of 1983 where the opposite occurred, fewer feeding fish in the middle region. Sampling in early spring occurred more frequently in the southern region, resulting from largers number of anglers present at that time. Practically 90 % of the early spring fish sampled were empty, contributing to a low southern region (mostly spring) percentage. Efforts to reduce the.bias included extending the spring sampling season by two weeks, in order to compensate for the two week difference in feeding intensity (Table 1). And secondly, sampling as soon as possible in the other regions during spring once their fishing began. Anglers pursued the larger, mature chinook salmon in the fall, mostly in.the northern region. Because of mature salmonids exhibiting spawning behavior, the frequency of empty stomachs greatly increased. In spite of spawning behavior salmonids were still caught by anglers. This may have resulted from a split allocation of energy by salmonids towards spawning and food location. In observing the deterioration of the digestive tract in snagged, mature 65 Table 14.--Re1ative percent of feeding salmonid species by reghm1 (sample sizes inparentheses), 1984. SPECIES SOUTH MIDDLE NORTH CHINOOK 8. 33.10 46.10 34.30 35.60 (492) (191) (581) (1264) COHO 3. 42.00 63.20 48.70 47.90 (286) (95) (308) (689) LAKE T. 52.90 63.30 70.00 62.90 (223) (89) (270) (582) STEELHEAD 63.80 95.00 81.30 73.40 (53) (20) (16) (94) BROWN T. 60.50 44.40 50.00 54.70 (38) (9) (28) (75) ATLANTIC S. 00.00 00.00 100.00 100.00 (0) (0) (l) (1) ALL SPECIES 42.00 56.90 47.00 46.50 (1097) (404) (1204) (2705) 66 chinook salmon it was presumed that the majority of mature salmonids, prior to spawning, allocated.more energy towards spawning while still opting for an available meal. A decline in fall feeding fish reflected a reduction in forage availability from 1983 to 1984. The relative percent of feeding salmonids by season followed the above pattern with the exception of trout (lake trout, steelhead and brown trout). They tended to feed more extensively in the spring, declining towards the fall (Table 15). This feeding habit was discussed in a previous section (Wright, 1968; Frantz and Cordone, 1970; Chiotti, 1973). Among feeding fish, smelt were the most abundant forage fish consumed followed by alewives, bloaters, and perch. However, lake trout continued to have alewife as its most numerous prey. Absence of a sufficient sample size for lake trout in the fall resulted from the imposed 0 creel limit after August 15, 1984. Consumption of forage species, alewives and smelt, by chinook and coho salmon, fluctuated with the season in relation to their distribu- tion. Alewife were more abundant when salmonids were primarily pelagic (middle region - summer) and smelt more abundant when salmonids were more littoral (spring and fall - southern and northern regions) (Tables 16 and 17). Nine-spine sticklebacks and sculpins were most numerous during summer in the northern region. Occurrence of various prey items within the diets 67 Table 15.-éRelative percent of feeding salmonid species by season (sample sizes in parentheses), 1984. SPECIES SPRING SUMMER FALL ALL SEASONS CHINOOK S. 28.10 46.60 32.20 35.60 (466) (432) (366) (1264) COHO 8. 42.50 72.50 44.70 47.90 (292) (102) (295) (689) LAKE T. 65.20 62.60 50.00 63.10 (89) (491) (2) (582) STEELHEAD 81.30 67.70 60.00 73.40 (48) (31) (15) (94) BROWN T. 62.10 45.90 66.70 54.67 (29) (37) (9) (75) ATLANTIC s. 0.00 0.00 100.00 100.00 (0) (0) (1) (1) ALL SPECIES 40.00 56.80 38.80 46.50 (924) (1093) (688) (2705) 68 .Hmsmo «m. NN.q mm. 00.0 Hm.H o©.H 00.0 MN.OH 00.0 mm.w mm.q nomon mm.m HH.H 05.6 mm.m Hm.H om.m mo.HH 00.0 mm. nw.ofi MN.H mHummzH m0.ma no.0H Om.NH oo.NH ©O.NH q©.wH oo.m mm.oH NO.HH mm.mH 0N.m .HzmaHz: mq.N mm. NH.N oo.N Hm.H mo.H oo.mH OH.H om.m 00.0 00.0 mmmeo wH.N oo.N mm.q oo.oH mm.¢ 00.0 00.0 00.0 00.0 00.0 00.0 mszmuo mq.m 00.0 cm.n m©.q 00.0 wo.m 00.0 00.0 mo.H 00.0 00.0 szADum mm.o om.o co.H mo.¢ No.m «m.om mo.am mm.m mc.n oo.mH qo.HH momma Nw.MH HH.¢N oq.oH m©.¢N 00.0N mo.NN w©.mm Nm.qq wo.m mm. om.Hm mme msu mo .COwau >n .mocmusooo mo >Ocmsqmuw ucwoummni.mtnmabmh 69 .Hmbmo «m. mo.m NN.< o 00.0 00.0. mm. 00.0 mo.m NN.H 00.x 06.x nomon mm.m 05.0 HH.H o mo.m mm. mN.m mo.q mo.H mq.m oa.oN 00.0 . mHUMmzH mo.MH mo.ma no.0H. o Ho.oa mo.n Hm.MH Hm.MH q¢.HH mo.NH mm.mH HN.NH .HzmQHz: m¢.N qw.m am. 0 mm.w Wm. oo.N mm.H om. NN.H 00.0 mm.H mmmHo wH.N mm.q oo.m o mm.a 00.0 oo.N mm.mH w¢.q 00.0 00.0 00.0 mszmum m¢.n NH.N 00.0 0 on. 00.0 mm.m HH.w 00.0 mfi.m 00.0 00.0 szADUm mm.o mm.ma mm.o o 00.0 qm.q qw.m mm.mm.wq.q mo.w oo.m wo.NH mommm No.MH om.oa HH.@N o om.om Hw.wm 0N.mH mm.mfi mm.qq ma.m Hm. qm.n mmH3 twesmcoo mama“ xmum mDOfium> man mo .sommwm >n fiwocmusooo mo zoomsqwum osmoummnu.ha CHLOE 70 (Tables 18 and 19) parallel observations from the number of prey species consumed (Tables 16 and 17). Further examina- tion of frequency of occurrence (Tables 18 and 19) showed alewife and bloaters were dominant prey species for chinook salmon, as contrasted to alewife and smelt for coho salmon and lake trout. Bloaters showed a marked increase of 300 % in the frequency of occurrence in combined species from 1983 to 1984. Regions were again employed to determine the percent that each prey item contributed to the various salmonid diets. Regions were used for reasons stated in 1983, with the exception of similarity between regions and seasons (Figures 14-16). The major observations from Figures 14- 16 were the smaller percentage of alewife and larger percent of bloater and perch in the 1984 diets. These percentages differed from.previous studies mentioned and that found in 1983. The major presumption for the marked changes in feeding habits were due to regional and seasonal differences among forage species' populations. Statistical differences between 1983 and 1984 were insignificant (p3.10) with the exception of larger numbers of zooplankton being consumed during spring of 1984. FISH SIZE AND PREY SELECTION No significant correlations were found in 1984 between predator and mean alewife lengths. Correlations (r) obtained included: .190, .064, and..297 for chinook salmon, 71 qqqm «OHN mum mom mmHommm AA< How omNH moNH qu mam oow moa Hmm omfi wmm mow mam A3 .COfiwmu >n .vwesmcoo mEmuH zone HauoHnn. .qwmfi .cmwflzow: Oxmu cumummm 6.569 72 qqqm HmmH oqu mmw mmHummm 33< How omNH moNH H mum Nmm m.> omN wmq qu omH mom 3mmm mZOm3 .commmm >3 .Owesmcoo mEmuH >mum HmuoHuu. OH3QH 73 .emmH .cmmflcoflz mead cumpmmm CH COEHmm xoocHnm mo OOHU O33 03 mcH353Huucoo EmuH >Oum HOHOE zoom 30 momucmoummln.VH musmHm '74 x~.o xo.u wbummzL ”wwuzcaeco~ xs.o . ou.1.ezuo.z= wwwumz. szLSOe 35¢ mwrmo zosoum zmuzuxoz musmshzmo_z: x. x6.o ...mw.xuchm mzsam-m muzho uo._ xm.m uzsam m z_¢5=um .wwddushm xm.m . gamma zueue .x..mm LL.2LL¢ _ "xe.ev x..s. umuzmmm «memoJQ xm.m~ «whconm xm.m.111, Nm.es 34mzm 1111 hdmzm Nn.m - zapxzcnmao~ x¢.m zowz2¢4¢oo~ zouomx zxmzhaom zoucwx wqoouz No.9 Nv.m ou—muhzmouza NN.oH xm.m~ :uzua mRHIMJc Nm.om mmuzwgc xv.o. qunw Nm.mm Np.pm x8. ambushm muhmomm egucm 75 CHmummw CH .vmmH .cmmflcoflz mxmq COEHmm 0300 30 ume 033 cu mcH353Huucoo EOOH >mum Ochoa comm mo mmoucmoummul.mH wusmHm 76 xm.NuHm.N zo»xz¢3mDON Nmuvo.m wpummzu wonomx 33¢ omummhzwo~za xm.Nuom.N muzho Nm.mumm.m xv.dum.. 2314206 u . m xm.mum.o mmwmw rumba Nv.HHqu.HH zwhcoqm xv.b mhummzu Nb.m GuamHHZMOHz: zouamm mgooaz Nm.bN mumzmqa NN._H pqmnm xm.w3 «upcoqm “mmgxumkw wzummum zomamm zmmremoz NH.PH mmuzm;¢ Nm.mm hquzm 2036mm zmmzhaom xN.Hv HJuzm Nb.N whammzu Nv.oH omumuhzw0323 Nm.o murbo xv.m .mugxuuhm m2~mwum NN.m zHLJDQm NM :umwm Nm.hH muecoqm xm.m zo>xz¢4moo~ Nm.vH ou—mnhzwomz: Nm.o map¢64m 77 cumumwm CH .emma .cmmflcoflz 0983 39033 mme 30 OOHU 033 03 mcH353HHOCOO EmuH >mum HommE 30mm 30 mompcooummnn.©H musmHm wzouauz 33¢ NN.wm uuuzw3¢ & zo—Guz u3oo~z '78 Nh.mv mmHIM3¢ NH.o zowxzc3moo~ Nb.H whuwwzu NH.m om~m~hzuoaz= NH mmzho Nm.H .mmu3xuubm wzummlm Nm.m z~m3=uw xv.v :umum Nm.m zwh¢o3m NoH h3wzm NH.D zowzzm3moo~ xw.m whomwz3 x3.m omummpzmouz: Xb.o mwzpo uH.m -umw3xumhm mszmlm zo~¢mm.zmu:pmoz xw.NH .xm.mm mm3¢OLm ULHzLL¢ N3.o xn.o zohxz¢3aoo~ phmzm . . Nee. Rm 3 mpummzs meummzs . Nm.o3 RN o. ousmshzuo322 zo36u¢ zmuzesom wwwmshzuoszs NNOD C muzho mmzho x~.¢ xu.. 2313=um . 2313=um xm.m comma N .m« xc :wmma m33¢93m xv.om “1.2L3q 4e.m. xm.33 ~3mnm «umeEm Na.» h3mrm 79 coho salmon, and lake trout, respectively. Scatterplots depicting lack of correlation for chinook salmon and lake trout are illustrated in Appendix D. Although no significant correlations were found, a few generalizations can be made from the number of salmonids (per size category) consuming alewives (Figures 17-18). Chinook salmon between 70 and 108 cm more frequently consumed alewives than those 50 to 70 cm, and especially those from 25.4 to 50 cm (Figure 17). The number of chinook salmon consuming alewives was shown to increase with increasing salmonid length. Lake trout showed fewer alewives being consumed between predator lengths of 50 and 75 cm (Figure 18) . Further examination of predator-alewife interactions reveal a large consumption of 120 to 200 mm alewife by all salmonids, with larger alewife (180-240 mm) being consumed primarily by chinook salmon (Figure 19). Lake trout, however, show a tendency to feed more on slightly smaller alewives (<200 nm0(Figure 20). In considering the size dis- tribution of predator species sampled, no difference in alewife size classes consumed.was evident. Predation on zooplankton increased.as salmonid size decreased (Figure 21). In many samples, the diets of smaller salmonids (25.4-45 cm) were packed with microcrustaceans, especially Daphnia spp. Stomach content analysis of steelheads showed that insects, while present throughout the size range, were preyed on primarily by larger individuals (60-95OHQ HOCHO UCO OHHBOHC CO COHumomum mmmHO ONHm COEHmm MOOCHCUII.>H musmHm 81 FEEDING 0N HNY GIVEN ITEI'I \R FEEDING 0N FILENIVES I R\\ R\\\\\\\ 1 \\\\\\\\\\\N (\\\\\\\N N\\\\\\\\ RV (\\\N (\m \ I 1 I I V \‘ I RN I I I O O C) O NOW'IVS )IOONIHO :IO HEBWflN 35.2 47.5 52.5 57.5 52.5 67.5 72.5 77.5 82.5 87.5 92.5100.0 CHINOOK SALMON LENGTH CLASS MIDPOINT(CM) 82 mg Chmwmmw CH memuH .qmmH .cmchon >mHQ HOCHO UCC OMHBOHO HO COHOOCOHQ mmeO ONHm 33033 @3031I.mH mHsmHm FEEDING 0N RNY GIVEN ITEM FEEDING ON RLENIVES [I [§_ \\\\\\\\\\\\\ N\\\\\\\\\\\\\\\\N- R\\\\\\\X\\\\\\\\ \\\\\\\\\\\\\\\\\\\\\V MN \ I 35.2 47.5 52.5 57.5 62.5 67.5 72.5 77.5 82.5 87.5 92.5100.0 C) C O O C) (D <’ N 10081. 3)W"l :IO HSGWI’IN LAKE TROUT LENGTH CLASS MIDPOINT (CM) 84 Cumummm CH .6663 .cmmHnon 6x64 mmmmmHO ONHm mcoHHm> HO COEHmm xOOCHCO >3 OOECmCOO OHHBTHO HO 330C03 cmmz11.mH chmHm 85 ‘ E Q“ I m I W QLL SPECIES ’zQ CHINOOK SRLNON NV RN 14D .. o “o o N o a: O (D %_ \_ N V \ \_. E o o o <- N SEAIMS'IV ONIWDSNOO HBSWDN 110 130 150 170 190 210 230 MEAN ALEWIFE LENGTH CLASS MIDPOINT (CM) 70 90 45 86 .6663 .26636032 mme Cumummm CH mowmmHO ONHm mCOHCm> 30 35033 @363 >3 COEDmCOO OHHEOHO HO 33mC®3 COOZII.oN mHsmHm 87 I I §_ I R\\\\\\\\\\\\\\r W— N\\\\\ s_ 110 130 150 170 190 210 230 MEAN ALEWIFE LENGTH CLASS MIDPOINT (CM) 1‘3.— 85’: \ 5'3 03— k COM 3:: 9’55 Q”? \ Q %- :2 I O O C O O O D C) V N O m (D ‘1' N fl SSAIME'IV ONIWDSNOO BBSWDN 88 .6663 .66636032 6963 cumummm CH >Hom®3mo COHRCCHQOON OCH CHCHHS meHCmmHO CO COHDOOOHQ mmmHO ONHm CHCOEHmm11.HN mHDmHm WALL PECIE . . // W- \\\\\\\\\\\\\\_g 00000000 c-a . c-n NOIMNV'IdOOZ NO ONIGBBd BESWDN 67.5 72.5 77.5 52.5 57.5 SALMONID LENGTH CLASS MIDPOINT(CM) 9O 22). Conversely, coho salmon between 50 and 60 cm in length showed a larger percent feeding on insects (Figure 23). FEEDING HABITS EX WATER DEPTH The majority of fish examined were caught between 0 and 40 meters of water. Lake trout and chinook salmon were mostly taken from deeper water correlated to their population dynamics and physiological needs (Sommers et a1. 1981; Eck and Wells 1982). The distribution of salmonids feeding on alewife and smelt followed its prey; chinook and coho salmon fed more extensively on smaller alewife (<85 mm) and smelt (30-180 mm) in the shallower regions than in deeper waters where more larger alewife (>85 mm) were consumed (Tables 20 and 21). Larger numbers of smaller fish were consumed in shallower waters because of their greater abundance and relatively larger numbers needed to satisfy a salmonid feeding. In shallower regions diversity among prey species ingested was highest resulting from numerous forage species inhabiting the region. Forage fish fed primarily on zooplankton and invertebrates which were most abundant in these areas. 91 .6663 .66633032 6363 6363666 63 6306663 so 663366636 66630 6636 66633663611.66 639636 92 who m.Nm mék. m.NH. m.bm m.Nm mixm m.Nw 9:. N.mm 2203 ...Z_On_O:2 mw<30 :._.G2m3 0.20.2.2...» &R& R\\ \ \ 03 m3 ON mN om SIOBSNI NO ONIGBBd HBSWDN 93 .6mmH .CmmH3on @363 CHmummm CH mCOOwCH CO COHCCUOHQ mmmHO ONHm COEHmm OCOUII.MN mHzmHm 94 2203.62.06.95. mm<3o :hwzm3 0.20.2335 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 _ - L _ _ H _ 3E zo=3¢m oroo munummm 33¢ _ \\\\ \k 6\\\\\\\\\\\- 6\\\\\\\\\\\\\\\ V6 . _ \\\\\ OH m3 ON mN om SIOBSNI NO ONIOBSd HBSWHN Table 20.--Salmonid distribution and predation on alewives 95 (ALW) by depth in eastern Lake Michigan, 1984 (numbers consumed in parentheses). WATER DEPTH CHINOOK fl/ALW COHO W/ALW LAKE T; WZALW 0-10 feet 2 0 5 0 0 0 (0) (0) (0) 11-20 19 1 0 0 0 0 (l) (0) (0) 21-30 64 6 87 15 2 1 (9) (35) (1) 31-40 79 16 33 6 9 5 (28) (25) (14) 41-50 78 12 21 8 39 12 (40) (14) (19) 51-60 85 11 18 7 71 38 (28) (12) (96) 61-70 68 14 11 4 71 34 (16) (7) (81) 71-80 85 10 46 0 68 36 (ll) (0) (73) 81-90 61 12 45 3 39 12 (18) (5) (22) 91-100 41 7 46 0 28 10 (15) (0) (17) 101-110 33 2 5 1 41 7 (3) (l) (11) 111-120 48 11 10 4 21 7 (13) (9) (13) 121-130 9 0 2 2 ll 5 (0) (4) (6) 131-140 0 0 0 0 l 0 (0) (0) (0) 141-150 11 2 0 0 5 2 (2) (0) (2) 151-160 14 1 0 0 5 3 (1) (0) (6) 96 Table 20 (cont'd). 161-170 2 0 0 0 1 0 (0) (0) (0) 171-180 13 2 1 0 8 3 (5) (0) (4) 181-190 4 0 1 1 4 0 (0) (1) (0) 191-200 0 0 0 0 1 0 (0) (0) ~ (0) 221-230 10 5 1 0 1 0 (16) (0) (0) 231-240 0 0 1 1 0 0 (0) (l) (0) 241-250 3 0 0 o 0 0 (0) (0) (0) 261-270 1 0 2 0 0 0 (0) (0) (0) 291-300 1 1 0 0 0 0 (l) (0) (0) 381-390 1 0 0 0 0 0 (0) (0) (0) 531-540 1 0 0 0 0 0 (0) (0) (0) UNKNOWN 531 89 342 27 157 62 (192) (49) (147) 97 Table 21.--Salmonid distribution and predation on smelt (SMLT) by depth in eastern Lake Michigan, 1984 (number consumed in parentheses). WATER DEPTH CHINOOK W/SMLT COHO W/SMLT LAKE T. * 0-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 101-110 111-120 121-130 131-140 141-150 151-160 49 19 64 99 78 85 68 85 61 41 31 46 11 2 (3) 1 (1) 6 (ll) 8 (266) 4 (6) 3 (67) 5 (6) 3 (5) 7 (20) 2 (2) 2 (30) 2 (4) 0 (0) 0 (0) 0 (0) 0 (0) 61 87 33 21 18 11 46 45 46 22 5 (7) 0 (0) 8 (30) 6 (83) 4 (10) 2 (4) 3 (6) 14 (227) 4 (39) 8 (63) 0 (0) 6 (46) 0 (0) 0 (0) 0 (0) 0 (0) O 39 71 71 68 39 28 41 21 11 W/SMLT 0 (0) 0 (0) 0 (0) 1 (2) 1 (2) 7 (9) 5 (14) 11 (17) 1 (1) 0 (0) 3 (ll) 0 (0) 1 (1) 0 (0) 0 (0) 1 (2) 98 Table 21 (cont'd). 161-170 2 0 0 0 l 0 (0) (0) (0) 171-180 13 1 l 0 8 1 (l) (0) (1) 181-190 4 O 1 0 4 1 (0) (0) (1) 191-200 0 0 O 0 l 0 (0) (0) (0) 221-230 10 O 0 0 l 0 (0) (0) (0) 231-240 0 0 l 0 0 0 (0) (0) (0) 241-250 3 l 2 l 0 0 (l) (l) (0) 261-270 1 0 0 0 O 0 (0) (0) (0) 291-300 1 0 0 0 0 0 (0) (0) (0) 381-390 1 0 0 0 0 0 (0) (0) (0) 531-540 1 O 0 0 0 0 (0) (0) (0) UNKNOWN * * * * * * (*) (*) (*) 99 GRAND TRAVERSE BAY GENERAL FEEDING HABITS, 1983-84 In 1983, the stomachs of 57 salmonids from Grand Traverse Bay, predominantly lake trout (94.74%), were examined for food contents. Smelt was the major prey item consumed (89.13%) while alewife and bloater occurred respectively inZUL87% and 2.7% of the salmonids representing 8.15% and 0.54% of total diet (Table 22). Analysis of 121 salmonids in 1984 showed similar feeding habits to those found in 1983 with lake trout again dominating the sample (93.39%) (Table 23). Smelt constituted the major portion of the diet (76.73%) occurring in 82.05% of the fish sampled in 1984 (Table 22). This value significantly correlated (r=.99) with 82% in 1983. The only noticeable differences between the two years were: 1) a larger proportion of sublegal lake trout taken in 1983 correlating with a larger percent of feeding fish in 1983 and; 2) the occurrence of a Whitefish in a 1984 lake trout stomach. A large percent of insects in the 1984 diets correlated with a Hexagenia hatch at the time of sampling. These comprised 95% of the insects consumed. Because of small sample sizes, and large variability, little can be postulated about feeding habits, with the exception of smelt dominated diets in 1983 and 1984. Furthermore, diversity of prey items ingested reflected the opportunistic nature of feeding salmonids. 100 Table 22.--Percent frequency of occurrence of various prey items consumed by Grand Traverse Bay salmonids, 1983-84 (percent of diet in parentheses). ALL SPECIES-1983 ALL SPECIES-1984 ALEWIFE 10.87 10.26 (8.15) (6-52) SMELT 89.13 82.05 (85-33) (76.73) BLOATERS 2.17 1.28 (0.54) (1-28) OTHER 0.00 1.28 (0.00) (0.43) UNIDENTIFIED 8.70 10.26 (5.98) (6-63) INsECTs 0.00 12.82 (0.00) (8-41) Table 23.--Species distribution of Grand Traverse Bay salmonids, 1983-84. CHINOOK SALMON COHO SALMON LAKE TROUT 1983 0 5.26 % 0 94.74 % 1984 2 4.96 % 1.65 % 93.39 % 101 LAKE SUPERIOR GENERAL FEEDING HABITS, 1983-84 Between 1983 and 1984, the stomach contents of 158 salmonids from Lake Superior were examined, with 113 (71.52 %) containing prey items. This value, however, does not illustrate the percentage of feeding fish in Lake Superior because of selective analysis on only feeding fish in 1983 and one major sample collection in the fall of 1984. The latter resulted in 45 (50.56 %) feeding fish and 44 (44.49 %) non-feeding fish (empty). A better estimate for the percent of feeding fish between 1983 and 1984 was 10-20 % (Kinnunen and local anglers, personal communication). Stomach contents from Lake Superior salmonids showed a high diversity among prey items consumed. The wide array and presence of prey items found in salmonid diets from 1983 and 1984 are illustrated in Table 24. High diversity among diets refelct Lake Superior's smaller forage base (Lawrie and Rahrer 1972). In 1983, 9-spine sticklebacks showed the highest frequency of occurrence in diets with 37.5 % followed by smelt 33.33 % and insects 29.17 % (Table 24); insects were most prevalent in coho salmon and steelhead. Chinook salmon and lake trout fed more on bottom dwelling species such as mysids, sculpins, and 9-spine sticklebacks. In 1984, smelt was the most frequently consumed prey item (53.93%) followed by insects (21.35%) and mysids (12.36%) (Table 23) . 102 Table 24.--Percent frequency of occurrence of various prey items consumed by Lake Superior salmonids, 1983- 84. ALL SPECIES-1983 ALL SPECIES-1984 ALEWIVES 12.50 0.00 SMELT 33.33 53.93 BLOATERS 0.00 2.25 LAKE HERRING 0.00 1.12 SCULPINS 12.50 10.11 9-SPINE STICKLEBACKS 37.50 8.99 BURBOT 4.17 3.37 LAKE WHITEFISH 8.33 2.25 ROUND WHITEFISH 0.00 2.25 COHO SALMON 4.17 0.00 BROWN TROUT 0.00 1.12 LAKE TROUT 8.33 2.25 UNIDENTIFIED 8.33 4.49 MYSIDS 4.17 12.36 INSECTS 29.17 21.35 103 Several sport and commercially fished species were represented in the diets, namely coregonids and salmonids. The data indicates that a more diverse forage base, as represented by the larger array of species consumed, exists in Lake Superior than in Lake Michigan. In Lake Michigan four fish species (alewife, smelt, bloater, and perch) constituted the majority of salmonid diets as compared to Lake Superior with seven species (alewife, smelt, sculpins, 9-spine sticklebacks, lake Whitefish, and lake trout). SUMMARY AND CONCLUSIONS Study of salmonid feeding habits in eastern Lake Michigan led to the following conclusions. Salmonids fed more often during spring and summer of 1983 than of 1984. The majority of fish found feeding in the fall were immature. From 1983 to 1984 there was a decrease in number of feeding salmonids; chinook salmon declined 72.1%; coho salmon, 43.3%, and lake trout, 1.08%. The major food of salmonids was fish, with the exception in spring 1983 for coho salmon and in 1984 for steelhead. Cohos fed primarily on zooplankton, Qaphnia spp., and steelhead on insects. Alewife and smelt made up the majority (73.2 %) of salmonid diets in 1983, while large quantities of zooplankton in coho salmon and in smaller chinook salmon constituted an additional 21.5 %. In 1983, alewife and smelt made up the bulk of chinook salmon diets (77.12 %), lake trout diets (82.47 %), and coho salmon diets (41.52 8;). ZooplanktOn in the spring represented 40.07 % of the coho diet. Four fish species (alewife, smelt, bloater, and perch) made up the bulk (78.1 %) of 1984 salmonid diets. Salmonids ate fewer alewife in 1984 than in 1983 with frequency of occurrence in chinook salmon being 64.24% during 1983 and 37.50% in 1984; coho salmon, 20.70% compared to 19.00%, and lake trout 69.72% compared to 51.30%. 104 105 Organisms within the zooplankton category showed a significantly higher frequency of occurrence in 1983 (23.72%) than in 1984 diets (2.63%). In 1983 and 1984, seasonal shifts of species consumed indicated more littoral fish and zooplankton being taken in spring (smelt and Qaphgia spp.), pelagic fish in summer (alewives, bloaters and perch), and a mixture of littoral, pelagic and benthic species in the fall (alewives, smelt, bloaters, perch, sculpins, 9-spine sticklebacks). A higher diversity of prey items consumed was seen in the fall. Regional shifts of species consumed correlated with seasonal shifts in 1983. Higher frequency of occurrence of sculpins, 9-spine sticklebacks, and Whitefish were reported in salmonids of the northern region. Larger numbers of smelt were consumed in spring of 1983 coinciding with a large spawning smelt run occurring at the same time in the southern region of Lake Michigan. From 1983 to 1984, an increase in the consumption of bloaters and perch was found to occurr throughout the lake with larger numbers being ingested in the middle region. Lack of correlation between salmonid length and alewife length reflects salmonids opportunistic feeding behavior, in addition to on-going changes in alewife population dynamics. Kitchelljv (1984) forage base models predict that with a decline in forage base, prey popula- tions will be dominated by smaller, younger fish. Further- more these fish will show up in salmonid diets giving predator - prey length correlations 0 to negative values. 106 Forage base structure in eastern Lake Michigan is believed to be moving towards a higher abundance of smaller prey species, i.e. more juvenile alewife. Chinook salmon in 1983 showed a higher affinity for consuming larger size classes (>67.5 cm) of alewife, while chinook salmon in 1984 consumed a broader size range of alewife. Lake trout showed no significant change in size range of alewife consumed from 1983 to 1984. Mean alewife lengths consumed in 1983 were not significantly different from those consumed in 1984 unnOB), although a larger abundance of smaller, juvenile alewife were consumed in 1984. In 1983, predation on organisms within the zooplankton category was primarily by coho salmon (50 - 60 cm) in spring (mostly southern region). An inverse relationship between the occurrence of zooplankton in salmonid diets and predator length was found. steelhead, regardless of size, ate insects; however, insects were found more frequently in steelhead from 65 to 90 cm. The majority of smelt and smaller alewife were consumed by salmonids caught between 3 and 15 meters below the surface of the water and in 7-40 meters of water. Larger alewife were consumed in deeper“waters correlated with their distribution. Grand Traverse Bay salmonids fed predominantly on 107 smelt and illustrated their opportunistic nature to feed on any available prey items. Coho salmon showed a quicker response (more opportunistic behavior) to the changing forage base than did lake trout, which may have been more selective on various species (e.g. sculpins and alewife). Study of salmonid feeding habits in southern Lake Superior led to the following conclusions. Salmonids reflected Lake Superiors' smaller forage base, in comparison to that of Lake Michigan. Smelt were the predominate prey item. Lake trout dominated the samples and reflected its benthic nature by consuming relatively large numbers of 9-spine sticklebacks and sculpins. Changes in salmonid feeding behavior appears to reflect changes in the forage base. The decline in alewife as a predominant prey item in eastern Lake Michigan salmonids may depict a reduction in the alewife population. Furthermore with increasing diversity and numbers of other prey items consumed. Fewer feeding salmonids were recorded in 1984 than in 1972 (Chiotti, 1973) and 1983. This reduction in predator- prey encounters may indicate a decline in forage availability. Thus the question of forage exploitation by salmonid predators arises and future redUCtion in salmonid size. Several biologists and I feel this has been the trend. Further study will be necessary to better assess the salmonid fishery as well as answering the question: Do we want a quality or quantity fishery? -// LITERATURE CITED Baldwin, N.S. 1968. Sea lamprey in the Great Lakes. Limnol. 1(3):20-27. Birge, ELA. and C. Juday. 1922. The inland lakes of Wisconsin. The plankton, Part 1. Its quality and chemical compostion. Bull. Wisc. Geol. Nat. Hist. Surv. 64: 1-222. ' Brandt, SJB.1978. Thermal ecology and abundance of alewife (Alosa pseudoharengus)inLakeMichigan.Doctoral dissertation. Univ. Wiscfl, Madison. Brandt, S.B., J33. Magnuson, and 16B. Crowder. 1980. Thermal habitat partitioning by fishes in Lake Michigan. Can. J. Fish. Aquat. Sci. 37:1557-64. Brown, EJL,Jr. 1972. Population biology of alewives, Alosa pseudoharengus, in Lake Michigan, 1949-1970. J.Fish.Res.Bd.Can. 29:477-500. Chiotti, 12L. 1973. A study of the effects of installing and operating a large pumped storage project on the shores of Lake Michigan near Ludington, Michigan. M.S. thesis, Mich. St. Univ., East Lansing, Mich. 62 pp. Christie, W61. 1974. Changes in the fish species composition of the Great Lakes. J.Fish.Res.Bd.Can. 31:827-854. Crowder, 16B., J.J. Magnuson, and S.B. Brandt. 1980. Complementarity in the use of food and thermal habitat by Lake Michigan fishes. Can. J. Fish. Aquat. Sci. 38: 662-668. Crowder, 168. and JuJ. Magnuson. 1982. Thermal habitat shifts by fishes at the thermocline in Lake Michigan. Can. J. Fish. Aquat. Sci. 39:1046-1050. x/Crowder, L.B. and F.P. Binkowski. 1983. Foraging behaviors and the interaction of alewife, Alosa pseudoharengus, and bloater, Coregonus hoyi. Env. Biol. Fish. 2:105- 113. Eck, G.W. and L. Wells. 1983. Biology, population struc- ture, and estimated forage requirements of lake trout in Lake Michigan. Technical papers of the‘UAL Fish and Wildlife Serv. No. 111. 108 v/ 109 Eshenroder, ILL. 1977. Effects of intensified fishing, species changes, and spring water temperatures on yellow perch, Perca flavescens, in Saginaw Bay. J. Fish. Res. Ed. Can. 34:1830-1838. Frantz, T.C. and A.J. Cordone. 1970. Food of lake trout in Lake Tahoe. Calif, Fish and Game, 56 (1): 21-35. Gailbraith, Jr. MQG. 1967. Size-selective predation on Daphnia by rainbow trout and yellow perch. Trans. Am. Fish. Soc. 96 (1): 1-10. Hatch, R.W. and E.H.Brown,Jr. 1978. Population Dynamics: Dynamics and ecology of Lake Michigan alewives. ProgressinFflshery Research 1978. Great Lakes Fisheries Lab, U.S. Fish Wildlife Service, Ann Arbor, Mich. Janssen, J. and S.B. Brandt. 1980. Feeding ecology and vertical migration of adult alewives (Alosa pseudoharengus) in Lake Michigan. Can. J. Fish. Aquat. Sci. 37:177-184. Kitchell, J.F. 1984. gleeting of salmonid diet survey participants- 114 AnnualAmericanFisheriesSociety Meeting, Cornell, N.Y. Lagler, K.F. 1956. Freshwater fishery biology. W.C. Brown, Dubuque, Iowa, USA. Lawrie, A.H. and J.F.Rahrer. 1972. Lake Superior: Effects of exploitationandintroductionson the salmonidcommunity. J. Fish. Res. Ed. Can. 29:765-776. McComish, T.S. and W.G. Miller. 1976. Notes on the biology of the lake trout and other selected salmonidae in Indiana Waters of Lake Michigan. Proc. of the Indiana Acad. Sci.(l975). Vol. 85. Mellors, VLK. 1975. Selective predation on ephippial Daphnia and the resistance of ephippial eggs to diges- tion. Ecol. 56: 974-980. Miller, R.R. 1957. Origin and dispersal of the alewife, Alosa pseudoharengus, andthegizzardshad, Dorosoma cepedium, in the Great Lakes. Trans.Am.Fish.Soc. 86:97-111. Murdoch, WRW. 1969. Switching in general predators: Experiments on predator specificity and stability of prey populations. Ecol. Monogr. 39:336-354. Murdoch, W.W., S. Avery, and M.E.B. Smyth. 1975. Switching in prdatory fish. Ecol. 56: 1094-1105. 110 Nelson, W.R. and C.H. Walburg. 1977. Population dynamics of yellow perch (Perca flavescens),sauger(Stizostedion canadense), and walleye (Stizostedion vitreum vitreum) in four main stream Missouri River reservoirs.J.Fish. Res. Bd. Can. 34:1748-1763. Paloheimo, J.F. 1979. Indices of food type preference by a predator. J. Fish. Res. Bd. Can. 36:470-473. Peterson, W.T. R.D. Brodeur, and W.G. Pearcy. 1982. Food habitsof juvenile salmon in the Oregon Coastal Zone, June 1979. Fishery Bulletin, 80(4):841-851. Richman, S. 1958. The transformation of energy of QEthgg pulex. Ecol. Monogr. 28: 273-291. Ricker, W.E. 1975. Computation and interpretation of biological statistics of fish populations. J.Fish.Res. Bd. Can. Bull. 191. pg. 280. Rottiers, D.V. and R.M. Tucker. 1982. Proximate composition and caloric content of eight Lake Michigan fishes Technical papers of the U.S. Fish and Wildlife Serv. No. 108. Rybicki, R.W. and M.Keller. 1978. The lake trout resource in Michigao _a|ers of Lake Michigan, 1970-1976. MichifanDepartme~t of Natural Resources, Fisheryes Research Report 1863. Ann Arbor, Mich. Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-hill Book Co., Inc. pp. 104-111. Smith, S.H. 1964. Status of the deepwater cisco population of Lake Michigan. Trans.Am.Fish.Soc. 93(2):155-163. Smith, S.H. 1970. Species interactions of the alewife in the Great Lakes. Trans.Am.Fish.Soc. 99:754-765. Smith, S.H. 1977. Walleye (Stizodeon vitreum vitreum) and yellow perch (Perca flavescens) populations and fisheries of the Red Lakes, Minnesota, 1930-75. J. Fish. Res. Bd. Can. 34:1774-1783. Smith, S.H., H.T. Buettner, and R. Hile. 1961. Fishery statistical districts of the Great Lakes. Great Lakes Fish. Comm. Tech. Rept. No. 2, 24 pp. Sokal, R.R. and F.J. Rohlf. 1969. Biometry. W.H. Freeman and Co. 859 pp. 111 Sommers,L.M., C.Thompson, S.Tainter, L.Lin, T.W.Colucci, and J.M. Lipsey. 1981. Fish in Lake Michigan-Distribu- tion of selected species. Michigan Sea Grant Program. MICHU-SG-81-600. Stewart, D.J., J.F.Kitchell, and L.B.Crowder. 1981. Forage fishesand theirsalmonid predatorsinLakeMichigan. Trans.Am.Fish.Soc. 110:751-763. Taylor, W.W. and S.D. Gerking. 1980. Population dynamics of Daphnia pulex and utilization by the rainbow trout (Salmo Gairdneri). Hydrobiologia 71: 277-287. Tody, W.H. and H.A.Tanner. 1966. Coho salmon for the Great Lakes. Mich.Dept.Cons., Fish Mgt. Rep. 1. 38 p. VanOosten, J. and H.J. Deason. 1938. The food of the lake trout (Cristivomer namaycush) and of the lawyer (Lota maculosa) of Lake Michigan. Trans. Am. Fish. Soc. 67: 155-157. Wardle, (LS. 1977. Effects of size on the swimming speeds of fish. p. 299-313 in T.D. Pedley, ed. Scale effects in animal locomotion. Academic Press, New York, N.Y. Wells, L. 1968. Seasonal depth distribution of fish in southeastern Lake Michigan. U38. Fish and Wildlife Serv., Fish. Bull., 67 (1): 1-15. Wells, L. and A.M.Beeton. 1963. Food of the bloater, Coregonus hoyi, in Lake Michigan. Trans.Am.Fish.Soc. 92:245-255. 'Wells, 16 and A.McLain. 1972. Lake Michigan: Effects of exploitation, introductions,andeutrophicationonthe salmonid community. J.Fish.Res.Brd.Can. 29:889-898. Wells, 16 and Hatch, RJL 1983. Status of bloater chubs, alewives, smelt, slimy sculpins, and deepwater sculpins in Lake Michigan, 1982. Presented in the Minutes from the Great Lakes Committee-1983. Werner, E.E. 1974. The fish size, prey size, handling time relation in several sunfishes and some implications. J. Fish. Res. Bd. Can. 31:1531-1536. Werner, E.E. and D.J. Hall. 1974. Optimal foraging and the size selection of prey by the bluegill sunfish (Lepomis machrochirus). Ecology 55:1042-1052. Wright, KJL.1968. Feeding habits of immature lake trout (Salvelinus namaycus) in the Michigan waters of Lake Michigan. M.S. Thesis, Michigan State University, .East Lansing. 42 pp. APPENDIX A To determine total length from caudal peduncle (CP) length the following derived equations were utilized: ALEWIFE For CP: SMELT For CP: BLOATER For CP: >70 <50 <70 >50 <50 <30 >50 <50 UNIDENTIFIED For CP: >50 <50 Total Total Total Total Total Total Total Total Total Total 112 length length length length length length length length length length 13.5068 + 1.482 (CP) 10 + 1.482 (CP) 6 +1.482 (CP) 3.61 + 1.38 (CP) 3 + 1.38 (CP) 1.5 + 1.38 (CP) -1.6 + 1.59 (CP) —3 + 1.59 (CP) 5.172 + 1.484 (CP) 3 + 1.48 (CP) 113 .16663 .60366 6:6 633631 mCmOCHCmHmm CH C3036 mum mHm>HUDCH wOCwUHHCOO “lem6mH 3363 .CmmHCOHz @363 CH 633633 503303 03 03363Hm>m mmHoQO mmMHOH HQ 36:03 OHHCUE HO mpcmmsonuv mmmeoH3 CmumsHumm11.m prCwmm< 114 AOHHIQCV mh How.o|hp.ov mm.c >c>.N10—._3 sm._ Aw.H¢1@.o—3 m.0m Am.QIN.H. m.m Am.hnl_.vN3 m.ov Nmap Ahmptmov mN— Ham.»1m—.cv ac.o Ham. 1mm.—3 om.N “H.3m1—om—3 3.5N Av.0mlo.mpv O.Nm ma.vo|m.zmv 0.35 Hom— AHu—Imm. mm ANm.—le.ov Nc.— ACN.mImh... w¢.m AC._N13.NHV 5.3— AO.NNIO.33 D.V. Am.mol$.pmv m.h¢ 0mm— A¢mplhhv mop Amm.—Imm.o. Ho.— ANo.Oxhv.ov 35.3 .Ac.3mlm.0—3 N.m— Am.hlo.N3 —.m "op—Ip.m®v m.wo mhmp Ammpcmhv wop Amm.oqu.ov No.0 Am6.01mv.o3 No.0 .c.am:m.__3 p.mp AN.oNlo.o. v.v3 Am.mmav.mm3 3.36 . mhmp 33xlm¢3 mo >m~.313m.03 no.0 “mm.OINH.o3 93.6 Ac.o—up.6v c._— 36.031H.m3 o.© 36.Nmzo.6m3 m.vv >6¢3 Hmmspmv 06 Hom.mtmm.—. ._.N 336.010m.ov.hm.c AN.¢_10.33 m.c. 36.9—1m.m3 N.6 36.Nm16.mm3 m.mv 06m— Amwplvopv mm. Hom.m10m._3 ¢M.N Hom.onmp.cv mm.o Ac.:.1m.mv p.m— Hp.m31m.63 m.m3 HNm—Io.mov a.mo— mhm— 303316033 6m, Acm.paom.ov 6m._. Hm6.016m.o3 om.o 30.6.1N.33 a.mp >6.stw.033 0.0m HocHIm.mhv 3.m¢ v6m3 3mm31mop. «mp Ham.31mm.ov m¢.o Hem.01mp.03 mm.o Hm.c~1r.c3 6.6— . >0.o1o.33 o.m A_m_16.6m3 v.63— mum— CmmuCu mCHQHsom Csm>103313017:so> 33:33 Cuo>1vrc1uciaczc> u3sw4 mCHHQEnm .III II tl.‘ l|'|i.| III...’ .I... I)..." II II")..- I. .I‘n Olil.....|.l.all‘ HauOH >EHHm uHusm nO>Hmmm< «0 Cam» 115 msmuw> 3:033 @363 vmmH pCm .665366 3606360 6663 .>Hm>HuommmOH .CumCmH mmHzmHm .COEHmm xoOCHCO mmmH mo muonHmuumomul.o xHUCOmQ< 116 DNH o: _ on: _ A203 IP02m3 ZOE-Em XOOZEO om _ om _ or _ om _ ow _ 9. cm _ _ 0N ON. ":09 cm mb 9: mNH omH m: cow mNN (WIN) HIONE'I 3:)IM3'IV 117 ONH OHH _ OOH _ OO _ OO _ — OO om _ =2“: IHOZM3 205.345 v_OOz=._O on O6 om ON cm... 6.200 U) Q l‘ LO Q 9. N31 3:!IM3'IV mN3 118 AEOV :h02m3 #30:... mv_<3 OHH OOH Om O0 OH. . OO om Ov on _ _ _ _ _ L _ _ _ ON D ID 0 U) D :9. 5 a 9 *0 (WW) HLONE'I 3:"M3'IV W F .— OON mNN 119 N O O O O O O O H H O O O H O O O O O O O O H O O O m O O O O O O O O H O H H O O O O O O O O H m O H H N O O O O H O O O O O O H q H H O O O H O O H O O O OH O O O O O H N H q 6 H O NN H H O O O N m N m 6 H O O O O O H O H m H O O O O HO O O O O N N m HH 6 N O N O6 O O O N m OH NH ON OH 6 N O OHH O O O m O OH mN mm NH O N N «MN O O H O m Om HmH Om N O O O NH O O O O O O m O O N H O OMH O O O O O .O N O mm HO O m me O O H O m Om OmH _OO mm mm O OH HmH ON mm Om OH O OH OH N N O O O we N m 6 6 6 m m m c H m O ONH O 6 m ON ON OH OH ON ON NH H OH 66m Om m3 m6 Om mm Nm ON 6N Om NN q NN 33< OOHIOO mOIOO OOumw mmaow Omumm mmnom OmumO OOIOO OOamm mmnom Omumq manq.mN AZOV OHOZNH MOH 633 HO CoHus3H336HO 66Hmnn.O xHOCOOQ< 120 O OM ON M6 OH ON OO .66 H OOO 6O NmO ONN OO NON MOO OOM MHO NOO O O O O O O O O O O O O O O O O O O O O O H H H M O O O M OH O O N M O O H 6 6 OH O O O O O H O H O HH O O H M M H O N 6 OH O O H O N O O O O H O O O H O O O H H M OO OO OO OO ON OH 6N OO OO 6N O ON OM NO OO OMNH MO OHH OOH NOH OHH o 63 63 66 66 66 MNHHOO NOH OO O O O6 OH OM OO OOH OOmov-Ix—Iov—IO OOH ON ONH nMH OM O ONH NH 6N Hm OHH ONMHMNQ’NH OMH 6 mMH OO OM HO ONH 6 OH OO NO O OH ON OH OH NO OO O O OOH OON O OH NH OM H OOWQ’ NH N NMHNOQ‘Q’O N O 66 66 63 66 6 66 v-I OMMQQHMOHHOOQ’NC‘C") C‘ H OO OO OOH HH< OOHIOO OOIOO OOIOO OOIOO OOION ONION ONIOO OOIOO OOIOO OOIOO OOIOO OO:0.0N Azov OBOZMH OOB 633 O0 COH3:3H336HO 66HO11.O xHOCOOO< 121 .mOOH .COOHCOHZ @363 Cumummm CH muOOmCH CO COHHOOOHQ ONHm UHCOEHmm11.m xHOCmmm< 122 AEOO 5.2.0630;— mw<30 Ih02m3 n:20-23