AD-All7 279 JONES EDMUNDS AND ASSOCIATES INC … · ad-all7 279 jones edmunds and associates inc...

AD-All7 279 JONES EDMUNDS AND ASSOCIATES INC GAINESVILLE FL FIG 6/6A IOASSAY EXPERIM4ENT TO DETERMINE WATER QUALITY IMPACTS RELATE--ETC(U)

MAR 82 M E LEHMAN, M R CREZEE, R H JONES OACAI1-80C-036A

UNCLASSIFIED NL2 ffffffffffffEEmhEmhhmhohhI

111 1111

1.8I~

FINAL REPORNTFOR

A BIOASSAY EXPERIMENT TO DETERMINE WATERGUALITY IMPACTS RELATED TO TROUT MORTALITY

AT A IIAT04ERY DOWNISTREAM OFLAKE SIONSEY LANIER, GEORGIA

Pwmntod to:

U.S. Army Conns of Enginme, Mobile DistrictP.O. Box 2286

Mobil., Alabama 36626

4 . Contract No. DACW1"C0368

husn ed by:

Jones, Edmunds & Auoclatm, Inc.730 North Waldo Road

Galnmlll., Florida 32601

March 19WAI

file IN

C=112r udk

FINAL REPORTFOR

A BIOASSAY EXPERIMENT TO DETERMINE WATERQUALITY IMPACTS RELATED TO TROUT MORTALITY

AT A HATCHERY DOWNSTREAM OFLAKE SIDNEY LANIBER, GEORGIA

Presented to:

U.S. Army Corps of Engineers, Mobile DistrictP.O. Box 2288

Mobile, Alabama 36628

Contract No. DACWO 1 -80-C-0368

Presented by:

Jones, Edmunds & Associates, Inc.730 North Waldo Road

Gaines,,ille, Florida 32601

March 1982

-Unclassified - -___ ~-___SECURITY CLASSIFICATION OF THIS PAGE Mlhon Data Entwoo _________________

REPORT DOCUMENTATION PAGE BEFORE COMPLE71NG FORM

- R ORT"UM1119t V AqSS NM 13. RECIPIENT'S CATALOG NUMBER

.4. ToTLE ka &aI to ) S YEO EOTaPROOEEFinal Report for a Bioassay 'gxopriment'to Deter- Fin Set.160Mac 1982mine Water Quality Ikpacts Wiafted to ?rottMortality at a Hatchery Dosmrat ta of- Lake.Sidney 6. PERFORMI NO ONO. *01POR NUMBERLanier, £Ceorgia 29700-195-00

7. AUTHOO~s) S. CONTRACT OR GRAW4T NUNSRi'.

Melvin E. Lehman, Michael R. Crezee, Richapd H. DACW1-8Q-C-0368Jones; Randy L. Schulze, Mary K. Leslie

9. PERPORM SG ORGANIZATION NAME ANC ADDRS r5 10 RGAM ELEMMENT. PROJECT, TASK

Jones, E:dmunds & Associates,' Inc. RA'OKNTNUESSciece Division730 Dortb Waldo Road, Gainesville, Floulfd. $2601

III. CONTROLLING OFFICE NAME AND ADDRESS ti. iiison dAlffDepartment of the Army March 1982Mobile District, Corps of Engineers -IT NUUMmE0 07 PAGESP.O. Box 2288, Mobile-, Alabama 36628 136

14. MONITORING AGENCY MAMIE & ADDRESSIf difi.,nt Irani Co-0Inbl Office) 1S. SECURITY CLASS. (of dals topsot)

Unclassified

ISO. :NW~JI(CAION/6011116111011NG

IS. DISTRIBUTION STATEMENT (of Dale Reort)

Approved for Public Release; Distribution:Unlimited

17. DISTRIBUTION STATEMENT (of th. abstract snowed f In. loc It Effffersnt Mum Report

1S. SUPIPLEMENTARY NOTES

IS. KEY WORDS (Contae on rover". oldi It neeeeein ad UfmUb' b Mok oL-. rITrout mortality, metals toxicity to trowtv. tor stances in reservoirdischarges of hypolimnetic waters, aquatic bi- _ 'ising trout, environmentalaspects of reservoir releases, bioassays..for siL i-ters, riverinebioassays, toxicity of iron and manganese to trrp-, rish hatchery water qualityproblems, water quality management, toxicity of water in'stratified reservoirs,fisherie s manastement

26 ASST*A? (ftedmoa rmm an oes fo aosw a"I uer ky block nombeA major trout kill occurred at Buford Trout Hatchery in northern

Georgia during its first year of operation in autumn 1976. Severalstudies linked toxicity to anoxic conditions in the hypolimnion of LakeSidney Lanier (the reservoir formed by Buford Damt, 1 1/2 miles up theChattahoochee River from the hatchery). No consensus as to the toxicagent(s) was 'reached as a result of the Initial studies. An~ tnteragencyTask Force recomended bioassays to Identify the toxic agent(s), and the

JAW ~S 0 EDITIbN OF f NOW3 IS 0 OSOLRTIE

SECUmWY CLAIM FICATION OF 15615 PASS (u.. ODeto Snows*d

Unclassified

SECUNITY CG. MICATION OF THIS PAOS(3tm Date s

U.S. Army Corps of Engineers, Mobile District, retained Jones, Edmunds& Assogtates, Inc. (JEA), to design and conduct bioassay teats on sitedurin the fall and winter of .1980., After ap roval of its Pltn ofSt'uf, JEA established a compounAt on top o' Bllrd Dam and drew test,.4atIr.di3rctj.y from 4pke Sidney. Lanier. ""Waters ;from dLfferent depth.l in.the reservoir Wie tested for toxicity to rainbow trout swim-up fry, and-emat wter was extensively analyzed chemically. Bottom water was Sasttoxic and had the highest levels of manganese (Nn),and iron (Fe)., Noother metals were detectable, no humics or pestic'des ,o~ld '"ahed,

and no hydrogen sulfide was found. In a second set of experiments, bo't-.. toe W^e.er as treated in various ways to remove potential toxicants. An

anion exchange resin to remove organics had no effect on toxicty of thewater. Na4 EDTA increased toxicity, but Ca2 EDTA gave full protec-tion. Removing Mn and Fe removed toxicity' if: the hardness (12 pom inlake) also removed by the cation exchange resin was added back. In-creasing hardness by 25 ppm without other treatment. alpo preye.adtrout fry mortality. Prolonged aeration decreased dissolved F, hut notMn, and 'increased toxicity. Activated carbon nmaowed alwi 1 a17 7-Obut.no Mn, and increased toxicity.

In static bioassays, Mn+ + or Fe+ + added to non-toxic epilim-netic. water was acutely toxic to trout fry at the concentrations mea-sured in the Chattahoochee River below Buford Dam.

Rainbow trout fry held at 6 stations in the Chattahoochee River upto 23 km below Buford Dam suffered heavy mortalities while the lake "Wasstill stratified (November), with only slight improvement at the far-thest station. Yearling trout showed the same pattern but with lowermortality rates. Bluegill sunfish, which occur naturally in the river,survived 4-day exposures at all stations.

Comparisons of liver contents of 24 metals between Buford TroutHatchery and Walhalla National Trout Hatchery were difficult to inter-pret. Of the eight detectable metals, six showed significant differ-ences between hatcheries, all higher in Buford trout even though some ofthe six are known to be in higher concentrations in water at Walhalla.

The study concluded that Mn and Fe explained all toxicity observedat the hatchery and demonstrated in the river, with no evidence for in-volvement of any other metal nor of any organic compound. Recommenda-tions were made with respect to possible solutions to seasonal toxicityproblems at Buford Trout Hatchery and to further understanding of theimpacts of Mn and Fe toxicity on biota of the Chattahoochee River belowLake Sidney Lanier.

Unclassified

SECURITY CLASSIICATION Of THIS PAGIl(Wl Date Int.,.l)

TABLE OF CONTENTS

Section Title Page

LIST OF TABLES iiiLIST OF FIGURES vDISCLAIMER viiABSTRACT viii

1.0 INTRODUCTION 1-1

1.1 PROBLEM STATEMENT 1-11.2 EVENTS LEADING TO THIS STUDY 1-11.3 THIS STUDY 1-31 .4 ACKNOWLEDGEMENTS 1-6

2.0 "LAKE, RIVER, AND HATCHERY SETTING 2-1

3.0 HISTORY OF THE PROBLEM AT THE HATCHERY 3-1

4.0 MATERIALS AND METHODS 4-14.1 INTRODUCTION 4-14.2 WATER MOVEMENT AND FLOWS 4-24.3 TREATMENTS AND TREATMENT MATERIALS 4-34.4 FISH USED IN BIOASSAYS 4-34.5 RIVERINE BIOASSAYS 4-34.6 METHODS FOR ROUTINE CHEMICAL MONITORING 4-44.7 METHODS FOR OTHER MONITORING 4-54.8 SPECIAL CHEMICAL SAMPLES 4-54.9 TISSUE ANALYSIS 4-64.10 HISTOPATHOLOGY 4-64.11 STATISTICAL HANDLING OF DATA 4-64.12 ANALYTICAL QUALITY CONTROL 4-7

5.0 RESULTS 5-15.1 TABULAR RESULTS FOR EACH BIOASSAY 5-15.2 EPILIMNETIC CONTROLS 5-15.3 TOXICITY VARIATIONS WITH WATER DEPTH 5-15.4 VARIATIONS IN UNTREATED BOTTOM WATER 5-15.5 ORGANICS REMOVAL WITHOUT EFFECT 5-25.6 FORM OF EDTA CRITICAL 5-25.7 PROLONGED AERATION INCREASES TOXICITY 5-25.8 METALS REMOVAL AN EFFECTIVE TREATMENT 5-35.9 HARDNESS ADDITION PROTECTS FISH 5-35.10 EFFECTS OF DILUTING BOTTOM WATER 5-45•1I MANGANESE ADDITIONS TOXIC 5-45.12 IRON ADDITIONS TOXIC 5-45.13 MANGANESE AND IRON TOXICITIES ADDITIVE 5-55.14 SUBLETHAL EFFECTS 5-55.15 ROUTINE PHYSICAL MONITORING 5-65.16 CHEMISTRY AND QUALITY CONTROL 5-65.17 METALS IN TROUT LIVERS 5-75.18 FEW HISTOPATHOLOGICAL EFFECTS 5-85.19 RIVERINE BIOASSAYS 5-95.20 LAKE PROFILE DATA 5-10

i;

TABLE OF CONTENTS

Section Title ?a2.

5.21 HYDRAULIC CALCULATIONS 5-10

6.0 DISCUSSION 6-1

6.1 MANGANESE AND IRON ARE TOXICANTS 6-1

6.2 NO OTHER METALS ARE INVOLVED 6-1

6.3 NO ORGANICS INVOLVED 6-26.4 RESULTS APPLICABLE TO HATCHERY 6-36.5 NEW QUESTIONS 6-4

6.6 RIVERINE DISCUSSION 6-56.7 WATER QUALITY IMPACTS ON THE RIVERINE SYSTEM 6-6

7.0 CONCLUSIONS 7-1

8.0 RECOMMENDATIONS 8-1

8.1 MANAGEMENT CHANGES 8-18.2 TREATMENT OF HATCHERY INTAKE WATER 8-18.3 ALTERNATIVE SOURCES OF WATER 8-28.4 TREATMENT OF WATER AT DAM OR RESERVOIR 8-28.5 RIVERINE CONSIDERATIONS 8-3

REFERENCES R- 1

APPENDICES

LIST OF APPENDICES TABLES A-iAPPENDIX A. CHEMISTRY DATA A-iAPPENDIX B. ROUTINE PHYSICAL MONITORING B-I

APPENDIX C. RIVERINE BIOASSAY DATA C-I

Aosssicn For

DTIC TABUnamnnoaced U

TSJust i f t o------

Vint ribuioin/_Avmflability Codes

Avafil arl4d/or--lot Speolal

ii

LIST OF TABLES

Table Title Page

3.1 Summary of Data from the 1976 Buford Trout Hatchery 3-5Experiments Reported by Noell and Oglesby, 1977

4.1 Sumary Comparison of Set I and Set II Flow-Through 4-8Bioassay Experiments Performed at Lake Sidney LanierGeorgia, Fall 1980

4.2 Summary of Treatments for Flow-Through Bioassay Experi- 4-9ments Performed at Lake Sidney Lanier, Georgia, Fall

1980

4.3 Methods Used for Chemical Analysis of Water Samples, Lake 4-12

Sidney Lanier, Georgia, Fall 1980

5.1 Summary of Set I Run 1 Bioassay Results, October 28 5-13through November 2, 1980, Testing Untreated Water Drawnfrom Four Different Depths in Lake Sidney Lanier, Georgia

5.2 Summary of Set I Run 2 Bioassay Results, November 7 5-14through November 10, 1980, Testing Untreated WaterDrawn from Four Different Depths in Lake Sidney Lanier,Georgia

5.3 Summary of Results of Set II Run I Bioassays, November 19 5-15

through November 23, 1980, Testing the Effectiveness ofVarious Treatments in Removing Toxicity from Bottom Water(35-Meter Depth), Lake Sidney Lanier, Georiga

5.4 Summary of Results of Set II Run 2 Bioassays, November 23 5-16through November 24, 1980, Verifying Some Results of Set IIRun 1, Lake Sidney Lanier, Georgia

5.5 Summary of Results of Set II Run 3 Bioassays, December 4 5-17through December 8, 1980, Further Testing the Effective-ness of Various Treatments in Removing Toxicity fromBottom Water (35-Meter Depth), Lake. Sidney Lanier, Georgia

5.6 Summary of Results of Set II Run 4 Bioassays, December 19 5-18through December 23, 1980, Lake Sidney Lanier, Georgia

5.7 Summary of Results for Static Bioassay 1, December 12 5-19through December 16, 1980, Lake Sidney Lanier, Georgia

5.8 Summary of Results of Static Bioassay 2, December 14 5-20through December 16, 1980, Lake Sidney Lanier, Georgia



iii

LIST OF TABLES

Table Title Page


5.12 Summary of Results of Manganese Additions in Static 5-24Bioassays, Lake Sidney Lanier, Georgia

5.13 Results of Metals Scan of Water from 23-Meter and 35- 5-25Meter Reservoir Depths Used in Set I Run 2 Bioassays,November 7-10, 1980, Lake Sidney Lanier. Analysisby EPA, Athens, Georgia

5.14 Loading Rate (Water Flow Per Gram Fish) Differences 5-26Related to Toxicity Variations in Untreated HypolimneticWater (35-Meter Depth) from Lake Sidney Lanier, Georgia,Fall 1980

5.15 Comparison of Hypolimnetic Iron and Manganese Levels in 5-27

Lake Sidney Lanier, Georgia, for Fall 1979 (ESE, 1981)and Fall 1980

5.16 Analyses of Livers of One-Year Old Wytheville Strain 5-28Rainbow Trout Taken from Buford Trout Hatchery, Georgia,November 11, 1980

5.17 Analyses of Livers of One-Year Old Wytheville Strain 5-29Rainbow Trout Taken from Walhalla National Fish Hatcheryon November 20, 1980

5.18 Comparisons of Liver Metals Contents for Yearling Rainbow 5-30Trout from Buford Trout Hatchery (B), Georgia, VersusWalhalla National Fish Hatchery (W) for Metals aboveDetection Limits, November 1980

5.19 Results of Riverine Bioassays. Summary of Fish Survival, 5-31Physical Parameters, and Metals Concentrations in theChattahoochee River Below Buford Dam, Georgia,

November 23 to November 26, 1980

iv

LIST OF FIGURES

Figure Title Page

1.1 Location of Lake Sidney Lanier, Buford Dam, and Buford 1-9

Trout Hatchery, Georgia

2.1 Daily Fluctuations in Flow, Dissolved Oxygen, and Temper- 2-2ature in the Tailrace of Buford Dam on October 29, 1972,During Stratification of Lake Lanier, Georgia

2.2 Comparison of Total Iron and Manganese Concentrations 2-3

During High Flow Versus Low Flow at Tailrace of BufordDam, Georgia, from August 3, 1977, to December 21, 1977

3.1 Rainbow Trout Mortality, Total Manganese, and Total Iron 3-6

Levels for Buford Trout Hatchery, Georgia, in Octoberand November 1977

3.2 Rainbow Trout Mortality, Total Iron, and Total Manganese 3-7at Buford Trout Hatchery, Georgia, October and November1978

3.3 Daily Manganese, Iron, and Mortality Levels for Buford 3-8Trout Hatchery, Georgia, from October 20, 1979, toNovember 21, 1979

3.4 Total Manganese and Dissolved Iron (Fe++) in the Chat- 3-9tahoochee River, Georgia, at Buford Hatchery Intake,Fall 1979 and 1980

4.1 JEA Buoy Location with Hypolimnetic Pump Approximately 4-13

200 Meters Upstream of Buford Dam, Lake Sidney Lanier,Georgia

4.2 JEA Experimental Bioassay Compound on Top of Buford 4-14Dam, Georgia

4.3 JEA Laboratory Test Water Intake Assembly for Bioassay 4-15Experiments at Lake Sidney Lanier, Georgia

4.4 Diagram of Water Flows for One Treatment of Set II 4-16Bioassays at Buford Dam, Georgia, Fall 1980

4.5 Diagram of the Flows for Individual Water Treatments 4-17

in Set II Run 1 Bioassays Performed at Buford Diem,Fall 1980

4.6 Locations of Riverine Bioassay Stations in the Chatta- 4-18

hoochee River, the Buford Trout Hatchery, and the JEAExperimental Compound, Georgia, Fall 1980

4.7 Cages Used to Hold Fish for Bioassays Performed in the 4-19Chattahoochee River Below Buford Dam, Georgia, November1980 and February 1981

V

LIST OF FIGURES

Figure Title Page

5.la Graphic Presentation of Survival Versus Time for Set I, 5-34Runs 1 and 2, Lake Sidney Lanier, Georgia, Fall 1980

5.1b Regressions of Survival Against Dissolved Iron Using 5-34Combined Run I and Run 2 Results at 72 Hours and 96 Hoursinto Bioassay, Lake Sidney Lanier, Georgia, Fall 1980

5.2a Graphic Presentation of Survival Versus Time for Set II, 5-35Runs 1 and 4, Bioassay Results, Lake Sidney Lanier, Georgia,Fall 1980

5.2b Graphic Presentation of Survival Versus Time for Set II 5-36Runs 2 and 3, Bioassay Results, Lake Sidney Lanier, Georgia,Fall 1980

5.3 Some Examples of Regressions of Survival Versus Hardness, 5-37Dissolved Iron, and Dissolved Manganese Combining Datafor All Four Runs of Set II Bioassays, Lake SidneyLanier, Georgia, Fall 1980

5.4 Graph of Percent Survival Versus Manganese Concentration 5-38in Static Bioassays, Lake Sidney Lanier, Georgia, Fall1980

5.5 Iron and Manganese Concentrations in the Hypoliunion of 5-39Lake Sidney Lanier (35-Meter Water Depth) About 600 Feet(180 a) Behind Buford Dam, as Measured in Water PumpedContinuously into Bioassay Compound on Top of BufordDon, Georgia, Fall 1980

5.6 Photomicrographs of Eye and Muscle of Trout Fry Taken from 5-40Set II Btoassays Performed at Buford Dam, Georgia, Fall1980

5.7a Survival of Rainbow Trout Fry in the November 1980 5-41Riverine Bioassays, Chattahoochee River, Georgia

5.7b Survival of Rainbow Trout Yearlings (15-cm Length) in the 5-42November 1980 Riverine Bioassays, Chattahoochee River,Georgia

5.8 Temperature and Dissolved Oxygen Profiles from Lake 5-43Sidney Lanier, Sampled at JEA Laboratory Intake BuoyAbout 600 Feet (180 Meters) Behind Buford Dam, Georgia,Fall 1980

5.9 Dissolved Oxygen and Temperature Profiles for Lake Sidney 5-44Lanier in 1978 and 1979 at ESE Station 1, About 200 MetersBehind Buford Dam, Georgia

6.1 Manganese Concentrations and Total Mortalities at Greer's 6-7Ferry National Fish Hatchery, Arkansas, a) November 8, 1976,to January 5, 1977, and b) November 1967 to January 1568

vi

"" .. . . '... . . . ... . .. . . .. . .. . .. . imum! ... . ... '1 I" '

... J - "-' '- " .. . ';'i.¢, . .'" , 1 ... "" - . . ...

DISCLAIMER

The contents of this report are not to be used for advertising, pub-lication, or promotional purposes. Citation of trade names does notconstitute an official endorsement or approval of the use of such com-mercial products.

v1i

- - ~ - - . . . - ~ -- w ~ -~. -4~.

ABSTRACT

A major trout kill occurred at Buford Trout Hatchery in northernGeorgia during its first year of operation in autumn 1976. Severalstudies linked toxicity to anoxic conditions in the hypolimnion of LakeSidney Lanier (the reservoir formed by Buford Dam, 1 1/2 miles up theChattahoochee River from the hatchery). No consensus as to the toxicagent(s) was reached as a result of the initial studies. An InteragencyTask Force recommended bioassays to identify the toxic agent(s), and theU.S. Army Corps of Engineers, Mobile District, retained Jones, Edmunds& Associates, Inc. (JEA), to design and conduct bioassay tests on siteduring the fall and winter of 1980. After approval of its Plan ofStudy, JEA established a compound on top of Buford Dam and drew testwater directly from Lake Sidney Lanier. Waters from different depths inthe reservoir were tested for toxicity to rainbow trout swim-up fry, andeach water was extensively analyzed chemically. Bottom water was mosttoxic and had the highest levels of manganese (Mn) and iron (Fe). Noother metals were detectable, no humics or pesticides could be measured,and no hydrogen sulfide was found. In a second set of experiments, bot-tom water was treated in various ways to remove potential toxicants. Ananion exchange resin to remove organics had no effect on toxicity of thewater. Na 4 EDTA increased toxicity, but Ca 2 EDTA gave full protec-tion. Removing Mn and Fe removed toxicity if the hardness (12 ppm inlake) also removed by the cation exchange resin was added back. In-creasing hardness by >25 ppm without other treatment also preventedtrout fry mortality. Prolonged aeration decreased dissolved Fe, but notMn, and increased toxicity. Activated carbon removed almost all Fe, butno Mn, and increased toxicity.

In static bioassays, Mn+ + or Fe++ added to non-toxic epilim-netic water was acutely toxic to trout fry at the concentrations mea-sured in the Chattahoochee River below Buford Dam.

Rainbow trout fry held at 6 stations in the Chattahoochee River up

to 23 km below Buford Dam suffered heavy mortalities while the lake wasstill stratified (November), with only slight improvement at the far-thest station. Yearling trout showed the same pattern but with lowermortality rates. Bluegill sunfish, which occur naturally in the river,survived 4-day exposures at all stations.


The study concluded that Mn and Fe explained all toxicity observedat the hatchery and demonstrated in the river, with no evidence for in-volvement of any other metal nor of any organic compound. Recommenda-tions were made with respect to possible solutions to seasonal toxicityproblems at Buford Trout Hatchery and to further understanding of theimpacts of Ma and Fe toxicity on biota of the Chattahoochee River belowLake Sidney Lanier.

viii

... .. . . .. .. . . .. . . . . .~ j ; .. . .

ABSTRACT

A major trout kill occurred at Buford Trout Hatchery in northernGeorgia during its first year of operation in autumn 1976. Severalstudies linked toxicity to anoxic conditicns in the hypolimnion of LakeSidney Lanier (the reservoir formed by Buford Dam, 1 1/2 miles up theChattahoochee River from the hatchery). No consensus as to the toxicagent(s) was reached as a result of the initial studies. An InteragencyTask Force recommended bioassays to identify the toxic agent(s), and the

U.S. Army Corps of Engineers, Mobile District, retained Jones, Edmunds

& Associates, Inc. (JEA), to design and conduct bioassay tests on siteduring the fall and winter of 1980. After approval of its Plan ofStudy, JEA established a compound on top of Buford Dam and drew testwater directly from Lake Sidney Lanier. Waters from different depths inthe reservoir were tested for toxicity to rainbow trout swim-up fry, andeach water was extensively analyzed chemically. Bottom water was mosttoxic and had the highest levels of manganese (Mn) and iron (Fe). Noother metals were detectable, no humics or pesticides could be measured,and no hydrogen sulfide was found. In a second set of experiments, bot-tom water was treated in various ways to remove potential toxicants. Ananion exchange resin to remove organics had no effect on toxicity of the

water. Na4 EDTA increased toxicity, but Ca2 EDTA gave full protec-tion. Removing Mn and Fe removed toxicity if the hardness (12 ppm inlake) also removed by the cation exchange resin was added back. In-creasing hardness by >25 ppm without other treatment also preventedtrout fry mortality. lrolonged aeration decreased dissolved Fe, but not*Mn, and increased toxicity. Activated carbon removed almost all Fe, butno Mn, and increased toxicity.

In static bioassays, Mn++ or Fe+ + added to non-toxic epilim-netic water was acutely toxic to trout fry at the concentrations mea-sured in the Chattahoochee River below Buford Dam.

Rainbow trout fry held at 6 stations in the Chattahoochee River up

to 23 km below Buford Dam suffered heavy mortalities while the lake wasstill stratified (November), with only slight improvement at the far-thest station. Yearling trout showed the same pattern but with lower

mortality rates. Bluegill sunfish, which occur naturally in the river,

survived 4-day exposures at all stations.


The study concluded that Mn and Fe explained all toxicity observedat the hatchery and demonstrated in the river, with no evidence for in-volvement of any other metal nor of any organic compound. Recommenda-tions were made with respect to possible solutions to seasonal toxicityproblems at Buford Trout Hatchery and to further understanding of theimpacts of Mn and Fe toxicity on biota of the Chattahoochee River below

Lake Sidney Lanier.

viii

SECTION 1.0. INTRODUCTION

EE

1.0 IW DUCTION

1.*1 PROLEM STATT

Lake Sidney Lanier, a multi-purpose reservoir In northeast Georgiaoperated by the U.S. Army Corps of Engineers, Mobile District, was form-ed by the construction of Buford Dam on the Chattahoochee River in themiddle 1950's (Figure 1.1). The lake elevation is normally between1,060 feet (323 m) mean sea level (msl) and 1,070 feet (325 m) al, atwhich level the lake covers 38,000 acres (15,378 hectares). BufordTrout Hatchery was built in 1976 approximately 1.5 miles (2.5 kilo-meters) downstream from Buford Dam, by the Game and Fish Division of theGeorgia Department of Natural Resources. The hatchery location waschosen because bottom waters released from Lake Sidney Lanier are coldenough to allow trout culturing during periods when water temperaturesso far south are generally too warm for trout.

During the fall of 1976, massive rainbow trout kills occurred at thehatchery, which had been placed in operation six months earlier. Ap-proximately 435,000 fish (60 percent of operational loadings) were lost.Initial studies by the Envirormental Protection Division of the GeorgiaDepartment of Natural Resrouces postulated both manganese and organicmaterial (some form of humic acid) as toxicants. The toxicants wererelated to the seasonal stratification of Lake Sidney Lanier and to therelease of hypolimnetic water into the Chattahoochee River, from whichthe hatchery draws its water.

In Hay 1977, Georgia Governor Busbee wrote General Mclntyre of theU.S. Army Corps of Engineers South Atlantic Division in Atlanta, bring-ing to his attention the serious problem at the Buford Trout Hatchery.Subsequent coordination resulted in the formation of a task force com-prised of Georgia Department of Natural Resources Game and Fish, andEnviroiental Protection Divisions; the U.S. Army Corps of Engineers,South Atlantic Division, Mobile District, and Savannah District; and theU.S. Environmental Protection Agency (EPA). The primary objectives ofthe task force were to better define water quality below Buford Dam andto investigate alternative solutions to toxicity problems at the haLch-ery.

1.2 ZVES LEADING TO THIS STUDY

Task Force members performed several investigations in the fall of1977 directed at understanding and solving the water quality problems inthe hatchery and at determining if similar impacts existed in the river.Georgia Department of Natural Resources monitored water quality and fishhealth in the hatchery and initiated a temporary treatment system fororganic material. EPA, at the request of the State of Georgia, per-formed an independent investigation of hatchery conditions. In additionto the hatchery studies, the U.S. Army Corps of Engineers funded severalstudies pertaining to conditions in the river. Water quality data werecollected by the U.S. Geological Survey (USGS), Georgia District Of-fice. Fish movement studies were conducted by the University of Geor-gia, and fish health studies were conducted by Auburn University. The

1-1

U.S. Army Corps of Engineers also funded a limited assessment of muta-genic activity by Morehouse College (Atlanta University) at the requestof the State of Georgia.

The 1977 studies provided better definitions of water quality condi-tions in the Chattahoochee River below Buford Dam and in the hatchery;however, the acute mortality problems in 1976 were not experienced in1977. No evidence of fish mortality was detected in the river. Fishhealth was impacted somewhat, but causes were undefined. The signifi-cance of any adverse water quality impact on the downstream fishery re-source or on other uses of the Chattahoochee River, either alone or inconjunction with other related environmental factors, such as short-termflow variations, remained unknown.

Disagreement resulted over interpretation of data gathered duringthe 1977 studies (U.S. Army Corps of Engineers, Mobile District,1980a.). The U.S. Army Engineer Wfterways Experiment Station, Vicks-burg, at the request of the U.S. Army Corps of Engineers, Mobile Dis-trict, reviewed the technical aspects of the fish kill studies. Water-ways Experiment Station commented that a definitive conclusion for the1976 kill could not be made based on available data. These commentswere rebutted by Georgia Department of Natural Resources, whose hypothe-sis of trout mortality due to humic substances was restated followingthe 1977 work. EPA, as a result of its investigations in 1977, postu-lated that copper toxicity was the cause of trout mortality (Mount etal., 1978).

Studies similar to those initiated in 1977 were continued throughout1978. The U.S. Army Corps of Engineers funded continuation of fishhealth and electron microscopy studies of fish gills by Auburn Univer-sity. The U.S. Army Corps of Engineers also funded periodic water qual-ity sampling in Lake Sidney Lanier and in an 8.5--ile (13.7 kin) reach ofthe Chattahoochee River downstream of Buford Dam. EPA performed a lim-nological study directed at selected heavy metals and conducted sometissue analyses of hatchery fish. Georgia Department of Natural Re-sources continued testing and monitoring water quality in the lake andriver. Fish movement studies were continued by the University of Geor-gia. The 1978 studies generally supported 1977 data concerning waterquality below Buford Dam, but the basic question of the primary causefor the hatchery mortality remained unresolved. Hatchery conditions in1978 appeared worse than in 1977, but not as severe as in 1976.

Studies as extensive as those conducted in 1977 and 1978 were notcarried over into 1979 because the task force felt that little addition-al information would be gained (U.S. Army Corps of Engineers, MobileDistrict, 1980a). The U.S. Army Corps of Engineers continued the base-line water quality study of the lake and river, and Georgia Departmentof Natural Resources continued its hatchery, lake, and river monitoringprogram. Evaluation of 1979 monitoring data indicated water qualityconditions similar to previous years. The hatchery was able to avoidserious trout mortality problems in 1979 by a 50 percent reduction inloadings and by recirculating water within the hatchery during periodsof low flow in the Chattahoochee River since low flow periods exhibited

1-2

worse water quality. Hatchery water could only be recirculated a limit-ed number of times, so special flow releases from Buford Dan were grant-ed by the U.S. Army Corps of Engineers to augment critical low flow per-iods.

Study efforts from 1976 to 1979 contributed significantly to the de-finition of water quality conditions within the lake, river, and hatch-ery, but no consensus was reached as to the cause of the toxicity nor asto whether the river biota was adversely affected by the autumnal waterquality deterioration. However, hatchery personnel were able to predictthe onset of toxicity by following the decrease of oxygen in the hypo-limnion of Lake Lanier and by monitoring the subsequent increase ofdissolved metals in the river.

Don Mount of EPA (Duluth) proposed the basic concept for a bioassayexperiment after his studies in 1977. In 1978, a technical committeefrom the U.S. Army Corps of Engineers, EPA, and Georgia Department ofNatural Resources was formed to develop a conceptual scope of work forthe experiment. The Task Force decided in mid 1979 that the experimentdesigned by the technical committee was the best next-step towardsdetermining the cause of the toxicity. Time did not allow initiation ofthe bioassay experiment in the fall of 1979, so the task force recom-mended that the U.S. Army Corps of Engineers, Mobile District, performthe studies in 1980.

In May 1980, Jones, Edmunds & Associates, Inc. (JEA) was selected byU.S. Army Corps of Engineers, Mobile District, to prepare a detailedstudy design for a bioassay experiment to investigate the cause of troutmortality at the Buford Hatchery. The detailed study design proposalwas approved in August 1980 by an Interagency Technical Review Committee(ITRC) composed of the U.S. Army Corps of Engineers, Georgia Departmentof Natural Resources, U.S. Fish and Wildlife Service, and EPA. JEA wasnotified by the U.S. Army Corps of Engineers, Mobile District, in Sep-tember 1980 to proceed with the plan of study and to conduct the bio-assay experiment.

1.3 THIS STUDY

Three contract documents generated during this project containedversions of the plan of study for the Lake Lanier bioassay experiment.JEA submitted a draft plan of study (DPOS) to the Interagency TaskReview Committee on August 6, 1980. The DPOS proposed three sets ofbioassays at the lake, one set at the hatchery, and benthic experimentsand in situ bioassays in the river. The DPOS document was reviewed anddiscussed at a meeting of the ITRC and JEA in Atlanta on August 19,1980. The ITRC generally concurred with the DPOS with some revisions.Benthic experiments in the river were eliminated because too extensive astudy would have been required to produce useful results. Time andbudget constraints eliminated Set 3 of the lake bioassays, and thehatchery set of bioassays were held in contingency. The Set 2 lake bio-assays were expanded from 96-hr to 14-day tests.

1-3

The second contract document, the Final Plan of Study (POS) (U.S.Army Corps of Engineers, Mobile District, 1980b.), outlined the con-tracted workscope. The POS included two sets of bioassays at LakeLanier and in situ bioassays in the Chattahoochee River. The hatcherybloassays were contingent based on completion of lake bloassays withsufficient time remaining before lake destratification to move to thehatchery.

The third document, this Final Report, describes the actual experi-mentation, including two major differences between the work accomplishedand the work proposed. The contingency bioassays at the hatchery werenot accomplished. The mos7 significant difference is that in additionto the scheduled bioassays in the POS, five sets of static bioassayswere completed which contributed further information in achieving thestudy objectives.

Objectives of this study were to:

1. Define the toxic constituents which have caused trout mor-tality at the Buford Trout Hatchery.

2. Define water quality impacts (related to the toxic constitu-ents) on the aquatic environment of the Chattahoochee Riverdownstream of the dam and evaluate their significance, and

3. Recommend future courses of action for evaluating potentialsolutions to the problems defined in Items I and 2.

The successful completion of all phases of this study was the resultof the collective efforts of several participants. JEA developed thedetailed experimental approach and plan of study and conducted the sci-entific experimentation. The U.S. Army Corps of Engineers, Mobile Dis-trict, initiated the project, had major funding responsibility, and co-ordinated the collective efforts of all participants. The Georgia De-partment of Natural Resources (GDNR), the U.S. Fish and Wildlife Service(FWS), and EPA all served as members of the ITRC. The ITRC providedreview, comment, and assistance with study design and experimentalexecution. In addition, the Game and Fish Division of GDNR and the FWSprovided test organisms; the Environmental Protection Division of GDNRprovided the majority of the required chemical analyses; and EPA provid-ed selected tissue and metals analyses. Representatives of these agen-cies and JEA comprised the Project Team as follows:

JONES, EDMUNDS & ASSOCIATES, INC.

Administration

Richard Jones, JEA President

Proj ect Management

Mel Lehman, Vice-President; Manager, Sciences DivisionFred Ramsey, Water Resource Engineer; Assistant Manager, Sciences

Division.

1-4

Project Staff

Project Scientists

Mike Crezee, Toxicity Bioassay

Richard Jones, Water Treatment

Project Assistant Scientists

Randy Schulze, Aquatic EcologyMary Leslie, Linnology and Aquatic ChemistryWilly Erikson, Chemistry

Jim Rosenbauer, Biology

Project Technicians

Ray Lewis, AquacultureChris Newman, Biological Technician

Lisa Grant, Biological Technician

Technical Support

Project Technical Supervisor

Don Lockard

General Services

Rob Puller, Electrical and Mechanical Engineering Design

Bob Edmunds, Hydraulic DesignDean Scott, Construction TechnicianEddie Taylor, Construction Technician

Document Prod uction

Deb Kramer, Document Production Manager

Brenda Brewer, Word ProcessingSue Patterson, Graphics

JEA Consultants

Expertise Name Affiliation

Hydraulics/ Dr. Wayne Huber Dept. Env. Eng. Sci.Hydrology University of Florida

Gainesville, Florida

Statistical Design & Dr. James McClave Info Tech, Inc.Analysis Gainesville, Florida

Histopathology Dr. Norman Blake Dept. of Marine Sci.Univ. of South FloridaSt. Petersburg, Florida

1-5 . ,.

U.S. ARMY CORPS OF ENGINEERS, MOBILE DISTRICT

Administration

Colonel Robert H. Ryan, Contracting OfficerWillis Ruland, Contracting Officer's RepresentativeGerry Penland, Negotiator

Project Management

Emery Baya, Project Manager (through 3/81)Diane Findley, Project Manager

U.S. ARMY CORPS OF ENGINEERS, SOUTH ATLANTIC DIVISION

John Rushing, Chief, Environment and Resources BranchMary CooperJim Bradley

U.S. ENVIRONMENTAL PROTECTION AGENCY

Bill Peltier, Athens, GeorgiaRon Weldon, Athens, Georgia

GEORGIA DEPARTMENT OF ENVIRONMENTAL PROTECTION

Environmental Protection Division

Roy Hervig, Atlanta, GeorgiaBill Kennedy, Atlanta, GeorgiaOtis Woods, Atlanta, GeorgiaKerry Wilkes, Atlanta, Georgia

Game and Fish Division

Wayne McCord, Buford Hatchery

George Engel, Buford HatcheryRussell England, Gainesville, GeorgiaRich Fatora, Lake Burton Hatchery

U.S. FISH AND WILDLIFE SERVICE

Frank Richardson, Atlanta, Georgia

Gene Braschler, Atlanta, GeorgiaDon Toney, Buford HatcheryJim Clugston, Clemson, South Carolina

104 ACDILID DhDTS

This report is based on studies sponsored by the U.S. Army Corps ofEngineers, Mobile District. Mr. hery Baya, the Project Officer throughMarch 1981, provided valuable historical and background information, inaddition to serving the vital role of coordinator of the Project Teamthrough contracting, study design, and field studies. Dr. Diane Findley

1-6

Now""

served an important technical advisory role to Mr. Bays, and has effec-tively served as the Project Officer during final analyses and report Iwriting. Mr. Gerry Penland, the project negotiator, was instrumental inaccomplishing the necessary contractual details within a compressedschedule so that adequate time remained to conduct critical field stud-Les.

Mr. Cecil B. Patterson (U.S. Army Corps of Engineers), Lake LanierResource Manager, and his staff provided timely and valuable logisticalsupport that was key to the success of field operations.

Mr. Don Toney and Mr. George Engle, Buford Trout Hatchery, providedinformation and use of hatchery lab facilities which were important tofield operations.

Mr. Sherman Stairs (U.S. Fish and Wildlife Service) sent shipmentsof rainbow trout swimup fry from Max Meedows Fish Hatchery (Virginia) asneeded for the bioassays, and Mr. Jack Trask (U.S. Fish and WildlifeService) provided yearling trout from Walhalla National Fish Hatcheryfor liver analyses.

Site visits and technical edvice by Mr. Russ England, Georgia De-partment of Natural Resources Game and Fish Division, Gainesville, andMr. Bill Peltier, EPA Athens, were beneficial to the project. Theiradv-ce in discussions of contingency experiments during the Set 2 bio-assays is reflected in types of treatments used and in static bioassaysperformed.

Several people reviewed a preliminary report of this study. Theircomments were appreciated:

Dr. Diane Findley U.S. Army Corps of Engineers,Mobile District

Mr. Emery Bays U.S. Army Corps of Engineers,Mobile District

Mr. Larry Aggus U.S. Fish and Wildlife ServiceDr. John Grizzle Auburn UniversityDr. Raymond K. Hart Pasat Research Assoc., Inc.Mr. Russ England Georgia Department of Natural

ResourcesMr. Bill Peltier EPA-AthensDr. Ron Raschke EPA-AthensMr. Paul Frey EPA-AthensDr. James P. Clugiton U.S. Fish and Wildlife ServiceMr. Don Toney Buford Hatchery, GeorgiaDr. Joseph Norton U.S. Army Corps of Engineers,

Waterways Experiment Station(Vicksburg)

Mr. Herb De Rigo U.S. Army Corps of Engineers,Savannah District

Mr. Dan M. Mauldin U.S. Army Corps of Engineers,South Atlantic Division(Atlanta)

1-7

Mfr. Jim Bradley andI his wife Jan, Lake Lanier residents, befrienidJEA project staff and! provided gracious hospitality, especially duringthe holiday season.

1-s

AtInawo8

4P

Atlant andnVeinit

Figure1.1. oin of Lk inLmr BufordDaendufdTrtHtceyGegi

tle

SECTION 2.0. LAKE, RIVER, AND HATCHERY SETTING

L

2.0 LUE R IVER, AND HATCRnE SETTING

Lake Sidney Lanier was formed in 1957 by the impoundment of theChattahoochee River by Buford Dam, located 35 miles (56 ka) northeast ofAtlanta, Georgia. Minimum lake elevation for power generation is 1 035feet (315 m) above sea level, and maxiaum flood level is 1,085 feet 1331i). Lake elevation is normally between 1,060 feet (323 a) and 1,070feet (326 i), at which level the lake covers 38,000 acres (15,378hectares).

Penstocks for power generation draw from 919 feet (280 a) elevation,at the bottom of the lake. Usually, only a small generator (6,000-kW)is in operation, and water released through the dam is approximately550 cfs; this is the low flow period in the river. Two larger genera-tors (40,O00-kW) are peaking units delivering power to Georgia PowerCompany. Usually, these generators operate for about two hours on week-day afternoons, but occasionally the peaking units are needed duringweekend afternoons or weekday mornings. During peaking, up to 9,000 cfsmay flow through the dam; this is the high flow period in the river.

Buford Trout Hatchery was built in 1976 approximately 1.5 miles (2.5km) downstream from Buford Dam. This location was chosen because, inlate summer and autumn, bottom waters released from Lake Sidney Lanierare cold enough to allow trout culture when water temperatures so farsouth are generally too warm for trout. Unfortunately, these bottomwaters are also the source of the problem at the hatchery because, inautumn, the hypolimnetic water becomes toxic to trout unless substan-tially diluted by epilimnetic waters.

After lake destratification (December) and until the lake once againstratifies beginning about late July, few or no chemical differencesexist between high and low flow waters each day. However, in late sum-mer and fall, as the chemical characteristics of the hypolimnion andepilinnion differ more and more, so too do low and high flow waters inthe Chattahoochee River. During low flow, the deep intake for the damdraws only hypolimnetic water. During high flow, however, the volumedrawn through is so great that lake stratification immediately behindthe dam is temporarily disrupted (Mount et al., 1978), and ChattahoocheeRiver water is a mix of hypolimnetic and-ep-Timnetic waters.

Thus, in autumn, the Chattahoochee River shows not only brief, largeincreases in flow rate but also associated increases in temperature anddissolved oxygen (Figure 2.1). During periods of high flow, total iron(Fet) and manganese (Mnt) concentrations decrease dramatically fromlow flow levels (Figure 2.2); lesser changes occur in other constitu-ents.

Approximately 4,200 gpm (gallons per minute) of water is continu-ously drawn from the Chattahoochee River and returned to the river afterflowing through the hatchery only once. In autumn, the hatchery some-times draws water in only during high flow and recirculates it, thusavoiding intake of toxic bottom waters during low flow. However, warm-ing of the water limits the duration of recirculation; furthermore,loading of fish in the hatchery must be less for recirculation than foronce-through circulation. Recirculation is neither an adequate norreliable solution to the hatchery problems.

2-1

CI-

00

.2

Z-

pir

LL

2-2"

qdd 'uW

CC

*. C SILLL

oi f0 09

le 0

on~0

0..

0 al

I E

2-3r

SECTION 3.0. HISTORY OF THE PROBLEM AT THE HATCHERY

3.0 HISTORY OF THE PROBLDI AT THE HATCHERY

In the autumn of 1976, its first year of operation, the Buford TroutHatchery lost approximately 435,000 trout of a total of 700,000 brook,brown, and rainbow trout. Ninety percent of the fish lost were rainbowtrout, which are most sensitive to metal toxicity (Affleck, 1952;Nehrig and Goettl, 1974; Chapman, 1978). Fish were excitable, swam incircles before death, and showed "gill deterioration" (Noel1 andOglesby, 1977). The cause of death is disputed. Declining water qual-ity appeared to be the triggering factor, but disease and overstockingof fish may have contributed to the severity of the problem. Consider-able iron floc was found in the hatchery raceways, and physical cloggingof gills was believed responsible for many deaths. Mortality problemsbegan September 12; when first measured on September 18, total Fe was0.9 ppm and Mn 0.5 ppm in the hatchery. All surviving fish were eithermoved to another hatchery or stocked into streams (Noell and Oglesby,1977).

In late October 1976, a team from the Georgia Department of NaturalResources, with the EPA, Auburn University, and the Army National Guard,used the entire fish hatchery in a large-scale experiment in whichdifferent raceways received different chemical additions to determinethe effects on rainbow trout delivered from Walhalla National FishHatchery (Noell and Olglesby, 1977). Results (Table 3.1) showed thatincreasing hardness to 70 ppm mostly eliminated mortality but that asimilar alkalinity increase (and some iron decrease) had a marginaleffect. Tetrasodium EDTA at approximately 100 ppm also markedlydecreased mortality during the 2 1/2-day experiment; fish appeared inpoor condition (England, personal communication) and showed nohistopathological differences from toxic controls. Most likely,mortality was somewhat delayed but would have occurred if tests hadcontinued longer. The other treatments did not affect water chemistrymeasurements or fish mortality. This study also included histologicalexamination of fish, blood hematocrits, and blood iron and manganeselevel determinations. The conclusion was reached that blood iron didnot rise significantly in any of the series but that mortality didcorrespond to variations of total manganese in the blood (Oglesby etal., 1978). However, only single or duplicate determinations of blood

were recorded and, for some treatments, live fish were bled, whilefor others (those with higher Mn) dead fish were taken for bleeding.Histopathological differences in gills, livers, and trunk kidneys werefound between treatments and between fish before versus after thetreatments. Few fish were examined, and the exposure time was too shortto expect any clear effects.

In 1977, daily mortalities went as high a 1.4 percent in November.Weekly chemical monitoring by USGS (data reported by Metrics, Inc.,1978) showed total iron (Pet) at the hatchery intake reached 1,700 ppband total manganese (Mnt), 870 ppb in November (Figure 3.1). Erraticswimming, convulsions, loss of equilibrium, tremors, rapid ventilation,and often darkened coloration were the same symptoms observed duringother years (1976-1980).

3-1

In 1977 several studies related to possible water quality problemsbelow Buford Dam were funded by U.S. Army Corps of Engineers, MobileDistrict. These studies included those of Gilbert and Reinert, Hess,Grizzle, Hart, and England.

Gilbert and Reinert (1979) stocked over 2,000 tagged trout into theriver and followed their movement, growth, and survival by biweeklyelectrofishing. Larger, untagged fish collected from the river weretagged and released. Stations were approximately 0.1, 1.9, 5.2, and 11miles (0.2, 3.1, 8.4, and 18 km, respectively) below the dam.

Ninety-two percent of recaptured fish were recaptured from the sta-tion where they had been tagged. Although results varied between sta-tions, generally a continual decrease in catch-per-unit-effort fromAugust through December (1977) seemed to occur. The majority of recap-tured rainbow and brook trout had lost weight, but most brown trout andall yellow perch had gained weight.

During 5 months of electrofishing, a total of 27 species of fishwere collected from 4 stations. This was considered a depauperatefauna. Yellow perch were most abundant, and the three species of stock-ed trout ranked second, third, and fourth in abundance. Centrarchids(sunfish) were most diverse with 11 species amd hybrids.

Artificial substrates were placed at the same 4 river stations toassess abundance of prey items for fish. The lack of replicates and thelack of any clear trend downstream make it difficult to draw any con-clusions based on these samples. Total numbers ranged from 6,386 ben-thic macroinvertebrates (39 taxa) at the station furthest from the dam(11 miles) to 185 (15 taxa) at the next station upstream (5.2 miles from

the dam).

Concurrent with the above study by Gilbert and Reinert (1979), Hess

(1978) electrofished 390 fish representing 21 species approximately 6miles (10 kn) downstream from Gilbert and Reinert's furthest downstreamstation but found no evidence of tagged fish moving downstream.

Grizzle (1981) histopathologically examined fish from the Chattahoo-chee River immediately below the dam. In 1977, he noted increases inthe numbers of lesions soon after an increase in iron and manganese lev-els and, in 1978, a continual increase in lesions throughout the studyperiod (August through December). No consistent differences were notedbetween fish taken immediately below the dam and near the hatchery (1.5miles below the dam), but Immediately below the dam, rainbow trout weremore affected than brook trout or yellow perch. Edema and aneurysms inthe gill lamellae, fatty change in the liver, congestion of the spleen,and vacuolization of the kidney- tubule epithelium were similarly commonin yellow perch and the trout species, while the gill lamellar hyper-trophy common in trout was observed only once in yellow perch in 1978.Such lesions could be due to low dissolved oxygen; however, they alsooccurred at a station where dissolved oxygen was near or over 5 ppm.

Hart (1979) used energy dispersive x-ray analysis to examine gillsof fish collected immediately below the dam and 1.5 miles (2.5 kin)

f 3-2

further downstream. He found that the iron content of lamellae in-creased with iron content of the water for rainbow trout, but not foryellow perch.

Grizzle (1981) also performed metals analyses on the livers of fishtaken from the Chattahoochee River shortly below Buford Dam in 1978.Average levels found In ppm dry weight were:

C1 Cu Fe Mn Zn

Rainbow Trout 1.1 416 821 30 137Brown Trout 3.4 610 580 27 96Brook Trout 1.7 162 585 15 177Yellow Perch 2.5 8.7 1,405 23 96

Because manganese toxicity had been suggested as the problem at Bu-ford Trout Hatchery, England (1978) performed manganese (MnC1 2 .4H2 0)bioassays at Lake Burton Trout Hatchery where pH (6.2) and hardness (2ppm) are comparably low but where no toxicity problems have occurred.The 96-hour LC50 for yearling rainbow trout was 24.7 ppm manganeseand was estimated for brook and brown trout as considerably higher, thussuggesting that the far lower Mn levels (up to 1 ppm) at the BufordHatchery should not be a problem.

Also during the fall of 1977, EPA (Mount et al., 1978) studlied thetoxicity problem at the hatchery. Hatchery rainbow trout exposed sever-al weeks to low flow waters showed no consistent pattern of necrosis inthe kidney, liver, or gill. The most consistent finding was dilation ofthe terminal vessels of the lamellae, with clavate-glovate lamellae con-gested with hemorrhagic exudate, blood, and/or Inflammatory cells. Somehepatic necrosis, and decrease in liver glycogen and in pancreatic zymo-gen granules also were observed in some distressed rainbow trout, butkidney tissue appeared normal.

Gas chromatographic-mass spectrometric (GC-MS) analyses of riverwater found no organics detectable at one ppb; even when organics wereconcentrated 550 X by an extraction process, no consistent effect onDaphnia could be demonstrated. Examination of all information arousedno suspicion of any toxicant from municipal or industrial discharges norfrom any non-point source. No abnormal levels of chlorinated hydrocar-bons were found in kidneys or livers nor did rigorous GC-MS analysis of100 grams of muscle tissue detect hydrocarbon contamination (Mount etal., 1978).

Single water samples at Gwinnet County and City of Atlanta waterworks and in Lake Sidney Lanier, but not at the hatchery, showed poten-tially toxic levels of copper. Livers of rainbow trout from the hatch-ery and from the edjacent Chattahoochee River contained 230-510 ppm cop-per (dry weight). These copper levels were considered to be high and toindicate exposure of the trout to toxic levels of copper; however, nocomparative data for copper levels In known unexposed rainbow trout werepresented to justify this conclusion of Mount et al. (1978). Furtheranalyses were performed In 1978 with similar results but still withoutany controls (U.S. Environmental Protection Agency, 1979).

3-3

~ *~. ':

In 1978 the hatchery used recirculation, supplemental aeration, andhydrogen peroxide in efforts to limit mortalities. Mortality rates(Figure 3.2) were lover than in the two previous years until late Novem-ber when they reached 0.6 percent per day. Few measurements of Fe andMn are available for 1978, but levels were apparently lower than for1976 and 1977, with Fet reaching only about 1,200 ppb and Mat about700 ppb.

In 1978, Grizzle continued histopathological examination of fishfrom the Chattahoochee River. This study, initiated in 1977, is dis-cussed earlier in this section. U.S. Army Corps of Engineers, MobileDistrict, also contracted for chemical and biological studies of LakeSidney Lanier; this work included occasional chemical sampling at fourstations in the Chattahoochee River between Buford Dam and McGinnisBridge (Environmental Science and Engineering, Inc., 1981; data pre-

sented in Figure 3.2).

In 1979, the hatchery was able to limit mortalities effectively byrecirculating water drawn in only during high flow. This was possiblebecause U.S. Army Corps of Engineers made special releases from BufordDam for this purpose on weekends. Mortalities and iron and manganeselevels are graphed in Figure 3.3. The only high mortality began twodays after low flow water (2,300 ppb Fet) was drawn on November 21.Unfortunately, manganese measurements had been discontinued, but a fewdays prior to low flow intake levels in the river at low flow had beenapproximately 600 ppb Iait and 2,400 ppb Fet.

In 1980, trout losses were limited by recirculation of high flowwater and by stocking mostly brook trout, which are more resistant tometals than rainbow trout. In spite of these measures and metals con-centrations lower than in 1979 (Figure 3.4), some losses were sustained,and trout showed such excitability that feeding had to be discontinuedfor several days. A mortality increase predicted from elevated ironlevels did not occur, possibly because manganese did not show the usualconcommitant rise. Later, when manganese concentration did increase,mortalities also increased (Don Toney and George Engle, personal commun-ications).

Note:

Table 3.] and Figure 3.4 present data for ferrous iron (Fe"+).Figure 3.4 also refers to Feq+ as dissolved iron. It may not becompletely accurate to assume all dissolved iron measured is ferrousIron, or conversely, that all ferrous iron is dissolved. However,in general, most of the ferrous iron will be measured in the dis-

solved fraction. Throughout this report, for discussion purposes,ferrous iron (Fe+ + ) and dissolved iron (Fed) are used inter-changeably. Where iron additions were made to test waters in thiOstudy, additions were made as ferrous iron and are reported asFe++ . Where test waters were analyzed for iron content, concen-trations are reported as dissolved iron (Fed) and/or total iron

(Fet)•

$3-4

Table 3.1. Summary of Data from the 1976 Buford Trout Hatchery ExperimentsReported by Noell and Oglesby, 1977

Treatment Mean Fe Fet pH Alk HardnessMortality (pp.) (ppm) (ppm) (ppm)

BT

Control 50.17 Z 2.28 2.88 6.71 18.30 11.50

CaSO4 & 1.432** 2.22 2.81 6.81 18.54 70.00NBSo 4

Na4 EDTA 2.962** 0.16 2.96 7.10 27.56 5.00lOoppo

NaHC0 3 & 38.292* 0.76 2.95 7.48 86.27 11.59Na2CO3

Ca(OH) 2 48.09Zn.s. 1.97 2.89 6.79 18.73 12.89

03 53.76Un.s. 2.30 2.92 6.70 17.50 11.50

Xotes:

* Significantly different from control (p - 0.05) one-tail test.* Significantly different from control (p - 0.01).

n.s. - Not significantly different from control, two-tail test (p - 0.05).INT - Rainbow trout.Alk - AlkalinityFe+ + - Ferrous IronYet - Total Iron

3-5

A- .

§ qdd 'Iuz ' 1

I 0

4.

4.0

*0.o% c

2 z r

i~4,.

zQddauja

Iu w

a2a

z

0

0 w

0 4

0 0 NB

a.O

a.NO

inoil moquieW jeqwnN 'A411ISJOW~ 10%01

CC

- I.

.0

MORN ,.

2z

C0

uh~ulil - U

qdd l 'uw '14"

3-8

500 K 8j, I I N

I N

&L7

a / l4S

.1 00/

I I200 t vI

100V

010/1 10/10 10/20 10130 11/9 11/19 11/29 12/9 12119 12/29

Time. days

20007

1500i

U. is

1000II

500

0iOu 10/10 10/20 10/30 11/9 11/19 11/29 12/9 12/19 12/29

Time., doys

Figure 3.4. Total Manganes and Disolved Iron (Fe++) in the Chattahoochee

River, Georgia, at Buford Hatchery Intake, Fall 1979 and 1960

(Dots Courrew of Don Toney)

3-9

SECTION 4.0. MATERIALS AND METHODS

4.0 MATERIALS AND ]ETHODS

4 .1 INTlODUCTION

The Interagency Task Force recommended use of several bioassaytreatments for the current study, each removing a different potentiallytoxic agent from water of demonstrated toxicity. Where toxicity was re-duced, the removed substance would be identified as toxic.

To avoid the complications of frequent toxicity fluctuations inriver water at the hatchery and to provide test water of relatively con-stant composition and toxicity during individual experiments, water wasdrawn from Lake Sidney Lanier immediately behind Buford Dam (Figure4.1).

A fenced bioassay compound (Figure 4.2) was established directly ontop of Buford Dam. The compound was staffed by JEA personnel 24 hours aday for the duration of the project. The compound consisted of a mobilebioassay laboratory, water and fish holding facilities, a utility shed,and a small Winnebago trailer which provided office space and accommoda-tions for personnel on duty.

The mobile bioassay laboratory was modified and equipped for thisproject. The lab was divided into two operational sections. The rearsection served as the bioassay laboratory, requiring additional insula-tion, temperature control equipment, and aquarium racks with fluorescentlighting. The front section served as a support laboratory for handlingchemical samples and monitoring equipment.

As an experimental expedient, rainbow trout swim-up fry were usedrather than the larger fingerling trout normally received into thehatchery. Fry were chosen both because of their smaller size and conse-quently lesser space requirements and because of their greater sensitiv-ity to metals (U.S. EPA, 1980), the class of toxic agents most stronglyImplicated by previous evidence from the hatchery. Two sets of experi-ments were conducted, each with more than one run. Set I determined thetoxicity of untreated waters from four different depths 20, 76, 95, and115 feet (6, 23, 29, and 35 meters) in the reservoir 600 feet (180 m)behind Buford Dam. The objectives were to choose the most toxic waterfor use in Set II bioassays and to attempt to correlate toxicity withchemical composition of the waters. In Set II, which was run four timeswith variations, water from the reservoir depth found most toxic in SetI was treated in various ways to attempt to remove the toxic factor.Designs for Set I and Set II experiments are compared in Table 4.1.

Initially, a third set was to be conducted at the hatchery to verifyconclusions based on Sets I and II. However, due to the limited timebetween project initiation and the cessation of toxicity when the reser-voir destratified, this set was not performed.

4-1

4.2 MRE NOVEMENT AM FLOWS

To facilitate retrieval of hypolimnetic lake water, the U.S. ArmyCorps of Engineers anchored a buoy 600 feet (180 meters) off-shore up-stream of the dam (Figure 4.3). JEA personnel attached a submersiblepump (STA-RITE 10-cm, 60-hertz, 2-horsepower, 2.5 liters per second) tothe buoy in a manner that allowed raising or lowering the pump to anydesired depth. An identical pump was suspended over the side of the dam20 feet (6 meters) below the water surface to provide a continuous sup-ply of non-toxic epilimnetic water for controls and for holding andacclimating fish prior to testing. Water from both sources was conveyedthrough 2-inch (5-cm) diameter linear polyethylene pipe (Driscopipe8600).

Warmer, epilimnetic water was cooled by passage through a kam carbonsteel heat exchanger (6 inches x 7 feet), using hypolimnetic water as acoolant.

Water sufficient for Set I bioassays was pumped from four depthsprior to the bioassay and stored in separate 250-gallon (950-liter)linear polyethylene (LPE) tanks. To prevent temperature fluctuations,these tanks were held in two aboveground swimming pools, measuring 12feet (3.5 meters) in diameter, through which cool hypolimnetic water wascontinuously circulated. Eighty-gallon (300-liter) LPE fish holdingtanks were also kept in these swimming pool water baths.

To prevent aeration during storage of the hypolimnetic water compos-ited for the Set I bioassays, a special system was designed to maintaina nitrogen atmosphere within the void space of the tanks during storageand pump-out.

The water source for the Set I experiments and all epilimnetic con-trols was the 250-gallon (950-liter) tanks outside the bioassay trailer.For the first three runs of Set II experiments, hypolimnetic water waspumped directly from the bottom of the lake into a 30-gallon (110-liter)header tank inside the trailer. Because turnover of the reservoir ap-peared imminent, all water for Set II Run 4 was drawn prior to the ex-periment and stored under nitrogen atmosphere in a 3,000-gallon (11,500-liter) iron tank.

For both Set I and Set II experiments, water movements into the ex-periments were controlled by timers on a 10-minute cycle (Figure 4.4).A separate pump filled a volume setting chamber for each treatment.After the first pump was off long enough for all excess water to over-flow and for water to be aerated, a second pump transferred water to aspecially designed plexig)ass splitter which divided the single inflowinto 8 equal outflows to , replicate aquaria (see Figure 4.4). Averageflow to each aquarium was 230 millillters/10 minutes, which provided onereplacement of the 4-liter standing v~lume approximately every threehours. Each volume setter was equipped with an air stone which providedthe only aeration for most treatments. Timers activated the air pumpsfor approximately 4 minutes during each 10-minute cycle. This aerationalso provided mixing for hardness or EKTA additions.

4-2

4.3 TREAThENTS AND TREAT2ENT MATERIALS

The various treatments provided in the Set II flow-through bioassaysare summarized in Table 4.2. Essentially, these treatments were accom-

plished by adding resin columns, granular activated carbon columns, orprolonged aeration before the volume setters; by adding chemicals (EDTA,hardness) via Dias Model BX55 peristaltic pumps to water in the volumesetters; by adding filtration (Teel filter holders and 5-vm cellulosecartridges) either before or immediately following the volume setters;or by combinations of the above (see Figure 4.5).

Anion exchange resin (Dowex 11), cation exchange resin (DowexHCR-S), and activated carbon were individually packed in well-leached1-meter (3.3-foot) lengths of 15-centimeter (6-inch) diameter PVC pipewith tightly fitting PVC caps; all tubing used in bioassay equipment andtreatment apparati was Silastic@ silicon tubing.

4.4 FISH USED IN BIOASSAYS

All fish tested in the static and flow-through bioassays were rain-bow trout swim-up fry from the U.S. National Fish Hatchery at MaxMeadow, Virginia, provided to JEA by the U.S. Fish and Wildlife Serviceupon request. Fry were approximately 2 to 3 centimeters (approximatelyone inch) in length and varied from 0.117 to 0.44 grams average weight.

Fry were packed with ice in Styrofoam* coolers and shipped by bus(14-hour trip). When received, the unopened plastic bag was placed in aholding tank with cooled, flowing, epilimnetic water to allow gradualwarming over 4 to 6 hours. Fry were then released into the holdingtanks and fed. Mortalities during shipping and holding were extremelyfew when fry shipments were received promptly. Fry were held at least

two days before being used in bioassays. During holding and acclima-tion, fry were fed trout chow provided by Buford Hatchery 4 times duringdaylight hours. During the bioassays, fry were fed sparingly twicedaily. No disease treatment was needed; handling was minimized.

4.5 3.IVERINE iIOASSAYS

The object of the riverine bioassays was to document fish mortalityin the river and to determine the extent of toxic effects downstream.

Two species of fish were used, rainbow trout (Salmo gairdneri) and blue-gill (Lepomis macrochirus), both of which are found in the river. Thein situ bioassays were run concurrently with Set II laboratory bioas-

says, thus facilitating a comparison between the riverine and laboratorybioassays.

Swim-up fry and 15-centimeter (6-inch) rainbow trout were used inthe riverine bioassay. Swim-up fry were also intended for use in his-tological examinations, but they were not examined because histologicaleffects were not observed in the laboratory bioassays.

The in situ bioassays were conducted simultanrously at six sites inthe Chattahoochee River (Figure 4.6). The six si, were the tailraceof the dam (300 meters downstream of Buford Dam), the sluice channel atthe dam (immediately below Buford Dam), in the river near the water

4-3

intake to the trout hatchery (2.5 km downstream of Buford Dam), Settle'sBridge (7.5 km downstream of Buford Dam), McGinnis Bridge (also known asLittle's Ferry Bridge) (13.6 km downstream of Buford Dam), and AbbottsBridge (22 km downstream of Buford Dam). In addition, cages were placedin the first raceway at the trout hatchery, and a control was held incooled epilianetic water at the bioassay facility at the dam (Figure4.6).

The first set of 96-hour bioassays was conducted November 22 through26, 1980, and the second set on February 5 through 9, 1981, after thelake had destratified.

At each of the eight sites, fish were placed in a 15-chamber holdingcage (Figure 4.7) and the 96-hour tests run. The test chambers weretethered to a stable post in the river and weighted by an attachedcement bag. After securing the cages in the river, a 2-day leachingperiod was allowed before fish were introduced. Each cage containedthree chambers with ten large trout each, eight replicates of ten swim-up fry each, and three chambers with ten bluegills each.

Dissolved oxygen, temperature, pH, and conductivity were measured inthe field on a daily basis during the 96-hour in situ bioassays. Totaland dissolved iron and manganese were determined on one mid-low flow andone id-high flow sample for three consecutive days at the hatchery in-take and at McGinnis Bridge. Samples for dissolved metal analyses werefiltered immediately in the field using a 0.1-micron Nucleopore membranefilter and then acidified. All metals samples were stored in polyethy-lene bottles.

4.6 METHODS FOR ROUTI CIICAL NONITORIC

Samples for chemical analyses were all composites of six equal ali-quota taken 4 hours apart. Samples were taken during the beginning,middle, and final 24-hour periods for Set I runs and daily for Set IIruns. Volumes for the separate samples in each set were 900 millilitersfor total metals and hardness (pH < 2 by 11103 addition), 120 milli-liters for dissolved metals (0.1 pa filter, pH < 2), 1,800 millilitersfor chlorinated hydrocarbons (Set I only), 240 milliliters for ammoniaand total organic carbon (pH < 2 with H2SO4), and 1,800 millilitersfor alkalinity, color, and total dissolved and suspended solids. Sam-ples for chlorinated hydrocarbons were stored in glass containers; allother samples were stored in LPE bottles.

Water samples were taken either from the volume setters or from thesplitters by siphoning water into graduated cylinders. For filteredsamples, water was drawn into a 50-milliliter plastic syringe and thenforced through an 0.1-um Nucleopore polycarbonate filter (47-millime-ter diameter) held in a syringe filter holder. To prevent cross-contam-ination between treatments, a separate siphon, graduated cylinder,syringe, and filter holder were used for each treatment.

Samples were analyzed according to methods referenced in Table 4.3.

4-4

4.7 ITODS FOR OTUIR NWITORIlC

Routine measurements of dissolved oxygen (DO), pH, conductivity, andtemperature were taken in three replicate aquaria for each treatmentevery four hours for the duration of each experiment, up to 120 hours(see Appendix B). DO measurements were made with Yellow Springs Instru-ment Company, Model 57 portable DO meters and conductivity with a YSIModel 33 SCT meter. For pH measurements, VWR mini pH meters with digi-tal readout were used; consistent problems with drift during readingsoccurred, with poor repeatability of readings. pH probes were laterfound to have a short life and to give erratic and low readings (up to0.5 pH units) as they failed. Because the meters continued to cali-brate, not giving a clear beginning for gradual failure, no correctionswere made for any pH values. Weston Model 2284 Mirroband stainlesssteel dial thermometers accurate to + 0.5*C were used for temperaturemeasurements.

DO meters were air calibrated before each set of readings, and cali-bration was checked at least once during and at the end of a series ofreadings. At the same intervals, the pH meter was calibrated using pH7.0 and 4.0 buffers.

Fish counts and removal of mortalities were performed every 8 hours.Early in the tests, mortalities were removed more frequently, but thisoften caused great distress to the commonly excitable survivors and wasdiscontinued.

4.8 SPECIAL MWICAL SAMPIZS

Supplemental to the routine chemical monitoring of the bioassaytreatments, several kinds of special samples were taken once or a fewtimes each. Hydrogen sulfide samples from the 250-gallon (950-liter)LPE tanks (Set 1), from the header tank in the laboratory (Set II) andfrom a Kemmerer grab sample from the bottom of the lake were carefullysiphoned into the bottom of one-liter glass jars. Approximatley onequart (one full volume) was allowed to overflow. Four milliliters of 2Nzinc acetate were immediately added, followed by addition of 6N NaOH andthe completely filled bottle was carefully sealed so as to trap a mini-mum of air.

Samples for volatile organic carbon (VOC) analyses were taken fromthe volume setters (Set I only) before aeration. Solvent-rinsed, thor-oughly dried (300*C), septum top, glass vials supplied by Georgia EPDwere filled and sealed underwater so that no air bubbles were trapped inthe vials.

During Set II, samples were taken simultaneously from the headertank in the laboratory and by Kemmerer grab from the bottom of the lakeadjacent to the pump which filled the header tank. Samples were ana-lyzed for total and dissolved iron and manganese, sulfides, TOC, andalkalinity.

Two 2-quart glass jars of hypolimnetic water taken from the 3,000-gallon (11,500-liter) storage tank, and placed in cold, dark storage,

4-5

were subsequently analyzed for humic substances by high pressure liquidchromatography (HPLC).

4.9 TISSUE ANALYSIS

Identical rainbow trout liver samples were taken at Buford TroutHatchery and at Walhalla National Fish Hatchery, where trout of the sameage and strain and originating from the same hatchery were available.At each hatchery, livers were taken from thirty 20-centimeter rainbowtrout and pooled into 10 separately analyzed samples of 3 livers each.Samples were frozen and delivered to EPA Athens laboratory where theywere analyzed.

4. 10 RISTOPATROLOGY

At the end of each bioassay run, two fry from each replicate (total16 fish per treatment) were fixed in Dietrich's solution and sent to Dr.Norman Blake (University of South Florida Marine Science, St. Peters-burg, Florida) for histopathological examination.

In Dr. Blake's laboratory, trout were cut longitudinally so as toexpose most of the internal organs. Each half was placed in a casetteand washed overnight in flowing tap water. The fish were then processedon an Autotechnicon through a series of S-29, UC-670, and Paraplast.Each fish was embedded in Paraplast in such a way that sections could bemade through the internal organs as well as through the eye, epithelialtissues, and musculature. The tissue blocks were cut at 6 um and theresulting sections were routinely stained with Harris- hematoxylin.Slides were examined and interpreted by Dr. Blake and selected slidessent to Dr. Paul Yevitch (EPA Narragansett) for confirmation ofinterpretations.

4.11 STATISTICAL EANDLIUG OF DATA

In the statistical analyses of bioassay results, each time periodwas analyzed separately rather than together because for each replicatesurvival at one time is highly dependent upon survival at precedingtimes. The arc sine square root transform of proportion survival wasused to obtain more stable variability than untransformed proportionsurvival (Mendenhall, 1968).

Analyses of variance (ANOVA) were conducted to compare stations(riverine) or treatments (laboratory) for each 24 hours. When ANOVAshowed differences between stations or treatments, a Duncan's multiplecomparison was employed to compare and rank the individual station ortreatment means. Eight replicates made the experiments so sensitivethat even seemingly small differences may be statistically significant.In riverine experiments, the station-age interaction in ANOVA was signi-ficant; therefore, fry and yearling tests were analyzed separately. Forlaboratory experiments, the run-treatment interactions were significantusually for Set I and always for Set II, so treatment comparisons weremade only within runs.

4-6

i4

For each set of experiments, mean survival (transformed) for alltreatments was regressed aainst dissolved iron (Fed), total iron(Pet), and total manganese (lnt), and hardness measured in thetreatments. Experiments were not designed specifically for this type oftreatment, and regressions can provide no more than suggestions ofpotential causality.

Comparisons of metal contents of trout livers simply used a stu-dent's t-test.

4.12 ANALYTICAL qUALITY (ODNTOL

The bulk of the samples was analyzed by the Georgia EnvironmentalProtection Division laboratory, Atlanta; additional QA/QC samples wereanalyzed by JEA's laboratory, Micro Methods, in Pascagoula, Mississippi.

Quality assurance and quality control (QA/QC) duplicates, spikes,and blanks were included with each set of chemical samples. For onetreatment chosen randomly, two additional sets of samples were taken tosend Georgia EPD triplicates for all but total metal samples. For totalmetals, duplicates were sent and the third sample was spiked at knownlevels of some or all of the metals to be analyzed. For a second treat-ment, also randomly chosen, two additional sets of samples were takenfor analysis by JEA's laboratory; one of these total metals samples wasspiked. A dissolved metals blank, prepared by filtering an aliquot ofdoubly de-ionized water each time one of the six aliquots was taken, wasincluded with each set of samples for Georgia EPD.

Spiked samples mere prepared using certified standards from FisherScientific and from Environmental Research Associates; the latter wasprovided by the U.S. EPA, Athens, Georgia.

To assure that spike values would fall within acceptable limits, 500ug/l Fe or Mn was added. Total metals samples were spiked with 50 al oflO-ug/l stock solution and brought to one liter total volume. For dis-solved metals, 5 ml of stock solution spiked 100 al total final volume.

All glassware used to prepare spiked samples and all sample contain-ers were washed with soap (Alconox), thoroughly rinsed with de-ionizedwater, and, if appropriate, finally rinsed in a 10 percent nitric acid(reagent grade) solution.

4-7

6--

Table 4.1. Summary Comparison of Set I and Set II Flow-Through BioassayExperiments Performed at Lake Sidney Lanier, Georgia, Fall 1980

Set I Set II

Date of runs 1. Oct. 28 - Nov. 2 1. Nov. 21 - Nov. 232. Nov. 7 -Nov. 10 2. Nov. 23 - Nov. 24

3. Dec. 4 - Dec. 84. Dec. 19 - Dec. 2i

Types of water none, water type varied several, to remove specific

treatments with reservoir depth potential poisons or to addEDTA or hardness (see Table4.2)

Hypolimnetic depths 23 m (76 ft), 29 m (95drawn from ft), 35 m (115 ft) 35 m (115 ft)

Non-toxic control cooled epilimnetic water same as Set I

from 6-m (20 ft) depth

Schedule for drawing drawn prior to bioassay drawn continuously from

water and stored under N2 reservoir

Treatments each run 4 5 to 8

Replicates per treatment 8 8

Rainbow trout fry per 20 20

replicate

Frequency of physical every 4 hours every 4 hours

monitoring

Chemical parameter intensive, including se- narrowed by elimination

monitoring eral metals, H2S , organ- of numerous substances

ics, pesticides, PCB's not detected in Set I

(see Appendix A)

Times of chemical 24 hr., 60 hr., 96 hr. 24 hr., 48 hr., 72 hr.,

sampling 96 hr.

Method of chemical each sample sum of same as Set I

sampling aliquots every 4 hr.for previous 24 hr.

4-8

.A

Table 4.2. Summary of Treatments for Flow-Through Bioassay ExperimentsPerformed at Lake Sidney Lanier, Georgia, Fall 1980

Set I Run 1, October 28 through November 2and Set I Run 2, November 7 through 10

A Epilinnetic water from 6-meter (20-foot) water depth, pump suspended fromBuford Dam

B Hypolimnetic water from 23-meter (75-foot) water depth, 180 meters (600 feet)behind Buford Dam

C Hypolinetic water from 29-meter (95 foot) water depth, 180 meters (600 feet)behind Buford Dam

D Hypolimnetic water from 35-meter (115-foot) water depth, 180 meters (600feet) behind Buford Dwm

A, B, C, and D - all water for the run was drawn just before the experiment andstored in 950-liter (250-gallon) LPE tanks.

Set II

General: All epilimnetic water was drawn by a pump suspended at 6-meter (20-foot) water depth off Buford Dam. Water was stored briefly in a 950-liter (250-gallon) LPE tank before use in experiments.

All hypolimnetic water was drawn by a pump suspended at 35-meter (115-foot) water depth about 180 meters (600 feet) behind Buford Dan. For

Runs 1, 2, and 3, water flowed continuously into a 110-liter (30-gallon) header tank in the laboratory. For Run 4, all water was drawnbefore the experiment and stored under nitrogen atmosphere in a

11,500-liter (3,000-gallon) iron tank.

For every treatment, water was aerated for about 4 minutes while inthe volume setter.

See Figure 4.5 for diagram of Run I set up.

Run 1, November 21 through 23

A TOXIC CONTROL - untreated hypolimnetic water.

B NON-TOXIC CONTROL - epilimnetic water.

C EDTA - hypolimnetic water with 10 ppm Na4 EDTA added.

D AERATION - hypolinnetic water continuously aerated during passage through ten16-liter LPE buckets, giving an aeration tine of 12 to 15 hours before flowinto the volume setter. The objective was to oxidize soluable Fe+ + toinsoluble Fe.. compounds.

(Continued)4-9

Table 4.2. Summary of Treatments for Flow-Through Bioassay ExperimentsPerformed at Lake Sidney Lanier, Fall 1980

E ORGANICS REMOVED - hypolimnetic water run through a 5-urn cellulose filterthen throngh a column of Dowex 11 resin which exchanges C1- for otheranions (which Include organic@).

F MEZTALS REMOVED - hypolanetfc water run through a 5-urn filter then DowexHCR-S resin which exchanges Nqat for other cations.

G ORGANICS AND METALS REMOVED - hypolimnetic water, 5-urn filter, activatedcarbon, 5-urn filter (to remove carbon particles), Dowex 11, Dowex HCR-Sresins.

H FILTER CONTROL - otherwise untreated hypolimnetic water run through a 5-uacellulose filter.

---------------------------------------------------------------------------------

Run 2, November 23 and 24

A TOXIC CONTROL - untreated hypoliumnetic water.

B YON-TOXIC CONTROL - epilimnetic water.

C EDTA - hypolimnetic water with 50 ppm Na4 EDTA added (vs. 10 ppm Run 1)

F METALS REMOVED FROM EPILIMNETIC WATER - run through 5-urn filter and DowexHCR-S resin. Objective was to see if removing hardness, as happened alongwith metals removal in SII IF and C, would adversely affect the trout fry.For unknown reasons, however, the resin did not remove hardness from theepilimnetic water.

C CHANGED ORDER OF RESIN AND CARBON COLUMNS - hypolimnetic water through 5-ufilter, Dowex HCR-S, Dowex 11, activated carbon, 5-urm filter. Objective wasto test possibility that an organic leached from Dowex HCR-S resin was toxic;Dowex 11 and activated carbon should remove any such organic.

-------------------------------------------------------------------------

Run 3, December 4 through 8

A TOXIC CONTROL - untreated hypolimnetic water.

B NON-TOXIC CONTROL - epilimnetic water.

C AERATION + FILTRATION + HARDNESS - hypolimnetic water aerated 12 hours as inRun 1, D, then allowed to settle for 2 hours, then run through a 5-u filterto remove iron floc, then 50 ppm hardness added as CaC12 (but expressed asCaCo 3 ).

D AERATION + FILTRATION - exactly as C but without hardness addition. Flows Cand D were split after 2 hour settling and before the filters.

(Continued)

4-10

Table 4.2. Summary of Treatments for Flow-Through Bioassay ExperimentsPerformed at Lake Sidney Lanier, Fall 1980

I METALS REMOVED THEN HARDNESS REPLACED - hypolimnetic water run through 5-umfilter, activated charcoal, 5-urn filter, HCR-S cation exchange resin, then 10ppm hardness added (as CaC12 but expressed as CaCO3) to replace hardnessremoved by the resin.

P METALS REMOVED - exactly as E but without hardness aidition. Filters andcolumns were completely separate for R and F..

G ACTIVATED CHARCOAL + HARDNESS - hypolimnetic water, 5-urn filter, activatedcharcoal column, 5-urn filter, the 10 ppm hardness added to replace what thecharcoal was expected to remove.

H ACTIVATED CHARCOAL - exactly as G but without hardness addition. Flows for Gand H were completely separate.

Run 4, December 19 through 23

A TOXIC CONTROL - untreated hypolimnetic water. All bottom water for Run 4 wasstored in a 3,000-Sallon (11,500-liter) iron tank.

B NON-TOXIC CONTROL - epillmnetic water.

C METALS PRECIPITATION - performed as a static. Hypolimnetic water, pH raisedto 11 with NaOH, aerated I hour, settled for 1 hour, floc removed by 5-urnfilter, aijusted to pH 6 with HC1. The acid pH ad justment step to pH 6 couldnot be performed reedily for a flow-through.

D METALS PRECIPITATION + HARDNESS - exactly as C with final addition of CaC12to increase hardness by 10 ppm.

E AERATION - hypolimnetic water aerated for about 6 hours in a setup similar tothat in SII1D.

F AERATION + 10 PPM HARDNESS - otherwise untreated hypolimnetic water with 10ppm hardness aided as Cad 2 (but expressed as CaCO3).

G AERATION + 25 PPM HARDNESS - similar to F.

H AERATION + 50 PPM HARDNESS - similar to F.

4-11

. ... .. .. . . .... .: ... .. . .. ..... ... .. .... ,, , , , ,,'- -- .. '

Table 4.3. Methods Used for Chemical Analysis of Water Samples, Lake SidneyLanier, Georgia, Fall 1980

Parameter Method Reference

Temperature Thermometer APHA p. 125Dissolved Oxygen Membrane electrode APRA p. 450pH Hydrogen ion electrode EPA p. 239Conductivity (@ 25 C) Condictivity bridge APHA p. 71Total Suspended Solids Gravimetric EPA p. 268Total Dissolved Solids Oravimetric, lab filtered EPA p. 266Ammonia Ammonia Electrode or APHA p. 412

Distillation, Nesslerization EPA p. 165Alkalinity Titrimetric method EPA p. 3Color Apparent Visual comparison method APHA p. 64Total Organic Carbon Beckman TOC analyzer EPA p. 236Aluminum, Total AAS EPA p. 92Arsenic, Total AAS APHA p. 159Cadmium, Total AAS EPA p. 101Chromium, Total AAS EPA p. 105Copper, Total AAS EPA p. 108Iron, Total AAS EPA p. 110Iron, Dissolved Field filtration (0.1 un), AAS EPA p. 110Lead, Total AAS EPA p. 112Manganese, Total AAS EPA p. 116Manganese, Dissolved Field Filtration (0.10 um), AAS EPA p. 116Mercury, Total AAS (Flameless-AA) APHA p. 156Nickel, Total AAS EPA p. 141Zinc, Total AAS EPA p. 155Sulfide, Total Methylene blue APRA p. 503Hardness EDTA titration APHA p. 202Calcium, Total AAS EPA p. 103Magnesium, Total AAS EPA p. 114PCIB's - Arochlor series GC-EC PAM, APRAOrganophosphorus Parathion, malathion PAM, APHA

Pesticides - methyl parathion, GuthionOrganochlorine DDE, DDD, DDT Toxaphene PAM, APHA

Pesticides - Chlorodane, Hsptachlor,Dieldrin, methoxychlor

Abbreviations:

AAS - Atomic Absorption Spectrophotometry, all total metals are digested by theEPA, 1974 method on p. 83, Section 4.1.4.

GC-EC - Gas Chromatography using an election capture detector.

APHA - American Public Health Association. 1975. Standard Methods for the Exam-ination of Water and Wastevater, 14th ad. AWWA, WPCP, APAA, Washington, D.C.

EPA - U.S. Environmental Protection Agency. 1974. Environmental Monitoring andSupport Laboratory. Cincinnati, Ohio. Methods for Chemical Analysis of Water andWastes.

PAN - Pesticide Analytical Manual, 1977. U.S. Department of Health, Educationand Welfare. Vol. I. Methods which Detect Multiple Residues. Washington D.C.

4-12

- - -- -------

Lke Sidney Ln~

SEA Oypol~ntjM Waler nkOBU

*r l t e m "N

F o

S~yLOinwf yolmtclunpApxisel 0 M~sUPNO

F ipe 41. L~N SdmvLan. cao Nof gufrd Dom

Figure 4.2. JEA Experimental Bioassay Compound on Top of Buford Dam, Georgia

A Elioassay nrohile lakxmratory with special cooling unit on left end.

jf vr'lvte foot rfiatrreter swinifin 0 pool usedi as a water baitI fo, four 250 ial Ion lnear Polyvethyvlene ( LP El water sto 01 d

tan ks and for two fish holdi ng tanks.

C Heat exchange for cooling epilirrnntic water.

Nitrogen tanks for maintaining N2 atniospheve in 2O-gailon (LPE) tanks

Field (tffiCi7

Sroraqv sthed.

I wii iri,:h Wim f,iuii hvpoliinioir.

hirftlow le tii lake-

Two ich owi 1,01eiloilrl pnof is lusiifilton, the dJani iost off the upppr left of i- phitii

4-14

LU

I0 4

LL.

CkC

2U

L-

59 8

JGAG W-UWW "V ":jUl U43XOIC

Mender Tank (inside Sioassay Treiierl

Treatments

(Resin Columns) b

Aerator

m

volume Setter

Splash cap

± Flow Slitter

Overflow from Standpipe

Drain Overflow from Aquaria

Hypoiimnetic Water Continuousiy Flowed Into and Overflowe Out of the Header Tank.

Flos for Set I and for $et 11 Run 4 Ware Identical Except That the Water Source Was an

Outdoor Storage Tank Mather Than a Continuous Flow M~edr Tank.

Figure 4.4. Diagramn of Water Flow$ for One Treatment of Set 11 Bioauays at Buford Dam, Georgia,Fall I'.0

4-16

"Nov.-

Fro

T I p

Sple Fltrtiler Fite

T o Aquff er e Air

P Pump Soc

Activw

IPerfrme Resuodiamna 18

4-17

JEA COMPn~Ptd

*3 0 1 2 3 4

SCALE (miles)

Buford Hatchery Raceways,.1

~ j t Buford 0i

3cinsB g ... 2

~~~~~~~~~~~Trout HatcheryanthJEExemetlCmon.eogaFll18

Bom-18Isan

1.0mm Nytqx Mesh

PV/C Coupling

Sliding Plexi' Door

cii,~~~~~~ O ie h 2 mle hmesw nece inC paisuin PVCc.omm Nytex Mesh Pl coated C able

Sliding Plexiglnts Door

River Flow

Each holding cage consisted of 3 large chambers made of 60-cm lengths of25-cm 0 PVC pipe and 12 smaller chambers made of 12-cm lengths of 10-cm 0I PVC pipe. The 12 smaller chambers were connected in pairs using PVCcoupling&. The chambers were held together with plastic coated cable. Thecages were secured in the river by tethering to a post and attaching a weight

to the front of the unit.

Figure 4.7. Cages Used to Hold Fish for Bioamys Performed in the Chattahoochee RiverBelow Buford Dam, Georgia, November 1980 and February 1981

4-19

SECTION 5.0. RESULTS

4

5.0 RESULTS

5.1 TABULAR RESULTS FOR RACE BIOASSAY

Tables 5.1 through 5.12 summarize mortality and chemical data foreach of the bioassays: Set I (SI) and Set II (SII) flow-through bioas-saye and five groups of static bloassays (Stat 1-Stat 5). Set I bloas-says were run twice (SIl and S12), and four different groups of experi-ments were included in Set II (Sill to SII4). Individual treatments foreach experiment are lettered (e.g., SIIlA through SIIIH). Graphic pre-sentations of mortality data and of selected regressions of mortalityagainst metals or hardness are presented in Figures 5.1 through 5.3.

Each experimental run included treatments testing several differenthypotheses. Because of this presentation of results is much clearerwhen organized by the ideas explored rather than by the temporal se-quence of experiments.

5.2 ZPILIMNETIC CONTROLS

Fry kept in epilimnetic water as non-toxic controls throughout theexperiments showed excelleat survival rates, thus demonstrating that frywere not seriously stressed by handling or by the experimental proce-dures. Of a total of 800 fry used in flow-through experiments (SI andSI1) only one died, and only two of 250 fry died in the static epilim-netic controls.

5.3 TOXICITY VARIATIONS WITH WATER DEPTH

As anticipated, Set I experiments showed that both iron and manga-nese content and toxicity increased with water depth. Therefore, bottomwater was drawn for Set II experiments, because bottom water was mosttoxic.

Toxicity was clear or suggested in those waters with dissolved man-ganese (Mnd) over 500 ppb and dissolved iron (Fed) over 200 ppb.Regression analysis showed that measured differences in Fed betweentreatments in Set I could explain 98 percent of the mortality differ-ences (Figure 5.1). No copper, arsenic, lead, cadmium, chromium, ormercury was detectable in any of the 36 routine chemical samples (TablesA.1 and A.2). EPA's scan for 27 metals detected only aluminum (390ppb), barium (20 ppb), manganese (1,150 ppb), strontium (20 ppb), titan-ium (10 ppb), calcium, magnesium, iron, and sodium (Table 5.13). Dis-solved aluminum was always below detection, but total aluminum reached250 to 390 ppb in bottom water. Although mostly below detection, zincwas inconsistently found at levels up to 32 ppb, but only in one ofthree sets of samples from a single source of stored water from Set I.No organochlorine compounds were detected in any of the samples.

5.4 VARIATIONS IN UNTREATED BOTTOM WATER

During the experiments, the dissolved metals levels of bottom watervaried approximately 50 percent for Mn and 500 percent for Fe, and or- Itality rate also varied in a pattern not clear at the time. The loading

5-1

-. .d(

rate (water flow per gram of fish) differences now help explain thesometimes large toxicity variations between runs. However, loadingrates were the same for all treatments within a single run, and theresults and conclusions of this study are in no way compromised by be-tveen-runs differences in loading rates (see Table 5.14).

With the exception of S114, the 48-hour survival rate decreases asthe flow per gram decreases both as Fed increases from 1,133 ppb to3,288 ppb and as Fed decreases back to 1,940 ppb. The exceptional lowsurvival rate for a light loading rate in 5114 was associated with highmanganese but relatively low Iron. Why lowered Fe levels should make Mnmore toxic is not understood; some completely unrelated variable may beof importance. However, in aeration treatments which lowered Fedwithout changing Mrd, toxicity was increased; and when a column ofactivated carbon removed all Fed but no lnd, toxicity was increased.

5.5 ORCANICS REMOVAL WITROUT EFFECT

Removal of organics by Dowex 11 anion exchange resin (SIIIE) yieldedexactly the same survival rate as no treatment (SII1A). Dowex 11 andHCR-S resins together with carbon (SII1G) produced exactly the same sur-vival rates as HCR-S resin alone.

Since these experiments demonstrated that organics did not influencetoxicity and no organics were detectW chemically (Tables A.1 and A.13),no further experiments were performed with Dowex 11 resin.

5.6 FORM OF EDTA CRITICAL

Addition of 10 ppm Na4 EDTA (SIIlC) left fewer fish alive than intoxic controls (59 percent versus 73 percent at 24 hours). IncreasingNa 4 EDTA to 50 ppm (SII2C) caused an even larger decrease in survivalrate (24 percent versus 50 percent). Neither addition caused any changein any chemical measurement except for a slight increase in alkalinity.A preliminary static test showed that 100 ppm Na4 EDTA also greatlyincreased mortality.

Although Na4 EDTA at 10 ppm or 50 ppm caused mortality increases,Ca 2 EDTA at 100 ppm and 10 ppm (Stat 5 C and D) was completely effec-tive In preventing mortality during the 2-day test.

5.7 )rOLONGOD AULATION MUMS TOXICITY

Aeration was attempted three times in Set I experiments without im-proving survival. In Run 1, 12-hour aeration decreased dissolved ironby 67 percent, yet survival rates were significantly less than in un-treated water (2 percent versus 14 percent at 48 hours).

In Run 3, settling and 5-us filtration was added after 12 hoursaeration, resulting in an 80 percent decrease In dissolved Iron; how-ever, after slightly Improving 24-hour survival, fewer fish remainedalive at 48-96 hours than in untreated bottom water (16 percent versus37 percent at 96 hours). A parallel treatment differed only in that

5-2

hardness was increased to 50 ppm. This added hardness substantiallyimproved survival over that in untreated water.

In Run 4, 6-hour aeration did not decrease dissolved iron levels,and survival rates were statistically identical to toxic controls.

5.8 METALS EMOVAL AN EFFECTIVE TREATMENT

Interpretation of the effects of metals removal by Dowex HCR-Scation exchange resin was complicated by the fact that the resin removednot only Fe and Mn but also hardness (Ca and Mg). Removing hardnessalso increased toxicity, but removing Fe and Mn and leaving or addingback hardness resulted in almost complete survival.

In Run 1, the HCR-S resin alone (SII1F) removed 79 percent of Fedand 69 percent of Mnd, but also 64 percent of hardness, and survivalwas significantly lower than in untreated hypolianetic water (SIIIA) (46percent versus 73 percent at 24 hours).

HCR-S, Dowex 11, and activiated carbon columns together (SHIG) re-moved 97 percent of Fed, 93 percent of lkxd, but 82 percent of hard-ness, and survival rates ware statistically identical to those for HCR-Salone. In Run 2 (SII2G), these three treatment columns were combined ina different order with Dowex 11 last so that any organic leached fromthe other columns would be removed by Dowex 11, but the new arrangementof columns yielded the same result as the first arrangement.

In Run 3, 1CR-S and activated carbon together (S113F) reduced Fedand Mnd below detection but also lowered hardness from 14.5 ppm to 4.5ppm. All fish died within 48 hours compared to 54 percent survival inuntreated water. A parallel treatment (SII3E) differed only in thatCaCl 2 was added to increase hardness back to 12 ppm. In this treat-ment, 94 percent of the fish survived the 96-hour test (versus 37 per-cent in untreated water).

In Run 4, Fed and Mnd were precipitated out by raising the pH to11 (NaOH), aerating, filtering, and adjusting back to pH 6 (HCl). Noneof the fish died in this treatment (SII4C), showing that removing Fedand Mnd removed all toxicity.

Activated carbon alone (SII3H) removed all dissolved Fe but did notdecrease Mud or hardness. Surprisingly, this selective removal of Fesubstantially decreased survival (16 percent versus 54 percent in un-treated water at 48 hours). In a parallel treatment (SI13G) differingonly in that hardness was increased to 19 ppm (versus 13 ppm S113H),survival was the same as in untreated water.

5.9 EARDNESS ADDITION PROTECTS FISH

In all cases, increasing hardness was effective in increasing survi-val of the fry. In Run 4, increasing hardness to 28 ppm (versus 12 ppmwithout treatment) by adding CaCI 2 increased survival rate to 84 per-cent after 96 hours (versus 24 percent). With hardness increased to 38ppm (SII4G) or 63 ppm (S114H), 100 percent of the fry survived the 96-hour tests.

5-3

In the static test also, adding 20 ppm or 100 ppm hardness was com-pletely effective in preventing mortality (Stat I J, K, and L).

An increase to 50 ppm hardness was less effective following 12-houraeration (SII3C), but the improvement in survival rate still was sub-stantial (77 percent versus 37 percent without treatment at 96 hours).

The detrimental effects of removing only iron with activated carbon(S113G and H) were ameliorated by increasing hardness from 13 ppm (R) to19 ppm (G) (7 percent versus 31 percent survival at 96 hours).

Adding hardness increased the survival rate associated with anygiven level of Mn or Fe addition to epilimnetic water. Adding 10 ppm,20 ppm, or 50 ppm hardness to 1,000 ppb Nn++ completely eliminatedthe 90 percent mortality (48 hours) of 1,000 ppb n++ alone (Stat 2B, d, I, J). In Stat 4, the 64-hour survival rate of 4 percent when1,000 ppb MO+ was added was increased to 94 percent if 10 ppm hard-ness was added. Survival in 3,000 ppb Mn+- plus 10 ppm hardness wasequivalent to 1,000 ppb na+ + without added hardness (Stat 4 F andD).

Adding 10 ppm hardness also markedly increased the survival ratesassociated with 2,000 ppb or 4,000 ppb Fe++ or 1,000 ppb Mn++ +2,000 ppb Fe++ (Stat 4 1, J, K, L, N, P).

5.10 EFFECTS OF DILUTING BOTTOM VATER

In Stat 1, 90 percent and 70 percent hypolimnetic water (dilutedwith epilimnetic water) were just as toxic as undiluted water, and 50percent dilution increased survival only to 26 percent (versus 8 percentat 42 hours). A repeat of the test with 50 percent dilution in Stat 2yielded an improved survival rate of 40 percent versus 5 percent instraight hypolimnetic water at 48 hours.

5.11 MANGANESE ADDITIONS TOXIC

In several static tests, manganese was added to epilimnetic water.WhEre chemically measured, concentrations agreed with calculated values,arid manganese toxicities were the same for Mn(N0 3 ) 2 as for kC1 2 .in every case of eight tried, 1,000 ppb was acutely toxic, leaving 10percent or fewer survivors after 48 hours, while 250-320 ppb show littleor no effect. From plotting these and intermediate values (Figure 5.4)and roughly sketching a line, the estimated 48-hour LC5 0 is approxi-mately 650 ppb VF", which is considerably lower than previouslyreported (Lewis, 1976; England, 1978; Lewis et al., 1979).

5.12 IRON ADDITIONS TOXIC

Ferric iron added to epilimnetic water had no effect either alone orin combination with manganese (Stat 1 and 2). Ferrous iron, in con-trast, was toxic at relatively low levels; however, poor correlationbetween calculated and the few measured values makes the exact levelsuncertain. A calculated value of 1,000 ppb Fe (measured 220-240 ppb)left only 22 to 28 percent of fish alive at 48 hours (Stat 4R and 51).A calculated level of 2,000 ppb (measured 260-290 ppb) produced 30 to 35

5-4

percent survival at 48 hours (Stat 41 and 5J). Calculated 500 ppbFe++ and measured 130 ppb Fed gave 62 percent survival at 48 hours(Stat 5H).

From these few data, straight line interpolation produces a 48-hourLC5 0 of 660 ppb iron (Fe+ + ) based on calculated values or 160ppb iron (Fed) based on values measured at the end of the exposureperiods. These values compare with the 480 ppb Fe 96-hour LC5 0 re-ported for yearling brook trout at pH 6.0 (Decker and Menendez, 1974).The 480 ppb Fe was an average of daily measurements in a flow-throughsystem, and this value represents actual exposure level over 96 hours.On the other hand, in the 48-hour static tests reported herein, concen-tration probably started at the calculated value and decreased at anunknown rate to the concentration measured only at the end of the48-hour test. Given the differences in duration of test, basis for nom-inal concentrations, trout species and age, results are similar.

The poor correlation between calculated iron levels and those mea-sured at the end of the bioassay most likely is due to adsorption of Feto aquarium walls or possibly to uptake by fish. Such losses of testsubstances are common 'in bioassays and are a major reason that flow-throughs are preferred over statics. While this discrepancy gives awide range for LC5 0 values based on statics, it does not affect anyconclusions based on the flow-throughs because Fed was measured dailywith consistent results and because constant replacement would preventiron adsorption to aquarium walls from causing a decrease in Fe concen-tration.

5.13 HA EGA5SE AND IRON TOXICITIES ADDITIVE

At moderate concentrations, Fe and Mn toxicities appear to be simplyadditive. Using calculated concentrations, 500 ppb Mn+ + in 24 hoursproduced 34 percent mortality (Stat- 5F), 1,000 ppb Fe++ 44 percentmortality (Stat 51), and the two combined caused 70 percent mortality(Stat 5L), in agreement with predicted 78 percent mortality for additiveeffects. In Stat 4, 500 ppb Mn++ produced 13 percent mortality(interpolated between B and C) in 18 hours; 1,000 ppb Fe + + 26 per-cent (H); for a projected combined mortality of 39 percent compared toan observed 36 percent (M). In Stat 5, 250 ppb Mn+ + caused no mor-tality (24 hours, 5E) and 500 ppb Fe + + 18 percent (5H), but the twotogether killed 36 percent of the fry (5K), in this case suggestingmore-than-additive effects for low concentrations.

At high Fe concentrations results were inconsistent, with calculated4,000 ppb Fe+ + not much different from 1,000 or 2,000 ppb (Station 4H, I, J), and calculated 1,000 ppb and 2,000 ppb Fe+ + measured sim-ilarly at 220 ppb and 260 ppb, respectively. because the results didnot give a consistent pattern, conclusions about interaction of 1k andFe at higher concentrations cannot be drawn.

5.14 SUBIITRAL UFCTS

behavioral changes and sublethal effects were similar for all toxictreatments, and the severity could be qualitatively correlated to thedegree of toxicity.

5-5

Frequently, fish were disoriented and poorly coordinated, swimmingslr ..y on the side or back with occasional bursts of rapid swimming.Und, observation, fish became nervous and excitable, often swimmingquickly and in a frenzied manner around the aquaria.

In the more toxic treatments, dip netting out the dead fish or put-ting DO or conductivity probes in the aquaria made the fish nervous andcaused some to go into shock and immediately die.

Numerous fish showed trailing feces and some had mucus strandstrailing from the gills; darkened color was common. In several treat-ments, the fish fed poorly. Dead fish often were curved in a C-shape.

5.15 ROUTINE PHYSICAL NONITORING

Conductivity was usually 30 to 35 umhos for bottom water and 20 to25 unhos for surface water. Hardness additions were apparent in theconductivity records.

Dissolved oxygen levels were always above 6 ppm except for briefaeration problems during S11.

Temperature was held near 12 to 13C for SIll, S112, and SI3, but adaily temperature range of 5C', sometimes more, could not be avoidedduring SII, S12, and S114 because of heat gain or loss in the waterlines running into the laboratory from outside water storage tanks.

pH mostly remained between 6.0 and 6.5, but pH as low as 5.1 wasrecorded. The low readings are likely erroneously low, since pH probeswere later found to have a short life and to give erratic and low read-ings as they failed, even though they continued to calibrate againststandards. Because the meters continued to calibrate, not giving clearsign as to when gradual failure commenced, no corrections were made forany of the pH values in Appendix B.

Daily temperature, DO, pH, and conductivity ranges for each treat-

ment are presented in Appendix B.

5.16 HNISTRY AND QUALITY ODY01 L

Throughout the study period, the levels of manganese and especiallyiron increased steadily (Figure 5.5). Initially, in early October, al-most all iron was particulate iron, but the percentage of dissolved ironrose to 65 in late November and to 88 in early December. Measurementsafter that were from water stored in the 11,500-liter tank, and oxida-tion and precipation during storage may explain why total and dissolvediron decreased while manganese did not. Nearly all manganese was indissolved form throughout the study.

Comparison with the few data reported for fall 1979 by ESE (1981)suggests that iron was five times as high in 1980 and manganese threetimes as high as in 1979 (Table 5.15), with dissolved to total ratiossimilar for both metals. However, hatchery records, which are sore

5-6

frequent for both years, show that manganese and dissolved iron (Fe+ + )

in the river were hgher In 1979 than In 1980 (Figure 3.4).

Chemical monitoring results for all runs of both sets of flow-through bloassays are presented in Appendix A. Interlaboratory compari-sons of quality control samples (also In Appendix A) shows that qree-ment between the two laboratories was usually closer than the maximumacceptable difference of + 25 percent (Dr. J.T. McClave, personal com-munication). Of the numerous spiked metals samples, the only recoveriesoutside + 10 percent were some low copper recoveries (22 to 77 percent),one hlgh--zinc (138 percent), and one high dissolved manganese (120 per-cent).

5.17 WTALS IN TROUT LIVERS

Of the 24 metals measured in livers of rainbow trout from BufordHatchery (Table 5.16) and from Walhalla National Fish Hatchery (Table5.17), 16 were below detection in all samples. Of the 8 detectablemetals, Al, Ca, Cu, It, Mn, and Zn all were significantly higher inBuford trout than in Walhalla trout (Table 5.18). Na averaged 42 per-cent higher at Buford, but variance was so high that this difference wasnot significant. Fe contents were the same in both sets of liver sam-ples.

Not all of these results can be explained based on known water chem-istry for Buord Hatchery (Appendix A) and the following analyses ofwater flowing into Walhalla National Fish Hatchery on November 1, 1977(the most recent analyses):

Ca 21 ppm Ft 0.33 ppm14 10 ppm Mn <0.01 ppm

Hardness 62 ppm Al 0.8 ppmAlkalinity 47 ppm pH 6.9

Ca, It, and Al appear to be in substantially higher concentration inwater at Walhalla than at Buford, yet trout from Walhalla show lowerliver Ca, 1%, and Al levels than do rainbow trout from Buford. Fe at0.33 ppm is relatively high at Walhalla and possibly comparable to Feconcentrations averaged from summer lows through November highs atBuford, so that equal liver Fe contents for the two hatcheries perhapsrequires no special explanation.

Mn content is significantly higher in livers from Buford trout, inaccordance with probable water chemistry differences. However, concern-ing higher Cu and Zn in Buford trout, the fact that Ca, ft, and Al arehigher in water at Valhalla yet lower in livers from Walhalla troutmakes accepting high Cu and Zn in Buford trout livers as evidence of ex-posure to high levels of Cu or Zn difficult. Other factors may explainthe differences in Cu and Zn levels between the two hatcheries as dis-cussed below.

The relationship between environmental levels of individual metalsand tissue levels of those metals has been insufficiently studied; therelationship is a complex one which can be Influenced by factors such as

5-7

age and reproductive stage of the animals, foods, and environmental lev-els of other elements and compounds. Age, strain, source, reproductivestage of trout, and food were controlled variables in choosing thesource of trout to compare with Buford trout, but controlling all envir-onmental variables was impossible. Increased hardness is known to de-crease toxicity of metals (Chakoumakos et al., 1979; Pagenkopf et al.,1974; Lloyd, 1960); perhaps hardness decreased availability of themetals at Walhalla and therefore decreased the tissue uptake of metals.

Alternatively, other types of research suggest that the known highMn levels at Buford might explain the elevated tissue Cu levels. Lambs

fed high levels of supplemental Mn showed significantly higher levels ofCu in livers than did controls (Watson et al., 1973). Similarly, ratsgiven manganese supplemented drinking water showed increased copper lev-els in brain tissue (Singh et al., 1979). While these data are for mam-mals rather than for fish, they do show that higher tissue concentra-tions of Cu (or other metal) need not indicate higher environmental ordietary levels of Cu (or other metal).

In view of the probable influence of known high Mn at Buford Hatch-ery on tissue metal levels and of cor.sistent failure to detect Cu or Znin river and reservoir water of demonstrated high toxicity, the factthat Cu and Zn are more concentrated in livers of Buford trout than inlivers of Walhalla trout cannot be accepted as evidence of Cu or Zn tox-icity at Buford hatchery.

5.e18 FEE EIsTOPATEOLOGICAL EFFECTS

Fish used in S11 (5 days exposure), S12 (4 days), and SIll (7 days)were examined. None of the 90 fish examined showed any abnormality inthe gills, liver, kidney, gut, central nervous system, or skin. Pathol-ogy was limited to proliferation of the pigmented layers of the eye,some separation of the retina, and somatic myodegeneration (Figure 5.6).

The lack of effects, particularly in the gills, is unexpected be-cause previous examinations of rainbow trout from the hatchery and fromthe river have reported gill, liver, and kidney pathology, with the sug-gestion that gill damage was the primary toxic action with the otherpathology a possible secondary result of poor oxygen exchange across the

gills (Noell and Oglesby, 1977; Grizzle, 1981).

In SIl, one fish exposed to water from 35 meters, 2 fish from 28meters, 3 from 23 meters, and none from 6 meters (10 examined from eachtreatment) showed a proliferation and thickening of cells in the pig-mented layers of the eye and showed small foci of degenerating musclebundles. In S12, the same symptoms were evident in one fish from 35meters of water, 3 from 28 meters, none from 23 meters, and one from 6meters.

For SIll, only 10 fish were examined, 5 exposed to untreated bottomwater (A) and 5 kept in epilimnetic water (B). One fish from B showedthe same early indications of pathology seen in SI. In A, three fishshowed only beginning myodegenerstion, but the other two showed exten-sive muscle degeneration with replacement by connective tissue.

5-8

The musculnr degeneration first would suggest inadequate diet. How-

ever, the first histological effects of inadequate diet should be seenin erosion of the intestinal columnar epithelium, yet the gut appearednormal even in those fish exhibiting the most advanced myodegeneration.

Myodegeneration with no suggestion of pathology in liver, kidney, orgut is unexpected because the internal organs usually are the first toshow histopathological effects. Neither Dr. Blake nor Dr. Yevitch (EPANarragansett) could recall seeing this phenomenon before nor could theyexplain it.

A similar muscular dystrophy without other obvious histopathologyhas been reported for Atlantic salmon fed a diet deficient in Vitamin Eand selenium (Poston et al., 1976), but such a dietary deficiency is un-likely in trout fed regular hatchery food.

5.19 RIVERINE BIOASSAYS

River water was acutely toxic to rainbow trout in November but wasnon-toxic in February after Lake Sidney Lanier reservoir had destratifi-ed. No rainbow trout died in February riverine bioassays. Bluegills

survived both fall and winter bioassays except for asphyxiation in tail-

waters (Station 3) from the dam in the fall.

Trout fry showed 24 percent survival at the last station, 13.5 miles

(22 ki) below the dam, but significantly lover (0 to 6 percent) survivalat other stations (Table 5.19, Figure 5.7). Yearling trout asphyxiatedwithin an hour immediately below the dam (Station 3) but showed 73 to 97

percent survival at other river locations except the sluice at the dam,where only 17 percent survived.

The increase in survival at Station 8, 13.5 miles (22 kia) below

Buford Dam in November, is statistically significant for fry but notclearly significant for yearlings (Table 5.19). Yearlings showed highsurvival rates. Given freedom of movement to seek most favorable condi-

tions, stocked yearlings might live through the fall. Survival is evi-denced also by the facts that live trout were seen in the river during

the work in fall and that fishermen catch rainbow trout between the damand the hatchery.

Fry mortality was still substantial 13.5 miles (22 k) below thedam, and it is unlikely that fry could survive in this section of theriver. Fe and Mn in the Chattahoochee River were measured at the hatch-

ery (2.5 km below the dam) and at McGinnis Bridge (13.6 km below thedam), 5 miles (8 ki) upstream from the last riverine station. Mean low

flow Fed was 1,222 ppb and Mnd 752 ppb at the hatchery. At McGinnisBridge, the mean low flow Fed was 267 ppb and Mnd 595 ppb, valuesnear the 48-hour LC50 estimated from laboratory static bioassays.

Projecting the maximum rate of decrease between the hatchery andMcGinnis Bridge a further 8 ka to Abbotts Bridge (22 km below the dam),estimated levels for Fed would approach zero; estimated levels for

Mnd would fall below the highest 48-hour laboratory "no effect" 320ppb levels for Mnd.

5-9

As expected, iron and manganese levels were low at both high and low

flow periods after lake turnover, and no trout died at any of the sta-tions during the February riverine bioassay (Table C.5).

5.20 LAKE PROFILE DATA

By the time JEA arrived on site in early October, Lake Sidney Lanierhad already been stratified long enough for the hypolimnion to have be-come anoxic (Figure 5.8). In 1978 and 1979 (Figure 5.9), the reservoir

showed some thermal stratification in April, and oxygen in the hypolim-

nion began declining in July and became depleted about early October.

The oxygen profile of December 11, 1980 (Figure 5.8), showed mixingof waters above the slight thermocline at 30-m depth, and by December 21

the reser,,oir was isothermic.

5.21 HYDRAULIC CALCULATIONS

Hydrodynamic aspects of water quality were considered in addressingoverall experimental considerations of this work. Hydraulic calcula-tions were necessary to determine placement of the water intake pumpsfor the experiments. Some indications of potential hydraulic effect onwater quality were also determined by calculations of residence times ofwater in Lake Lanier. Other hydraulic calculations provided informationon the shape of the isotherms in Lake Lanier during low and high flows,and the immediate zone upstream of Buford Dam influenced by withdrawal.

Residence time is defined by the volume of the reservoir divided bythe inflow rate (Fischer et al., 1979). In large, deep reservoirs,residence time is often se;era- years, whereas in some run-of-tbp-riverreservoirs formed by small dams, the residence time may be only a week.A short residence would suggest that water quality may be primarilydetermined by inflow to the reservoir; a long residence time would allow

time for significant effects on water quality from surface or bottominputs or from biological activity. Calculations of residence time weremade for three cases based on the following:

1. Volume at normal pool elevation - 2,368 x 106 m3

(1.92 x 106 acre-ft) (Noell and Oglesby, 1977)

2. Minimum and maximum flows: minimum flow - 550 cfs; maximum flow;- 9,000 cfs (Personal Communication, U.S. Army Corps of Engi-neers, Mobile District, 1982).

3. Average annual flow - 2,064 cfs (25-year average, includingwater year 1981) (Personal Communication, U.S. Army Corps of

Engineers, Mobile District, 1982).

Residence time at minimum flow

Residence Time - volume of reservoir - 4.8 yearsminimum flow rate

5-10

Residence time at maximum flow

Residence Time - volume of reservior . 0.3 yearsmaximum flow rate

Residence time at average annual flow

volume of reservoirResidence Tie - vrg annual flwrt- 1.3 yearsaverage nna flow rate

The residence time calculations suggest that Lake Sidney Lanier isinfluenced, but not dominated, by the run of the river, and the poten-tial for effects on water quality from surface or bottom inputs is high.

Estimates of withdrawal layer thickness were made using appropriateequations from Fischer et al., 1979 and Brooks and Koh, 1969, for out-flow dynamics. The efforts to estimate withdrawal layer thickness werelimited and should not be considered as a full treatment of selectivewithdrawal in an elongated reservoir. The results represent a prelimin-ary look at withdrawal during low and high flow situations with respectto approximate location of isotherms during October 1979 stratification.Results of the calculations were as follows:

1. Estimated withdrawal layer thickness at high flow (9,000 cfs) isapproximately 75 meters using Fischer, et al., 1979, equations,or approximately 44 meters using Brooks and Koh, 1969, equa-tions.

2. Withdrawal layer thickness at low flow (550 cfs) is approximate-ly 24 meters using Fischer et al. equations, or is approximately11 meters using Brooks and ih, 1969, equations.

U.S. EPA (1979) data show the October isotherms were located between15-20 meters from the surface. Similar data were obtained during 1980(see Figure 5.8). The estimated thicknesses of withdrawal layers indi-cate that certainly under high flow conditions, withdrawal would occurfrom both the epilimnetic and the hypolimnetic layers. The calculationsindicate the potential for withdrawal from both layers under low flowconditions as well. These calculations also indicated that over a day,average volume withdrawal will occur up to 914 meters (3,000 feet) up-stream from the dam outlet, and up to 20 meters above the outlet centerline. The influence of withdrawal on the isotherms would increase fromOctober to destratification in December of January, as the thermoclinedescended. A more rigorous analysis would be necessary to define thespecific behavoir of the isotherms under varying flow regimes; however,these calculations presented evidence that the isotherms will definitelytilt during high flow, but will tilt only slightly during low flow.

The final hydraulic consideration focused on the isotherm tilt underhigh flow situations and the necessity for locating the water intakepumps upstream of the zone where this would occur.

5-11

A rough drawdovn zone (where the discharge would be strong enough toovercome buoyancy forces both in the hypolimnion and at the thermocline)was determined to exist up to 152 meters (500 feet) from the dam. Thisestimate was compared to U.S. EPA (1979) Isotherms, which generally con-firmed this calculation. Subsequently, the vater intake assembly forthe experiments was located between 150-180 meters from the dam.

5-12 .

o Uho

w4

C45 V.w aDA

w41 '.

"44

9: C

o1tO0 .0 go ~

- .0 ' bo.Q U 4 0

410 0 a 441 4141 C

4J. W ) to 0 4)I0. C-0 v b.~

4.1 0 0U

0 b"4. 0 0 t-. ~ ba z1. CI. f 04 04

0 "4 co A 416

I A

w :1 0 '%o - jU Q446w N

c le ov C4 to 4) W

4144 1 41 Aigo0 to04

-4 4) w4 W

000 -f Ai .e 0.

"4i "4N v0 wVU4 .1P ' 4 bo 0". h 5.e4 a

-P4 "4 41 41 m-

14 0 IA ~ ~ ~ 0 aa -

ad o A 4

w4 0~ q414

*0 A i vq a. r41UI%4 U) 04 Ito44a141 Ai a 4 '

if 04

41 Q41O.*~ &j

to4

547

co -4 -d"

--- 1

Aj a

0

w 4

0 4)-

06 cn

0 wr 01

00) - a44 ~ ~ ~ ~ ~ CL.a-T'0.) 0 r 0 0 aD X ~ a,

A0 ua0 - 41 4

(-. a, C .Q.w 1OA 04 41 0)4 0 4

W0 0. >~r 4 b

0 0 m -,w 0)0 00 5

0.' CD r- W4 .0 .00

aa I a, j 4

0 44

000

CD 05554.40 60 14 C) Ln

0 C4 41 %4 4,4 %6004c, -ra % 0 w v

-j 4) 04 00 14 w 0) 61 .0 UU IdIVv )

I 0 05 0 a U 4.4

54't &d 0,) 1t o0&a~aa

;p~0 51 00 a,- $

4 = 0 0 %D W A o-% 4 . u-4"0.50 % 0n .0 W. 0.4 W V

a.) 0 0 a a 0%'a 1- 4A ira 0q 0W a,0). 0

IV S m5 4 44 4'50)oo0odov ) 0 . ONI

Ol4.4. tic ~ ,-4

'a144 1'' .444 U(>00 0 a

S4.4 0) * 5 % ' 41 4A 44 0000c

4444 0 wa 4100v 4

.0a %.05 6 I1"

5-4

4)~

0

0 1.

4'-

> V44A

0 61 en Ot' C" v n in 0 v+ 0 00w

A0 0. cooen 0 V0 >0 0..w0iI -) - .0 " 4 r.A 0

0% 0 p. 0 fV~ A if@0% M O C 00 -4- 0 cc go

o 6Q %D %~U~ 00 aDL) ot

*~W 0 iSV c. Id bOb C

(n A N O 0 r-. en+ P5 to 6wt00. In a 4 %DC 0% w 1+ 4 0"

'a~~~4 v.. 0 - I. i 0 .~ C ~ N 0 bOCJ U'0~ to

wO 0- V4~ Li 0 0D

'A i 0 04 tor vQ.cJ-A40 0 WW04Lii W44

*0~~ 00. i

k-4 w fa a 4 0 0 w04 444. 00 0 vCW

6J4 30 "4 0-4 0 i5

r.4p Vi'aJ

00 20 '0 0hCP . *

> 41

q*A41 u~i Al'o.6 .

2 -4 ODf M nV410 0 "1 1

C* I L, JC4 $4 4) f4 -4 01-15

Table 5.4. Sumary of Results of Set II Run 2 Bioassays, November 23 throughNovember 24, 1980, Verifying Some Results of Set II Run 1, LakeSidney Lanier, Georgia

No. Survivors Fet Fed Mdd Hardness AlkTreatment 24 hours (ppb) (ppb) (ppb) (ppm) (ppm)

A 80 3,200 1,940 1,000 13 15B 144 50 <50 <5 9 9C 38 3,200 2,040 1,025 11 24F 159 <50 <50 <5 13 8C 42 400 100 110 5 8

Notes:1. Treatments:

A Hypolimnetic waterB Epilimnetic waterC A+ 50 ppm Na4 EDTAF B + 5-um filter + HCR-S resinG A + 5-an filter + HCR-S + Dowex 11 + 5-ua filter

2. No dissolved and total were always within measurement error of beingequal (complete chemical monitoring data in Appendix A)

3. Statistical grouping of treatments listed in order from highest to lowestmean survival (underlining groups together those treatments not statis-tically different from each other): 24 hr B F A G C

4. Fet - total iron; Fed - dissolved iron; Mnd - dissolved manganese; Alk- alkalinity

5. Treatment B started with 144 fish; treatments A, C, F, and C with 160 fish.

5-16

S.' 4.M

QD61C%%a') ~IOU

00 3CA

44 4.-

Aja -u. -% 0 Li(Immmk

x low 0 ItU0t r en0 V-. >.fl4 0 O 0 ' 0 00 O "4

000 I G 0 aD Go~ 0 04 0 C0cc O1.%D- . - b** ci ''

C* w ( 0 0) 0

01 -1 bo W 0-LM '0. 0't V.% W m 0 .00 m0 b

41 60. 0%i0 %D~U " 4 00 %DU +0i 4

9 m N n %D m m. 0 to~

01 Wl

.-I A00$.-S 1. 4.0

000V4 "04. Colo r 00 A q 41Z0 ) .

"4 9L v. 0 .0000 u- -4- .4 tJw- ) 0.4 .00 + .0 0

404 16 A0

W 0 0

04 0 Vw6jJ 0 0 4 v-a4)Aj

0- Ad. .0 x00' w4'0I

P.. 4 a

*4 C-4 4 +-++

w o 0 0-4)10-4~ &a+ -4 w * 6 0. 61 1.

.0 4m

in 0 -- 1--4 -

Table 5.4. Sumary of Results of Set II Run 2 Bloassays, November 23 throughNovember 24, 1980, Verifying Some Results of Set II Run 1, LakeSidney Lanier, Georgia

No. Survivors Fet Fed ld Hardnesr. AlkTreatment 24 hours (ppb) (ppb) (ppb) (ppm) (ppm)

A 80 3,200 1,940 1,000 13 15B 144 50 <50 <5 9 9C 38 3,200 2,040 1,025 11 24F 159 <50 <50 <5 13 8G 42 400 100 110 5 8

Notes:1. Treatments:

A Hypolimnetic waterB Epilimnetic waterC A+ 50 ppm Na4 EDTAF B + 5-uim filter + HCR-S resinG A + 5-nm filter + HCR-S + Dowex 11 + 5-ua filter

2. Mn dissolved and total were always within measurement error of beingequal (complete chemical monitoring data in Appendix A)

3. Statistical grouping of treatments listed in order from highest to lowestmean survival (underlining groups together those treatments not statis-tically different from each other): 24 hr B F A G C

4. Fet - total iron; Fed - dissolved iron; Mnd - dissolved manganese; Alk- alkalinity

5. Treatment B started with 144 fish; treatments A, C, F, and G with 160 fish.

5-16

I-S

1.4 GoU4V)-4

.0- . c b lv1 2 0 r4.

ON ow0 0 0 0A~~~ ~ ~ GoI "% l nW u

I. IV 1.a1. 1

-H4 A in U, Na Vo* *4 04 M -4 WW ,s

00.-

0 0 0 x A0 0L cc H .

0 +. . . 4 0

In M 4 C % L ,-4 10 r40

0. 4 41 I l

1.' 0.0 044

cc 0 C: c 0i - 4)'8a

%4~' 1.' 0 -F .' I wg-J CP d00 r. 0W00 s 0 + 0 + 0 goLp 0 0

41 0 %C 4n 4 C&A W) "4N.-l 10 -41 W4 4 to. 00a4# 41ZIC ~ - 14 4 -4 a0j 4 . 4j4 4 )A a a4 o

>.. 0 4 -4 . 4 w 4 04+0 1' a 0 4

00 v 4) 4 + IA+ U0" 41.'

o1c% a W ~ l 'l% '4 4 0 04U- "0 -4.-4 -4 VV 4) * 4V 00 PC 1 CA

ow. ad4 -4 0 40r

wA I 41'm 0+ 0 + 00 P. .0 C.lf C.(

ow 1. 0 '0a N If"- 4- 1-~ C- 1.4 & v0 aV 0jai004 bN -- F-4 0 W~4 41 0 &j V-14 0 4 g~ W4 r a 0

aU- r A -04 + +4' -4 Q..S.h : W

S.. hiU 0 0 0 .44.i 41V~~~s .0 U . 4 0 0

I ~ Q.4 0 0 0 0 14. .

5-17 ~ ~ +~ U~00 0 .

coi

40

.0A C4 r4 c - Go en .04 CV0 66

in 1. w. 0

iO Li'- c" 4 4 ~0 -

m LA

r--4

-0 inO OOU~ O 0u 0 O S C

U Ai 0r-4 6. "

nU1i nr -4 s 0 4)

0 'e C 0> 0. *a

r* 44 0 4cr k a 0

go I'I o U $4$ S j

".4

41 00% UW 0 0 w 366 V0 ? 4M )wa

4) %4~ oo'i nO u0 '.v14 w 4 s al a"

-4W u. -4 v44- 9. 0 '~41

.0 k 4.14e 0 ,- - A 4 4.816~~~~V 0410 0 0 6 ~ ~ -

1-44V- OQ 664 KN 60.ow cc 1OAA O ~ ~0 0 0 aU0 0 &A I0

U4O Go-0% tn.' No 6 o '-i40 )L - VAi0 . G q%0 L 0 CID r4.-44 ,-e s%' N1 4NN -4 c 0.44ra

CAW6ds c.00 L 0 0Mo..4 so4 4 41 so cc, 0 .44IVo4 .01 C 0 C C C 0 4.1o s

00 0 0.lIL o 00.0 u 0 .4 LP 1- 4

inD.4%'0% a0 44 ip .0 ... 0 44 W'49. 04 - 4 00C n 4 0 w 6 1 0

0o *41. .. 5w I &,I +6604"mI

IAA

5-18

Table 5.7. Summary of Results for Static Bioassay 1, December 12 throughDecember 16, 1980, Lake Sidney Lanier, Georgia

Number Surviving Out of 50

Treatment 18 hours 42 hours 96 hours

A ypolimnetic control 50 4 *

B Epilimnetic control 50 50 49

C 90 percent A + 10 percent B 50 1 *

D 70 percent A+ 30 percent B 50 4 *

E 50 percent A + 50 percent B 50 13 *

F B + 1,000 ppb ?fnt++500 ppb Fe++" 49 3 *

G B + 1,000 ppb Mn+++l,00O ppb Fe++ + 48 8 *

H B + 1,000 ppb Mn+++2,000 ppb Fe+++ 49 2 *

1 B + 2,000 ppb Fe+++ 50 50 50

J A + 20 ppm Hdn 50 50 50

K A + 100 ppm Hdn 50 50 49

L A+ 70 ppm Hdn + 30 ppm ?% Hdn 50 50 50

M A + Ca(O) 2 at 50 ppm Hdn, aerate 50 50 501 hour, filter 5-u, + H2SO4 to pH 6

N A + NaOH to pH 8 50 1 *

Notes:1. Hdn - hardness added as CaC12 (or MgCl2) but expressed as ppm

CaCO3 -2. Mn+ + added as Mn(N0 3 ) 2 .3. Fe + + added as FeCl 3 , which formed a floc.4. * - treatment discontinued because of complete or almost complete

mortality.

5-19

Table 5.8. Summary of Results of Static Bioassay 2, December 14 throughDecember 16, 1980, Lake Sidney Lanier, Georgia

Number surviving out of 10 per replicate24 hours 48 hours

Treatment 1 2 3 1 2 3

A 2,000 ppb Mn+ + 9 5 1 0 0 0

B 1,000 ppb N ++ 7 6 6 0 2 1

C B + 500 ppb Pe+++ 8 7 7 1 1 0

D 500 ppb Mn + 10 10 9 8 6 8

E D + 500 ppb Fe.. 10 9 10 7 7 6

F D + 1,000 ppb Fe+ + + 10 9 10 8 7 9

G D + 2,000 ppb Fe+ 8 10 10 3 3 7

H B + 10 ppm Hdn 10 10 10 10 10 10

I B + 20 pm Hdn 10 10 10 10 10 10

J B + 50 ppm Bdn 10 10 10 10 10 10

K Hypolimnetic control 9 9 8 1 0 0

L Hypoltmnetic control 9 9 8 2 0 0

H 50 percent Hypolimnetic + 9 10 10 1 4 650 percent Epilimnetic

N 50 percent Hypoli netic + 10 9 10 5 5 350 percent 1pilimnetic

Note:

1. All metal additions were to epilimnetic water. Mn++ added asHn(NO3)2 , Fe

+ ++ as FeC1 3. Run was concurrent with Static 1control (no orta-lities). All levels are calculated; none was measured.Extensive floc formed on bottoms of aquaria where Fe++ + was added.

2. Hdn - hardness added as CaBC 2 but expressed as CaCO3 .

5-20

Table 5.9. Stmmary of Results of Static Bioassay 3, December 19 throughDecember 22, 1980, Lake Sidney Lanier, Georgia

Number Surviving Out of 50

Treatment* 24 hours 48 hours 72 hours

Static 3a December 18-19, 1980

A Epilimnetic Control 50(3a was run

B A + 750 ppb Mn+ + 37 for 24 hr only)

C B + 1,000 ppb Fe++ 17

D B + 3,000 ppb Fe++ 10

Static 3b December 19-22, 1980

A Epilimnetic control 50 50 50

B A + 1,000 ppb Mn++ 40 5 2

C A + 500 ppb Mo++ ppb Fe+ + 44 21 14

D A + 1,OOO ppb Fe++ 47 37 35

*Notes:

1. Mn++ added as MmC12.2. Fe+ + added as FeC1 2 , no floc formed.3. Mn+ + Fe+ + values all calculated only. When measured (Tables

5.10 and 5.11) Mnt levels were near the calculated values of n++

addition, but Fed levels were substantially lower than calculatedvalues for Fe++ addition.

4. Fed = dissolved iron; Mnt - total manganese

5-21

alla

0

ao & 'T'T en 'T P 4 en~ 'T 'T fn t000N

bIo ~ w.

0 03

htoen 0ada

0 0 .460 0 0 0 s

0% C4 o 0-40 . - Sr c ~ N * ''C- P C10 CN 0

0

.0 %.,

C) W)0 0 00 . M0 0

0o Al ON C" C% V4P 0 '7% P % % 0U 0

L0 -444-

41 03 m 03.0 to swQ

bo 0.r w0. -a ~ C-4 -4a--

ui ' 0 It to

0 03,

03 * 0 Ub 0AA .0 '741+4 4CL hm m L M *

6.. a.) CL 0 6 .1 L ."0 C0. I* v03 11

0 ai 00 0 800 0 0c 0 0 0 0 0 I j4

-4 4 wi-0

0 60o b3Aw 0 xi maw A_: 4C4 03;03

5-22.

r7

L

Table 5.11. Summary of Results of Static Bioassay 5, December 27 throughDecember 29, 1980, Lake Sidney Lanier, Georgia

Treatment Measured Number Surviving Out of 50

Conc. 24 hours 48 hours

A Hypolinmetic Water - 37 7

B Epilimnetic Water - 50 49

C A + I00 ppm Ca EDTA - 50 50

D A + 10 ppm Ca EDTA - 50 50

E B + 250 ppb Mn++ 320 ppb tbd 50 49

F B + 500 ppb Mn++ 660 ppb Ma d 33 3

G B + 1,000 ppb Mn++ 1,020 ppb Mnd 24 1

H B + 500 ppb Fe++ 130 ppb Fed 41 31

I B + 1,000 ppb Fe++ 240 ppb Fed 28 11

J B + 2,000 ppb Fe++ 290 ppb Fed 33 15

K E + H 360 Mkd + 140 Fed 32 11

L F + I 580 Mnd + 240 Fed 15 2

Notes:1. All measurements on water from end of test, filtered 0.4-Um.2. Mn++ added as MkC1 2 , Fe++ as FeC1 2 .3. Had - dissolved manganese; Fed - dissolved iron.4. Fe, when added was added as Fe++, when measured was measured as

Fed-

5-23

inI mm l

44ao

V44no

09 0 Go 0 a

41

0 ~V4

41 4) V V.

0~ 41

0

U 0 44&j Q 0. 0- ~a. .0 0 8- P 0 8A 0 IA '0 % A0 y-I co

0

m m

N -At1 Ln 0 in

05-2

Table 5.13. Results of Metals Scan of Water from 23-Meter and 35-MeterReservoir Depths Used In Set I Run 2 ioassays, November 7-10,1980, Lake Sidney Lanier. Analysts by EPA, Athens, Georgia

Element Concentration

35ma 23m

Silver Ag <10 ppb <10 ppbAluminum Al 390 <100Arsenic As <45 <45Barium Ba 20 <20Beryllium Be <10 (10

Cadmium Cd <10 ppb <10 ppbCobalt CO (20 <20Chromium Cr (10 <10Copper Cu <110 (10Manganese ma 1,150 160

Molybdenum MD <20 ppb <20 ppbNIckle Ni <35 <35Lead Pb <40 <40Antimony Sb <25 <25Selenium Se <40 (40

Tin Sn <60 ppb <60 ppbStrontium Sr 20 16Tellurium Te <40 (40Titanium Ti 10 <10Thallium TI <100 <100

Vanadium V <10 ppb (10 ppbYttrium Y <10 <10Zinc Zn <10 <10

Calcium Ca 2.7 ppm 2.2 ppmMagnesium )~1.1 0.9Iron Fe 3.3 0.2Sodium Ha 1.4 1.4

5-25

Table 5.14. Loading Rate (Water Flow Per Gras Fish) Differences Related toToxicity Variations In Untreated flypolletic Water (33-MeterDepth) from Lake Sidney Lanier, Georgia, Fall 1980

liter/ gram/ 48-hourRun hour/gram fish Percent Survival Fea(ppb) Mnd(ppb) Mnt/Fea

S114 0.618 0.117 56 535 1,310 2.4

S11 0.439 0.157 96 613 833 1.4

S12 0.307 0.225 81 1,133 1,010 0.9

S11 0.203 0.342 54 3,288 1,178 0.4

SMl 0.157 0.440 14 2,690 1,090 0.4

S112 <0.157 >0.440 projected 0 1,940 1,000 0.5

S114 -Set II flow-through bioassay, Run 4S11 Set I flow-through bioassay, Run 1Fed -dissolved IronMnd dissolved manganese

5-26

as

-C4

* D %VN -4 Go0 L0 O0 0%, .- )

as%0 C4 Go0 4

.0 a O

44 t* 0 1 O 0 r-

heh

1-4 r400a% 00 ~ - Y S

0OD 4J4 -4 >A1. .

Q % 0 00

0~ tn coo .u1- 0 0C4 0' en-~ %C 00 04 0

C% 0%0% -I CD C4 0to 0%0% Ln ,P C0 co 4 0

'-1 GoP s 1he h

&40 -4 .0 %V. **.4,- *a

*0 V4 U 0 4 -4-4 0 4 Cf I V4

0

t4 N4 tnt~ C-4

r,4

M C4 a a a a4

0 0J OD O-* 0%-44s h fo

Ch Go04G-w Go 000 000 0C c ,

U A C4 I-. - U & - 00 0 a " A --40 oNM a %. a a 5Ccc 400 go C44

V- -44 C 4C"4 a1

5-27

Table 5.16. Analyses of Livers of One-Year Old Wytheville Strain Rainbov TroutTaken from Buford Trout Hatchery, Georgia, November 11, 1980

Reelicate No

Element 13 2B 3B 4B 5B 6B 7B 8B 9B 1OB

Silver (Ag) <.5 <.5 (1.8 <.4Arsenic (As) <1.1 <1.2 <4.5 <1Barium (Ba) <1.1 <1.2 <4.5 <1 see Note 4Beryllium (Be) <.5 <.5 (1.8 (.4Cadmium (Cd) <.5 <.5 <1.8 <.4Cobalt (Co) <1.1 <1.2 (4.5 <1Chromium (Cr) <.5 <.5 <1.8 <.5Copper (Cu) 84 108 174 65 57 73 65 59 116 85Molybdenum (No) <1.1 <1.2 <4.5 <1Nickel (NI) (1.1 <1.2 <4.5 <1Lead (Pb) <1.1 <1.2 (4.5 <1Antimony (Sb) <1.1 <1.2 (4.5 <1Selenium (Se) (1.8 <2 <7.1 <1.7 see Note 4Strontium (Sr) <.5 <.5 <1.8 <.4Thallium (Tl) <.5 <.5 <18 <4Vanadium (V) <.5 <.5 <1.8 <.4Yttrium (Y) <.5 <.5 (1.8 <.4Zinc (Zn) 27 29 37.6 20 22 21 13 16 30 23

Calcium (Ca) 54 60 - 48 90 - 34 39 50 54

Magnesium (Mg) 194 225 250 131 182 140 88 119 214 174Aluminum (Al) 14 14 - 8.7 9 - 4.5 7.8 8.4 12Iron (Fe) 27 30 - 35 36 - 19 23 42 31Manganese (Mn) 2.2 3.1 - 2.4 2.7 - 1.3 2 3 2.6Sodium (Na) 811 950 - 419 436 - 316 390 924 814

Note:

1. All values ppm (ug/g) wet weight.2. Each analysis was performed on a pooled sample of three trout livers.3. - no analysis4. Blank - EPA reported as below detection, giving only the first four columns

to show approximate detection limits.5. Replicate code, e.g., replicate no. 18 - pooled sample number one fram Buford

Trout Hatchery.

5-28

....................................

Table 5.17. Analyses of Livers of One-Year Old Wytheville Strain Rainbow TroutTaken from Walhalla National fish Hatchery on November 20, 1980

Replicate No.

Metal 1W 2W 3W 4W 5W 6W 7W 8W 9W -lW

Silver (Ag) <.2Arsenic (As) <.6Barium (Ba) <.6 see Note 4Beryllium (Be) <.2Cadmium (Cd) <.2Cobalt (Co) <.6Chromium (Cr) <2.3Copper (Cu) 15 23 30 20 20. 25 15 21 14 20Mnlybdenum (Mo) <.6Nickel (Ni) <.6Lead (Pb) <.6 2Antimony (Sb) <.6Selenium (Se) <.9Strontium (Sr) <.2 see Note 4Thallium (TI) <2.3Vanadium (V) <.2Yttrium (Y) <.2Zinc (Zn) 12 11 12 9.5 16 i2 8.6 7.4 11 12

Calcium (Ca) 49 27 35 36 - 49 28 44 43 36Magnesium (Mg) 107 91 98 84 100 98 75 58 95 105Aluminum (Al) 3.6 7.4 6.3 2.1 13 5.5 <4 <5.5 7.7 6.0Iron (Fe) 30 44 41 30 29 47 20 33 22 46Manganese (Mn) <1.2 <1.2 <1.6 <1 - <1.1 <1 <1.4 <1.3 <1Sodium (Na) 512 420 442 400 494 558 344 275 461 545

Notes:

1. All values ppm (ug/g) met weight.2. Each analysis was performed on a pooled sample of three trout livers.3. - - no analysis4. Blank - EPA reported as below detection, giving only the first four columns

to show approximate detection limits.

5. Replicate code, e.g., replicate 1W pooled sample number one from Walhalla

National Fish Hatchery.

5-29

0 9 0.

4.1

41 s4

0 0i 0

0 0 1 06l u

1- 0404

40 3 4. 1

84 6 4 6

00 0 a.-O 00

- 44 '4

-4 -4 00 0 0 0 0 04

A.% 4 aj 1036 C4r -4 A4

14-4 ) 48 L00 0 0 4

0E' 04b 0j g op00

0 0D 0 01010 411 to4 u."4 4

r. 6.

04 q bC "4 6-4.0301.1

u1 a'

410 0-4bO

0 0 4 bO 4 Av 604 0% V4m

U44

a 4. .0 0 4

sa v v 0 a

U-4. a.f 0-4 C4 en0 w 4 -4 W

w1- 41 M In I~N - N - -0 6 0'0 . 0 44 be b4 44i - .U U U

C6 4 lo~ W4 P

0. 0 0 V A

60114&V0 %0 I0 . 4- . I O

hi~ . A8 8 ' A N 0 *

5-30 14bO O4

1

V% fw 0

5-3

%0 C14-

-2 -w a .2

O C4 1

I -*1 j ~5-32?

CA °l

I

13

Iv %.6j I to A.

Sj I1

1ACAJA A j Ai IZ5-33

Treatments TREATMENTS(10% laAB & CAb

0.9. 0.9 C

0.8 -0.00

0.7 0.7->~ ABIC

aD 0.6-CAzD

2 as0.5 0 3

1 AA 1AS

0.3 -0.3-

0.2. 0.2-I

0.1 Run 1 0.1 Run 2

0 24 46 72 96 0 24 48 72 9 6HOURS HOURS

Figure 5.1 a. Graphic Presentationi of Survival Versus Time for Set 1, Runs 1 and 2Lake Sidney Lanier, Georgia, Fall 198048-Hr. Prooortlons Surviving are Listed Vertically In Some order as GraPh Lines Repremantingthe Treatmeants. Vertical Lines to the Left of Letters Join Together Thoae Treatments NotSignificantly Different From Each Other

4100%) 1.0. 1: .0

0.9- % 0.9-

0.7 .5 0.7

%% %%5~~

0.6 0.8 N5

2 0.5 CIA.

0.

;0.73N 0.7

~0.6.7 ~js 0< .0001 0.6 6 or P <0.00

2 0. 0

0.1un2Reutsa 72 Hours andoooi . 96 Hours t lesy

Lake Sidney Lanier, Georgia, Fall 1980At Both 72 Hour and 96 Hour Ohawolvad Iron concentration ExplainsN Ptewet of the Variability in Survival Rates Wth Narrow ConfidenceLbnit. (ahed Lines) end High Probability ID 0.0001 I

Oaurlptions of Each Lettered Treatmentt we Given in Tables 5.7. end 6.2.

5-34

U.c

mI~w~I*

LL -

7t&

&- to

- ~ . I-,x cc

cc

CC a.

> e

2 : E

-~~C 0- 0r

% L v

0 -l

5-35

FU

m w UwU~OZ ae~~- -

Z

r '

- -- - -

c-i .i~0> ~~L to-

-~L -40 0 0 0 a 0 0 0 0 rgo

* SiLLC, 04

Ow40" UOIOO,

5-36i-

AD-Al17 279 JONES EDMUNDS AND ASSOCIATES INC GAINESVILLE FL F/6 6/6A BIOASSAY EXPERIMENT TO DETERMINE WATER QUALITY IMPACTS RELATE--ETC(U)MAR 82 M E LEHMAN, M R CREZEE, R H JONES DACWO-AO'C-0368

UNCLASSIFIED NL22fllllfmlflmlml

E hhEhNhhIIIIIIIIIIIIIIffllfllf

*flflflflflflIII

1.25114 1U8

M I 0 I I[ I IO I: I f

S 0.

/ aS 0

Pa a/ 0 0

os a

Ic8

I -cc

gi x

CL-

10ft .2

facIv I ' ac: . a~'I

0 .0 0 0

5-37-~

OEM =* ~

2000

0-24 Hours1750 X 48 Hours

1500-

M-.gMb 1250

1000. e as .

750-

24 Hr.

2501 48 Hr.

025% 50% 75% 100%

PERCENT SURVIVAL

Figure 5.Gaho Percent Survival Versus Manganese Concentration in Static Bioassiys,

5-38

4500-

4200 ,

3M00

3600 Tot Fe3M' / /330 / \

3000.

2700 1 \.

c 2400 /

2100

1500.

1200...........,Tr,,. M,1 .*-. / .... - - - --. \.

..w .... I, .O". .Y.-.., ,,n, . v.,,

600

300 1

10/23 11/7 11/23 12/4 12/8 12/23 2/5

Time, days

Figure 5.5. Iron and Mangenese Concentrations in the Hypolimnion of Lake Sidney Lanier (35-Meter Water Depth)About 00 Feet (180 m) Behind Buford Dam, as Measured in Water Pumped Continuouslyinto BOlomay Compound on Top of Buford Dim, Georgia, Fall 1980

Now:

I. Ow. for Pabrwry War. Extrouu FMM MMwnnat the HOaW Wy (Low Flow) NMIrkl iANWIns Slow.

5-39

Figure 5.6. Photomicrographs of Eye and Muscle of Fry Taken From Set 11 Bkoasays Performed atBuford Dam, Autumn 1980

KEY

A. Section Through Normal Eye.a. Section Through Eye of Trout Fry from SII1 1 A (Toxic Control) Showing Proliferation of Pigmented Layers (Arrow).

C. Muscle Bundles of Normal Trout.0. Degenerating Muscle Bundles (Arrow) of Trout Fry from S11l1 A.

5-40

g a .

v 01-s N~ sto

u L~~ .

C4C

LL)

S~ c

~ V8

a a';

5-41

C4C

In

5-42

10-

e 20-

4-Bottom

0 21 6 8 10 10o 15 20 25Dissolved Oxygen, ppm Temperature,oc

Figure 5.8. Temperature and Diusolved Oxygen Profiles from Lake Sidney Lanier, Sampled at JEALaboratory Intake Buoy About 600 Feet 0180 Meters) Behind Buford Dam, Georgia, Fall 1980

4131

qr

toC

- CO

cce0 CL.

C 0

Ca- Sb

6 Q

'S *(

a rj a"M Om o

4-

SECTION 6.0. DISCUSSION

U

I

* ~* 7

6.0 DISCUSSION

6, 1 K AUMRhSE AND IRON ARE TOXICANT S

Evidence that Mn and Fe are responsible for trout mortality at thehatchery is compelling. Only Fe and Mn have been found in high concen-trations every autumn when mortality increases at the hatchery. Hatch-ery personnel have been able to predict the time of first mortalities byprojecting date-concentration curves forward to Fe concentrations of 750ppb or Mn concentrations of 500 ppb (Don Toney, p.c.). Approximatelythese same levels of Fe and Mn were found toxic in the laboratory bioas-says.

Further, when Fe and Mn were removed by a resin column and hardnessreturned to its normal level, bottom water was no longer toxic to rain-bow trout fry. Also, precipitating Fe and Mn by aeration at pH 11 andthen filtering them out and returning to pH 6 removed all toxicity.

Finally, completely non-toxic epilimnetic water from Lake Laniercould be made acutely toxic by adding MnC12 or FeC12 to give theconcentrations found in low flow river water in autumn. For epilimneticwater, the complicating possibility of some unknown toxicant or syner-gist produced by anoxic conditions or released from bottom sedimentsdoes not exist, yet Mn and Fe still are at least as toxic as in theriver.

*6.2 30 OTHI T IALS ARE INVOLVED

Copper, zinc, and aluminum have at various times been reported athigh concentrations in the lake, river, and/or hatchery. Mount et al.(1978) concluded that copper was the poison, because rainbow troutlT7iv-ers contained 400 to 500 ppm copper. U.S. EPA (1979) reported highlevels of zinc (430-733 ppb) at the hatchery, and aluminum was once re-ported at 6 ppm at the city water intake, Atlanta (Mount et al., 1978).

However, no metals other than Fe and Mn were above detection limitsin the current study, yet the water was always toxic to rainbow troutfry. The water was frequently analyzed; stored water analyzed severaltimes during storage was toxic, yet contained no metals except Fe and Mnat detectable levels. In this case, no occasional toxic pulses of Cu,Zn, or Al could have been missed.

Cu was always below detection. Zn was sporadically only slightlyabove detection, and the fact that all dissolved Zn values were higherthan total Zn suggests that even these low levels may be due to an uni-dentified source of Zn contamination in the sampling procedure. TotalAl never exceeded 250 ppb, and dissolved Al was seldom above the 25 ppbdetection limit. Arsenic (As), lead (Pb), cadmium (Cd), chromium (Cr),and mercury (ag)were not detected in any of the water samples.

Since these several metals were never detectable in water of demon-4. strated acute toxicity, the toxicity cannot be attributed to Al, As, Cd,S, Cr, Cu, Vg, Pb, or Zn, either singly or in combination.

6-1

6.3 30 OWGAICS IMOLVED

Since the 1976 hatchery experiments reported by Noell and Oglesby(1977) and Oglesby et al. (1978), toxicity has often been attributed tosome organic interacting with manganese. After discussing results,Oglesby at al. (1978, p. 290) concluded that "in order to account ade-quately for these observations, it is necessary to postulate presence ofan additional active material with the manganese... We postulate thatthis additional active substance is some form of huric material." How-ever, poetulating the presence of a humic material was not necessary.Their results (see Table 3.1) showed that increasing hardness to 70 ppmprotected fish, that decreasing Fe+ + by 67 percent and increasingalkalinity to 86 ppm protected fish, and that complexing metals withNa4 EDTA gave at least temporary protection to fish. No humic mater-ial is necessary to explain these results.

Furthermore, explaining toxicity as manganese interacting with ahumic from sediments or from decomposing organic material on the lakebottom is inconsistent with the demonstration in the static bioassaysthat Mn added to surface water is at least as toxic as an equivalentlevel of Mn in bottom water.

A conceptual problem with the humics hypothesis is that total organ-ic carbon levels in Lake Sidney Lanier are low and that analyses havebeen unable to identify any humics. Analyses of the most toxic watersin Set I for volatile organic carbons, breakdown products of hunic acid,found none. Examination of S114 water by high pressure liquid chromato-graphy, the appropriate technique for identifying humics, found no traceof hwmic substances.

Even if some humics were present, only a low level of complexeswould be formed. In Lake Sidney Lanier, total organic carbon usuallydoes not exceed 3 ppm. After experimenting with 3 ppm humic substances,Wilson (1978) concluded that "total metal concentrations... are com-prised primarily of inorganic species. Rlunic complexes are present, butnot in appreciable mounts". In Lake Erie sediments with approximately2 percent humics less than 10 percent of total Ma may be bound to thehumics (Nriagu and Coker, 1980). Thus, research suggests that even ifall organics in Lake Lanier were humics, few setal-humic complexes wouldbe present and that Ma-humic complexes would be few even at higher humicconcentrations.

Furthermore, comparison of total and dissolved (0.1 us filtered)Mn values shows no evidence of Hn-humic complexes. When Poldoski (1979)added cadmium and hamic acids to Lake Superior water, he found a sharpdrop in the mount of Cd passing a 0.1-un filter as opposed to a 0.45-un filter, showing that metal-humic complexes do not pass an 0.1-ufilter. In the current study, all amples for dissolved metals werefiltered through a 0.1-un filter, yet dissolved and total Mn almostalways were within measurement error of being identical. When 0. 1-urand 0.4-u filters were compared, both filters gave similar Fed and

Mad concentrations (Table A-14), showing that there were no huaic-mtal complexes.

6-2

Finally, if humics were present and if Mn-humic complexes were form-ed, the result would be to decrease rather than to increase im toxicity.The results of Oglesby et al. (1978) showed that metals complexing byEDTA decreased toxicity. Even though copper is highly toxic, "it is nowfairly well established that the organocupric complexes... are relative-ly nontoxic to fish and algae" (Wagemann and Barica, 1979, p. 519). Forcadmium, five of six complexing agents decreased Cd uptake by Daphnia(Poldoski, 1979). The exception formed a hydrophobic complex retainedby a 0.1-um filter (see above paragraph). In general, "the more strong-ly a metal is complexed, the lower the toxic effect" (Sposito, 1981, p.396).

In summary, postulating the presence of a humic to explain resultswas not necessary. Few or no humics are found in Lake Sidney Lanier.At low levels of humics, only a small proportion of metals would bebound. The humics postulate is inconsistent with the facts that iM add-ed to surface water (no humics) is as toxic as Mn in bottom water (sug-gested humics) and that all Mn passes a 0.1-um filter while Mn-humiccomplexes are retained on an 0.1-um filter. Finally, Mn-humic com-plexes would almost certainly be less toxic than free Hn++, not moretoxic, as postulated.

6.4 RESULTS APPLICABLE TO HATCHERY

The large differences in Mn toxicities between Lake Sidney Lanierwaters and waters at Lake Burton Trout Hatchery (England, 1978) provethe necessity for having performed experiments on-site and using thesame source of water used by the hatchery. However, several parallelobservations at the hatchery and laboratory demonstrate that the shortdistance between Buford Dam and Buford Trout Hatchery is not importantand that results of the bioussay studies on water from Lake SidneyLanier are still directly applicable after the water flows 2.5 kilo-meters from the lake to the hatchery.

In the laboratory experiments where Mn levels varied, the acutelytoxic level was found to be close to the 500 ppb Mn which hatchery per-sonnel have found correlated with the onset of hatchery mortalities.(In round-robin tests, results are considered good if the highestLC50 is not more than twice the lowest LC5 0 , (Schimel, 1982).)

Low flow water in the river is drawn from several depths in thehypolimnion so that metals are more dilute than in full strength bottomwater used in laboratory bioassays. However, even with this dilution,the river water is approximately as toxic as bottom water. The factthat toxicity does not decrease in proportion to dilution was also dem-onstrated in static bloassays in the laboratory, and the mortality ratein the river is roughly as would be predicted from the static bioassays.

In the laboratory, oxidation by prolonged aeration was found tocause an increase in toxicity. At the same time, oxidation by addinghydrogen peroxide to hatchery water caused an increase in mortality

4 there.

6-3

In the laboratory, adding hardness was found to prevent mortalities.Noell and Oglesby (1977) reported the same results for the hatcheryexperiments in 1976.

The similarities between laboratory, hatchery, and river in metalslevels found toxic, in dilution effects, in oxidation effect, and ineffects of hardness additions can leave no doubt that the laboratoryresults are fully valid for both river and hatchery.

6.5 NEW QUESTIONS

In the course of answering the questions for which it was designed,any investigation finds new questions to pose. The primary new enigmasarising from the current study relate to why Mn and Fe are so toxic, whyremoving iron increases toxicity, why hardness removal so strongly af-fected the fish, and why experimental fish placed in the hatchery quick-ly died while hatchery fish did not.

The estimated 48-hour LC5 0 of 650 ppb Mn is far lower than the24.7 ppm 96-hour LC50 for yearling rainbow trout (England, 1978) atLake Burton Trout Hatchery (Georgia) where hardness was only 2 ppm orthe 3.25 ppm Mn which over 72 hours had no effect on rainbow fry (Lewis,1976) in distilled water. On the other hand, toxicity problems atGreer's Ferry National Fish Hatchery, Arkansas, were attributed to Mnconcentrations of 1-2 ppm (Figure 6.1) at hardness of 12 to 35 ppm. Nixand Ingols (1981) reported no other metals in elevated concentrations,but few were measured. Toxicity problems at Greer's Ferry were neverthoroughly investigated, and some toxic agent other than Mn remains apossibility. However, results of the current bloassays add credence tosuggestions than Mn alone killed trout. Nix and Ingols (1981) suggestedthat differences in reported Mn toxicities could be explained if Mn weremore toxic as it became more highly oxidized. However, it does not seemlikely that Mn recently released from the anoxic hypolimnion of LakeSidney Lanier would be more oxidized than Mn added in laboratory bioas-says by Levis (1976) and England (1978), nor does it seem likely thatoxidation states would differ substantially between the current labora-tory experiments and those of Lewis and of England. Since differencesin the oxidation state of Mn seem unable consistently to explain thelarge differences in Mn toxicities, the differences must for now remainunresolved, for no other explanation is apparent.

In treatment SII4H, an activated carbon column removed all Fe fromhypolimnetic water but left hardness and Mn levels unchanged. This Feremoval increased toxicity and the toxicity of the Mn alone was nearthat predicted from the static experiments. Prolonged aeration de-creased Fed concentrations in two cases out of three and in those sametwo cases also caused an increase in toxicity (SIIID and SII3D versusS114E). Static experiments indicated that at moderate levels, Mn and Fetoxicities were additive. This finding makes a protective effect ofhigh Fe against high Mn seem Improbable, but such a protective effect isthe only readily apparent explanation for toxicity increase associatedwith Fe removal. Such a protective effect also was suggested by thecomparison of different untreated waters (Table 5.6) because survivalwas anomalously low in S114 where Fe but not Mn also was low.

6-4

When resin columns decreased hardness from 12 ppm down to 2-4.5 ppm,fish died quickly even though all Fe and Mn also were removed. The ex-cellent survival rate obtained by adding back the hardness (SII3E) showsthat the low hardness does explain the high mortalities. However, LakeBurton hatchery, where England (1978) found the 96-hour LC5 0 to be24.7 ppm Mn, raises rainbow trout at a hardness of 2 ppm, and Lewis(1976) found no effects transferring rainbow trout fry from 5 ppm hard-ness into 3.25 ppm Mn in distilled water. An obvious explanation wouldbe that it was not 4 ppm hardness that was lethal but rather the abruptchange from 12 ppm to 4 pp.. The reservations in accepting this explan-ation are that the absolute change is small and that, while the relativechange is large, it seems no greater than in Lewis' transfer of fishfrom 5 ppm to distilled water.

It may seem inconsistent that experimental rainbow trout held in thehatchery died quickly at a time when the hatchery recirculated water andlost few fish. However, the hatchery was raising mostly brook trout,which are considerable less sensitive to metals than rainbow trout(Chapman, 1978), and the fewer rainbow trout at the hatchery were largerand therefore less susceptable to metals than were even the yearling ex-perimental trout.

Metals levels at the hatchery could have been near those in theriver because the previous high-flow period, during which water had beendrawn from the Chattahoochee River, was one of little flow increase andwould have given less than usual dilution of hypolimnetic water. Also,the hatchery was not on complete recirculation; low flow water was drawncontinuously at 10 percent of the normal intake rate.

6.6 RIVERINE DISCUSSION

The intent of the riverine bioassay was primarily to document fishmortalities in the river and secondarily to determine the extent of theeffects downstream. The results of the in situ bioassays conducted inlate November showed that the conditions I- t-E --Chattahoochee River pro-duced trout mortalities in an area from Buford Dam downriver to AbbottsBridge. Seventy-five percent of the swim-up fry and 13 percent of thelarger trout died at Abbotts Bridge, the bioassay station located furth-est donriver from the dam (13.5 miles). As expected, no mortalitieswere observed during Run 2 (February) after Lake Lanier had completelydestratified resulting in a sharp decrease in the concentration of Feand Mn in the river.

In the riverine bloassays, the swim-up fry were considerably moresensitive to toxicity than were the larger trout (Figure 5.7). Similarresults were obtained in a study by Howarth and Sprague (1978) in whichsmaller fish were found to be more sensitive to metal toxicity than werelarger members of the same species.

Introduced trout and native yellow perch both show histopathology4 probably attributable to metals (Grizzle, 1981). However, all yellow

perch and some trout tagged and recaptured from the first 18 km belowBuford Dam showed weight gains, although many trout, especially rain-bows, lost weight (Gilbert and Reinert, 1979). The 28 species of fish

6-5

taken by electrofishing in fall 1977 were considered to represent adepauperate fauna (Gilbert and Reinert, 1979). Ten times as many inver-tebrates settled on artificial substrates 18 ka below the dam as at anyof the 3 stations nearer the dam (Gilbert and Reinert, 1979), but diver-sities were the same at all four stations. All four stations were up-stream of the point where a toxicity decrease was found in the in situbloassay experiments reported herein, and no comparisons were ade itother streams.

Available evidence suggests that fauna in the Chattahoochee River isa depauperate and species-poor fauna with animals in sub-optimal healthfor an undetermined but significant distance below Buford Dam. However,the severity of the impact is not established, and how much impact isdue to toxicity, how much to low DO, and how much to other factors isnot known. Other aspects of hypolimnetic releases shown to impact faunainclude large and rapid changes in flow (Fisher and LaVoy, 1972; Kroger,1973; Minshall and Winger, 1968), alteration of annual temperatureregime (Lehmkuhl, 1972; Ward, 1974, 1976; Gore, 1977), and decreases inthe amount and variety of suspended matter (Ward, 1976).

6.7 WATER QUALITY IDACTS ON TIE RIVERINE SYSTI

Since completion of Buford Dam and creation of Lake Sidney Lanier in1957, the Chattahoochee River for 50 miles below Buford Dam has beenstocked with trout and has become a popular "put and take" trout stream.No water quality impacts were obvious until a comercial trout fish-outoperation and later Buford Trout Hatchery commenced operations 1.5 to 2miles below the dam and suffered heavy losses of trout in the fall ofthe year.

Riverine bioassays showed that in autumn the Chattahoochee Riverwater is acutely toxic to trout, which are stocked into the river, butthat bluegills, which occur naturally in the river, were much more re-sistant and survived the 96-hour riverine bioassays. Similarly, Gilbertand Reinert (1979) found that many tagged trout, especially rainbowtrout, lost weight between recaptures below Buford Dam but that all ofthe native yellow perch gained weight.

Gilbert and Reinert (1979) also examined invertebrates colonizingartificial substrates collected monthly September through December afterone month in the Chattahoochee River either 100 m, 3.1 ki, 8.6 ki, or 18km below Buford Dam. Ten times as many invertebrates colonized the 18-km station as colonized the other three stations, but diversities werethe same at all stations. All stations were upstream of the point wheretoxicity decrease was detected in the riverine bioassays.

The sparseness of the fauna below Buford Dam and the results of theriverine bioassays strongly suggest impact of the hypolimnetic releaseson the Chattahoochee River. However, without comparisons with data fromsimilar rivers (where impacts are not suspected), and considering thelimited aount of comparative literature applicable to this case, firmconclusions cannot be made.

6-6

. . . . . . , ,,, -_ _ _-_ _.__ _ _ _ _ _....._:'.. . . : . . .

90004

I aI a

! a

00C Daily Mortalities

U. I0030.1

IL 0000

-lS; -

300 Toal'np e

- 10002000 8-V * us te ,ll J

8 3e 28 1 is 28 1 7

Nov 1976 Dc 1976 Jan 1977a) Daily Mortalities 1976-1977

Water H~ardnes was IS ppmnA Total of 126.151 Trout were Lost

Data CoDurtesy of James CamPerGreer's Ferry Naktional Fish Hatchery

a '' a I

£I

20000

20 aFo Mix 1L

Is It MnISC

10 001 0 h -

Nov 1976 Dece 197 Jn97196 Dal ortwta1967

b)t Courtyof iaes aPer 0 ih16-1

Figure 8.1. Mangnese Concenaions and Total Mortalities at Grer's Ferry Nationall FishHte

Hatchey 5. a) Noeb 8, 1976, to Januar 5,1977, and b) Noeb

1 t-1

SECTION 7.0. CONCLUSIONS

7.0 CONCLUSIONS

1. In numerous chemical analyses of toxic waters from Lake SidneyLanier, iron and manganese were the only potential toxicants found,and they were consistently present in high concentrations in toxicwaters.

2. Experiments adding iron and/or manganese to epilimnetic water demon-strated that the levels of iron and manganese in Lake Sidney Lanierand in the Chattahoochee River can explain all observed toxicity.

3. Removing metals from toxic water rendered it non-toxic if hardnesswas not also decreased, but removing organics had no effect.

4. Adding 25 ppm hardness to bottom water was completely effective inpreventing trout mortality for at least 96-hours.

5. At low-flow in autumn Chattahoochee River water is acutely toxic torainbow trout especially yearlings and fry but not to bluegill sun-fish. The toxicity is moderately reduced 23 km downstream from thedam.

6. Of the eight detectable metals out of the 24 for which trout liverswere analyzed, two (Na, Fe) did not differ between hatcheries whilesix were significantly higher at Buford. Results were consideredinconclusive and perhaps explainable on the basis of high Mki atDuford Trout Hatchery.

7-1

SECTION B.0. RECOMMENDATIONS

Wo.

8.0 CTIOnS

The results of this investigation have led to several potential al-ternatives to solving the toxicity problems at the Buford Trout Hatch-ery. Some of these alternatives have already been implemented at BufordTrout Hatchery or at other das around the country.

Potential alternatives include managesent changes at the hatchery,treatment of hatchery intake water, use of an alternate source of water,treating water at the dam or reservoir, or creating non-toxic mixes ofhypolinetic and epilimnetic water.

S•.1 NAIMhGIKNT CBhUIS

Past modifications to management practices at Buford Trout Hatcheryhave been partially satisfactory. By drawing in water during high flowconditions and recirculating during low flow, the hatchery has been ableto limit losses of fish. The U.S. Army Corps of Engineers have cooper-ated in the past with special high flow releases even when power was notneeded. Recirculation requires reduced loading rates, so the hatcherymust operate below capacity.

In 1980, the hatchery stocked mostly brook trout rather than rain-bow trout because brook trout showed less sensitivity to low flowwaters. Brook trout mortalities in 1980 were above normal; at times thetrout became so excitable feeding had to be discontinued for severaldays.

Changes in management practices have provided interim, temporary,and partial reduction in trout mortality problems at Buford Trout Hatch-ery. Recirculation during high flow, special releases (when feasible),raising brook trout rather than rainbow trout, and operating at reducedloading should be continued on an interim basis until feasibility stud-ies can be completed for a permanent alternative. Management changesalone do not provide a totally satisfactory cost-effective, long-termsolution to the toxicity problem.

8*2 TX&AIM1N OF Nh2GKW INTAKE WATER

The removal of Fe and Mn from water enterinb the hatchery is an al-ternative. Fe and Mn may be removed by oxidation followed by eithersedimentation or filtration of the oxidized metals. The rate of oxida-tion is pH dependent, especially for Ma; pH adjustment, adequate deten-tion time and filtration facilities would have to be constructed. Pre-liminary engineering design and cost estimates would be necessary todetermine the cost-effectiveness of this alternative.

The addition of hardness to the water entering the hatchery can pre-vent toxicity. Because the problem is a seasonal one, the addition ofhardness (in this case, calcium) could possibly have the lowest initialcost as well as operating cost. Potential chemicals include, but are

8-1

not limited to, CaCO 3 , CaC12, and CaSO4. Engineering design and

cost estimates would have to be performed for this alternative.

0.3 ALTKRELTIVE SOURCES OF WATER

A new source of water for the hatchery could be created by installa-ting a multi-level intake structure on Buford Dam so that all waterreleased would be a mix of epilimnetic and hypolimnetic water (similarto mix of water currently released at high flow).

Another source that should be considered includes. a pipeline tocarry a non-toxic mix of cool water from the reservoir to the hatchery.The pipeline could possibly be constructed to siphon water from thereservoir to the hatchery.

Temperature studies and engineering conceptual design and cost stud-ies would have to be performed for the multi-level intake and pipelinealternatives.

Installing a well is another feasible alternative if a suitablegroundwater source is available. Water quality and quantity studies ofthe local aquifiers would have to be conducted in conjunction with engi-neering design and cost studies for this alternative.

8.4 TREATHENT OF WATER AT DAM OR RESERVOIR

Aerating river water at the dam or in proximity to the penstockswould add oxygen to the water and reduce the problem of low dissolvedoxygen in the river immediately downstream; however, this aeration mayexacerbate toxicity problems at the hatchery, as evidenced by the toxi-city increases associated with prolonged aeration without sufficientsedimentation or filtration. If sedimentation and filtration systemswere constructed at the hatchery, then aeration at the dam could be con-sidered feasible. Possibilities for aeration include but are not limit-ed to: pulsed injections of oxygen close to penstocks, turbine modifi-cations to draw in air, and rip-rap in the tailrace to create turbu-lence. Each alternative would have to be evaluated through conceptualengineering design and cost studies in conjunction with similar studiesfor sedimentation and filtration of hatchery intake water.

Seasonal aeration of a portion of Lake Sidney Lanier to prevent orto disrupt oxygen stratification and avoid creation of an anoxic hypo-limnion could prevent reduction and solubilization of M and Fe. Thisalternative could reduce or eliminate problems in both river and hatch-ery. Air or pure oxygen can be introduced into the hypolimnion suchthat no temperature destratifying turbulence is created. Air is lessexpensive than oxygen, but hypolimnetic aeration produces nitrogensupersaturation which can be toxic to fish. Studies would have to beconducted to compare feasible approaches to this alternative, such asoxygen injection into a hypolimnetic diffuser system compared withsurface aerators. Some studies of this type have been conducted by the

8-2

k4

U.S. Army Corps of Engineers, Savannah, in Clark Hill Lake. Conceptualengineering design and cost studies will also be necessary.

8.5 rERINK COESIDBRATIONS

Additional studies are necessary for the Chattahoochee River systemto determine specific measures to mitigate riverine impacts due to Feand Mn. The extent, nature, and cause(s) of Impacts must be identifiedand clarified. The available evidence suggests that fauna in the Chat-tahochee River is a depauperate and species-poor fauna, with animals insub-optimal health for an undetermined distance below Buford Dam.

Additional studies will be needed to address the severity of impact;how much of the impact is due to metals toxicity, how much is due to lowdissolved oxygen, and how much to other factors is not known. The firstof such studies should be a continuation of in situ riverine bioassayexperiments similar to those conducted previously. The experimentsshould also include other faunal organisms and sufficient water qualityand histopathological data to provide data for cause-effect determina-t ions.

6-3

REFERENCES

REFERENCES

Affleck, R.J. 1952. Zinc Poisoning in a Trout Hatchery. Australian Jour-nal of Marine and Freshwater Research 3:142-169.

American Public Health Association. 1975. Standard Methods for the Examin-ation of Water and Wastewater. 14th Edition. Washington, D.C.

Brooks, N.H. and R.C.Y. Koh. 1969. Selected Withdrawal from Density Strati-fied Reservoirs. Journ. Hyd. Div. ASCE, HY 4, Vol. 95.

Chakoumakos, C., R.C. Russo, and R.V. Thurston. 1979. Toxicity of Copperto Cutthroat Trout (Salmo clarki) Under Different Conditions of Alkalin-ity, pH, and Hardness. Environmental Science and Technology, 13(2):213-217.

Chapman, G.A. 1978. Toxicities of Cadmium, Copper, and Zinc to Four Juven-ile Stages of Chinook Salmon and Steelhead. Transactions of the Ameri-can Fisheries Society, 107:841-847.

England, R.H. 1978. Toxicity of Manganese to Rainbow Trout. Report to theMobile District, U.S. Army Corps of Engineers. Contract No. DACW01-77-C-0166.

Environmental Science and Engineering, Inc. (ESE). 1981. Water QualityManagement Study, Lake Sidney Lanier, Georgia. Final Contract Report tothe U.S. Army Corps of Engineers, Mobile District (Contract No. DACW01-78-00105), Mobile, Alabama.

Fischer, H.B., E.J. List, R.C.Y. Koh, J. Imberger, and N.H. Brooks. 1979.Mixing in Inland and Coastal Waters. Academic Press. New York.

Fisher, S.G. and A. LaVoy. 1972. Differences in Littoral Fauna Due toFluctuating Water Levels Below a Hydroelectric Dam. Journal of theFisheries Research Board of Canada, 29:1472-1476.

Georgia Department of Natural Resources. Game and Fish Division. 1978.Sequence of Events During Water Quality Problems at Buford Hatchery,1978. Handout at 13 December 1978 Meeting of the Task Force on WaterQuality Problems Below Buford Dam. Atlanta, Georgia.

Georgia Department of Natural Resources, Game and Fish Division, and Envi-ronmental Protection Division. 1980. Water Quality and Aquatic Commun-ity Investigations in the Chattahoochee River Below Buford Dam. Reportto the U.S. Army Corps of Engineers (Contract No. DACW 01-77-C-0166).

Gilbert, R.J., and R.E. Reinert. 1979. A Study of the Effects of WaterQuality on Movements of Fishes in the Tailwaters of Buford Dam, Georgia.August through December, 1977. Final Report to the U.S. Army Corps ofEngineers, Mobile District (Contract No. DACW 01-77- C-0160).

4, Gore, J.A. 1977. Reservoir Manipulations and Benthic Macroinvertebrates in4, a Prairie River. Hydrobiologia, 55:113-123.

R-1

Grizzle, J. 1981. Effects of Hypolimnetic Discharge on Fish HealthBelow a Reservoir. Transactions of the American Fisheries Society,110:29-43.

Hart, R.K. 1979. An Electron Microscopic and X-Ray Fluorescence Exam-ination of Fish Gills from the Tailwaters of Buford Dam, Georgia.Pasat Research Associates, Inc., Atlanta, Georgia. Report to MobileDistrict, U.S. Army Corps of Engineers. Contract No. DACWO01-77-C-0166.

Hess, T.B. 1978. Collection and Analyses of Fisheries Data at OldJones Bridge on the Chattahoochee River. Report to the Mobile Dis-trict, U.S. Army Corps of Engineers. Contract No. DACWO1-77-C-0166.

Howarth, R.S., and J.B. Sprague. 1978. Copper Lethality to RainbowTrout in Waters of Various Hardness and pH. Water Research 12:455-462.

Kroger, R.L. 1973. Biological Effects of Fluctuating Water Levels inthe Snake River, Grand Teton National Park, Wyoming. The AmericanMidland Naturalist, 89:478-481.

Lehmkuhl, D.M. 1972. Change in Thermal Regime as a Cause of Reductionof Benthic Fauna Downstream of a Reservoir. Journal of the Fisher-ies Research Board of Canada, 29:1329-1332.

Lewis, M. 1976. Effects of Low Concentrations of Manganoua Sulfate onEggs and Fry of Rainbow Trout. The Progressive Fish-Culturist,38(2) :63-65.

Lewis, M.A., D.W. Evans, and J.G. Wiener. 1979. Manganese. Pages 137-144. In: A Review of the EPA Red Book: Quality Criteria forWater. R.V. Thurston, R.C. Russo, C.M. Fetterolf, Jr., T.A.Edsall, and Y.M. Barber, Jr. (Eds.) Water Quality Section, Ameri-can Fisheries Society, Bethesda, MD.

Lloyd, R. 1960. The Toxicity of Zinc Sulphate to Rainbow Trout. An-nals of Applied Biology, 48:84-94.

Mendenhall, W. 1968. The Design and Analysis of Experiments. Wads-worth, Belmont, California.

Metrics, Inc. 1978. Lake Lanier, Chattahoochee River, and Lake LanierTrout Hatchery In Situ and Chemistry Measurements. Report to Stateof Georgia Department of Natural Resources. G.B. Oglesby, Con-tracting Officer. Atlanta, Georgia.

Minshall, G.W., and P.V. Winger. 1968. The Effect of Reduction inStream Flow on Invertebrate Drift. Ecology, 49:580-582.

Mount, D.I., W. Peltier, T. Bennett, R. Weldon, and T. Norberg. 1978.Investigations of Fish Mortality at Lake Sidney Lanier Fish Hatch-ery, November, 1977. U.S EPA, Duluth, Minnesota.

R-2

Nehring, R.B., and J.P. Goettl, Jr. 1974. Acute Toxicity of a Zinc-Polluted Stream to Four Species of Salmonids. Bulletin of Environ-mental Contamination and Toxicology, 12:464-469.

Nix, J. and R. Ingols. 1981. Oxidized Manganese From HypolimneticWater as a Possible Cause of Trout Mortality in Hatcheries. Pro-gressive Fish-Culturist, 43:32-36.

Noell, W.C., and G.B. Oglesby. 1977. Investigations Regarding Causesof Fish Kills at Lake Lanier Trout Hatchery. Report of Findings.Georgia Department of Natural Resources, Atlanta, Georgia.

Nriagu, J.O., and R.D. Coker. 1980. Trace Metals in Humic and FulvicAcids From Lake Ontario Sediments. Environmental Science and Tech-nology, 14:443-446.

Oglesby, G.B., W.C. Noell, H.O. Delo, W.A. Dyer, R.H. England, J.R.Fatora, J.M. Grizzle, and S.J. Deutsch. 1978. Toxic Substances inDischarges of Hypolimnetic Waters from Seasonally StratifiedImpoundment. Environmental Conservation, 5:287-293.

Pagenkopf, G.K., R.C. Russo, and R.V. Thurston. 1974. Effect of Com-plexation on Toxicity of Copper to Fishes. Journal of the Fisher-ies Research Board of Canada, 31:462-465.

Poldoski, J.E. 1979. Cadmium Bioaccumulation Assays. Their Relation-ship to Various Ionic Equilibria in Lake Superior Water. Environ-mental Science and Technology, 13:701-706.

Poston, H., G.F. Combs, Jr., and L. Leibovitz. 1976. Vitamin E andSelenium Interrelations in the Diet of Atlantic Salmon (Salmosalar) : Gross, Histological, and Biochemical Deficiency Signs.Journal of Nutrition, 106:892-904.

Schimel, S.C. 1982. Proposed Standard Practices for Conducting ChronicToxicity Tests with the Early Life Stage of Fishes. ASTM Draft No.2, E47.01. U.S. EPA, Narragansett, R.I.

Singh, S., G.S. Shukla, R.S. Srivastava, and S.V. Chandra. 1979. TheInteraction Between Ethanol and Manganese in Rat Brain. Archivesof Toxicology, 41:307-316.

Sposito, G. 1981. Trace Metals in Contaminated Waters. EnvironmentalScience & Technology, 15:396-403.

Strain, G.M., 1980. Environmental Aspects of Reservoir Releases. Pro-ceedings of a Seminar on Water Quality Evaluation. U.S. Army Corpsof Engineers Committee on Water Quality. Washington, D.C. January22-24, 1980, Tampa, Florida.

U.S. Army Corps of Engineers, Mobile District. 1974. Final Environ-mental Statement. Buford Dam and Lake Sidney Lanier, Georgia.Colonel Drake Wilson, District Engineer, Mobile, Alabama.

R-3

U.S. Army Corps of Engineers, Mobile District. 1980a. General Back-ground Information on Water Quality Problems Related to the LakeLanier Trout Hatchery. Attachment I to Scope of Work-Phase I,SAMPD-EE Baya, June 3, 1980, Mobile, Alabama.

U.S. Army Corps of Engineers, Mobile District. 1980b. Bioassay Experi-ment to Evaluate Water Quality Impacts Related to Trout Mortality ata Hatchery Downstream of Lake Sidney Lanier, Georgia. Plan of StudyContract DACW01-80-C-0368. Prepared by Jones, Edmunds & Associates,Inc., Gainesville, Florida.

U.S. Department of Health, Education and Welfare. 1977. PesticideAnalytical Manual. Volume I. Methods Which Detect Multiple Resi-dues. Washington, D.C.

U.S. Environmental Protection Agency. 1974. Methods for Chemical Anal-ysis of Water and Wastes. Environmental Monitoring and Support Lab-oratory. Cincinnati, Ohio.

U.S. Environmental Protection Agency. 1979. Limnological Investigationof Lake Sidney Lanier With Special Reference to Heavy Metals. EPA904/9-79-046.

U.S. Environmental Protection Agency. 1980. Bioassays for SurfaceWaters, Wastes and Sediments. Draft Prepared by the Bioassay Sub-committee, Biological Advisory Committee. Aquatic Biology Section.Cincinnati, Ohio.

Wagemann, R., and J. Barica. 1979. Speciation and Rate of Loss of Cop-per From Lakewater with Implications to Toxicity. Water Research,13:515-523.

Ward, J.V. 1974. *A Temperature-Stressed Stream Ecosystem Below a Hypo-limnial Release Mountain Reservoir. Archiv fuer Hydrobiologie.74:247-275.

Ward, J.V. 1976. Comparative Liunology of Differentially RegulatedSections of a Colorado Mountain River. Archiv fuer Hydrooiologie,78:319-342.

Watson, L.T., C.B. Aerman, J.P. Feaster, and C.E. Roessler. 1973.Influence of Manganese Intake on Metabolism of Manganese and OtherMinerals in Sheep. Journal of Animal Science, 36:131-136.

Wilson, D.E. 1978. An Equilibrium Model Describing the Influence ofHumic Materials on the Speciation of Cu2+, Zn2+ , andMn2+ in Fresh Water. Limnology and Oceanography, 23(3):499-507.

R-4

F-,

'4

APPENDICES

I

4

4,'

... , .'

LIST OF APPENDICES TABLES

Table Title

A.1 Lake Lanier, Georgia, Bioassay Set I Run 1 A-I(October 28-November 2, 1980)

A.2 Lake Lanier, Georgia, Bioassay Set I Run 2 A-2(November 7-10, 1980)

A.3 Lake Lanier, Georgia, Bioassay Set II Run I A-3

(November 19-23, 1980)

A.4 Lake Lanier, Georgia, Bioassay Set II Run 1 A-4Additional Metals Analyses on Selected Samples(November 19-24, 1980)

A.5 Lake Lanier, Georgia, Bioassay Set II Run 2 A-5

(November 23-24, 1980)

A.6 Lake Lanier, Georgia, Bioassay Set II Run 3 A-6(December 4-11, 1980)

A.7 Lake Lanier, Georgia, Bioassay Set II Run 3 A-8

Additional EPD Analyses and Micro Methods Data

(December 4-11, 1980)

A.8 Lake Lanier, Georgia, Bioassay Set II Run 4 A-9(December 19-23, 1980)

A.9 Summary of QA/QC Data for Lake Lanier, Georgia, A-10

Bioassays, Fall 1980

A.10 Lake Lanier, Georgia, Bioassay Set II Run 3 A-13Special Chemical Samples (December 4-11, 1980)

A.11 Lake Lanier, Georgia, Bioassay Simultaneous A-14

Sampling from Header Tank and Lake Hypolimnion(December 1980)

B.1 Summary of Results of Routine Physical Monitoring B-I

of Set I Run 1, October 28 through November 2,

1980, Lake Lanier, Georgia

B.2 Summary of Results of Routine Physical Monitoring B-2of Set I Run 2, November 6 through November 10,


B.3 Summary of Results of Routine Physical Monitoring B-3of Set II Run 1, November 21 through November 23,1980, Lake Lanier, Georgia

A-i

- - - - - - - -- - -

LIST OF APPENDICES TABLES

Table Title Page

B.4 Sumary of Results of Routine Physical Monitoring B-4of Set II Run 2, November 23 through November 24,1980, Lake Lanier, Georgia

B.5 Sumary of Results of Routine Physical Monitoring B-5of Set II Run 3, December 4 through December 8,


B.6 Summary of Results of Routine Physical Monitoring B-6of Set II Run 4, December 19 through December 23,1980, Lake Lanier, Georgia

C.1 Physical Chemical Data for Riverine Bioassay C-1

Stations, Run 1, November 22-26, 1980, ChattahoocheeRiver/Lake Lanier, Georgia

C.2 Physical Chemical Data for Riverine Bioassay C-2

Stations, Run 2, February 5-9, 1981, ChattahoocheeRiver/Lake Lanier, Georgia

C.3 Survival Data (Salmo gairdneri Swim-Up Fry) From C-3Riverine Bioassay Run 1, November 22 through

November 26, 1980, Chattahoochee River/Lake Lanier,Georgia

C.4 Survival Data (Salmo gairdneri 15-Centimeter Trout) C-5from Riverine Bioassay Run 1, November 22 ThroughNovember 26, 1980, Chattahoochee River/Lake Lanier,Georgia

C.5 Daily Survival Data for Rainbow Trout in February C-61981 and Bluegills in November 1980 During RiverineBioassays in the Chattahoochee River Below BufordDam, Lake Lanier, Georgia

C.6 Iron and Manganese Data From Two Riverine Bioassay C-7Stations on the Chattahoochee River and from theHypolimnion of Lake Sidney Lanier, Georgia, November

23 through November 26, 1980

C.7 Iron and Manganese Data (Recorded from Run 2) from C-8Two Riverine Bioassay Stations on the ChattahoocheeRiver, February 5 through February 9, 1981, LakeLanier, Georgia

A-ii

•7 *

APPENDIX A. CHEMISTRY DATA

V~VVVvv~ v v11222222V- v V V-VVV vvv VVvVVV" v CII

(r n) uPrnl v 0v v v'vv v .v v 0 0 -

.'V..' VVVV' v /vV VVVV v

v v VVV VVVV

vvv

t/2n) Py 6K~h IQRRcJ to ~~

(1/2) 3 gdsRS

%itC1 -4 -E 4 . CI - -

1j) 0.t~ -C. L" In C1 I In~

MN 22) C0

Lld - - U(IN-. Z:it v g CC- s A oo~oooo~

IA v AAA 0I I I I -C LA I n % % -.4 - WgmV

"I'D

(r/)LPrrUM) v gi am I EIC'dII 1ig Ia'ii

A-i

IAU~s~&M- IAUl Ln MI -. $ in wI

. . . .. .II(/01) a sa -0 Q.. Q Cv al* Sao vR .01 vvv vv

(ii' ~~Yvvvvvv vvlvvv v v VvVVv

vvvv vvvv v v VVVV\0

Q/A)zpw v v vu VV Vol Vn QQV

V vv v vv v

(B)Pq MUVR 0 dv aoOQv'. "wQOMS Q wQt AmP

v v

N ~(T/Sn)~.

(T/O) rf :'01mv~ isnA ORIn

CA... eniim 4e

000000 Go0 ~ co GoGoooUN U, V) V-

u~ k C5;6 s eoee

(1/9) OU -4 6- 1

(Tih'kL)nru v -4 C4 n - o &a 6

LApq m)a -9(TI O-LIT! 01-6/11!W

h-2

70-40 Li 0 0000INLM~ 00 &A000000uN0 I gOo'nino

r 8" "O~ a" rN 0 IV -A O NWL %D0- nr r

N0

NN~6O. z 0

P44

* 000 00tlU0 0 00o~r.Ar%&~0 LIOOOVIfloO0

a a~ a a ame a a ra an aD a an an amr n a a 0 "ow %.w - --* D a a O n o 4 N 47- 0% -4 -4 N 0 m 00 c D

V;~~ ~ ~ C4c -; .

0. CV 4 V

go c

- 0.ic 0)

N1 .1- Olt

10 0 I la

Al I.-

-.544

0v%000000 POOOOOOV 0 COCOV 0 I

0 caI W DMO(Q % Di C nV.* Cl 4 L

0* S141 .0 At

0j 0

I00

oj (0-to 1

0~ v5'-W4

a 0'ON v.0@0% ba b4 0 w 6, 1N@'1 1 1

* 4.1

A-3..o

.4P4

U p. 0' Icool I0ti so gn0

-4>

*41 06

00U.. 0 01

I~IL

41 a 04of4P

64 s41*ra g;-4 IA 3

A %.0 to 0

C.I III

IW N01.0 1 0.41

P4 V-4

".41

- VI4

100

A-4

.0 In u' LA 0

6-4 ~VeyV

P-

P-4 Le 6-1 O'-4a

-4

0-" 0V4 0U4 V

Or.r~

C14 0

*"4 00 0

W -a -4

- N 0>-#44

A0a 0j

0- XI40 pq &I-"

0 it

*p4 S c 9

FA544 4)

N ~l CC.%... 0-tI .

.0 w-

0 w

ma 4.44

.yW4 a5

P"4

0)--5

41 mvmSr M&M00 %DQM 000 0 "&no" 0 D -4 OD

%D ~ tnN Nf 0% cnL 1 nL M % 1 VCr nL nL 1 Dt Me n LNV 40 nO! V -C4 N a%- V V V4 V4 C4 V- V- vv -deLv v

4

to 00000 0-0 0000 00000 oooo~rsln&' 000 0 u ODL = - 0@DV 000r.4' - -6@ CO 7 4C eM%

- ~ U'i %D %D C- - =ti J in Mf C4 NNC4 D

L. -~~~ - a -i~ 4- - 4 - &n~ IA - - J- - -

02 Ru.i moo0 iL n t nUt n nLW % W i06%L nu Mu k-4vv- 4MC a 4 V TvUL

P44

'-4 %. 4V N

40-z

CA c

* 4 __%%DDD OI %D fIa00% m ~~00 P-0 "4 WOI -

o U1

"4 O 0e-~ -- '4 0f~-c~,t I~N . 6--INI-A-6--

%~~~ 00 0 0 0 0.

bo Cmost acNc0N Iw a0 41 SW

40 lpw C V ovvv* 000

00

'0

-. M a 1 J N

C 0-

-7w 0- .r o * 0

* U, 0 V-4 000 6i - .0 0

c P.- Q som%C C4 %0 %Qir %D *4 o.

V ~ -t V i j0

0Uhv0h 0

0 U)2*

00

0 00 0 0 0 O 0

P6 L4

00

0 4 .0

-s m . I J

Ln -4 0 w

0 ~~- ~: rh

--

&o rIn L64 N C4 " C-

0 3

4t C1 0aOC;

A AjV4 0 -.. "

-4C4 0 4U

$-446o "4 -

v44 a r

V--4

0. -a Ln0 4 Y

In0. (An 44

8 w 0

o0 0

W440 0 W

0 I0.

40~ 0 Aj aV -4

-4-

A-8A

-4M

en C1 C4 ena a a a V ena

-a-j

0 t 4f9 Q 0 . Vri Ln C4 0000000A1A00000 0.

-4 C

~-4 0.0 0

-40-4 0.0u121 ! C r~lr, rl 11. . Z.50W

4) C A.0 ' 0-4-. c4 4 4U~~~~~~ lwO 0,if 'C'400 AU00 di~00 . ~ L

64 ?1 mJ. co 0 ?Z C40 Ul c ra,-. C:) %0 41 '0 . 0ad en~ a an a0 en a %0 4. 1 0 I

C%4 4 Z

S4v\1 vvVv\1v \1vV% 30 .. *U4

*~4 4)4~ 0 4

0f O g o -" OO C1 4.401 C4

40~0

6.0 00bo S.Q

La L'

-4 4 -

1114

to0

14.'I I I I I..4a. C4- (U 04 C4C4C

A-9

0'V-

I0 Ln1

-C -4 i4

A1 Q0% 00 t-) 0% M

01 41 II-4I

0 C

01

104 41

en go e0n 'f coo0040

-4 -4:1 1

W 41

14 u

41 -41 i410 4

Aj %I

91 1N 0%

A-10

w) -4

0~. 0,% 0%0

-4

D0

to I -o

> 00 0 Ln0%0 (ct - 00

04 0-

0%r- 0- U

cc C-4 -

100 00

be pa 0

n %000

0: .

-4 a4

14 0.41 00 4 00N

aD 0% 00

~~ 04) 4J miU~ 0

v-4

00 1

'4~C '-d%

Sd

0 P--

6)6)OD Ii I

-4

IdO

bo 00 000

0? -V~ I - k * %0

-4 0

0 00 to 00,o

* Y 00

fa 0 0 O1 1 A

iOII 0 o0 0 0' ODI A8A

* In

Ad

U ~ 0 -400e

- 0

00 0 9 ~

ap a acr

41 t%

A-1

0

0 0-4 %D I co-

bo - a

*~ 4J )

0) . 0 vj.

A4 -J4 -0

0 o* o v

W,.

4%4

0 LMa tic Go I'

0.o~

4Si

00k vv 0 wVA

q..4

* KA1K

tio 00

V40

b -- 0 ~

*i 000

o ~ -4 'D

w*b.r-4 -r4

I. -

0 LP S

-4 O

41 00 %O6j

00

4r4

* -4-

bOA 14~

4

APPENDIX B. ROUTINE PHYSICAL MONITORING

I

4,

4-

Table B.I. Sumary of Results of Routine Physical Monitoring of Set I Run 1,October 28 through November 2, 190, Lake Lanier, Georgia

RangesTreatment* Day of Temperature DO pH Conductivity *

Test 6C pPI unhos/ca

A 1 13.5-17.0 6.9-8.1 6.1-6.4 30-332 11.0-14.1 7.1-9.1 5.9-6.4 22-253 12.0-17.0 6.4-8.7 5.6-6.3 31-344 11.5-17.0 4.4-9.2 5.5-6.5 32-355 10.0-12.0 6.3-10.0 5.7-6.4 30

B 1 13.5-17.0 6.2-8.0 6.1-6.4 29-302 12.0-13.8 6.9-8.9 5.8-6.5 26-283 11.5-17.0 7.2-8.5 5.7-6.4 26-274 11.5-17.0 3.8-8.6 5.9-6.6 29-325 10.0-11.8 7.0-9.3 5.3-6.5 25

C 1 14.0-17.0 6.8-8.2 6.2-6.7 24-282 11.5-14.4 6.7-8.5 5.9-6.4 21-223 12.0-17.0 6.2-8.0 5.7-6.3 25-274 11.9-17.0 6.1-8.0 6.0-6.5 27-315 10.5-12.5 6.0-8.6 5.6-6.5 22

D 1 14.0-17.0 7.5-10.1 6.5-6.7 21-232 12.0-14.0 8.9-10.0 6.0-6.6 22-243 12.0-17.5 7.7-9.0 5.8-6.5 234 12.0-17.5 6.3-9.4 6.0-6.6 25-275 10.5-12.7 7.4-9.6 5.8-6.6 22

*Treatments are described in Section 4.0,-Table 4.2.**Conductivity readings made only once per day for SII.

51-1

Table B.2. Stmmary of Results of Routine Physical Monitoring of Set I Run 2,November 6 through November 10, 1980, Lake Lanier, Georgia

Treatment* Day of Thaperature DOR pH ConductivityTest "C pp umhos/c.

A 1 8.5-14.0 6.1-8.4 5.8-6.5 29-332 11.5-16.1 6.9-8.0 5.7-6.4 30-353 12.2-15.0 7.0-8.4 5.5-6.4 31-354 13.0-16.0 6.9-8.7 5.9-6.5 32-38

B 1 8.5-14.0 6.4-8.4 6.1-6.5 22-292 11.5-14.9 7.1-8.4 5.9-6.6 24-293 12.3-14.0 6.9-8.3 5.8-6.6 25-304 12.5-15.5 6.4-8.4 6.0-6.6 26-30

C 1 8.5-15.0 6.8-8.4 6.0-6.5 22-302 11.5-16.6 7.0-8.0 6.0-6.6 22-273 12.3-15.5 6.8-8.0 5.9-6.5 24-294 13.0-16.5 6.1-7.9 5.9-6.6 24-28

D 1 8.5-13.5 7.4-9.9 6.1-6.7 20-272 11.0-16.9 7.9-9.4 6.1-6.7 21-263 12.5-15.5 7.4-8.6 6.0-6.6 21-274 12.6-16.0 6.8-8.8 6.1-6.6 22-25

*Treatments are described In Section 4.0, Table 4.2.

5-2

Table D.3. Smmary of Results of Routine Physical Monitoring of Set II Run 1,November 21 through November 23, 1980, Lake Lanier, Georgia

Treatment* Day of Temperature DO pH ConductivityTest "C pp. ushoo/cm

A 1 11.5-12.5 6.4-8.4 5.8-6.4 33-572 11.8-13.0 7.0-8.7 5.7-6.2 35-39

3 1 11.0-14.0 7.2-9.7 6.0-6.4 22-272 11.5-13.0 6.7-9.7 5.7-6.4 24-27

C 1 12.0-13.0 6.7-8.3 6.1-6.6 37-402 11.5-13.2 6.6-9.4 5.9-6.5 36-40

D 1 12.0-14.0 6.6-9.3 6.0-6.4 31-352 12.0-13.5 7.1-9.8 5.7-6.4 30-35

E 1 12.0-13.0 5.8-7.6 5.8-6.2 39-412 12.0-13.5 6.5-8.6 5.7-6.4 39-42

F 1 12.0-13.0 5.6-8.5 5.8-6.3 32-372 12.0-12.5 6.6-9.6 5.8-6.5 31-36

G 1 11.5-13.0 5.6-8.3 5.9-6.2 34-382 12.0-13.0 7.1-9.6 6.0-6.6 34-37

H 1 11.5-12.0 6.4-7.6 5.9-6.5 34-382 11.8-12.5 6.5-9.1 5.5-6.5 35-39

*Treatments are described in Section 4.0, Table 4.2.

b-3

A:

Table 3.4. Suoary of Results of Routine Physical Monitoring of Set II Run 2,November 23 through November 24, 1980, Lake LAnier, Georgia

Treatment* Day of Temperature DO PH ConductivityTest 0C PPM mhos/ca

A 1 12.0-13.0 6.0-8.2 5.9-6.3 31-35

3 1 11.5-12.5 6.9-9.4 6.1-6.4 22-25

C 1 12.0-13.0 7.0-7.9 6.1-6.5 44-50

F 1 12.0-12.5 6.6-8.4 5.9-6.6 20-25

G 1 12.5-13.0 6.8-8.4 5.8-6.6 29-37


1'-4

Table B.5. Sumary of Results of Routine Physical Monitoring of Set II Run 3,December 4 through December 8, 1980, Lake Lanier, Georgia

RangesTreatment* Day of Temperature DO pH Conductivity

Test 6C ppm iUhos/cn

A 1 12.0-13.5 7.4-8.4 5.6-6.5 35-392 12.0-13.5 7.8-8.8 5.5-6.4 35-403 12.5-13.0 7.8-9.2 5.5-6.5 36-404 12.5-13.2 8.3-8.8 5.1-6.0 35-40

B 1 10.2-13.5 8.3-10.4 5.5-6.6 21-252 10.0-13.0 7.8-10.4 5.5-6.2 22-293 10.2-14.0 8.0-10.6 5.4-6.9 22-284 11.0-15.0 8.1-9.5 5.5-6.4 24-28

C 1 12.0-14.5 7.9-9.6 5.3-6.2 102-1092 12.4-13.5 7.9-9.4 5.8-6.4 104-1153 12.5-13.7 8.4-9.5 5.6-6.3 105-1154 12.4-13.8 7.9-9.1 5.6-6.3 110-115

D 1 12.3-14.8 7.6-9.0 5.5-6.4 32-382 11.5-13.5 8.4-9.6 5.7-6.5 32-463 13.0-13.7 8.0-10.4 5.8-6.4 35-394 13.0-13.8 8.6-9.9 5.6-6.3 35-39

E 1 11.8-14.0 6.7-9.0 5.9-7.0 42-502 11.5-13.0 7.9-9.4 6.0-6.5 47-583 12.5-12.8 7.7-9.1 6.0-6.6 48-514 12.5-13.0 7.5-8.3 6.0-6.5 47-49

F 1 11.8-14.0 6.9-9.2 5.9-7.1 29-322 12.0-12.5 7.9-8.8 6.2-7.0 32-33

G 1 11.9-13.5 7.1-8.9 5.8-6.7 41-422 11.5-13.0 7.9-9.3 5.6-6.4 42-493 12.3-13.0 8.0-10.5 5.7-6.5 42-484 12.5-13.0 7.7-9.0 5.7-6.3 43-48

H 1 11.9-13.5 7.2-8.8 5.6-6.5 29r322 12.0-13.0 7.9-9.6 5.7-6.6 29-333 12.3-13.0 8.2-10.3 5.7-6.4 30-354 12.4-13.0 8.2-9.0 5.7-6.4 31-33


Ow

1-5

mov

Table B.6. Summary of Results of Routine Physical Monitoring of Set II Run 4,December 19 through December 23, 1980, Lake Lanier, Georgia

RangesTreatment* Day of Temperature DO pH Conductivity

Test 6C pp. Uahos/ca

A 1 11.0-13.0 9.1-10.0 6.2-6.7 31-392 8.0-12.0 9.8-10.5 6.0-6.6 31-433 7.7-9.7 9.9-11.2 5.8-6.8 29-374 7.9-10.1 10.0-10.9 6.2-6.5 31-37

B 1 9.8-13.0 9.1-10.4 6.4-6.8 21-252 6.0-12.2 9.8-11.5 6.1-6.7 20-243 4.5-8.0 10.5-12.8 6.1-6.6 21-234 6.2-9.6 10.2-11.8 6.2-6.5 22-23

E 1 11.0-12.1 8.9-10.1 6.2-6.8 31-382 8.5-11.4 9.8-11.0 6.2-6.5 31-373 7.0-9.8 10.5-11.8 5.9-6.9 32-354 8.8-11.7 9.9-11.4 6.0-6.4 31-36

F 1 10.5-12.0 9.0-10.2 5.8-6.7 49-512 8.0-10.0 9.8-10.5 6.2-6.4 40-523 6.0-8.6 9.8-11.2 6.1-6.6 32-504 8.0-9.6 9.8-11.3 6.0-6.4 49-52

G 1 10.0-11.2 8.7-10.1 5.9-6.7 72-802 7.5-10.2 9.6-10.4 6.2-6.6 35-85**3 6.0-8.7 9.8-11.1 6.2-6.6 32-70**4 7.0-9.3 9.6-11.3 6.1-6.5 70-75

H 1 10.0-11.0 9.0-10.0 6.2-6.7 103-1122 7.0-10.1 9.8-10.8 6.3-6.9 60-118**3 6.0-9.1 9.8-10.9 6.2-6.6 39-110*4 8.0-10.3 9.4-11.4 6.2-6.4 102-120

* Treatments are described in Section 4.0, Table 4.2.**The large variations in conductivity for these days are due to single low

values when the stock reservoirs for hardness additions ran dry. Stockdepletion occurred once only, but for the last reading of day 2 and the firstreading of day 3.

5-6

APPENDIX C. RIVERINE BIOASSAY DATA

Table C.1. Physical Chemical Data for Riverine Bioassay Stations, Run 1,November 22-26, 1980, Chattahoochee River/Lake Lanier, Georgia

Date Temp DOStation (oC) (ppm) pH

November 22, 1980 1 12.5 7.2 6.42 10.4 7.5 6.13 11.0 1.3 5.94 11.0 8.9 6.25 11.0 5.1 6.06 11.0 8.1 6.37 10.5 8.5 6.68 10.5 9.2 6.6

November 23, 1980 1 12.0 8.4 6.42 10.5 8.4 6.13 11.0 3.0 6.44 10.8 9.2 5.35 10.5 5.4 5.36 10.5 6.5 6.57 10.0 7.3 6.48 9.5 8.8 6.2

November 24, 1980 1 13.0 7.4 6.72 10.2 8.0 6.53 11.0 2.0 6.24 12.0 9.0 6.45 11.5 5.1 6.46 11.5 6.0 6.57 11.5 6.5 6.38 12.0 8.4 6.0

November 25, 1980 1 12.0 8.0 6.72 10.0 7.2 6.23 12.0 1.1 6.24 12.2 10.2 6.25 12.2 5.4 6.36 11.0 7.2 6.37 11.0 7.7 7.08 12.0 7.6 6.5

November 26, 1980 1 12.0 7.8 6.52 10.0 8.2 6.43 11.0 3.0 6.04 10.0 9.9 6.25 10.0 7.4 6.36 11.0 6.8 6.57 10.0 8.0 6.08 11.0 8.9 5.6

4

c-I

Table C.2. Physical Chemistry Data for Riverine Bioassay Stations, Run 2,February 5-9, 1981, Chattahoochee River/Lake Lanier, Georgia

ConductivityDate Station Tep -C DO ppm pH (umohs/cm)

February 5, 1981 1 5.0 12.5 6.7 302 6.0 14.0 7.1 303 6.0 13.5 7.4 254 6.0 12.2 7.4 275 6.0 11.2 7.3 266 6.0 11.0 7.3 267 6.0 11.0 7.3 258 5.3 11.2 7.3 25

February 6, 1981 1 6.3 11.5 7.2 3102 6.0 12.0 7.3 3503 6.5 11.0 7.2 5004 6.0 11.5 7.2 5005 6.0 11.5 7.2 5506 5.5 11.5 7.3 8007 5.0 11.0 7.3 1,6008 5.0 11.0 7.1 1,600

February 7, 1981 1 5.0 12.0 7.3 902 6.0 11.0 7.2 973 6.0 11.0 7.25 974 6.1 11.2 7.3 935 6.2 11.2 7.15 1006 6.0 11.0 7.15 1127 5.8 11.5 7.2 988 5.5 11.5 7.2 110

February 8, 1981 1 6.3 11.0 6.9 1082 6.2 12.0 7.4 1023 6.1 11.4 7.3 1014 6.2 11.0 6.2 1025 6.5 11.0 7.3 1046 6.2 11.0 7.3 1087 6.9 11.2 6.9 1098 6.8 11.2 7.2 110

February 9, 1981 1 6.5 11.0 7.3 1372 6.0 11.0 7.3 1303 6.0 11.0 7.3 1324 6.2 11.0 6.5 1205 6.9 11.0 7.2 1216 5.2 11.5 7.3 1127 5.0 11.0 7.2 988 5.0 11.5 6.9 26

C-2

t Table C.3. Survival Date (Salmo gairdneri Svim-Up Fry) from Riverine BioassayRun 1, November 22 through November 26, 1980, Chattahoochee River/Lake Lanier, Georgia

Station I Day I Day 2 Day 3 DayReplicates

1 10 10 10 102 10 10 10 103 10 10 10 104 10 10 10 105 10 10 10 106 10 10 10 107 10 10 10 108 10 10 10 10

Station 2 Day I Day 2 Day 3 Day4Replicates

1 8 2 1 12 7 1 0 03 7 2 0 04 8 1 1 15 6 3 2 26 9 0 0 07 8 2 1 08 8 3 3 1

Station 3 Day 1 Day 2 Day 3 Day 4Replicates

1 0 0 0 02 0 0 0 03 0 0 0 04 0 0 0 05 0 0 0 06 0 0 0 07 0 0 0 08 0 0 0 0


1 8 2 1 12 7 1 0 03 7 2 0 04 8 1 1 15 6 3 2 26 9 0 0 07 8 2 1 08 8 3 3 1

4

(Continued)

C-3

Table C.3. Survival Data (Salao gairdneri Swim-Up Fry) from Riverine BioassayRun 1, November 22 through November 26, 1980, Chattahoochee River/Lake Lanier, Georgia


1 6 2 0 02 8 0 0 03 7 1 1 04 7 1 0 05 7 0 0 06 7 0 0 07 8 0 0 08 8 0 0 0

Station 6 Day I Day 2 Day 3 Day 4Replicates

1 6 0 0 02 7 2 1 03 7 0 0 04 9 0 0 05 10 1 0 06 9 0 0 07 7 1 0 08 10 0 0 0


1 8 0 0 02 8 2 1 13 8 2 1 14 9 0 0 05 9 1 0 06 9 2 2 27 9 1 1 08 9 0 0 0


1 8 2 1 12 9 3 2 23 9 3 3 34 10 2 2 25 10 2 2 26 10 4 4 37 10 2 2 28 10 4 4 3

Note:

1. Values represent accumulative number live by day.2. Species: Salmo gairdneri (Fry).3. Each station started with 8 replicates of 10 fish each.

C-4

_ - -. ... - ,... -... - - - - -- - 77777., - _-. ... . -M, .- -.... . T T , ., ,,

Table C.4. Survival Data (Salmo g4ardneri 15-Centimeter Trout) from RiverineBIloassay Run 1, mber 22 through November 26, 1980,Chattahoochee River/Lake Lanier, Georgia

Station I Day 1 Dy2 Day 3 Day 4e-plIcates

1 10 10 10 102 10 10 10 103 10 10 10 10


1 7 6 3 12 7 4 2 13 10 7 4 3


1 0 0 0 02 0 0 0 03 0 0 0 0


1 10 10 9 62 10 10 10 83 10 10 10 6


1 9 8 8 82 9 9 8 63 10 10 10 8


1 8 8 8 72 10 10 10 83 10 10 10 10


1 10 10 10 92 10 10 10 103 10 10 10 10


1 6 6 6 62 10 10 10 103 10 10 10 10

Notes:I. Species: Salmo gairdneri (15-c. size).2. Each station"ta- ea IK 3 replicates of 10 fish each.

C-5

Table C.5. Daily Survival Data for Rainbow Trout in February 1981 and Blue-gills in November 1980 during Riverine Bloassays in the Cbattahoo-cbee River Below Buford Dam, Lake Lanier, Georgia

Nmber AliveStation 24 hour 168 hour 72 Hour Yb four

Bluegills (began with 30) November 22-26, 1980(no bluegill bioasaay in February)

1 30 30 30 302 30 30 30 303 8, 9, 10* 27 27 274 30 30 30 305 30 30 30 9, 9, 10*6 30 9,1 0, 10* 29 297 30 30 30 308 30 30 30 30

Rainbow Trout Fry (began with 80) February 5-9, 1981

1 80 80 80 802 80 80 80 803 80 80 80 804 80 80 80 805 80 80 80 806 80 80 80 807 80 80 80 808 80 80 80 80

Rainbow Trout Yearlings (began with 30) February 5-9, 1981

1 30 30 30 302 30 30 30 303 30 30 30 304 30 30 30 305 30 30 30 306 30 30 30 307 30 30 30 308 30 30 30 30

Notes:*Three ambers list survival in each of 3 replicates.Station numbers:1. Control2. Sluice at dam3. 300 a below dam4. latchery raceway5. River at hatchery6. Settle's Bridge7. McGinnis Bridge8. Abbott's Bridge

C-6

r.,. L ,

16 oCD. 0

0'0 SIW *00 U-40 In 0 &n 0 0 0%- 0 n m q

I~e - 0 -''-40 ha

Mi' 0 0Crih

* C" 0 r %a -4 C C 0 " & two C4 'P, e aD Go N WIm 0we 0 X

D.~

CWU4 r 401~~4 .nO)O r.enm 0

W 4) lu r

60 w "W .ha '.4p

a0 CS. 0D %Dm0 T% 4C 4

"4.4. 0. 060 in a a0

01. 0 fl0 0 0 0If -4 W ha 41

0S 0 w,41 -4

.Ff 0 0 Ln n In W% C W 141 0 v

14 0 0 .

~A -4 4

08 0 0 0n 80 0 n -4%a 0% r4CM P r IICi ~Cg',.O4 i C'~dh01 -4 4

0p CI z 0 Vl J2 Wh0 10~~ ~ 00 0 0000 W. . 0-s

a~~. -W W

W~ .

1a. ~A 6 OW

~080W4 aA0 A

Aj ap

C-7~negj

00. 4 -4 . . . .4

d0%

V &Ja1

6O 0.

D.

061 0.0

a

C-

AD-All7 279 JONES EDMUNDS AND ASSOCIATES INC … · ad-all7 279 jones edmunds and associates inc...

Documents

Transcript of AD-All7 279 JONES EDMUNDS AND ASSOCIATES INC … · ad-all7 279 jones edmunds and associates inc...