CLEAR TECHNICAL REPORT NO. 295 Effects of Toxic Substances on Growth, Mortality and Pathology of Larval Fishes in the River Raisin, Michigan E nvironmental Protection Agency Large Lakes Laboratory Grosse lie, Michigan T!f OHIO STATE UNIVERSITY CENTER FOR LAKE ERIE AREA RESEARCH COLUMBUS, OHIO Prepared by Laura A. Fay Mary Gessner Paul C. Stromberg John 1-Jageman and Prepared for: August 1985 ------- TABLE OF CONTENTS List of Tables . . . Author Page No . . . . . . . . . I I I I S I S List of Figures . . S S S . . . I S I I S S Acknowl edgments I I S I S I I S I I S I I S I I Executive Summary Introduction Background . . . . . . . . Program Objectives Study Objectives . Site Description . . . . . Basin Description H y d r o 1 o g y . . . . . . . . . . . . Industrial Contaminants . . Contaminant Sources Methods . I . . . . . . . . Field Methods Sampling Plan Flow Calibration . . . Laboratory Methods Larval Fish Sorting . Larval Fish Identi fi cation Identification Problems Pathology . . . . . . Chemical Extraction Procedures . Cleanup . . . . . Analysis and Quantification Data Analysis . . . . . . . . Results , Distribution/Abundance . . . Growth Rates Gizzard Shad Emerald Shiner Mortality Rates Pathology . . . . . . . Laboratory Exposure -Fathead Body Burdens . . . . . . . . Discussion Distribution/Abundance Growth Rates . . . . Mortality Rates . . Pathology . . . . . Body Burden Gizzard Shad . Emerald Shiner . Gessner I I I S I I I . . I I S I I I I I I I Ha ge man I I S I I . . Hageman I I I I ..Stromberg Gessner I I I S I I : : : : : : Fay . Fay . . Fay I I I I : : : : : : . . Stromberg Minnows. . Gessner & Lemon Fay I I I I . I I I I I I I I I I I I I Gessner & I I I I Fay I I S I I S Lemon S ------- TABLE OF CONTENTS (Continued) Page No . References Cited Tables . . . . . . . . . . , Figures . . . Appendix A Larval Fish Collected for Body Burden Ana1ys s . . . . . . . . . . . . . Appendix B - Field Tow Calibration Data Appendix C Growth and Mortality Rate Program Documentation . . . . . . . . . . . . . ------- LIST OF TABLES Page no . 1. Larval Fish Species List for the River Raisin 1983 and 1984. 2. Abundance of Larval Fish Collected in the River Raisin, 1983 3. Abundance of Larval Fish Collected in the River Raisin, 1984 4. Distribution of Larval Fish Collected in the River Raisin, 1983. 5. Distribution of Larval Fish Collected in the Raisin River, 1984. 6. Ranking of Species Abundance determined in the River Raisin 1983/1984 study compared to Judes 1982 study. 7. Estimated numbers of Fish Larvae entrained from February 13, 1982 through February 12, 1983 at Monroe Power Plant (Data taken from Jude et al 1982). 8. Gizzard Shad Simple Growth Rates , River Raisin 1983/ 1984. 9 Emerald Shiner Simple Growth Rates, River Raisin 1983/ 1984. 10. Larval Fish Growth Rate Coefficients, River Raisin 1983 11. Larval Fish Growth Rate Coefficients, River Raisin 1984. 12. Ranking of 1983 Larval Fish Growth Rate Coefficients. 13. Ranking of 1984 Larval Fish Growth Rate Coefficients 14. River Raisin 1983 Larval Fish Mortality Coefficients. 15. River Raisin 1984 Larval Fish Mortality Coefficients. 16. Macroscopically Observed Deformities in Larval Fish from the River Raisin during 1983. 17. Larval Fish by Species and Station Evaluated Pathologically. ------- LIST OF TABLES (Continued) Page No . 18. Lesions In Gizzard Shad from Raisin River, 1983. 19. Gizzard Shad Larvae by Size and Station Indicating Lesions. 20. Hi stopathologi cal Lesions in Gizzard Shad From the Control Station (#7). 21. Histopathological Lesions in Fathead Minnows. 22. Gizzard Shad Larval Fish Density of the River Raisin, 1983. 23. Review of Lake Erie Gizzard Shad Larval Density, Peak Abundance. 24. River Raisin Qualitative Sediment Survey (Data supplied by Michigan Department of Natural Resources). 25. Gizzard Shad Larval Fish Density of the River Raisin, 1983 (< 5 mm). 26. Review of Growth Rates. 27. Comparison of Histopathologic Lesions in Larval Gizzard Shad at Station 4 and the Control Station (7). ------- LIST OF FIGURES Page No . 1. River Raisin Larval Fish Sampling Stations 1983 and 1984. 2a. Aerial View of Station 1 and 2,River Raisin. 2b. Aerial View of Station 2 and 3,River Raisin. 2c. Aerial View of Station 4 ,River Raisin. 2d. Aerial View of Station 5,River Raisin. 2e. Aerial View of Station 4,5 and 6,River Raisin. 3. Ri ver Ral Si ri Drainage Basin (Taken from Michigan Water Resources Commission,1965). 4. Raisin River 11 Year Average Daily Flow (CFS) 12 KM Upstream from Lake Erie (USGS data). 5. Peak Flows in the RI ver Rai si fl at Monroe Since 1938 (Data from the USGS gage 04176500). 6. Lake Erie Level at Gage 3087 (NOAA data). 7. Larval Fish Density Calculation Procedure. 8. Gizzard Shad 1983 Simple Growth Rates. 9. Gizzard Shad 1984 Simple Growth Rates. 10. Gizzard Shad Simple Growth Rates, Station 1, 1983. 11. Gizzard Shad Simple Growth Rates, Station 2, 1983. 12. Gizzard Shad Simple Growth Rates, Station 3, 1983. 13. Gizzard Shad Simple Growth Rates, Station 4, 1983. 14. Gizzard Shad Simple Growth Rates, Station 5, 1983. 15. Gizzard Shad Simple Growth Rates, Station 6, 1983. 16. Gizzard Shad Simple Growth Rates, Station 7, 1983. 17. Gizzard Shad Simple Growth Rates, Station 1, 1984. 18. Gizzard Shad Simple Growth Rates. Station 2, 1984. 19. Gizzard Shad Simple Growth Rates, Station 3, 1984 ------- LIST OF FIGURES (Continued) 20. Gizzard Shad Simple Growth Rates, Station 4, 1984. Page No . 21. Emerald Shiner 1983 Simple Growth Rates. 22. Emerald Shiner 1984 Simple Growth Rates. 23. Emerald Shiner Simple Growth Rates, Station 1, 1983. 24. Emerald Shiner Simple Growth Rates, Station 2, 1983 25. Emerald Shiner Simple Growth Rates, Station 3, 1983 26. Emerald Shiner Simple Growth Rates, Station 4, 1983. 27. Emerald Shiner Simple Growth Rates, Station 5, 1983. 28. Emerald Shiner Simple Growth Rates, Station 6, 1983 29. Emerald Shiner Simple Growth Rates, Station 7, 1983. 30. Emerald Shiner Simple Growth Rates, Station 1, 1984. 31. Emerald Shiner Simple Growth Rates, Station 2, 1984. 32. Emerald Shiner Simple Growth Rates, Station 3, 1984. 33. Emerald Shiner Simple Growth Rates, Station 4, 1984. 34. Gizzard Shad 1983 Instantaneous Growth Rate Coefficients by Station. 35. Gizzard Shad 1984 Instantaneous Growth Rate Coefficients by Station. 36. Emerald Shiner 1983 Instantaneous Growth Rate Coefficients by Station. 37. Emerald Shiner 1984 Instantaneous Growth Rate Coefficients. ------- ACKNOWLEDGMENTS The authors of this report would like to express their appreciation to the following personnel involved in some aspect of this project completion: Julieanne Barth Cranbrook Institute of Science Shaio ChengMu Doug Frantz Gerald M. Gerber John Hageman Sue Lemon Kevin McGunagle Todd Parfitt Ken Rygwelski Andrea Wilson Special thanks are awarded to David Jude for verifying larval fish vouchers specimens, discussions regarding growth rate methodology and for reviewing the data. The authors would also like to thank NOAA for the use of their storage facility at Monroe, Michigan and to the crew of the R/Y Bluewater for laboratory access. Finally, we would like to acknowledge the financial support of the United States Environmental Protection Agency Large Lakes Research Station (Grosse Ile) and the Environmental Research Laboratory (Duluth). ------- EXECUTIVE SUMMARY 1. Gizzard shad were the most predominant larval fish collected during the 1983/1984 River Raisin study. Although the number of larval fish collected during 1984 was almost 101000 greater than in 1983 the relative abundance of gizzard shad remained stable over the two year period (72%). 2. The relative abundance of the remaining species varied over the 2 year study as a result of temporal differences in the initiations of the field programs. The 1983 field program began on May 30th and ended on September 12th, while the 1984 program was initiated on April 2 and lasted until July 19. Many of the sport fish and early spring spawners (white suckers) were caught as a result of the early Spring sampling program. 3. The major storm event of February 14, 1984 had no apparent negative effect on the fish spawning habitat based on the increased numbers of larval fish captured during 1984. 4. Simple growth rates (dl/dt) for gizzard shad were the highest at station 3 (adjacent to the Monroe Sewage Treat- ment plant); Gizzard shad growth rates at the remaining stations were substantially lower (0.25 0.78 inn/day). Possi bly thi s increased rate can be simply explained by an increased availability of food (i.e. plankton). 5. There is a discrepancy between results obtained for growth when comparing the simple growth rates and the growth rate coefficients for both gizzard shad and emerald shiners. The reason for this disparity has not been resolved. 6. Clearly, the instantaneous mortality rates are much higher during 1984 (stations 1 through 4). The reasoning for the increased mortality may be explained by something routine like weather, water temperature, lake level, food availability, or any one of the numerous variables accounting for natural mortality or by something more unorthodox like introduced chemicals or toxic contamina- tion. Food availability (phytoplankton and zooplankton) and contamination level data for the two years should be compared. 7. Real lesions compatable with acute toxicity were observed in organs in contact with the environment as well as for the intestine and the kidney. 8. Lesions appeared to affect primarily gizzard shad in all size classes. 9. Gizzard shad from all collection stations had lesions. ------- 10. Real lesions compatable with acute toxicity were observed in gizzard shad and alewife from the control station. 11. Observed lesions were identical in quality, distribution, and range of severity to those found in gizzard shad from the river stations. These lesions most likely indicate similar, adverse environmental conditions at both the riverine and lake control station. 12. The attempt to reproduce lesions, experimentally, in fat- head minnows was not successful. However, differences in metabolism between species, bioavailability of toxic substances and duration of action might account for such failure and does not mean toxicants were absent. 13. It is possible that a serious health problem exists for gizzard shad in Lake Erie, based on the number of lesions observed in gizzard shad at the control station. ------- RECOMMENDATIONS 1. Future growth rate studies involving gizzard shad should utilize data limited to the largest larvae captured each sampling period due to the ability of gizzard shad to spawn over wide temporal ranges. This ability results in a continual influx of newly hatched larvae skewing the growth rate downwards (Gordon 1982). 2. Comparisons should be initiated for the 2 year database to define the probable cause for both the Increased density observed in 1984 and for the increased mortality rates. 3. Additional fish submitted for histopathological evaluation should be between 20 mm and 50 mm. Fish smaller than this are not sufficiently differentiated to allow complete analysis of tissues. Fish larger than 50 mm create technical problems resulting in poor specimen quality. 4. Consider an investigation of spontaneous lesions in gizzard shad from multiple Lake Erie localities. Correlate observed lesions with water chemi stry data and toxicologic analysis of whole gizzard shad. ------- INTRODUCTION BACKGROUND Although the role of marine estuaries as spawning and nursery areas for economically important fish populations has been the subject of considerable research during the last 20 years, investigation of the role of ri Yen ne habi tats in the Great Lakes has long been neglected. Half of the approximately 175 fish species occurring in the Great Lakes basin are believed to be dependent on riverine habitats as spawning, nursery, or adult concentration areas. Approximately 50 of these species are currently important commercial, recreational, or forage species. Few of the species residing in the Great Lakes themselves, as opposed to those restricted largely to tributaries, are thought to be independent of niverine, coastal wetlands, or coastal shallows as spawning and nursery areas (Trautman, 1981; Hubbs and Lagler, 1964; Van Meter and Trautman, 1970). The riverine areas of Lake Erie have long been recognized as major breeding grounds for many species of fishes. These areas have tradi ti onally exhi bi ted greater species diversity and numbers of fishes, especially larval fishes, than the remainder of the lake (WI ckli ff, 1931; White et al . , 1975; Cooper et al ., 1981a, b, C,; Mizera, et al ., 1981). The cultural stresses placed on rivers mouth areas are quite intense. Since they are located at the mouths of tributaries, they are subject to inputs of toxic substances from agricultural, industrial, and municipal sources. Alterations in the flow of tributary water into the nearshore area by agricultural and storm water runoff can significantly affect the characteri sti Cs of the mixing zone ecosystem. The Lake Erie Basin is the most densely populated and heavily industrialized within the Great Lakes Basin and therefore the most seriously impacted. In 1981, the International Joint Commissions Water Quality Board identified 39 Areas of Concern within the Great Lakes Basin. The River Raisin was identified as an area exhibiting signficant environmental degradation and impairment of beneficial uses. This designation of the River Raisin was based on: a substantial number of violations of water quality objecti yes sediments highly polluted by heavy metals, and high concentrations of PCBs and industrial and agri cul tural organic chemicals in fi sh. PROGRAM OBJECTIVES In the spring of 1983, the U. S. Environmental Protection ------- Agencys Large Lakes Research Station at Grosse lie, Michigan, selected the River Raisin as a site to address the issue of transport, exposure and effects of contaminants In the tributaries and nearshore areas of the Great Lakes. The primary objective of the study was to develop a predictive capability whereby effects of con tami nants could be esti mated, given their loadings, transport and fate characteristics. Secondary objectives of the study were: 1) to investigate the longevity and Importance of inplace pollutants, 2) to provide input to surveillance databases, and 3) to develop a protocol for assessing ecological effects of toxic substances. In order to address these objectives, an integrated analysis and modeling framework was developed which included: 1) exposure modeling (via fate and transport), 2) food chain modeling (in the form of bioaccunulation/biocoflCefltratiOfl) and 3) toxicity modeling (based on correlations between chemical concentrations and bioassay results). The field and laboratory research, which was desi gried to provide input into model development and calibration included analysis of selected chemical residues in water, sediment and biota and measurement of toxic effects on various components of the ecosystem. STUDY OBJECTIVES As part of the biological effects work, we undertook a study to investigate the effects of toxic substances on growth, survival and pathology of larval fishes. The primary objectives of this work were: to identify species of larval fish present in the River, and determine spatioternporal differences in density and species composition of the ichthyoplankton of the River; determine the spatiotemporal dose patterns of toxic substances in fish larvae; determine spatiotemporal differences in instantaneous growth and mortality rates of the most abundant species of fish larvae and relate those to exposure and dose patterns; and determine the incidence of pathol ogi c 1 esi ons in the most abundant species. Inasmuch as excessive concentrations of toxic substances are a major problem in the waters, sediments, and biota of the Great Lakes, they are particularly so in rivers due to source proximity and lack of open lake dilution. Moreover, the coincidence of high ambient environmental concentrations of toxic substances with the early life history stages of many fish species (some of considerable economic importance) represents a potential hazard ------- to the growth, survival, and health of those stages and ultimately to recruitment and maintenance of adult populations. Thi s is parti cularly true In view of the rapid growth, cell proliferation, and cell differentiation which occurs during egg, larval, and juvenile stages. An initial approach to field determi nati on of the biological effects of toxi c substances on larval fishes in polluted riverine ecosystems Is to determine spatlo-temporal exposure (i.e. concentrations In water and food organisms) and dose patterns (i.e. residues in larval fish ) and attempt to relate these to instantaneous growth and mortality rates of larval fishes of different species present at various points along environmental and toxicity gradients In the river system. SITE DESCRIPTION To accomplish our objective, seven (7) sampling stations were established, five (5) in the River and two (2) in the nearshore areas of western Lake Erie (Figure 1). Station descriptions for the two sampling seasons are described below. Raisin River Station Locations (1983) 1. 300 meters downstream of the RI 50 dam, midstream, 100 meters upstream of the northwest tip of Sterling Island. Average depth during the study was 2 meters. (Figure 2a). 2. Approximately 50 meters downstream of the River Front Marina, at the electrical substation. Approximately 200 meters upstream of the RI 75 overpass. Average depth during the study was 3 meters. (Figures 2a and 2b). 3. Midstream, even with the mouth of a cove slightly down- stream of the Monroe wastewater treatment plant. Approximately 340 meters downstream of the RI 75 overpass. Average depth during the study was 4 meters. (Figure 2b). 4. Midstream, downstream of the ship turning basin, near the Port of Monroe Terminal building. Hear buoy #11. Average depth during the study was 8 meters. (Figures 2c and 2e). 5. Midstream at the Monroe power plant intake canal. Average depth during the study was 8 meters.(Figures 2d and 2e). 6. In the Raisin River mouths outermost region. Approximately 150 meters outbound of buoys #9 and #10 and 225 meters from the mouth. Average depth was 8 meters. (Figure 2e). 7. 200 meters beyond cans #7 and #8 in the shipping canal. Average depth was 9 meters. ------- River Raisin Station Locations (1984) In 1984, longitudinal, oblique tows were taken at: replicate A at 0.3 of river width, replicate B at 0.5 river width, and replicate C at 0.7 of river width. The transects consistently covered approximately the same distances in each 6 minute tow. Station 1 - Tow started at the downstream tip of Sterling Island and ended in the vicinity of the upstream tip of the same island. Average water depth was 2 meters. Station 2 Tow started at a lighted, white garage on the south s de of the river near the downstream edge of the boat slips and and ended in the vicinity of the electrical substation on the south side of the river across from the Riverfront Marina. Average water depth was 3 meters. Station 3 Tow started under the overhead high tension wires between the turning basin and the Monroe wastewater treatment plant and ended approximately 50 meters downstream of the 175 bridge over the river. Average water depth was 4 meters. Station 4 Tow started at the ship mooring post on the south side of the river and ended in the vicinity of buoy #11. Average water depth was 8 meters. Station 5 Tow started at a cove across from the Monroe Power Plant and ended in the vi ci ni ty of wood posts protruding from water on north side of river. Average water depth was 8 meters. The station pattern during both years of study is comparable, the only di fference being that in 1983 circular tows were made and hence the station location was more resricted than in 1984 when a range was sampled. ------- BASIN DESCRIPTION The River Raisin drains an area of 1,070 square miles (2,771 square km) and di scharges into the western basin of Lake Erie at Monroe, Michigan (Figure 3). A portion of Michigans southeastern lower peninsula and the northeastern portion of Fulton County, Ohio lie within the boundaries of the basin. The drainage basin narrows down to a 2.5 mile (4 kmwide) strip for the last 15 miles (24 km) of the river. The area consists of clay till reworked by glacial lake water and veneered by lacustrine sands, silts, and clays. Twothirds of Monroe County is covered by a layer of this glacial drift that is less than 50 feet (15 m) in thickness. The underlying bedrock is mostly carbonate in composition (Mozola, 1970). HYDROLOGY Monroe County is essentially flat terrain. There is a gentle slope southeastward from a maximum elevation of 730 feet (223 m) in the northwest corner to 572 feet (114 in) at Lake Erie. This gradual decline of only 158 feet (48 m) in nearly 26 miles (42 km) explains the low velocities of streams located in the county (Mozola, 1970). Runoff in the drainage basin i s Si gnfi cant due to the clay till. The runoff during rain events creates rapid stream fluctuations and very turbid waters. Relative to other areas in Michiagn, erosion in the River Raisin basin is considered to be high. The U. S. Department of Agriculture estimated that 8.3 to 10.8 tonnes of topsoil per hectare per year are lost (Michigan DNR, 1979). The Li. S. Department of the Interior (1967) reported that the average annual precipitation for the drainage basin area is 31.52 inches (80.1 cm). Of this amount, approximately one third runs off through the river system. Much of the area adjacent to the River Raisin is prone to flooding. A large portion of the eastern fringe of the city of Monroe was once marshland. Over the last thirty years, approximately 8O of the marshlands have been filled in for industrial and recreational uses. The river banks and surrounding areas at the mouth of the River Raisin are man-made (Monroe County Drain Commission, 1984). The U. S. Geological Survey (USGS) maintains a stream flow guage (Station #04176500) in the River Raisin near Monroe. It is located in Monroe County, 1.3 km downstream from the bridge on the Ida Maybe Road, at latitude 41 57 38 and longitude 83 31 52. The drainage area above this point in the river is 1,042 square miles (2,699 square km). The average discharge for the period of record 19371981 was 709 cubic ft/sec (19.9 cubic m/sec). The maximum and minimum discharge for the period of record was 14,500 cubic ft/sec (407.3 cubic m/sec) and 2 cubic ft/sec (0.06 cubic m/sec), respectively (U.S.Geological Survey, ------- 1982). River flows for an 11year period are displayed in Figure 4. Peak flow frequencies for the period of record since 1938 are presented in Figure 5. The City of Monroe maintains a stream flow guage in the River Raisin at Dam #1 (second low head dam relative to the river mouth). This guage Is located in the City of Monroe approximately 152 m downstream from Maple Avenue (Petty, 1984). Daily readings are recorded by the Monroe Waste Treatment Plant (WWTP). The lake level is monitored hourly by a National Oceanographic and Atmospheric Administration (NOAA) guage located near the study area. Water stage readings for Gage 3087 in the turning basin (station 4) are presented in Figure 6 (January 1, 1975 to March 31, 1983). The port of Monroe is served by a dredged shipping canal 15,800 feet (4.8 kin) long, 300 feet (91.2 m) wide and 21 feet (6.4 in) deep from Lake Erie to the mouth of the River Raisin. From the river mouth to the turning basin, there is a dredged channel 8,100 feet (2.5 kin) long and 200 feet (60.8 m) wide (Michigan DNR, 1979). INDUSTRIAL DEVELOPMENT Most of the River Raisin is in areas of agricultural production. Over 70% of Lenawee and Monroe Counties is farmland. Urban development of the basin is centered around three cities: Monroe, Adrian, and Tecumseh. Monroe, at the river mouth is the most populous and industrialized city in the basin. Much of the industry is associated with automobile manufacturing in nearby Detroit. Additional industries in the area are primary metals, fabrication of metal products, machinery and transportation equipment, manufacture of paper products, chemicals, furniture, food processing and dairy related industries (Michigan DNR, 1979). Several paper product companies are located on the River Raisin within the study area. Consolidated Packaging Corporation, South and North Plant closed on February 1978 and July 1975, respectively, produced corrugated and solid fiber containers. Time Container Company, a paper products industry, is located upstream of the study site near the Chesapeake and Ohio Railroad. Union Camp Corporation on the north shore of the River Raisin produces corrugated paper board and containers. The effluents from the primary treatment facilities of both Time Container and Union Camp are sent to the Monroe WWTP for secondary treatment (Michigan Department of Public Health and the Michigan Water Resources Commission, 1969). The Detroit Edison Monroe electric generating plant, located ------- near the mouth of the River Raisin, Is the largest coal burning plant in the United States. Up to 85 cubic rn/sec of river/lake water is pumped for cooling purposes. During spring runoff, the River Raisin makes up more than 95% of the cooling water. However, during low flow in the summer, the river makes up less than 5%, the balance of water coming from Lake Erie. Water enters the cooling system through a 100meter long intake canal that Is located about 650 meters upstream from the river mouth. The water passes through a condenser and is then released into a 350meter long, concrete conduit where water velocities are approxi mately 1 rn/sec at full operati on. The water is then discharged through a rockwalled 175-meter wide canal. Plum Creek joins the discharge canal, but contri butes less than 1% of the volumetric flow to Lake Erie. The average annual river discharge is equivalent to 20% of the total cooling water demand the rest is drawn from Lake Erie (Cole. 1978).In essence almost all the river water is funnelled through the power plant. The Monroe Metropolitan Pollution Control Facility is an acti Va ted sludge treatment plant with a design capacity of 24 MGO (90,800 cubic mid). The plant receives raw wastewater from the City of Monroe and the Frenchtown and Monroe townships. Industrial di schargers contri bute approxi inately 70% of the daily flow (Horvath 1985). The treated effluent is discharged into the RI ver Raisin. Under severe runoff condi ti ons, high flows in the collection system exceed plant capacity. During this time, untreated wastewater is pumped directly into the river from the flood pumping station. The Ford Motor Company Stamping Plant at Monroe draws its process and cooling water from Lake Erie. The water is treated with chlorine, lime and ferric sulfate prior to being used (Boerson. 1984). Waste cooling and process waters and sanitary wastewaters are treated by the company. The combined wastewaters are discharged to a polishing lagoon, with overflow discharged to the River Raisin (Horvath 1985). CONTAMINANT SOURCES Both toxic contaminant reserves in sediments and current toxic industrial, muninci pal and landfill effluent loadings to the River Raisin were considered as potential sources of toxins in the Monroe Harbor study. Copper, chromium, and zinc were analyzed during this study because of the relatively high concentrations of these materials found in sediments in the River Raisin and because of the toxic nature of these metals to cladocerans and other freshwater invertebrates. Relatively high levels of toxic heavy metals in the navigation channel have been reported in the literature. The U.S. Environemtnal Protection Agency (1975) recommended that the contaminated dredged sediments from the navigation channel should not be disposed in the open lake. Analysis of contaminants, ------- revealed high levels of copper (1450 mg/kg), zinc (970 mg/kg), and chromium (530 mg/kg). Based on atomic absorption spectroscopy (AAS) by Cranbrook Institute of Science and neutron activation analysis by the University of Michigans Phoenix Memorial Laboratory (Jones, 1983), concentrations of these metals were relatively high when compared to mean sediment levels in southern Lake Huron. Concentrations of some other metals were also found to be relatively high in these studies, but their toxicity at the current levels to freshwater biota was negligible or unknown. In addition to reserves of metals in the sediment, there is an existing potential for heavy metal discharge from primary metal production, plating, and metal machining industries in the Monroe Harbor area. Pol ychi on nated bi phenyl s (PCB s) were included in the study of Monroe Harbor because high levels of PCBs in fish were found in the area. In 1971, the Michigan Department of Natural Resources collected fish in the River Raisin and found up to 6.45 mg/kg of Aroclor 1254 in northern pike (wet weight) and up to 3.08 mg/kg of Aroclor 1254 in carp (wet weight). The results of a 1979 survey included a single carp with 77.2 mg/kg of total PCB (Bunby et al, 1983) PCBs have been linked to industrial activity that use the persi stant compounds in lubni cants and coolants for electri cal equipment. PCBs have also been found to be a by-product in paper recycling plants. These industrial uses and processes exist (or existed) in the River Raisin study area; therefore, it is possible that the sediment and fish contamination observed originated from local industrial activity. ------- METHODS FIELD METHODS Sampling Plan Larval fish samples were collected at night (45 minutes after sunset) twice weekly, towing a .75 meter diameter conical oceanographic plankton net of .571 millimeter mesh behind an outboard motorpowered boat travelling at 45 knots. Flow rates (i .e. volume of water sampled) were measured via a center mounted General Oceanic Model MKII flowmeter. From 30 May to 12 September 1983, 7 stations in the lower Raisin River and adjacent Lake Erie were sampled using 4 minute circular, oblique tows. Raisin River water temperature data was obtained from the Monroe Wastewater Treatment Plant and the Detroit Edison Monroe Power Plant. From 2 April to 19 July 1984, stations 15 in the lower Raisin River were sampled. In an attempt to insure parity among replicates by sampling new water during each tow, and to increase the number of species that could potentially be statistically analyzed, tow times were increased to 6 mintues and were made travelling upstream longitudinally at .3, .5, .7 of the width of the river. While returning downstream to begin the next replicate, special effort was made to travel around the pending transects. Refer to Figures 1 and 2 for the locations of the 1983 and 1984 stations. In 1984, Raisin River surface water temperature was measured at each station with a VWR Scientific thermometer. In both years, three replicate samples were collected from each station. In the field, replicates A and B were preserved with a 5 percent volume of 37 percent buffered formaldehyde solution, and replicate C was preserved with a 100 percent volume of Dietricks fixative for future pathologic analysis. Dietricks fixative was made using the following reci pe: 30 parts distilled water 15 parts 95 percent ethyl alcohol 5 parts 37 percent buffered formaldehyde solution 1 part concentrated, glacial acetic acid When larvae were determined to be sufficiently abundant, additional weekly samples were collected for body burden assessment. A single 10 minute tow was made at each body burden station, but was not chemically preserved. The target species of Gizzard shad ( Dorosorna cepedianum ) and Emerald shiner ( Notropis atherinoides ) were each: 1. Separated from the rest of the raw sample 2. Patted dry 3. Frozen whole for pickup by Cranbrook Institute ------- Refer to Appendix A for a list of fish larvae provided for body burden analysis. Additionally, 3 - 4 liter amber bottles were filled with 1 liter the surface water from each body burden station (4, 5, 7, in 1983; 4, 5 in 1984) using the following procedure for pickup and analysis by Cranbrook Institute: 1. Ri n se bottle wi th sta ti on water; Di sca rd ri n se 2. Submerge bottle and fill to I liter 3. Add 100 ml methylene chloride to bottle 4. Cap and shake vigorously for 3 minutes Water samples were not collected on nights when larval abundance was too low for body burden analysis. Flow Calibration Before the first larvae sampling date, once each month, and after the completion of the sampling season, the flowmeter was calibrated by towing the meter on the net frame (without the net) for a known distance (500 meters) for 10 repetitions (Appendix B). In order to sample all levels of the water column, our oblique tows were adjusted to conform with water depth, as determined with a weighted depth chain. LABORATORY METHODS Larval Fish Sorting 1. Record the sample date and station from the raw sample bottle and enter, along with sorters initials and date sorted, into the sample log book 2. Pour entire raw sample into a sand sieve (USGS #40). 3. Rinse with low pressure tap water using tygon hose to remove fine sedi ments from the raw sample. 4. Remove an aliquot (approximately 1 cubic cm) and dilute it with tap water into enamel or pyrex pan. 5. Place pan in sorting chamber, under a lamp, or other well lit area and search for larvae. 6. If larvae are found, label a tag with the station number and date, and insert it into a vial containing 70 percent Ethanol . Conti flue searching until all larvae in the pan have been found and removed. ------- 7. Repeat with additional aliquots until raw sample is finished. 8. Dispose of extraneous zooplankton, invertebrates, detritus, etc. 9. Rinse seive thoroughly. 10. Store larval sample in appropriate box for later I den ti fi ca ti on. Larval Fish Identification Using a Bausch and Lomb stereo dissecting microscope with a polarized stage, rheostatic light source, and magnification range of 6x to lOOx, larval fish were identified to species (when possible), developmental stage noted (as defined by Snyder, 1976), and total length measured to the nearest 0.5 mm . Gross morphology was examined for pathological defects using the criteria of Drummond(undated) The following taxonomic keys, relevant papers, and the CLEAR larval fish archive collection were utilized to facilitate identification. 1. Auer, N.A. (ed.) 1982. Identification of larval fishes of the Great Lakes basin with emphasis on the Lake Michigan drainage. Great Lakes Fishery Commission, Ann Arbor, Michigan 48105. Special Publication 823:744 p. 2. Drummond, R. A. Guidelines and terminology for using fish behavior checklist. Environmental Laboratory Duluth, Minnesota 55804. Unpublished. 6 p. 3. Hogue, J. J., R. Wallus, and L. K. Kory. 1976. Preliminary guide to the identification of larval fishes in the Tennessee River. Tennessee Valley Authority, Div. of Forestry, Fisheries, and Wildlife Dept., Norris, TN. 67 p. 4. Nelson, D. Working key to the larval fishes discovered near the west shore of Lake Erie. Michigan State University, Dept. of Fisheries and Wildlife. Unpublished. 12 p. 5. Norden, C. R. Key to larval fishes from Lake Erie. University of Southwestern Louisiana, Lafayette. Unpublished. 4 p. 6. Olney, J. E., G. C. Grant, F. E. Schultz, C. L. Cooper, and J. Hagernan. 1983. Pterygiophore Interdigitation Patterns in larvae of four Morone species. Trans. Amer. Fish. Soc. 1983, No. 4: 52 553 1. ------- 7. Siefert, D. E. 1976. Terminologies for intervals of larval fish development. Pages 4160 in Borrman (ed.); Great Lakes Fish Egg anrLarvae Identification. Li. S. Dept. of the Interior, Fish and Wildlife Service, Washington, D.C., FWS/ OBS76/23. Upon completion of the identification of a sample, the final columns of the sample log were filled with date of identification, identifiers initials, and number of the samples vials. All fully processed samples were preserved with 70 percent ethanol and stored in the CLEAR biological archive. Additionally, one voucher specimen of each species encountered at each developmental stage (1IY) observed were archived in the CLEAR reference collection. identi fication Problems As noted above, all larvae encountered were identified to species when possible, but several closely related species among families are difficult or impossible to positively identify while in the early larval stages. Problem families were treated in the following manner: CLUPEIDAE Alewife ( Alosa pseudoharengus ) and Gizzard shad ( Dorosoma cepedianum ) are separated only by meticulous measurements and/or muscle segment (myomere) counts. Gi zzard shad overwhel mi ngly dominated our catch, thus whenever damaged CLUPEIDAE were encountered, they were expressed as Gizzard shad. Sped mens in good condition were always keyed to proper species. CATOSTOMI DAE CYPRINIDAE Poor specimen condition occasionally called for an individual to be expressed as Unidentified Catostomidae or Unidentified Cyprinid. Carp/ Goldfish were expressed as carp due to the difficulty of separating wild caught specimens of these species made worse by their propensity to hybridize with each other (Crunkilton, 1977, personal communication) CYPRIN 0 0 0NTIDAE Fundulus spp. are poorly represented in the literature, thus no attempt was made to assign ------- our wild caught, Fundulus specimen to species. PERCICHTHYIDAE Morone spp. cannot be separated using morphological features until anal ray pterygiophores become evident at approximately 13 mm, thus Morone spp. less than 13 mm are usually expressed as Morone spp. and those over 13 mm In good - condition were separated to White perch ( Morone americana ) or White bass ( Morone chrysops) . CENTRARCU IDAE Lepornis spp. are virtually impossible to separate while in their early life stages due to similar morphology and widespread hybridization, thus almost always were expressed as Lepomis sp. Pomoxis spp. are also difficult, but attempts were made when possible to separate the two species using Seifert (1969), otherwise were expressed as Pomoxis sp. PERCIDAE There were occasionally darters, Etheostoma spp. that could not be assigned to species. Pathology Fish preserved in Dietrichs fixative were delivered to a certified histology technician. These fish were dehydrated and embedded in paraffin blocks. Smaller larvae (410 mm) were embedded at a density of five fish per block. Larger fish (1225 mm) were embedded one per block. Fish were oriented so that longitudinal, mid-line sections, cut at 5u could be produced. Sections were mounted on glass slides and stained with hematoxylin and eosin. The following tissues were examined for histologic lesions: skin, oral epi thel i urn, bronchial epi thel i urn, gills, thymus, brain, spinal cord, eye, otolith organ, thyroid, i nterrenal organ (adrenal ), pancreatic islets, heart, skeletal muscle, excretory kidney, urinary bladder, head kidney (hemopoietic organ), liver, exocrine pancreas, stomach, intestine, peritoneal fat, air bladder, cartilage and bone. A list of observed lesions from each fish examined was kept and a table of lesion frequencies was generated. This data was analyzed and compared to a table of lesions generated from similar fi sh larvae from the control collecti Ofl site. Several larval fish that had observed spinal deformities or tumored growths were sent directly to the pathologist for ------- observation before they were prepared for histological analysis. CHEMICAL METHODS Extraction Procedures Details of the extracti on procedures used for the biological samples may be found in Rathbun (1985), for the water samples In Smith et al. (1985) and for the sediment samples in Filkins et a]. (1 5T Brief descriptions are given below. Biological sample tissue (approxi mately 20 g) was mixed with anhydrous sodium sulfate and Soxhiet extracted for 48 hous with a 1:1 mixture of nhexane and dichloromethane. When less than 20 g of tissue was available, the total sample was extracted. Composite larval fish samples ranged in weight from approximately 5 g to 47 g. The extract was partitioned into nhexane and its volume reduced to 10.0 ml over a steam bath. A one ml sample was air dried in a tared aluminum weighing dish for lipid determination. Water samples were liquidliquid extracted with dichioromethane (10 parts water to 1 part 0CM) in clean 4 1 amber glass solvent bottles. The extract was partitioned into n hexane, dried through a sodi urn sul fate column and its vol ume reduced to 2.0 ml on a steam bath. Sediment samples (about 20 g) were mixed with anhydrous sodium sulfate and Soxhiet extracted for 48 hous with a 1:1 mixture of acetone and nhexane. The extract was paritioned into nhexane and its volume reduced to 10.0 ml on a steam bath. Clean-Up All extracts were cleaned of lipids and other interferring compounds with Florisil. Details are provided in Smith et al. (1985). Briefly, columns were packed with 20 grams of FT rT il ; rinsed with 50 ml nhexane; 1 ml of extract was injected onto the column followed by 250 ml of 4% 0CM in nhexane. The solvent volume was reduced over a steam bath to less than 10 ml, and to 1.0 ml under a stream of dry N2 gas. The extract was sealed in a glass ampule until analysis. Analysis and Quanitation The analytical procedures used in this study are described in great detail in Smith et a]. (1985). Briefly, samples were analyzed on a VARIAN ModeE3700 gas chromatograph equipped with a 63N1 electron capture detector. The chromatographic column was a 50 m fused silica column (0.2 mm i .d. ) coated with SE54 (HewlettPackard). Sample vol uine was 4.5 ul and the carrier gas was hydrogen. ------- DATA ANALYSIS Seven calculation steps were performed on the ichthyoplankton database obtained from the 19831984 Raisin River Study. Step 1: Calculate larval fish density (#/1000 cu meter) for all samples (A, B, and C) for each station and sampling period. See Figure 7 for density calculation procedure. Step 2: Average A, B, and C density replicates by species, size, station, and sampling period. Step 3: Sum each species total density by station on an individual sampling period basis and over the total season. Step 4: Calculate the average length (mm) of each species by station and sampling period. Step 5: Calculate the date when each species population length (IL) by station equals 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, and 70 mm. Step 6: Calculate the instantaneous growth rate coefficient (G) for each species at each station. Step 7: Calculate mortality (Z) for each species at each station. Instantaneous Growth Rate Coefficients were calculated for all fish that appeared for a sufficient period of time during the sampling season to acquire a rate (i.e. n > 3). The instantaneous growth rate equation utilized was Lt = Lto e G Ct-to) where: L(to) = Initial length Gt = Growth rate t = time final to = time zero Lt = Length at final time Length (Lo) and Growth (G) were determined by regression techniques using time (t) for x and length (L) for y. This ------- procedure is outlined by Hackney and Webb (1978) in the Proceedings of the National Workshop on Entrainment and Impi ngement. Mortality was calculated using the equation Nt = N(to) e z(t to) where: N(to) = Numbers of larval fish at an initial time Z = Mortality rate t = time final to = time zero Nt = Number of larval fish at final time Further information on the development and use of this equation can be seen in Hackney and Webb (1978). ------- RESULTS DISTRIBUTION/ABUNDANCE The predominant species of larval fish found during the 1983 1984 EPA survey are reported in Tables 2(1983) and 3 (1984). A total of 15,849 larval fish were collected from May 30th to September 12th, 1983. A substantially larger population (25,583) was collected from April 2, to July 19,1984. A feasible explanation for the increased larval fish catch is the improved sampling design utilized in 1984. The most abundant fish captured during both field seasons was the gizzard shad (11,410 in 1983 and 18,853 in 1984). The major portion of the 1984 larval fish Increase (9734) is accounted for by gizzard shad, however gizzard shad represented approximately 72 Z of the entire larval fish population collected during each field season. The predominance of the remaining species varied slightly between the two field years with the largest shift found in the white bass population (13th in 1983 and 3rd in 1984). This is explained by the spawning season of white bass which extends from late April to June. Remember that the field season in 1983 did not begin until May 30th. Analysis of the distribution of larval fish resulted in some conflicting results between the two years. For example, the top ten most predominant fish were generally collected at each station in 1983 except for freshwater drum and yellow perch. Freshwater drum larvae were not collected from either station 1,2 or 3 in 1983 and yellow perch were not observed at either station 2 or 3. However both species were represented at all stations during the 1984 field season (Tables 4 and 5). Once again this might be explained by our improved sampling design in 1984. Jude et al . 1983, studied the Monroe Power Plant from February of 1982 thru February of 1983 to assess the entrainment and impingement of fish larvae. The abundance of each species collected during the study has been ranked and compared to the two separate field seasons of the present study (Table 6). The major differences in species predominance are as follows: Alewife not found by Jude White Bass not found by Jude Brook silverisde not found by Jude Rock Bass not found by Jude Burbot not found by OSU Largemouth Bass not found by OSU Northern Hogsucker not found by OSU Estimated abundance of the predominant species and relative percentages were calculated for Judes 1982 study (Table 7). As ------- observed in the current study gizzard shad were the most abundant (4.08 and 10 ) for a total of 86.8 Z of the fish population. This percentage slightly exceeds the percentage calculated by OSU (73Z). GROWTH RATES Two different calculations were performed to assess the growth rate of pre and post larval fish from the River Raisin study area. The first method involved the simple ratio of differences in length (dl in mm) to differences in time (dt in days). The second calculation (instantaneous growth rate coefficient) involved the use of linear differential equation Lt= Lto e G(to). Each of the methods employed in these calculations are described previously in this report. The 1983/1984 data for simple growth rates (dl/dt) is shown in Table 4 for gizzard shad. Values for the two years range from 0.27 to 0.98 mm/day. A summary of the 19831984 rates can be seen in Figure 8 and 9, respectively. mdi vi dual plots of simple growth rates by station can be seen in Figures 1020. Simple growth rates were also calculated for emerald shiners at all stations in both 1983 and 1984 (Table 9). Summary plots of the 1983 and 1984 rates can be seen in Figures 21 and 22 respecti vely . The individual plots of the simple growth rates by station can be seen in Figures 2333. The instantaneous growth rate coefficients for Gizzard shad and Emerald shiners are presented in Tables 10 and 11 and Figures 34-37..Growth is represented by the variable 11 and ranges from 0.016 to 0.125. The 1983 rates for both gizzard shad and emerald shiners are summarized in Table 12. Each station was ranked giving the lowest ranking (r=1) to the station with the highest instantaneous growth rate coefficient and highest ranking to the station with the lowest instantaneous growth rate coefficient (r=7). The rankings for both species were combined and then re ranked. The initial rankings for both species were similiar for all stations except station 2. Station 2 gizzard shad exhibited the lowest instantaneous growth rate while station 2 emerald shiners exhibited the highest instantaneous growth. This discrepancy will be discussed later in this report. The 1984 rates for gizzard shad emerald shiners are ranked in Table 12, similiar to the method previously described. However,the 1984 data did not exhibit the discrepancy of the station 2 data in 1983. MORTALITY RATES Mortality rates were calculated according to the equation proposed by Hackney and Webb (1978) and outlined in the methods section. The mortality equation involves the use of the ------- initial and final density of larval fish resulting in the determination of Z, the mortality rate. The mortality rates for 1983 and 1984 are presented in Tables 14 and 15 respectively. Z should be positive under normal circumstances, indicating decreasing larval fish density thru time. Several of the species reported have negative estimates of the variable Z, indicating increasing larval fish density. The initial larval fish population size (No) in these cases is always small (<10 fish per 1000 cubic meters). These values should not be considered in further discussion. Values based on data from species when the initial population densities (No) are sufficiently large have mortality rates (Z) which range from 0.011 to 0.200 (Tables 14 and 15). PATHOLOGY Stati on 4 fi sh were sel ected for hi stopathol ogi cal anal ysi s because it had been selected by the USEPA as a master station and therefore would have a corresponding weekly database of organics and metals. Gizzard shad were selected from station 4 because of their high density and frequency. Fish were selected from each sampling period beginning on June 16. A list of samples taken for pathological analyses is included in Appendix C. In addition to station 4 gizzard shad, 15 samples were sent to the pathologist because of observed spinal defects or possible internal tumors (Table 16) Gizzard shad from every station on August 8 were analyzed due to tumors observed on August 4 and 8. Finally, at least one specimen from every species found at station 4 was analyzed. The results of the first group of fish revealed that the smaller fish (i.e. <15 mm) were too difficult to interpret. The second set of samples were selected so that the size exceeded the 15 mm limitation. A total of 104 blocks of fish collected from six stations between June and September 1983, were evaluated for histopathologic lesions. Twelve different species of fish were submitted but only gizzard shad were numerous enough for significant analysis (Table 17). The majority of fish were collected from stations 4 and 5. The quality of fixation of the specimens was generally good with autolysis impeding histopathologic interpretation in only a few cases. The quality of the prepared slides was excellent. Twenty three organs and tissues were present with sufficient frequency to permit si gni fi cant analysis. A total of 64 fishes were received for histopathologic diagnosis from the control station (#7). There were 39 gizzard shad, 4 alewife, and 3 yellow perch collected from the control lake station to be compared to gizzard shad in the riverine stations (16). In addition, 18 fathead minnows were submitted which were exposed to potentially toxic substances from station ------- 4. The technical quality of these specimens was fair to good. While consistent evaluation of 24 tissues was possible, fixation of tissues was clearly less satisfactory that the previous lot of fish. Many tissues had autolyzed or were distorted due to the fact that the samples were shipped to OSU in formaldehyde and then transferred to Dietrichs. On the whole, however, a significant number of specimens of good quality permitted adequate interpretation of lesions with suffi ci ent consi stency to validate the results. Lesions were consistently observed only in gizzard shad from the river stations. Basically, lesions consisted of acute epithelial necrosis characterized by picnosis, coagulation and separation of cells from the basement membrane and often accompanied by sloughing into the lumen of the organ. These changes ranged from mild to severe and from a focal to diffuse distribution. Acute epithelial necrosis was observed with a high frequency in the olfactory organ (94.6%), lateral line organs (94.4%), the oropharyngeal epithelium (96.2%), esophagus (91.5%), gills (91.5%), renal tubules (94.3%) and intestine (70.4%) (Table 18). Two of the most severely affected, important organs were the gills and kidney. Gill tissue was present in 47 of the 77 (61%) gizzard shad. Besides frank necrosis of branchial epithelium, a high percentage (80.8%) of gizzard shad had separation and ballooning in the gill tissue interpreted to be branchial edema. Seventy two percent of gizzard shad had gill parasites. The most common paratsites were the protozoans Ichthyophirius sp. and Trichodina sp. In addition, agents compatable with Epistylus , monogenetic trematodes and glochidia of fresh water mussels were occasionally observed. In all cases, the branchial epithelial changes associated with these agents were localized. The kidney was evaluated in 53 of the 77 (69%) gizzard shad. In addition to acute tubular epithelial necrosis observed in 94.3% of gizzard shad kidneys, there was a significant number of kidneys (32.1%) with hyaline droplet degeneration. The lesion was manifest as one to several circular, eosinophilic inclusions in the cytoplasm of renal tubular epi theli um. Sample sizes of the other species of fish were not large enough to provide signficant interpretation. However, carp, logperch, catfish, yellow perch and walleye had no lesions. The spottail and emerald shiners, troutperch and Morone sp. were too small for significant analysis. One freshwater drum was normal and one had a questionable lesion in the olfactory organ and intestine. The single specimen of alewife had lesions similar to those in gizzard shad. Table 19 compares the distribution of affected gizzard shad (exhibiting histopathological lesions) among four different size groups and five collection stations. Although the majority of fish were collected from stations 4 and 5, fish from all stations had lesions. Fish in all size classes had significant ------- histopathological lesions. The relatively low percentage of fish less than 20 mm long exhibiting lesions is an artefact. Many fish in this size class were too small or not sufficiently differentiated to permit pathological evaluation. Several specimens were submitted with severe spinal curvature but no hi stologi cal basis for this lesion was observed. In addition, several fish specimens were submitted with grossly evident tumors. Histological evaluation revealed these to be nonneoplatic, microsporidian cysts, probably of the genus Glugea . A large number of fish from the control station had lesions (Table 20). Acute coagulation necrosis of epithelial cells was observed in 11 of 12 olfactory organs, 1 of 28 otolith organs, 10 of 11 lateral line organs, 37 of 40 oropharynxs, 31 of 38 esophaguses, 41 of 43 gills and 15 of 16 intestines. Acute renal tubular epithelial necrosis was observed in 35 of 42 fish. Hyaline droplet degeneration occured in the kidneys of 16 of 42 fish. Thymic lymphoid necrosis occured in 2 of 29 fish. Gill parasites were observed on 10 of 43 (23.3%) of these fish. All gizzard shad evaluated had lesions. The most consistent lesion was epithelial necrosis in the gills and kidney, observed in 35 of 35 fish. Four of four alewives had gill and kidney lesions similar to gizzard shad but less severe. Three alewives had necrosis of the oropharyngeal epithelium. Only one yellow perch had lesions. Mild necrosis was observed in the olfactory and lateral line organs as well as the oropharynx and gills. A total of 18 fathead minnows were evaluated for histopathologic lesions. Twentythree tissues were examined but no lesions or abnormalities were noted. The three different groups of fish could not be distinguished in any way by microscopic evaluation (Table 21). ------- DISCUSSION DISTRIBUTION/ABUNDANCE Average density for 1983 gizzard shad larval fish ranged from a low at station 2 (49.6 fish/1000 cubic meters) to a maximum at station 7 (828.2 fish /1000 cubic meters) (Table 22). Miller (1960) reported that gizzard shad were abundant throughout the western basin of Lake Erie particularly In protected bays and at the mouths of tributaries. Gizzard shad are particularly attracted by warm water flowing from industrial plants and able to withstand temperatures up to 35 C. River Raisin should be an ideal location for gizzard shad due to the heat introduced from the oncethru cooling power plant located at the mouth of the river. Suprisingly, the densities found in the River Raisin and the surrounding portion of the western basin are low compared with densities of peak abundance reported in the literature for Lake Erie. Literature values of gizzard shad peak abundance (Table 23) for the Maumee River were recorded at 16,349 fish /1000 cubic meters (Snyder, 1978). The peak abundance of larval gizzard shad was recorded at station 6 at 5,596 fish /1000 cubic meters (Table YY) or approximately 35 Z of peak density at the Maumee. Literature values for the open lake area surrounding Davis Besse fluctuated greatly (1104 10,369 fish 1000 cubic meters) over a 3 year period. Data from Sandusky Bay (3812/1000 cubic meters) seems to be more in the range of the values reported for the open lake area near Monroe and at the mouth of the RI ver Rai sin (Snyder, 1978). Data provided by the Michigan Department of Natural Resources based on a qualitative sediment survey undertaken in field year indicates that station 4 and 5 both represent poorer sediment quality due to the presence of oil or oil odors (Table 24). This is incongruous with the larval fish density data reported for these stations. Average densities for fish < 5 mm (indicating they were hatched within the immediate area) indicate that station 4 and 5 contribute 10.9% and 15.2%, respectively of the systems larval gizzard shad. Stations 13 contribute less than 10% combined (Table 25). By far, stations 6 and 7 produce the major portion (31.9% and 30.7%) of the population. A spring rain event on February 14, 1984 resulted in the river stage level rising 6.2 feet above the previous day (577.20). Although the river level had subsided by February 16th to 578.3 feet (+1.1 feet) it took over one week for the river to return to the level prior to the rain storm. The river level was accentuated during this storm due to large chunks of ice blocking the river mouth. It was believed that this event would have disrupted much of the spawning habitat but it appears to have had no negative effect. ------- GROWTH RATES Gizzard Shad Simple Growth rate data for gizzard shad data collected by Carlander (1970) demonstrate that throughout their distribution, shad exhibit a higher growth rate in Lake Erie (1.0 mm/day) than elsewhere. These rates are presented in Table 26 (Carlander,197 0 and Bodola,1955). Growth data from the recent River Raisin study and surrounding lake area ranged from 0.25 to 2.20 mm/day (Table 8). The highest growth rates (0.94 2.20 mm/day) occurred at station 3 , adjacent to the Monroe sewage treatment plant. Growth rates from the remaining stations are substantially lower (0.25 0.78 mm/day). Growth rates following yolk sac absorption is dependent on food abundance and availability, ability of the larvae to capture food and water temperature (Gordon, 1982). Gizzard shad larvae are planktivores, switching from zooplankton to phytoplankton after the first few weeks (Miller 1 1960). Possibly the differences in simple growth rates between stations can be explained by the analysis of the distribution and abundance of plankton. The growth rate coefficient data presents a different picture than that of the simple growth rates (Tables 12 and 13). The fastest gizzard shad growth rates predi cted from growth rate coefficient data Occurred at station 1 during both 1983 and 1984. The second most productive station was station 3 in 1983 and station 2 in 1984. The discrepancy obtained from utilizing the results of the two different growth rate techniques has not been resolved to date. Part of the problem with utilizing the Hackney and Webb (1978) equation to calculate growth rate coefficients is that gizzard shad are wide temporal spawners and that the presence of newly hatched larvae over several months biases the actual growth rate. In the future, calculations for wide temporal spawners might be calculated by simply using data limited to the largest larvae captured as suggested by Gordon, 1982. The differences in rate coefficients that she obtained when using the entire popul a ti on (0.028) was lower than that obtained when data for only the largest larvae was utilized (0.034). The growth rate coefficients found by Gordon (1982) found for 1978 - 1980 gizzard shad at Davis Besse (0.028 0.034) are within the range of those found for the River Raisin 1983 1984 study (0.017 0.090). Emerald Shiners Simple growth rates (dl/dt) calculated for emerald shiners (Table 9) ranged from 0.19 to 1.06 mm/day with the highest value occurring at station 3 in 1984. Similiar data presented in ------- Carlander (1970) indicates growth for emerald shiner larvae was 5.6 mm/week or 0.8 mm/day, slightly higher than the average simple growth rate calculated for 1983 1984 (0.48 (mm/day). Growth coefficient data again indicates different results when compared to simple growth rate results. The 1983 coefficient data shows that station 2 has the fastest growth and station 5 the slowest. The results for 1984 indicate a wider range of growth rate coefficients (-.009 to 0.142) with station 4 having the highest rate and station 3 the lowest. - Once again the discrepancy between the two methods may be explained by the wide temporal spawning range. The number of days over which 5 mm larvae were collected is used to demonstrate the seasonal range for spawning. Length of the spawning season in days for 1983 gizzard shad and emerald shiners is as follows: STATION SPECIES Gizzard Shad Emerald Shiner 1 35 22 2 14 45 3 39 31 4 49 43 5 56 42 6 46 25 7 73 6 The number of potential spawning days ranges from 6 to 73. This variability results in an irregular flux of newly hatched larvae masking the actual growth rate results. MORTALITY RATES Gizzard Shad Data resulting from instantaneous mortality calculations is more difficult to interpret thatn the growth data. Z, the esti mated mortali ty coeffi ci ent should be positive under normal circumstances. In many cases (32 ) of the data points for both 1983 and 1984, many of these data points exist for species when the ml tial population size (No) is sma1l (i.e. < 10 fi sh/1000 cubic meters). One case of negative mortality (an increase in population size through time) occurred for gizzard shad at station 3, 1983, with an initial population density of 45.5 fish per 1000 cubic meters. This data will be deleted from further discussion due to a lack of suffi ci ent densi ty. The 1983/1984 mortality results for all species are presented in Tables 14 and 15 respectively. Mortality data for gizzard shad are as follows: ------- YEAR STATION Z 83 1 .018 2 .001 3 .010 4 .043 5 .044 6 7 84 1 .073 2 .082 3 .049 4 .049 Mortality rate coefficients ranged from .010 to .082. As with the instantaneous growth rates, station 2 mortality rates were the most inconsistent between the 2 years ranging from a low mortality (.001) in 1983 to the highest mortality rate observed in 1984 (.082). Station 2 growth data demonstrated the lowest growth (.017) in 1983 and the highest in 1984. In summary, 1983 station 2 data had the lowest growth rate and also the lowest mortality. In 1984, when the growth rate was high, mortality was also high. StatIon 1 gizzard shad mortality data also presented a dichotomy between the two field years due to the large increase in mortality during 1984. Little is written in the literature about the calculation of larval fish instantaneous mortality data. Hackney and Webb (1978) present only one example of intstantaneous mortality for larval crappie in which Z = 0.1067. This represents higher mortalities than those observed during this study. Hackney and Webb were also dealing with much more dense fish populations (i.e. No = 7.6 x 10 ) which exceeds any of the population sizes encountered in the Raisin. In general, mortality rates are much higher for the 1984 field season and unless data can be correlated on a yearly and a station basis for food availability, toxic contamination it will be difficult to interpret the significance of these results. PATHOLOGY Although the preliminary analysis did not include any fish from the control station the hi stologi cal al terati oris observed were determined to be real and in most individuals, severe. The lesions of acute epithelial necrosis in tissues in contact with environmental water (sensory organs,oropharynx,proximal esophagus and gills)are compatable with acute toxicity due to direct action of an environmental contaminant. Similar lesions in the intestine and excretory kidney are compatable with concentrations of a toxic substance at sites of absorption, metabolism and/or excretion. There was no observable evidence of carcinogenic ty. ------- Probably the most significant lesions were those observed in the gills and kidney. Significant necrosis in these tissues may impair gas exchange,electrolyte concentration,nitrogen metabolism and osmotic regulation, which might adversely affect performance. Hyaline droplet degeneration in renal tubular epithelium is often correlated with excessive proteinuria. The extent, frequency and severity of these lesions in the gizzard shad might reasonably be expected to have a negative effect on the exposed local fish population. In addition, if gizzard shad retain any toxic substances, predation by piscivorous fish, birds and mammals might cause accumulation and potentially cause lesions at higher trophic levels. The histological changes observed in the control station gizzard shad are interpretted to be real and significant pathological lesions. The lesions were characteristic of acute coagulation necrosis and ranged in severity from mild to severe. These lesions are almost identical to those found in the river shad, and while not diagnostic, are compatable with toxic etiology. The tissue distribution of lesions is strikingly similiar to that observed in the gizzard shad from the river stations. As observed in the river shad, the tissue pattern is consistent with an environmental toxicant which is concentrated or transported in the intestine and kidney. Although the numbers of alewife and yellow perch are too few to draw conclusions, it appears that alewife were simuliarily but less severly affected than gizzard shad. Likewise, the data suggests that yellow perch seem more resistant. The finding of fish with lesions similiar to those from the river and the control station was unexpected (Table 27). There are several possible explanations. One explanation is that the fish move between the two localities and that the two samples represent a single fish subpopulatjon. An argument against this hypothesi S may be found in the histologic observations of the two groups of shad. Gill parasites were observed on 34 of 47 (72.3%) of the river shad. However, only 10 of 36 (27.8 Z) shad from the control lake station had gill parasi tes. Thi s seems to be a large difference and suggests that the samples are drawn from either separate shad subpopulatjons or that exchange between the two localities is very slow. A second explanation for the pathologic changes in the control shad is that the control shad station is contaminated with similiar toxicants to those in the river system. Comparison of water chemistry data from the two localities not only will be helpful in answering this question but may also suggest which substance(s) may be involved. A third explanation is that the lesions might be caused by an unaccounted for variable common to both localities but unrelated to pollution (i.e. viral disease). In the authors opinion, the most likely explanation is contamination of the control station. If thi s is correct, it might indicate that a serious health problem exists for gizzard shad and perhaps alewife over a wide range of environments in Lake Erie. ------- Failure to observe lesions in the experimentally exposed fathead minnows might be explained by any of several hypotheses. It is possible that the fathead minnows are either more tolerant or resistant to the exposed toxic material than the naturally exposed fish (gizzard shad). Alternatively, there may have been an insufficient level of toxic material In the experimental system or low bioavailability of material which was present. It is also possible that there was insufficient time for lesion development. ------- REFERENCES CITED Boersen,G. 1984. Report of an industrial wastewater survey conducted at Ford Motor COmpany, Monroe Stamping Plant, Michigan Department of Natural Resources. 15 p. Bodola, A. 1955. The Life History of the Gizzard Shad, Dorosoma cepedianum (LeSeur), in western Lake Erie. Ph.D. Dissertation , The Ohio State University, 130 p. Burby,B.G., Barnes, M.D., and Herdendorf,C.E., 1983. Organochiorine contaminant concentrations and uptake rates in fishes in Lake Erie tributary mouths. The Ohio State Uni versi ty, Center for Lake Erie Area Research, Columbus, CLEAR Tech. Rep. No. 241. 185 p. Carlander, K. D. 1970. Handbook of Freshwater Fishery Biology. Volume 1. Life History Date of Freshwater Fishes of the United States and Canada, Exclusive of the Percjformes. The Iowa State University Press, 752 p. Cole,R.A. 1978. Entrainment at a oncethrough cooling system on western Lake Erie. Institute of Water Research and Department of Fisheries and Wildlife Michigan State University. EPA600/3-78070. pp. 110. Cooper, C.L., Bartholomew,W.C., Herdendorf,C.E., Reutter, J.M., and Snyder,F.L., 1981a. Lirnnetic larval fish of the Mauniee and Sandusky river estuaries. J. Great Lakes Res. 7(1): 5154. Cooper, C.L., Heniken,M.R., and Herdendorf,C.E., 1981b.,, Limnetic larval fish of the Ohio waters of western Lake Erie, 19751976. J. Great Lakes Res. 7(1):6264. Cooper, C. L., Mizera, J. J., and Herdendorf, C. E. 1981c. Distribution, abundance and entrainment studies of larval fishes in the western and central basins of Lake Erie. The Ohio State University, Center for Lake Erie Area Research, Columbus, CLEAR Tech. Rep. No. 222, 149 p. CrunkiltonR, 1977. Personal Communication. Filkins,J.C., Mullin,M.D .,, Richardson,W.L., Smi th,VE., Rathbun,J.E,, Rood, S.G., Rygwelski,K.R.an Kipp,T. ,1985. A report on the surficial and vertical distribution of pol ychi on nated bi phenyl s in the sedi ments of the lower River Raisin , Monroe Harbor,Michigan1983 and 1984. Report to the USEPA LArge Lakes Research Station, Grosse lie, Michigan. Gordon, 1. C. 1982. Instantaneous Growth Rates, Spatial and Temporal Distributions of Abundant Larval Fish in the Western Basin of Lake Erie at Locust Point, Ohio. M.S. ------- Thesis, The Ohio State Unvierslty, 51 p. Hackney,P.A., and Webb,J.C., 1978. A method for Determing growth and mortality rates of icthyoplankton. Pages 115 124 in L.D. Jensen (ed.) , Fourth National Workshop on Entrainment and Impingement. EA Communications Melville, N.Y. Horvath,F., 1985. Monroe Harbor Report. SWWQS CLP3-27002. Michigan Department of Natural Resources Unpublished Report, 9 pg. Hubbs,C.L. and Lagler,K.F ., 1964. Fishes of the Great Lakes region. The University of Michigan Press, Ann Arbor. 213 p. Jones, J.D., 1983. Personal communication. University of Michigan, Ann Arbor, Michigan. Jude,D.J., Mansfield,p .J ., and Perrone,M.,Jr.1983. Impingement and entrainment of fish and effectiveness of the fish return system at the Monroe Power Plant, Western Lake Erie, 19821983. Special Report No. 101 Great Lakes Research Division, University of Michigan, Ann Arbor, Mi Michigan Department of Natural Resources, Water Quality Division, Biology Section. 1979. River Quality in the River Raisin basin. pp. 34. Michigan Department of Public Health and the Michigan Water Resources Commission. 1969. The River Raisin basin. p. 74 Miller, R. R. 1960. Systematics and biology of the Gizzard Shad (Dorosoma cepedianum) and related fishes. Fishery Bulletin 173, Volume 60, U. S. Fish and Wildlife Service. Mizera,J .J., Cooper,C.L., and Herdendorf,C.E., 1981. Ljmnetjc larval fish in the nearshore zone of the western basin of Lake Erie. J. Great Lakes Res. 7(1):6264. Monroe county Drain Commission. 1984. Environmental Assessment Monroe Metropolitan area. pp. 1-6. Mozola,A.J., 1970. Geology for environmental planning in Monroe County, Michigan report investigation 13. Geological Survey Division, Department of Natural Resources. pp. 18. Petty,S.M.,, 1984. Personal communication. City of Monroe Smith,V.E., Rathbun,J.E., Rood.,S.G., Rygwelski,K.R., ------- Rathbun, J. 1985. Richardson, W.L., and Dolan, D.M., 1985. Distribution of Contaminants of Monroe Harbor (River Raisin), Michigan and Adjacent Lake Erie. USEPA, Large Lakes Research Station, 154 p. Snyder, 0. E. 1976. Terminologies for intervals of larval fish development. pp. 4160 In J. Borrrnan (ed.), Great Lakes Fish egg and larvae Identification. U. S. Department of the Interior, Fish and Widlife Service, Washington, D.C., FWS/OBS76/23. Snyder, F. L. 1978. Ichthyoplankton studies in the Maumee and Sandusky River Estuaries of Lake Erie. The Ohio State University, Center for Lake Erie Area Research, Columbus, CLEAR Tech. Rep. No. 92, 140 p. Tin, H. T. and Jude, 0. J. 1983. Distribution and Growth of Larval Rainbow Smelt in Eastern Lake Michigan, 1978 1981. Trans. Amer. Fish. Soc. 112:517524. Trautman,M.B., 1981. The fishes of Ohio. The Ohio State University Press, Columbus. 782 p. U. S. Department of the Interior. 1967. U. S. Environmental Protection Agency. 1975. Monroe, Michigan. Report on the degree of pollution of bottom sediments. 1975 Harbor Sediment Sampling program, April 9, 1975. U. S. Environmental Protection Agency, Region 5, Great Lakes Surveillance Branch, Chicago, 9 p. U. S. Geological Survey. 1982. Water resources data Michigan water year 1981. Water Data Report MISii, 435 p. Van Meter,H.D., and Trautman,M.B., 1970. An annotated list of the fishes of Lake Erie amd its tributary waters exclusive of the Detroit River. Ohio J. Sd . 70:6578. Wickliff. ,E.L., 1931. Fishery research by the -Ohio Department of Conservation. Trans. Amer. Fish. Soc. 61:199207. White, A.M., Trautman,M.B., Foell,E.J., Kelty,M.P., and Gaby,R., 1975. Water quality baseline assessment for the Cleveland area Lake Erie. Vol. II. The fishes of the Cleveland metropolitan area Including the Lake Erie shoreline. U.S. Environmental Protection Agency , EPA905/975O01. 181 p. ------- TABLES ------- Table 1 Larval Fish Species for the 1983 1984 River Raisin SPECIES # CODE 101 102 103 104 105 106 107 108 109 110 201 202 203 204 COMMON NAME carp goldfish shiner or minnow spottail shiner emerald shiner central stoneroller bluntnose minnow golden shiner creek chub silverjaw minnow whi te sucker lake chubsucker quillback carpsucker uni denti fled sucker SCIENTIFIC NAME Cyprinus carpio Carassius auratus Cypri ni d Notropis hudsonius Notropi s atheri noi des Campostoma anomalum Pimephales notatus Notemigonus crysoleucas Semotilus atromaculatus Ericymba buccata Catostomus commersoni Erimyzon sucetta Carpiodes cyrpinus Catostornus sp. 301 302 401 402 403 404 405 501 601 602 603 701 702 703 704 705 706 al ewi fe gizzard shad channel catfish stonecat madtom yellow bullhead tadpole madtom unidentified catfish troutperch wh.bass or wh.perch white bass whi te perch green sunfi sh unidentified sunfish white crappie rock bass wh. or bi. crappie bluegill Alosa pseudoharengus Dorosoma cepedianum Ictalurus punctatus Notorus flavus Ictalurus natalis Notorus gyrinus Ictalurus sp Percopsi S omi scomaycus Morone sp. Morone chrysops Morone americana Lempomi s cyanel 1 us Lepomis sp. Pomoxis annularis Ambl opi I tes rupestri 5 Pomoxi $ sp. Lepomis macrochirus ------- TABLE 1 (Continued) SPECIES # CODE 801 802 803 804 805 806 901 1001 1101 1201 1301 1401 1501 1901 COMMON NAME yellow perch logperch sauger wall eye johnny darter perch or darter freshwater drum rainbow smelt brook silverside killifish or topminnow northern pike Brook stickleback lake whitefish Unidentified SCIENTIFIC NAME Perca flavescens Perci na caprodes Sti zostedion canadense Stizostedion v. vi treum Etheostoma nigrun Perci dae Api odi notus grunni ens Osmerus mordax Labidesthes sicculus Fundulus sp. Esox lucius Culaea inconstans Coregonus ci upeaformi s Unidentified ------- TABLE 2 Abundance of Larval Fish Collected in the River Raisin, 1983 Total # Larvae % of Total Collected, 1983 Collected, 1983 1 Gizzard shad 11,440 72.1 2 Emerald shiner 919 5.8 3 Carp 814 5.1 4 Morone sp. 701 4.4 5 Freshwater drum 512 3.3 6 Spottail shiner 345 2.2 7 Channel catfish 245 1.5 8 Yellow perch 215 1.4 9 Lepomis spp. 114 0.7 10 Unidentified Cyprinid 99 0.6 11 Alewife 63 0.4 12 Walleye 67 0.4 13 White bass 49 0.3 14 Logperch 44 0.28 15 Brook silverside 49 0.27 16 Rock bass 29 0.18 17 Rainbow smelt 25 0.17 18 White sucker 16 0.101 19 Troutperch 16 0.100 20 Bluntnose minnow 14 0.088 21 Pomoxis sp. 12 0.080 22 Tadpole madtom 9 0.057 23 White crappie 8 0.050 24 Sauger 7 0.040 25 Stonecat madtorn 6 0.040 26 Johnny darter 4 0.020 27 Unidentified 5 0.020 28 Unidentified percid 2 0.020 29 Silverjaw minnow 3 0.019 30 Quillback carpsucker 2 0.013 31 Golden shiner 2 0.013 32 Green sunfish 2 0.010 33 White perch 2 0.010 34 Goldfish 1 0.006 35 Central stoneroller 1 0.006 36 Lake chubsucker 1 0.006 37 Creek chub 1 0.006 38 Unidentified catostomid 1 0.006 39 Yellow bullhead 1 0.006 40 Ictalurus sp. 1 0.006 41 Bluegill 1 0.006 42 Unidentified Fundulus 1 0.005 TOTAL 15,849 99. 834 ------- TABLE 3 Abundance of Larval Fish Collected in the River Raisin, 1984 Total # Larvae % of Total Collected, 1984 Collected 1984 1 Gizzard shad 18,853 73.7 2 Carp 1,849 7.3 3 White bass 976 3.8 4 Morone sp. 952 3.7 5 Channel catfish 907 3.5 6 Freshwater drum 532 2.1 7 Lepomis sp. 457 1.8 8 Emerald shiner 365 1.4 9 Spottail shiner 225 0.9 10 Rainbow smelt 124 0.5 11 White sucker 83 0.3 12 Yellow perch 44 0.17 13 Walleye 40 0.16 14 Troutperch 39 0.15 15 Logperch 29 0.11 16 Tadpole madtom 25 0.10 17 Pomoxis sp. 12 0.047 18 White crappie 10 0.039 19 Rock bass 10 0.039 20 Stonecat madtom 9 0.035 21 Lake chubsucker 8 0.031 22 Alewife 8 0.031 23 White perch 5 0.0195 24 Johnny darter 5 0.0195 25 Unidentified 5 0.0195 26 Northern pike 4 0.0156 27 Green sunfish 2 0.0078 28 Unidentified cyrpinid 1 0.0039 29 Bluntnose minnow 1 0.0039 30 Brook stickleback 1 0.0039 31 Lake whitefish 1 0.0039 32 Yellow bullhead 1 0.0039 TOTAL 25, 583 100. 0 134 ------- TABLE 4 Distribution of Larval Fish Collected in the River Raisin, 1983 Station # 1 2 3 4 5 6 7 Gizzard shad Emerald shiner Carp Morone sp. Freshwater drum Spottail shiner Channel catfi sh Yellow perch Lepomis sp. Unident. cyprinid Al ewi fe Wall eye White bass Logperch Brook silverside Rock bass Rainbow smelt White sucker Trout-perch Bluntnose minnow Pomoxjs sp. Tadpole madtom White crappie Sauger Stonecat madtom Johnny darter Unidentified Unident. Percid Silverjaw minnow Quiliback carpsucker Golden shiner Green sunfi sh Whi te perch Gol dfi sh Cent. stoneroller Lake chubsucker Creek chub Unident. catostomid Yellow bullhead Ictalurus sp. 91 uegi 11 Unident. Fundulus x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ------- TABLE 5 Distribution of Larval Fish Collected in the River Raisin, 1984 Station # 1 2 3 4 1 Gizzard shad x x x x 2 Carp x x x x 3 White bass x x x x 4 Morone sp. x x x x 5 Channel catfish x x x x 6 Freshwater drum x x x 7 Lepomis spp. x x x x 8 Emerald shiner x x x x 9 Spottail shiner x x x x 10 Rainbow smelt x x x x 11 White sucker x x x x 12 Yellow perch x x x x 13 Walleye x x x x 14 Troutperch x x x 15 Logperch x x x 16 Tadpole madtom x x x 17 Pomoxis sp. x x 18 White crappie x x x 19 Rock bass x x 20 Stonecat madtom x x x 21 Lake chubsucker x x x 22 Alewife x x x x 23 Whi te perch X 24 Johnny darter x x x 25 Unidentified x x x x 26 Northern pike x x 27 Green sunfish x 28 Undentified cyprinid x 29 Bluntnose minnow x 30 Brook stickleback X 31 Lake whi tefi sh 32 Yellow bullhead x ------- TABLE 6 Ranking of Species Abundance Determinined in the River Raisin 1983 Study Compared to Judes 1982 Study Jude osu osu 1982 1983 1984 SPECIES Gizzard shad 1 1 1 Emerald shiner 7 2 8 Carp 5 3 3 Morone Sp. 3 4 4 Freshwater drum 2 5 6 Spottail shiner 9 6 9 Channel catfish ii 7 5 Yellow perch 4 8 12 Lepomis 17 9 7 Cyrpinid (unid) 12 10 28 Alewife NF 11 22 Walleye 15 12 13 White bass NF 13 3 Logperch 18 14 15 Brook silverside NE 15 NE Rock bass NF 16 19 Rainbow smelt 8 17 10 White sucker 16 18 11 Troutperch 14 19 14 Bluntriose minnow 23 20 29 Pomoxis Sp. 20 21 17 Tadpole madtom 24 22 16 Damaged larvae 6 NF HF Quiliback carpsucker 10 30 HF Burbot 13 HF HF Largemouth bass 19 NE HF Lake whitefish 21 NE 31 Northern hogsucker 22 NE HF ------- TABLE 7 Estimated Numbers of Fish Larvae Entrained from February 13, 1982 through February 12, 1983 at the Monroe Power Plant (Data Taken from Jude, et al., 1983) Total Impinged % of Total 0 4.08 x 10 86.8 8 1.58 x 10 3.4 8 1.56 x 10 3.3 8 1.28 x 10 2.7 7 8.0 x 10 1.7 7 3.8 x 10 0.8 7 2.3 x 10 0.5 7 1.1 x 10 0.2 6 5.0 x 10 0.1 6 4.9 x 10 0.1 6 4.1 x 10 0.09 6 2.8 x 10 0.06 3 2.8 x 10 0.06 6 2.4 x 10 0.05 6 2.1 x 10 0.04 6 1.2 x 10 0.03 5 9.2 x 10 0.02 5 6.0 x 10 0.01 5 6.0 x 10 0.01 5 5.8 x 10 0.01 5 1.9 x 10 0.004 3 1.2 x 10 0.003 99.987 Species Gizzard shad Freshwater drum White bass and White perch Yellow perch Common carp Damaged larvae Emerald shiner Rainbow smelt Spottail shiner Quiliback carpsucker Channel catfi sh Uni denti fled Cypri ni d Burbot Troutperch Wal 1 eye Whi te sucker Lempomis spp. Logperch Largernouth bass Pomoxis spp. Unidentified Coregonid Northern hogsucker TOTAL ------- TABLE 8 Gizzard Shad Simple Growth Rates River Raisin 1983/1984 Data Taken from Printout Step 4 Species Station Year Initial Final Initial Final dl (mm) Day Day Size Size ai (day) (mm) (mm) Gizzard 1 1983a 199 220 40.5 49.2 0.41 shad b 202 244 31.2 48.2 0.40 c 192 237 14.7 35.3 0.46 d 171 181 4.0 6.5 0.25 2 1983a 160 209 3.5 41.6 0.78 b 171 227 3.5 40.4 0.66 c 174 230 4.3 33.5 0.52 3 1983a 160 195 3.5 38.0 0.98 b 164 209 3.9 47.4 0.97 c 167 216 3.8 49.9 0.94 4 1983a 160 223 3.0 45.3 0.67 b 209 251 12.3 41.7 0.70 5 1983a 150 234 6.7 37.3 0.36 b 160 227 6.8 31.5 0.37 c 171 230 6.0 27.2 0.36 d 188 241 10.1 30.0 0.38 e 209 216 13.9 16.4 0.36 6 1983a 150 227 8.0 32.9 0.32 b 167 234 8.9 30.8 0.33 c 174 230 9.7 28.3 0.33 d 160 181 4.0 9.7 0.27 e 199 213 11.7 16.9 0.37 7 1983a 164 216 7.8 24.9 0.33 b 167 227 7.5 26.0 0.31 c 171 213 5.2 19.8 0.35 d 195 220 11.4 20.4 0.36 e 199 234 6.8 17.8 0.32 ------- TABLES (Continued) Gizzard Shad Simple Growth Rates Data taken from Printout Step 4 Species Station Year Initial Final Initial Final dl (mm) Day Day Si ze Si ze i ( iy) (mm) (mm) Gizzard 1 1984a 145 201 3.3 43.4 0.72 shad b 159 194 3.1 29.0 0.74 c 163 191 3.3 23.2 0.71 d 166 180 3.3 9.9 0.47 e 170 184 3.3 11.2 0.56 2 1984a 163 187 3.5 21.6 0.75 b 166 184 3.4 16.1 0.70 c 173 198 7.6 24.6 0.68 d 170 194 3.4 19.8 0.68 3 1984a 170 176 3.8 16.4 2.10 b 180 187 8.6 23.7 2.20 c 184 194 13.4 32.8 1.90 d 170 191 3.8 18.4 0.70 e 163 198 3.4 25.8 0.64 f 159 187 3.2 23.7 0.73 4 1984a 149 201 3.3 35.4 0.62 b 156 180 3.2 18.1 0.62 c 166 198 3.9 22.6 0.58 d 170 194 3 .9 15.2 0.47 e 170 198 3.9 22.6 0.68 ------- TABLE Emerald Shiner Simple Growth Rates, River Raisin 1983/1984 2 1983a 171 b 195 mi tial Size (mm) Fl nal Si ze (mm) dl (mm) at caiy) Species Station Year initial Final Day Day Erneral d shi ner Emerald shi ner 1 1983a b c d e 171 192 171 192 174 230 234 216 241 223 5.9 6.0 5.9 6.0 6.0 31.3 14.0 31.0 33.8 30.8 0.43 0.19 0.56 0.57 0.51 241 255 6.0 6.5 34.3 36.0 0.40 0.49 3 1983a b c d 160 171 174 227 230 209 251 255 6.5 7.5 5.5 11.1 39.9 22.9 38.4 34.9 0.48 0.41 0.43 0.85 4 1983a b c d e 167 171 171 206 209 202 241 192 244 227 5.0 5.0 5.0 11.8 7.8 23.1 31.8 11.0 25.2 13.5 0.52 0.38 0.29 0.35 0.32 5 1983a b 181 164 230 241 14.0 4.0 26.0 20.5 0.24 0.21 6 1983a b c 185 181 206 195 216 230 17.3 5.5 11.0 19.4 20.2 19.0 0.21 0.42 0.33 7 1983a b c 185 202 206 195 251 230 13.4 13.3 13.0 19.0 32.7 23.0 0.56 0.40 0.42 1 1984a b c 166 163 184 198 187 191 5.5 7.0 8.2 25.8 16.5 12.5 0.63 0.40 0.61 2 1984a b 176 170 198 194 13.5 5.5 29.0 23.3 0.70 0.74 3 1984a 166 201 3.8 40.9 1.06 4 1984a 163 180 5.5 14.4 0.52 ------- TABLE -10 Larval Fish Growth Rate Coefficients, River Raisin 1983 Species Code: Lo: Li: Lo :L1: Convergence: For species identi Length at initial Growth rate coeffi Correlation Yes or No ty see Table 1 time fish observed ci ent = Slope of Growth STATION # SPECIES CODE Lo Li Correlation Lo Li CONVERGENCE 1 1 1 1 105 301 302 1101 3.5 55.7 4.3 30.7 0.037 0.032 0.051 0.022 .9919 .9074 .9923 .9399 yes yes yes yes 2 2 105 302 2.7 15.5 0.048 0.017 .9926 .9826 yes yes 3 3 3 3 105 301 302 1101 3.0 52.0 5.4 29.0 0.036 -0.006 0.036 0.016 -.9892 .8354 .9920 -.8943 yes yes yes yes 4 4 4 4 4 4 4 101 103 105 302 702 802 901 8.6 3.0 10.1 10.0 2.8 11.9 10.0 0.017 0.023 0.023 0.024 0.031 0.053 0.047 .9620 .9735 .9551 -.9724 .9787 -.7970 -.9306 yes yes yes yes yes yes yes 5 5 5 5 5 5 5 . 105 302 702 801 802 901 1001 9.2 10.8 0.5 7.0 4.1 12.9 18.6 0.016 0.019 0.092 0.037 0.023 0.038 0.013 .9324 .9488 .9961 .9209 .9850 .8923 .9081 yes yes yes yes yes yes yes 6 6 105 302 13.8 23.4 0.022 0.017 .7701 .8542 yes yes 7 7 105 302 12.1 14.1 0.021 0.025 .8939 .9401 yes yes ------- TABLE :i i Larval Fish Growth Rate Coefficients, River Raisin 1984 STATION # SPECIES CODE Lo Li CORRELATION Lo : Li CONVERGENCE 1 101 3.7 0.035 .9520 yes 1 104 193.0 -0.328 .9962 yes 1 105 0.34 0.125 .9970 yes 1 302 1.45 0.090 .9974 yes 1 401 8.80 0.205 .9819 yes 1 402 15.40 0.026 -.7528 yes 1 404 0.53 0.031 -.8789 yes 1 601 1.13 0.116 -.9918 yes 1 602 11.77 0.047 .9673 yes 1 702 56.15 0.08 .9729 yes 1 804 10.0 -0.022 -.4472 yes 2 101 1.35 0.069 .9741 yes 2 104 1.27 0.057 .9903 yes 2 105 3.69 0.081 .9808 yes 2 302 2.76 0.059 .9942 yes 2 401 6.34 0.225 -.9914 yes 2 601 9.56 0.066 -.8813 yes 2 602 11.99 0.072 .9960 yes 2 702 1.08 0.079 -.9913 yes 2 805 5.00 0.046 .9486 yes 3 101 0.87 0.077 .9853 yes 3 104 5.92 -1.776 .0000 yes 3 105 20.34 0.009 .1995 yes 3 201 12.34 0.020 .7646 yes 3 302 4.4 0.057 .9870 yes 3 401 48.6 -0.394 .9900 yes 3 404 5.0 0.101 -.0000 yes 3 601 4.3 0.032 .9225 yes 3 602 12.5 0.031 .9765 yes 3 702 0.04 0.220 .9987 yes 3 704 22.34 0.055 .8771 yes 3 801 5.89 0.055 .9722 yes 3 802 23.20 0.021 .6406 yes 3 901 20.45 0.055 .5986 yes ------- TABLE 11 (Continued) 4 101 10.93 0.039 .9804 yes 4 105 16.57 0.142 .9683 yes 4 302 6.84 0.056 .9735 yes 4 601 3.12 0.064 .9824 yes 4 602 16.25 0.031 .8666 yes 4 603 14.47 0.051 .8942 .yes 4 702 15.35 .131 .6710 yes 4 801 2.03 0.062 .9927 yes 4 802 11.68 0.007 -.9004 yes 4 901 4.75 0.098 .9752 yes 4 1001 33.62 .176 .7241 yes ------- TABLE 12 Ranking of 1983 Larval Fish Growth Rate Coefficients Growth Species Station Li Rank Mortality Gizzard Shad (302) 1 0.051 1 3 2 0.017 7 4 3 0.036 2 5 4 0.024 4 2 5 0.019 5 1 6 0.017 7 7 0.025 3 Emerald Shiner (105) 1 0.037 2 2 0.048 1 3 0.036 3 4 0.023 4 5 0.016 7 6 0.022 5 7 0.021 6 Re Rank Combination 1 3 1 = highest GS+ES 2 8 3 3 5 2 4 8 3 5 12 7 = lowest 6 12 7 = lowest 7 9 5 ------- TABLE 13 Ranking of 1984 Larval Fish Growth Rate Coefficients Species Station Li Rank Mortality Gizzard Shad (302) 1 0.090 1 2 2 0.059 2 1 3 0.057 3 4 4 0.056 4 4 5 Emerald Shiner (105) 1 0.125 2 2 0.081 3 3 0.009 4 4 0.142 1 Re Rank Combination GS + ES 1 3 1 = highest 2 5 3 3 7 4 = lowest 4 5 3 5 ------- TABLE 14 River Raisin 1983 Larval Fish (Taken from Step 7) Mortal i ty Station N0:Z Convergence Species Code Estimate No Estimate Z 1 105 301 302 1101 4.63 4.74 40.65 1.38 .039 .138 .018 .007 2 105 302 1.73 10.36 .048 .001 3 105 301 302 1101 10.04 1.16 45.54 5.00 .007 .027 .010 .049 4 5 6 0.9905 0.6805 0.9314 0.9028 yes yes yes yes 0.9924 0.9568 yes yes 0.9673 0.9435 0.9657 0.6882 yes yes yes yes 101 103 105 302 102 802 901 1072.56 292.97 33.46 269.04 239.41 3.31 5.28 .102 .068 .032 .043 .075 .063 .032 0.9909 0.9872 0.9376 0.9052 0.9966 0.5106 0.7824 yes yes yes yes yes yes yes 105 302 702 801 802 901 1001 32.80 1916.35 2.66 16.47 98.72 73.10 1.54 - .024 .044 .035 .012 .060 .207 .006 0.9606 0.9740 0.9935 0.7682 0.9887 0.9614 0.8795 yes yes yes yes yes yes yes 105 601 602 702 801 802 804 901 1001 53.80 821.00 2.91 11.84 18.25 6.71 6.84 6.32 2.36 - .037 .094 .054 .106 .013 .018 .130 .350 .430 0.9352 0.9666 0.1043 0.8770 0.7798 0.9837 0.9700 0.9867 0.9499 yes yes yes yes yes yes yes yes ------- 7 TABLE 14 (Continued) 301 302 601 602 702 802 901 1001 4.07 6355.70 48.01 7.27 0.62 5.73 12.15 1.40 .043 .072 .110 .062 .191 .018 .076 .006 0.6613 0.9868 0.9669 0.7963 0.9826 0.8740 0.8800 0.7384 yes yes yes yes yes yes yes yes No: Initial density over day by station and species Z: Estimated mortality rate ------- TABLE 15 River Raisin 1984 Larval Fish Mortality Coefficients (Taken from Step 7) No: Initial density over day by station and species Species No Z Correlation No:Z Convergence Station Code 1 2 3 4 105 302 404 601 602 .26 737 6.29 18.1 62.1 .128 0.073 0.027 0.058 0.096 0.9972 0.6541 0.6507 0.6574 0.9639 yes yes yes yes yes 104 105 302 602 2717 7.8 2182 36.2 0.184 0.005 0.081 0.023 0.9997 0.9638 0.9202 0.9167 yes yes yes yes 104 105 201 302 601 602 801 802 7.23 18.3 10.2 987 7.1 18.15 3.97 1.18 .022 .005 0.104 0.049 0.108 0.007 0.052 0.004 0.9559 0.411 0.451 0.769 0.9931 0.9713 0.541 0.7746 yes yes yes yes yes yes yes yes 101 302 901 107 2709 198 0.078 0.049 0.074 0.6676 0.7023 0.5414 yes yes Z: Estimated mortality rate ------- Macroscopi Date Station Jul-4 4-A Jun16 4C Jul21 4C Aug-8 3A Jun23 5C Jul4 5A Jul 4 5C Aug4 2A Jun-20 4-C TABLE 16 cally Observed Deformities in Larval Fish from the Ri ver Rai si n dun ng 1983 Name Age Length Deformity (mm) Gizzard shad III 19.0 irregular spine curvature Yellow perch II I 12.5 spinal deformi ty Gizzard shad IV 36.0 abnormal growth mass on stomach Gizzard shad IV 54.0 stomach tumor Gizzard shad II 11.5 6 specimens 15.0 with severe spine curvature Gizzard shad III 17.0 severe spine defect Gizzard shad III 15.5 many with spine 22.0 curvatures Rock bass II 7.0 tumor near tail Gizzard shad II 15.0 2 specimens with severe spine curvature Gizzard shad II 14.0 spinal deforoii ty Jul 7 4-C ------- TABLE 17 Larval Fish By Species and Station Evaluated Pathologically Species Station 1 2 3 Gizzard shad Yellow perch Spottail shiner Emerald shiner Carp Logperch Trout-perch Channel Catfi sh Al ewi fe Freshwater drum Wal 1 eye Morone sp. 2 3 48 24 4 4 1 1 2 1 1 - 2 1 2 2 - - - 3 - - - - 3 77 8 1 1 2 1 1 2 1 4 3 3 4 5 Total Total = 104 ------- TABLE 18 Lesions in Gizzard Shad from River Raisin,1983 Ti ssue Eye Brai n Spinal Cord 01 factory Organ Otolith Organ Lat. Line Organ Oropha rynx Esophagus (anteri or) Gill $ Gill s Gill s Heart Stomach Intestine Li ver Pancreas Excretory Kidney Excretory Kidney Hemo. Kidney Spi een Swim Bladder Thy mu s Ski n Skeletal Muscle Cartilage Bone Edema Epi thel I al Parasj tes Tubular epithelial necrosi S Hyaline droplet degenera ti on No. Affected 0/50 0/59 0/53 35/3 7 3/46 34/3 6 50/52 43 / 47 38/47 43/47 34/47 0/42 0/47 38/54 0/53 0/51 50/53 17/53 0/50 0/26 0/54 0/41 0/60 0/71 0/64 0/45 Z Affected 0.0 0.0 0.0 94.6 6.5 94.4 96.2 91.5 80.8 91.5 72.3 0.0 0.0 70.4 0.0 0.0 94.3 32.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Lesion Epi thel I al Epi thel i al Epi thel I al Epi thel i al Epi thelial necrosi S necrosi S necrosi S necrosi $ necrosi S necrosi s Epithelial necrosis ------- TABLE 19 Gizzard Shad Larvae By Size and Station Indicating Lesions. Collection Fish Length Station < 20 mm 21-30 mm 3140 mm > 41 mm C A C A C A C A 1 0 0 1 1 0 0 1 1 3 0 0 1 1 0 0 2 0 4 14 1 13 13 14 13 7 7 5 14 5 3 3 7 7 0 0 Total 28 6 18 18 21 20 10 8 C No. fish collected A = No. fish with lesions ------- TABLE 20 Histopathological Lesions in Gizzard Shad from Control Station (#7) Ti ssue Lesion No. Affected Z Affected Eye Brain Spinal cord Olfactory organ Otolith organ Lat. Line organ Oropha rynx E s op ha g u S Gill s Gill s Heart Stomach Intestine Li ver Pancreas Excretory kidney Excretory kidney Hemo. kidney Spi een Swim Bladder Thyrnus Ski n Muscle Cartilage Bone Epi thel i al Epi thel i al Epi thel I al Epi thel i al Epi thel I al Epi thel I al Parasi tes Epi thelial necrosi S necrosi S necrosi S necrosi s necrosi S necrosi S 0/23 0/34 0/32 10/11 1/24 9/10 33/3 4 3 1/34 3 6/36 10/3 6 0/30 0/30 15/16 0/36 0/27 3 5/35 13/38 0/34 0/13 0/25 2/28 0/37 0/37 0/38 0/26 0.0 0.0 0.0 90.9 3.6 90.0 97.1 91.2 100.0 27.8 0.0 0.0 93.8 0.0 0.0 100.0 34.2 0.0 0.0 0.0 7.1 0.0 0.0 0.0 0.0 necrosi s Tubular epithelial necrosi S Hyaline droplet degenera ti on Lymphoid necrosis ------- TABLE 21 Lesions In Fathead Minnows STA 4 STA 12 STA 12 No. H20 + sed H20 + sed H20 + sed Affected Affected (n=6) (n=6) (n=6) (n=18) Eye 0/2 0/4 0/4 0/10 0 Brain 0/5 0/6 0/6 0/17 0 Spinal cord 0/5 0/5 0/5 0/15 0 Olfactory org 0/4 0/3 0/5 0/12 0 Otolith org. 0/1 0/1. 0/2 0/4 0 Lat. Line org 0/0 0/0 0/0 0/0 0 Oropharynx 0/5 0/5 0/6 0/16 0 Esophagus oio 0 ,1 0/0 0 ,1 o Gills 0/6 0/6 0/5 0/17 0 Heart 0/0 0/0 0/0 0/0 0 Stomach 0/1 0/1 0/0 0/2 0 Intestine 0/4 0/6 0/6 0/16 0 Liver 0/6 0/6 0/6 0/18 0 Pancreas 0/2 0/4 0/6 0/12 0 Ex. Kidney 0/5 0/6 0/6 0/17 0 Hemo. Kidney 0/5 0/6 0/6 0/17 0 Spleen 0/0 0/1 0/0 0/1 0 Swim Bladder 0/3 0/6 0/4 0/13 0 Thynius 0/2 0/2 0/0 0/2 0 Skin 0/6 0/6 0/6 0/18 0 Skeletal Mus. 0/6 0/6 0/6 0/18 0 Cartilage 0/6 0/6 0/6 0/18 0 Bone 0/6 0/6 0/6 0/18 0 ------- TABLE 22 Gizzard Shad Larval Fish Density of the River Raisin, 1983 All Sizes Station 1 Species Gizzard Shad 280.0 2 160 255 199 9.5 414.7 3 160 255 199 8.7 844.4 4 160 255 195 10.2 662.6 5 150 255 202 8.5 4723.0 6 150 251 202 7.5 5595.7 7 164 - 251 188 8.5 2101.8 Date of First Capture Date of Last Capture Period of Peak Abundance Minimum Mean Density #/1000 m Maximum Mean Density #/1000 m Average of Density Means 171 244 199 9.2 74.0 49.6 129.4 119.1 406.4 781.7 828.2 2388.4 Relative Abundance 3.1 2.1 5.4 5.0 17.0 32.7 34.7 100 ------- TABLE 23 Lake Erie Gizzard Shad Larval Density Peak Abundance Date June 8,78 May 31,79 June 6,80 June 3,76 May 31,76 June 4,77 June 19,78 Location Davi sBesse Davi sBesse Davi sBesse Maumee River Sandusky River Western Basin Central Basin Density #/1000 m 1104.4 2004.4 10369.3 16348.9 3811.7 8000.0 1070.0 Gordon, 1982 Gordon, 1982 Gordon, 1982 Snyder, 1978 Snyder, 1978 Cooper et al,1981c Cooper et al,1981c Study ------- TABLE 24 Raisin River Qualitative Sediment Survey (Data Supplied by Michigan Department of Natural Resources) STATION TRANSECT SEDIMENT DESCRIPTION 1 1 No description 2 7 Hard rocky bottom along the central portion Fine gravel/sand along north shore.Silt along south shore. 3 10 Hard rocky bottom along south shore. Silt sand and gravel along central and north portion. 4 43 Silt dark gray color,slightly oily odor,some detritus,rocky along the north shore. 5 48 Black silt,oily, some detritus, sandy silt along north shore. 6 50 Silt,gray-brown,some detritus, no unusual odor 7 NS Not sampled ------- TABLE 25 Gizzard Shad Larval Fish Density of the River Raisin. 1983 < 5 mm Stati on 1 Species Gi zzard Shad Date of Date of Date of No. of Minimum Maximum Avge of First Last of Peak Spawning Density Density Density Relative Capture Capture Abundance Days #/1000 m #/1000 m Means Abundance 171 206 174 35 9.2 101.4 38.0 5.3 2 160 174 160 14 10.3 16.5 12.5 1.7 3 160 199 199 39 8.4 51.4 31.2 4.3 4 160 209 174 49 10.2 285.2 78.2 10.9 5 150 206 174 56 7.8 590.1 109.7 15.2 6 160 206 202 46 18.0 543.3 230.1 31.9 7 164 237 206 73 10.7 1140.7 221.3 30.7 721.0 100% ------- TABLE 26 Review of Growth Rates SPECIES Gizzard Shad Gizzard Shad Gizzard Shad Gizzard Shad Yellow Perch Yellow Perch Emerald Shiner Emerald Shiner Smel t Smel t SIZE RANGE 6.2 29.0mm 36 185 mm 4.0-49.2 mm pro larvae post larvae 4- 40.9 5.3 15.7mm 5.3 4 1. lmm GROWTH RATE 0.034 (I) l.01mm/day 0.99mm/day . 2 52.2mm/day 0.018 (I) 0.038 (I) 0.80mm/day l9.85mm/day 0.35mm/day O.39mm/day LOCATION L.Eri e L.Erle I. Erie I. Erie L.Eri e L . En e L.Eri e L.Erie L.Mj chi gan L.MIchl gan AUTHOR Gordon Bodol a Carlander Fay Gordon Gordon Carl an de r Fa y Tin Ti n ------- TABLE 27 Histopatho ogic Lesions in Larval Gizzard Shad Excretory ki dney Excre tory ki dney intestine Gill Tubular epithelial necrosi s Hyaline droplet degenerati on Epithelial necrosis Parasi tes Lesion Epithelial necrosis Ti ssue Olfactory organ Lateral line Oropharynx Esophagus (ant) Sill s % Affected STA4 STA7 95 92 94 90 96 97 92 91 92 100 94 83 32 34 70 72 93 28 ------- FIGURES ------- JOIN 1$ SNI,I I.M IN 5 HIO STATE UNIVERSIfl? / LARVAL FISH STUDY RIVER RAISIN 1983 7 P MUNKflI ,//i/ 7 S r \1I I I. iii - j if - - 13 - - WEST END OF LAKE ERIE 7. 1846 / -, I/f )) 6th ed., Dec 24/17 ------- Aerial Vie i of Station 1 and 2, River aisin. Figure 2a. Figure 2b. Aerial View of Station 2and3,RiverRaisin. ------- Figure 2c. Aerial View of Station 4, River Raisin. - v )_ _ _- :1 Figure 2d. Aerial View of Station 5, River Raisin. ------- -. . . ;. . .. ;. . .; . . . . ;.. . . F I. . .. . .. . 1 . . .,. . . . I - , ..; . . - - - .. Figure 2e. Aerial View of Stations 4,5 and 6,River Raisin. ------- I is.. ,.. .a. e.tI 1 t F-% r -r4- . 4 i .e. . .,,q SSS ? iuS .at(. .* ,eI. Sues U 5e sm. s iv s as,s . S 5 ,U . I . .4s. fd / tilt / / ..._,t_._ - - - Figure 3. River Raisin drainage basin. (Taken from Michi9an Water Resources Comnjission,1965). ------- 5000. 11-Yr. Average F1 s (197O - )) JAN GflGtt O1176500 RIVER RAISIN NCRR MONROE. tl0 00. 3000. 2000. 1000. 0. 1(0 MAR APR MAY JUN JUL hUG SEP OCT NOV DEC IC - . Figure 4. Raisin River 11-year averaqe daily flow (cfs) 12 km upstream from Lake Erie (USGS data). ------- GAGES 0417650 Flqure 5 PEAK FLOWS IN THOUSANDS OF CFS Peak flows In the River Raisin at Monroe since 1938. (Data from the U.S.G.S.) F R E Q U E N C I , 14 12 10 8 6 4 2 0 0 2 4 6 8 10 12 14 ------- 1976 1977 1978 1979 1980 1981 1982 1983 AR 574 573 572 571 U A I E p S I A G E F E E T 570 569 568 567 Figure 6. Lake Erie level at gage 3087. Data Is from NOAA. ------- cIO(JJ U L IQj d 1 c O 44 ac or 1000 or fM \JOtQrfl. x 2G,B73 ( . 3) SQrnpk . ) Vo) e rn IocAor >( O gi-c e r umbQ- er / I \ o(r.\ ( d .75r Figure 7. Larval Fish Density Calculation Procedure. ------- 1.0 0.8 : . 0.6 0.4 S S 0.2 0.0 1 2 3 4 5 6 7 STATIOU # Figure 8. 1983 Growth Rates Gizzard Shad (Taken from Step 4). ------- 1.0 0.8 0.6 S S 4 0.4 S 0.2 0.0 2 3 rATIoN# 4 5 6 7 Figure 9. 1984 Growth Rates - Gizzard Shad (data taken from computer Step 4). ------- F 27, 19E - M TION 1 SPEC!ES 3O2 - - PLLLLVEAVG34DAY {r4D A .11 085, B .2 0eS ETC . ___________ ______________- -4 4 + 4 4 4 4 $ + + 4 4 4 4 + ____ 17C. 135 i c c I°5 2CC ____ 2 52i)_____ !2r. -- P. _____ - QAY Figure 10. GIzzard Shad Sunp e Growth Rates Station 1, 1933. ------- . FVLI AUC U% JrJ V UF.T UT IAI1UN t. uu 5T. T1Oh 2 SPEC1E5=3O2 PLOT OF avt .n y LfCENfls A 1 OeS. B 2 OBS. FTC. W DNEZDAT,wARCH Z7, 19 + + + + + 4 + 4 4 + 4 + + + 4+ 4 + 4 4 _16O _3 5 370 12L__L9IJ! _1 ) 9Co fl _________ -1 -1- , Gzar4-Shad -41i pIe DLY 11 nn ------- -_____ ____ - ____ AvERlcELE cT rv D.1 yey T T1 nnr PEcTEs tt o Tq r WEtNE DAT7iU RCIrZTT3 91. STAT1OP 3 SPECIES=302 ____________ __ JLGT_OF_AQC3*CAY L3 A I OILS, B 2 _ BBS, ETC. - A 0 1 f f 4 + 4 + 4 + 4 4 4 4 + 4 4 4 4 + 4 1c o 1 5 11j..... 75 I j J9p_, 5 ?CO Q5 240 245 O 25 . Figure 12. Gizzard Shad SImp e Growth Rates Stailon 3, 1983. DAY ------- t..,. t____ i_ ;_._. : s .. Z - - - A z i*l jfl?rrSPEt it cUDEc- I fl iCHtVPItSDATyhA RCIITZ7rtvnt C k tηl i L !2_t_/_en i PLOT_PF_AVη3 tOAV -1_OBS ,_U__V Des. ETC. 4&tt c r a i ,:- 2 . - I a S S I a ai , , . j I I . : t - : : ; i , . ____ 2 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - t 4 4 + 4 4 4 4 4 4 4 + 4 4 $ + + 4 4 $ - i69j4J?J15_1fl_W 1!O 195 _ 2 _ OQ q5 210 2LS 22 _ 1? __ 2 j _ 24Q 245 2S3 2 1 - DAY Figure 13. Gizzard Shad Simple Growth Aates 1 Station 4, 1983. AZ S I 4 r - S 40 4 , . iS to 10 A S A 5- A A ------- t.;. : i . YERitEtEftGTH vsvntl-JrrAnoN 1. SPECflSηCODE 1 jnq 34EpflflpnjKftRclrt1. 196r - 4 I 1 t l i 1 5TAfICri . SPECttS=3O2 , I a J Ό 1 4c t 1. M t 1 Ά4 ( , ks t J \1 wre t , 4 - \I . th + r r - itt , /r t ..J ? r PLOT W *VC3OAY - f ECEUDz DOS, 6 2UBS . ETC, .t r .4 & - 4 t I 4 . . . M j: 1 . : : 1 k 5 , η : Ό eo . I i I dr I c ItI I I : I . ;: t,v2:;i ?_ :t :2 .I : . 1 A dT5j?: ;. i i ;. ; 4 η 1 c 1 t ., , a_ so . . : 1 I Ά t I 1 ;sn i ss 160 165 11Q 115 I S O i 5 190 195 200 2)5 210 215 220 225 23) 2?5 24C 245253 2S5 2 2tS 2lt i DAY Figure 14. Gizzard Shad Simple urowtn nates, station b . s as. ------- ,T .,.,.t ) a. r f *V TRaur J .ERGTIIV5 DAYr3TATTUFU(5FECT&S CUtJU . .; 7T5ISb UE0t 1E5UATT1iA1l%IFiZ1 (.19b Ill. - it \4,iη STA 1 10Nc4 5P C1ES 302 * - rff I tI Oa i i c, flAy a itEf1Q 4 ?sA Q6t ,f5caZ D85 fJCr / r1 j + t . 4 . 1 . . ar a ,4;- _ 2,. 1 I. , & . ;j . t ;! . :Pil at: LC_a ..liv. 4 £ I- , , , - p .. .4 . - c 1 :: A :. .4. 4. I I . 1 ,. I. f t _A nit 441 a I I 1T t..I t V t 4 1 a t 1 L.. ,r1 b . ?r p I a JIt t rr * I 4% P .,, c £ $ I r r a-: a, e -4 _ 4 -V t , V . - ,Z . 4 4 4. t I l_ c t. t_aII $ ..I C _. < -n I$ , . . A . i i Ji1 . ..- . .b .. , S I I . . - , ,. , . 1 . t .11 , _ 4 - . . .,, . -fI,... . 4 1 4 4. I . 4 1 , - t f .C 14 :IAc c ; ;t : : 1 k I c kcA..wf- f :v t b . , ,.: : .. :i 1 r ; ; .j .S j., s ,,t 1 9 1 i 1c p 9 i ui ,, ηn r a 1 l- .. 14- , c s . 1 It f . 43%.4Ii _,η 4 ik,.., 1± .,t j 4 ? 1 -4 jj,, r5D.rA..j,,,4 4 44 a , SI, , j- , ,, 44 1 % ,L 1 a C p I I __I 4 1 4. I - p. 4 I I I 1 I. -4 I___ -__....4_-_._ - b t t r o P4 1tti: 4,j ,)r r - . 4 H H aHH :t 30 2.5 C) 5 C t I 4 + + 4 + 4 + 4 + 4 + + + + + + 4 4 + . 150 LSS....J60 1a5 170 115 lEG 1F5 190 US N 235 240 24C t DAY - Figure 15. ,Gizzard SWad Simple Growth Rates, Station 6, 1983. ------- I jflAVCffZ I. Ji i ._ ++ 4 + 4 . 4 4, I. .. 162 ____JJ IEL. J12_______ _____21)fl.22222e ____ ___________ 1gure ]b. G1iz rd Shad S inp1e Growth Rates, StatIon 7, 1983. OLY 0 AVttC jt LtM J?I V UAT bY 1 IIUN ( η1t LUUt SIATJCN 7 SPC ES 3O2 PLDL0E A!GMPAY LEGEUC AZQBS. 2f36S. (IC. ------- 40 F A - 25 P A 20 1 o_ 5 04 4 4 4 4 4 4. + 4 i-.t + + + 4. 4. 4-. __Jl0, 13 _J 1 Q ___ ]t - I C _._ _a10 2Q _____ _____ 0 2 _ 260 74 t n T A CRAGE LII .CTH VS OAT .y TATI0I C SPECIES CC0E - O WE01 ESD IJGUST 2e. 1 e SIAIIUN.1 5IECIES 02 riot CF AVG3OAT LEGfl r: I CI S, 2 OhS, ETC. -. - - - . --- _ t.. / - - - - A b Figure 17. GIzzard Shad SImp e Growth Rates, StatIon 1, 1984. ------- A A h A / 4/ - -- - - AVERACE U G1H VS HAY Y S1ATI , PLCIES CLOE - - c:M wE0P ESUAY. AU SJ 28 1 S1 1IUu 2 SPLCjE! O ILOI F AVGY LECIt : 1 [ 15 , L 1185, ETC. - - - __: - - -- - - - - - -- - - - - I. - - - - -- - _ ::ii i.. i i :i::::::i_ _ _1 .a --- - ____ 0 - F 12 - _____ _ _ ±1T..II1. 10.0 1.5 ______________ ____________ ________________ _________________________________ - ---- - -- - -_ _ -______ _______- -_____ + + 4 4 + + + + 4 4 4 _________ 12 13C 13 _l4Q_U - 0 Y FIgure 18 . Gizzard Shad Siniple rowtb Rates, Station 2. L9B4 ------- s0 A 20 15 £ A * AVIRACE LEI.GTH VS VA LY 1 TATIIIII C SPECiES Ctt)E o: o wEo ESU* , AUtIU T 2t, 19b S IIUPl 3 SIECIE =3O2 P1,01 η AVC3vCAY LEGEP 0: a. (t S, b 2 085, Tc ,__ - h 4 4 4 4 4 $ 4 4 $ 4 4 4 __________ ______ 1 5 U _ _ 1 q_....J&5 19P ______I - DAY I 1lgure19-----6fuar Shad51nip1e-6rowtIr1 etesrStat1en3 -4984 ---- _________ -- ------- 36 33 30 27 24 Pt E 21 A o 18 F A V G VERACE LENG1H VS DAY BY SIATION SPECiES CODE STATIUN 4 SPE(1ES 302 PLOT OF AVG3*OAY LEGEND; A 1 085, B - 2 085. EIC. 22:55 TUESDAY, OCTOBER 1. 199 A -A- (s A 31 A A A A - 0 4 4 4 + 4 4 4 4 4 4 4 $ 4 4 4 130 - .135 140 145 150 155 160 -- _165 - 170 175 1 . 1 5 - 390 - 95 200 2 DAY Figure 20. GIzzard Shad Simpe Growth Rates, Statlo 4, 1984. 15 12 -- (ki A A - - ------- . S . S . . . S S S S S S V S S I I I I 1 2 3 4 5 6 7 STATION # Figure 21. 1983 Emerald Shiner Simple Growth Rates (data taken from computer Step 4). ------- S .8 . S .6 .4 .2 1 2 3 rAT1ON# 4 Figure 22. 1984 Emerald Shiner Simple Growth Rates (data taken from computer Step 4). ------- -- VFRACELENG1U Y5 C4UUT S1ATIfl:1 CSPECtESCC - 1 ; W D (SDAY. r.AF Ch-2?; SIAIIfP1 SPEC1ES 1O! FLDT OF AYC DAY LFGEIW A 1 015, B 2 OCT, ( IC. . i 0 2 A V21 - -_____ 18 - .: 12 >7 1 6 __ 4 16L... 171 177 183 1 195 2C1 207 213 219 225 233 237 243 249 . - DAY FIgure 23. Emer& d Shiner Simjile Growth Rates,Statlon 1, 1903. - ------- TV 1E1 CT rV YST £T rprrs p EC1ETCo r STATIUN PECIES 1O pLflLflfAyQ Y Lg In: £ t , FTr.. ]5-: 4 6---H E ClIE SD AYT PfAR CW27 1583 64 4+ 4 4 4 4 4-. ..- .. 4 4 4-4 I . 111__1111 3.1e9____I95..___3 _;l _ I9_. 23j_ 2 J__ 2 .3 DAY S2 + c1rn .- flatp c Stiitthn 1Q ------- AVtH bEL NUTWV VAT WU I JJ ONCSPEC1ESCOD STATIQP 3 PECIES=1O PEflY flF AVC $fl Y Icr.FFJfl! 1 flPS. 8 2 08S. EIC 15 ;q 14EDut5DAy(A cIr2?1 4 )e5 .+ 4 4 4 + 4 4 +4 + 4 4 $ + - .4 16O...16 13 1flIfl2 ..1 IV 1°5 Z ! __ 2.P5 L) 1_ 2 ?4! 2 2 t DAY Figure 25. Emerald Shiner Simple Growth Rates, Staign 3. 1983. ------- AVEF LGE LEl C1IVV! r:Y !Y TATIOI C P CIE5 CflDE 4U WEPrIESDAY,iaKCH T1 STATIrP 4 PECIE5=jL 5 - - L T_PE VC AV I&c -E :pL B 2 Ce!, FTC._ -- - -I- - --- ------ -----. ---.- -----.--- -- S.---. - . . - - --- ----T - 32.5 I 11111± _ 5.04 4 - + +-, -44+ + 444 + 4. I. 1f S A11 117 - 1F3 18 i?5 201 2G7 _____ _____ 2 5 231 237 243 DAY FIgure 26. Emerald Shiner S mp e Growth Rate Statloti 4,1983 . - ------- 36 4 33 4 30 AVERL6E LEl CTh VS DAyBY STAT1O,Ur-!PEc E5 CCD I ; 46 WtDIESDAYT PIARC1r27 STA11CK 5 SPEC1 SaI05 LPLJ3F B .PE S F ;- - 6 1 3 1 I - . T: 7ti: iit;i: . -. ,i r _ . r p & ..- 1 - I . - * - I,. L t .5- n, qy i:. :; . , I I 1% . = ;. - - - : -. 0 - + + $4 + 4 + 4 + + 4- - 162 168 174 -193 192 198 2C4 230 216 222 228 234 240 . 246 ; S 2 I , 1- . - s I t DAY 4 - - S t * 5 5* . * I_l. F1 ura 7. . 1 [ mθral4 Shiner Simple Gr 1th Rat s, Sta 1Qh 5p:L9 3. -r Y 27 j A y - a -. _ _ _I II ) S 2 * 9t - , - ,? It 2 1 2 ..I . I- , - C s_ - . -,. - _L ,. . , - - ------- - AVEAIEt CTH V YBr TATWN PEC1ESCL!DE 4C14ttNESOATi-hAECIr7 . i185 .1- . . - STAIhJN 6 . S C1ES 1O5 ,. - . . - _ ,,. . I zj I I . - - ±_I PLDtDF AVG3*DAY ctEGEUDz A-a 1 (lBS. B 2 flR . TC.- . - . 1. S - - - -. - . £ __________ .1 -, - , - .1 . - 1 -. ..- , L ; , ; . - . - A / II IS* ., H-- - -,- b - J t: Figure 28. E nera1d Shiner Simple rowth Rates 1 Station 6. 1983. DAY 22.5- ii .- 20.3 - 1: I A + + + + 1BO 19) 195 200 205 21 Z . 220 __ ?33 235 ------- . . vrin LtM 1H v U4I br ,ATTuNrsPEcTEsCot t fl - : 46-RtDfltSDA . hAU 1, . .z.. - I PLflTflF AVC34DAY IIGEUD A fDeS.8 ;2 0BS..E1C. . -: : - I / i & ( . 25 .oI:: :u1 ..Y: : - : - l : - : ; r * , : 0 20.0.! :-. . . . G 2 *7 I C io.o - . . - .. - . - - - 5.01 1BD tas 190 195 2QD 205 210 215 220 2&5_ 230 235 240 245 - . . . -: - 0Ay Ftgure 29. Emerald Shinnr Simple I rc wth W te , Station ? 1 R4 .. - .v . % . I , i_ . . I 3 ------- 4 - A _____z-it1 I - A EqAc LEI.Clh V j y py ΆTt 11CI C SP C1ES ccn , fl,j 5 Ay;AuGusT2t I9r S1A1IQN.1 SFCC1E5105 - fLCT. OVG3 tM __.L t, r . A - I 17.5 1 - - -. 15.0 - ___e __ _ - - 12.5 + - 10.0 751 - : > < - p , - . I, - . - , . : - . 5.0$ i...l ...... ....+ + + 4 + + 4 4 4 4 + 4 4 - 1 f )2 164 166 1 8 I7 .J32 11 .176 Ll _ J 0 182 184 I6 188 190 192 19t 196 193 DAY Figure_30 Emerald_Shiner_Simnie_Growth_Rates._Station_I._ 984- _______________________ ------- 27 . 25.0 I 22.5 AVERAGELE,GTI SOAYeY TATI {t PEC1E5CCOE a: co WEOr1ESOAY AUGIJSr 28.19B5 S A1IOP42 ScEC2E 1O I .J E 3!.( AT LEGEND A1fl S, s2IThS 1 ETC.______________________ A A A 1.5 - - + 4 4 $ 4+-+ + + $ + - - _J L I1A_ 11 _. 1 Q 2 . .18JI . - 19J - 19 lAY F .Emera1d5hlnerS1mp eGrOwthRdteS.St tiOfl?.i984. - - - H E -J I t ------- AVERAGELENG1H VSDAY BY TAT1fl SPECIESCI DE DQi1EOhESQAT,&UCU5T 19L 5 343 S1A11ON 3 SIECIESa1O?, pJ.OiUFAVG3 flfl _ G !p; 4 1 bs B E ETC __________ _____ - ----r-- - _______ ___ 36 j ( ) 1--- - - --- ------ -: - --- - - - 214 .- - I --I - - 249 - Fl I 12j 9+ - - - A A : .L3 __ . 1 5 - 1 O J 5 - l lc - Ipu 1 _. 19O - D Y FIgure 32. t p! !2wth_Rates, Station 3, 1 ------- -. !TATION 5PEC E CIJDE . D OThEDNESDAY.AUGIJST28.1985 3L SIAIUJN4 SP C S,1O5 PJJIJ_OF_AYc3 p y L GU LAP LQP Ic,...___________________________ . O. J L__ 18 A - _______ -- 5_+ II I F I A 44 1i p 13 A V C 9+ I -- _ -- - - . - - --__ 5+ - ,_ +_ + , 4+ 4 i + 4fi 4 - I6 167__AbS... . 111 __173 11_ . 177 179 1b1 . -__.I 3 1 ___ .1U1 - .191 13 - DAY Ffgure 33. Emerad Shther Simple Growth Rates, Station 4, 1984, 10 -- ---- A -.- - ----- - -- - ------- S 1 Figure 34. 2 3 4 5 6 ATION # Gizzard Shad 1983 Instantaneous Growth Rate Coefficients .05 .04 S .03 .02 S .01 7 ------- .09 :08 .07 .06 .05 U, C) . . .03 0 5 - .02 .01 1 2 3 4 5 Figure 35. Gizzard Shad 1984 Instantaneous Growth Rate Coefficients. . ._ I ------- 05 N.v I I I 1 1 I 1 2 3 4 5 6 7 & ATION # 1983 Instantaneous Growth Rate Coefficients - Emerald Shiners. I . .03 .02 .01 - Figure 36. ------- 2 3 4 5 STATION # Figure 37. Emerald Shiner 1984 Instantaneous Growth Rate Coefficients by Station. 1.4 1.3 1.2 .8 .05 .04 .03 .02 .01 1 ------- APPENDIX A ------- 1983 APPENDIX A Larval Fish Body Burden Samples Collected 14 July Station 4 Station 5 Gizzard shad Gizzard shad 1984 Station Station Station Station Stati on - Emerald - Emerald - Emerald Station 5 - Gizzard Station 7 Gizzard Emerald Station 5 Gizzard Emerald Station 7 Gizzard Emeral d shi ner shi ner shi ner shad* shad*, shi ner shad, shi ner shad, shi ner 21 June 28 June Station 4 Station 5 Station 4 Station 5 Gizzard shad Gizzard shad - Gizzard shad Gizzard shad 4 Gizzard shad* 5 Gizzard shad* 4 5 7 21 July 28 July 4 August 18 August 1 September 8 September Station 5 Emerald shiner Station 7 Gizzard shad Station 4 Gizzard shad* * Body burden samples analyzed by Cranbrook Institute ------- APPENDIX B ------- APPENDIX B Raisin River 1983 Flowmeter Calibration DATE REPLICATES REVOLUTIONS DISTANCE REV/METER 53083 1 17988 500 m 2 17724 500 m 3 18787 500 m 4 18886 500 m 5 18828 500 m 6 18581 500 in 7 18293 500 m 8 18283 500 in 9 18045 500 m 10 17522 500 m X 18293 500 in 36.6 707-83 1 21184 500.m 2 18621 500.m 3 19855 500 m 4 18623 500.m 5 20667 500 m 6 18962 500 m 7 19996 500 in 8 18239 500 in 9 19631 500 m 10 18545 500 in X 19432 500 in 38.9 8-04-83 1 18273 500 in 2 17542 500 in 3 17315 500 m 4 16682 500 in 5 17558 500 in 6 17439 500 in 7 17595 500 in 8 16811 500 in 9 16560 500 in 10 9422 500 in X 17308 500 in 34.6 9-08-83 1 18886 500 in 2 18813 500 m 3 16501 500 in 4 5070 500 in 5 15996 500 in 6 17740 500 m 7 18378 500 in 8 17707 500 m 9 18402 500 m 10 18248 500 in X 17852 500 in 35.7 OVERALL 18225 500 in 36.4 ------- APPENDIX C ------- Documentation for Project 1 I. Brief interpretation of this program (1) Lines 17 are the job control statements IJCL ) In these lines we input two data files: RAISIN83.XFRO and RAISIN3B.XFRO then we rename them as PETER and CHOKE respectively. (2) Lines 11-73 are for the step of input and proof data . (a) Purpose : In these lines we want to input the data sets and transform all lengths into standard lengths 0, 5, 10, ..., or 70 m. Also we cal cul ate the difference of final flow and initial flow for our density calculation. (b) Procedures : I *Lj 1139 We form the SAS data set JJ1 by using PETER as the input data file and drop some useless variables from the input data file. Note that we compute the difference of final flow and initial flow at line 21 and convert the lengths into the standard integer lengths 0, 5, 10, ..., or 70 mm. denoted by SYMBOL, at line 24. e.g. all lengths in the interval 2.67.5 are denoted by SYMBOL = 5 and so on. As to these lines 25-39 we assign to each standard length SYMBOL from 5 to 70 a corresponding notation SIZE from A to N and * for otherwise lengths. * jfl 4042 - We convert all missing data (values) in the variables DIFFLOW and F into the SAS standard form .. Then we define the obtained new data set as Ji. *Lfnes 4373 We repeat the same procedures as we did in lines 1142 for the input data file CHOKE and denote the obtained new data set as J2. (c) Some variable notations : DAY = Julian day PD = Period of station (e.g. in station 3A we mean station = 3 and PD = A) SP = Species codes ST = Larval Stage (14) L = Length (cm) INFLOW = Initial flow (revolutions) OUTFLOW = Final flow (revolutions) DIFFLOW = OUTFLOW - INFLOW F = Ave per stage = Frequency (3) Lines 7995 are for step 1. ------- PAGE 2 (a) Purpose : In these lines we perform the procedures of data reduction for data sets Ji and J2 and obtain a new data set COMBI which is going to be used to compute larval density. (b) Procedures : t Lines 7986 We perform data reduction and merge related data sets together. *Ljnes 8795 We compute the larval density and obtain a new data set COMB. (c) Some variable notations : SV = sample volume (m3) FACTOR = 1000 m3/Sample volume (m3)1 DENS = Density/100 0m3 = Factor x Ave. per stage TOTAL Ave. per stage = variable F. (4) Lines 102-116 are for step 2. (a) Purpose : We average replicates A, B and C densities in these lines by station, species code and size, then we plot the density vs Julian day. (b) Procedures : * Lines 102-104 -- We average A, B and C density to form a new data set TEMP2. * jfl 105 We delete those data with SP = 0. *Lines 115116 - We plot the density vs. Julian day. ------- PAGE 3 (c) Some variable notations : DENS = The density obtained by DAY STATION PD SP and SIZE. MOEN = The mean density obtained over PD by STATION DAY SP SIZE. (5) Lines 123144 are for step 3A. (a) Purpose : We want to average A, B and C density by station and species code. (b) Procedures : *Lines 123126 - We perform data reduction to obtain a new data set TEMP3. *Ljnes 127-131 - We calculate A, B and C density separately by station and species code. *Ljnes 132134 We average A, B and C density to obtain a new data set TEMP5. *Lines 143-144 -- We plot average A, B and C density vs Julian day. (c) Some variable notations : 1011 = Total frequency by DAY STATION PD and SP DENI = Density by DAY STATION PD and SP AVG1 = Average density by DAY STATION and SP (6) Lines 151175 are for step 38. (a) Purpose : We want to calculate the total seasonal density by station and species code. (b) Procedures : *Llnes 155159 We calculate the density for each Julian day. 160162 We average A, B and C density for each Julian day to obtain a new data set TEMP7. *Ljnes 163165 We obtain total seasonal density by summing up all Julian ------- PAGE 4 days density. *ljnes 174175 We plot total seasonal density vs station. (c) Some variable notations : TOT3 = Total frequency by DAY STATION SP and PD. DEN4 = Density by DAY STATION SP and PD AVG7 = Mean density over PD by DAY STATION and SP. TOT4 = Season total density by STATION and SP. (7) Lines 181191 are for Step 4. (a) Purpose : We calculate average length by STATION SP. (b) Procedures : *Lines 181-183 We average all sizes of larval fish by DAY STATION SP and ST (stage). These results, denoted by AVG, form the new data set COUNT. *Lines 184-186 - We average A, B and C density by DAY STATION and SP. *ljnes 190-191 - We plot average length vs Julian day by station and species code. (c) Some variable notations : AVG = Mean length over SIZE by DAY STATION PD and SP. AVG 3 = Mean length over SIZE and PD by DAY STATION and SP. (8) Lines 198-219 are for Step 5. (a) Purpose : We calculate the date when total length of the population is 5, 10, 15,... to 70 mm. (b) Procedures : *Lines 198-200 We sum the frequencies by STATION SP SYMBOL for every ------- PAGE 5 Julian day, denoted as 1, and form the data set Clii. *Ljnes 201203 -- We sum total seasonal frequencies by STATION SP SYMBOL, denoted as TI, and form the data set CH2. *Ljnes 204-207 - We calculate the relative frequency for each Julian day by STATION SP and SYMBOL, denoted as AVERAGE. *Lines 208210 We calculate the mean Julian day by using the relative frequency (AVERAGE), denoted as IL, and form the data set CH4. *Ljnes 218219 - We plot the SYMBOL vs TI (mean Julian day) by STATION SP. (c) Some variable notations : I = Frequency by DAY STATION SP and SYMBOL. TT = Total frequency over DAY by STATION SP SYMBOL. Prob = Relative frequency (i.e. TITT) by STATION SP SYMBOL (9) Lines 226-238 are for Step 6. (a) Purpose : We use the nonlinear regression method to estimate the slope of growth for specified STATION and SP. (b) Procedures : *Ljne 226 Suppose that we want to estimate the slope of Growth for STATION = 1 and SP = 302. i.e. We specify STATION = 1 and SP = 302. * jfl 5 229235 -- We use the MARQUARDT method as our tool for the non-linear regression. This method represents a compromise between the linearization (or Taylor series) method and the steepest descent method and appears to combine the best features of both while avoiding their most serious limitations. It is good in that it almost always converges and does not slow down as the steepest descent method often does. *Line 237 - We plot SYMBOL vs TI. ------- PAGE 6 (c) Some variable notations : 10 = Initial Length. 11 = Slope of growth. YHAT = V = Estimated length YRESID = V - V = Residual for V *We note that SYMBOL = LO*EXP(L1*(TL_150)) at line 232 means L=1.eG(tto) as it appears in Hackney and Webbs paper where hatching date to = 150. (d) How to find the estimated slope of growth for Station A Species B? We first replace 1 by A and 302 by B at line 226. Then we examine whether 150 (lines 232234) is a suitable initial value (Julian day) if it is not, replace all 150 in lines 232234 by a suitable initial value. Initial values can be determined by presence of larval fish in earlier plots. (e) If we want to obtain the results for more combinations of stations and species simultaneously, we can copy whole lines 226-238 repeatedly as many as desired stations and species and then follow Step (d) to make suitable modifications for stations, species and(or) initial (Julian) day. (10) Lines 244263 are for Step 7. (a) Purpose : We use a nonlinear regression method to estimate the mortality for specified STATION and SP. (b) Procedures : *Line 245 - We specify STATION = 1 and SP = 105. *Ljnes 254-260 - We use the MARQUARDT method as the tool for analyzing nonlinear regression. *Line 262 - We plot TI vs IL. (c) Some variable notations : NO = Initial (frequency) number over DAY by STATION SP SYMBOL. Z = Mortality rate. ------- PAGE 7 (d) How to find the mortality rate for station A species B? First, we replace 1 by A and 105 by B at line 245. Next, we examine whether 173 is a suitable initial Julian day. If it is not, replace all 173 in lines 257-260 by a suitable initial value, determined by initial presence of larval fish in previous plots. (e) Use the same steps as (9)(e) we can obtain results for more combinations of stations and species simultaneously. *Note that TI = N0*EXP(_Z*(TL_173)) means this formula Nt= Nto eZ(tto) which appears in Hackney and Webbs paper ir. How to use this program . (a) We can use this program to obtain the results for separate steps or some combination of steps. The basic procedures for establishing a desired subprogram are as follows: (i) Use CHENPJ1 on IRCC93. (ii) Lines 173 must be included in any subprogram(s). (iii) Keep those lines for corresponding steps desired in the subprogram and delete the rest of lines. (iv) Substitute suitable station, species or initial values when subprogram contains Step 6 or Step 7. Therefore the diagram for above procedures is: lines lines for examine ST.SP 1-73 + corresponding initial values if = subprogram steps steps 6 or 7 is concerned (b) Some examples : (i) Suppose we want to get the results of Step 3. The subprogram should contain lines 1-73 and 123175 (lines for Step 3) only, so we delete the other lines from main program and then run this subprogram. (ii) Suppose we want to get the results of Step 4 and Step 6 simultaneously and consider station Al species Bi and station A2 species B2 instead of station 1 species 302 in Step 6. In order to obtain the subprogram we first keep lines 173 and lines 181191 (for Step 4) and lines 226238 (for Step 6) then delete other lines, since we consider 2 combinations of station and species (A1,Bl), (A2,82), we need to copy lines 2 6238 once. Suppose these latter lines are renumbered as 239251 (Note that these lines 239251 are not the original lines 239-251 in our main program). Now, we replace 1 by Al an 02 by Bi at line 226 ------- PAGE 8 and examine the initial values for lines 232-234 to see whether the value 150 is suitable; replace 1 by A2 and 302 by B2 at line 239 and again examine the value 150 in lines 245247. After doing these, we finally obtain our subprogram which includes lines 173 and 181191 and 226238 and the new lines 239251. (c) Remark : If we choose different initial values for the same station and species in Step 6 and Step 7, the estimated values may be different but the estimated growth rate and mortality rate are still the same. i.e. the estimated values depend on the choice of the Initial values (its not important since we consider their corresponding confidence intervals) while the estimated growth rate and mortality rate do not depend on the choice of the initial values. However if we can choose a good initial value, the iteration times will be reduced. ------- |