Evaluation of Environmental Stress Imposed by a Coal Ash Effluent Wisconsin Power Plant Impact Study


? B t; 5 - 2 1 3 f4 2 9

EPA/600/3-85/045
June 1985

AN EVALUATION OF ENVIRONMENTAL STRESS IMPOSED BY A COAL ASH EFFLUENT
Wisconsin Power Plant Impact Study

by

Katherlne I, Webster
Anne M. Forbes
John J. Magnuson
Department of Limnology
University of Wisconsin-Madison

Grant No. R803971

Project Officer

Gary E. Glass
Environmental Research Laboratory-Duluth
Duluth, Minnesota

This study was conducted in cooperation with
Wisconsin Power and Light Company,
Madison Gas and Electric Company,
Wisconsin Public Service Corporation,
Water Resources Center
and Wisconsin Department of Natural Resources

ENVIRONMENTAL RESEARCH LABORATORY-DULUTH
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
DULUTH, MINNESOTA 55804

-------
TECHNICAL REPORT DATA
fPleae read insrmcnons on the reverie before completing)

1, P EPORT NO. 2.

EPA/600/3-85/045

3. RECIPIENT'S ACCESSION NO.

4. T.TLc AND SUBTITLE

An Evaluation of Environmental Stress Imposed by a Coal

Ash Effluent ;

Wisconsin Power Plant Impact Study

5. REPORT DATE

June 1985

6. PERFORMING ORGANIZATION CODE

7.AUTHORIS)

K. E. Webster, A. M. Forbes, and J. J. Magnuson

B. PERFORMING ORGANIZATION REPORT NO.

3. PERFORMING ORGANIZATION NAME AND ADDRESS

Center for Limnology
University of Wisconsin-Madison
Macison, Wisconsin 53706

10. PROGRAM ELEMENT f»0.

1 1. CONTRACT/GRANT NO.

803971

12, SPONSORING AGENCY NAME AND ADDRESS

Environmental Research Laboratory
U.S. Environmental Protection Agency
Office of Research and Development

DuLuth, Minnesota 55804

13. TYPE OF REPORT AND PERIOD COVERED

14. SPONSORING AGE NCV CODE

EPA-600/03

IS- SUPPLEMENTARY NOTES

16. abstract

Effluent discharged from the coal ash settling basin of the Columbia Generating
Station (Wisconsin) modified water chemistry (increased trace metal concentrations,
suspended solids and dissolved materials) and substrate quality (precipitation of
chemical floe) in the receiving stream, the ash pit drain. To test the hypothesis
that habitat avoidance could account for declines in macroinvertebrate density
observed after discharge began, drift rates of two species were measured in laboratory
streams containing combinations of reference and coal ash modified substrate and
water. Contrary to the hypothesis, drift was uniformly lower in laboratory streams
containing modified substrate and/or water compared to the reference condition for
Gamnarus pseudolinmaeus and Asellus racovitzai. No preference for the modified
condition was shown in a separate substrate-choice experiment.

1 7.

KEY WORDS AND DOCUMENT ANALYSIS



J DESCRIPTORS

b. IDENTIFIERS/OPEN ENDED TERMS

c. cosati Field/Group







'.E Z STRIBUTION' STATEMENT



T9 SECUFtlTV CLASS ITius Report,

Unclassified

21- NO. OP pages

66

Release to public



20 SECURITY CLASS fjmspaxt/

Unclassified

22. PRICE

E Pa 5220-1	4-77) previous edition j * oo^OLC t£

i

-------
DISCLAIMER

This report has been reviewed by the Environmental Research Laboratory-
Dulutb, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views
and policies of the U.S. Environmental Protection Agency, nor does mention
of trade names or commercial products constitute endorsement or recommenda-
tion for use.

ii

-------
FOREWORD

The U.S. Environmental Protection Agency (EPA) was designed to

coordinate our country's efforts toward protecting and improving the
environment. This extremely complex task requires continuous research in a
multitude of scientific and technical areas. Such research is necessary to

monitor changes in the environment, to discover relationships within that
environment, to determine health standards, and to eliainate potentially
hazardous effects.

One project, which the EPA is supporting through its Environmental
Research Laboratory in Duluth, Minnesota, is the study "The Impacts of Coal-
Fired Power Plants on Che EnvironmentThis Interdisciplinary study,
centered mainly around the Columbia Generating Station near Portage, Wis.,
involves investigators and experiments from many academic departments at the
University of Wisconsin and is being carried out by the Environmental
Monitoring and Data Acquisition Group of the Institute for Environmental
Studies at the University of Wisconsin-Madison. Several utilities and State
agencies are cooperating in the study: Wisconsin Power and Light Company,
Madison Gas and Electric Company, Wisconsin Public Service Corporation, and
Wisconsin Department of Natural Resources.

The research presented in this report examined the hypothesis that
habitat avoidance could account for declines In macroInvertebrate density
observed after effluent began being discharged from the Columbia Generating
Station. Drift rates of two species were measured in laboratory streams
containing combinations of reference and coal ash modified substrate and
water.

Norbert A. Jaworsk.1
Director

Environmental Research Laboratory

Hi

-------
ABSTRACT

Effluent: discharged from the coal ash settling basin of the Columbia
Electclc Generating Station (Wisconsin) modified water chemistry (resulting
in increased concentrations of trace metals, suspended solids and dissolved
materials) and substrate quality (precipitation of a chemical floe) in the
receiving stream, the ash pit drain. To test the hypothesis that habitat
avoidance could account for declines in macroinvertebrate density observed
after discharge began, drift rates of two species were measured in
laboratory streams containing combinations of reference and coal ash
modified substrate and water. Contrary to the hypothesis, drift was
uniformly lower in laboratory streams containing modified substrate and/or
water compared to the reference condition for GcuvmaruB pseudolimrtaeus and
Asellus -racovitzai . No preference for the modified condition was shown in a
separate substrate-choice experiment.

In a partial translation of the laboratory experiment to the field,
changes in macroinvertebrate distributions at two ash pit drain sites were
measured following resumption of coal ash discharge after a 2-week pause.
Findings were contradictory. At one site, densities increased after
discharge began, while at the other site farther downstream, densities
declined. The upstream site may have provided a refuge for Invertebrates
dislodged from upper reaches of the ash pit drain.

Data on macroinvertebrate community structure from this and other
studies was used to evaluate environmental stress at a site receiving coal
ash effluent. The following sequence was documented: (1) decline in total
abundance and taxa 4 months after effluent discharge began in 1974; (2) loss
of most macro invertebrates in 1977; (3) recovery of the community by 1980,
although a shift in dominance to more tolerant species, lower total
abundance and a slight decline in diversity, suggested continued sublethal
influence of the coal ash effluent.

The variation in the severity of stress suggested by macro invertebrate
community responses was attributable to a series of generating station
activities related to coal treatment, effluent discharge characteristics,
and dredging of accumulated floe in the upstream drainage ditch. When
macro invertebrate communities were most severely impacted in 1977, effluent
discharge and effluent concentration, indicated by conductivity, were at
their maximum. The effluent response threshold (1000 umbos conductivity)
proposed in an earlier study, accurately predicted the recovery of ash pit
drain macroinvertebrates when effluent concentrations fell below threshold.

iv

-------
CONTENTS

Page

Foreword	»»			ill

Abstract				lv

Figures.						vi

Tables. .............................	,.			vii

Acknowledgments 								ix

1.	Introduction					.........	1

2.	Site Description....						3

Columbia Electric Generating Station......				3

Alterations of Stream Habitats...............................	5

3.	Behavioral Responses of Stream Macroinvertebrates to the

Coal Ash Effluent..........................			8

Introduction.						8

Methods 						9

Results							 . 				14

Discussion.											17

4.	Macroinvertebrate Responses to Intermittent Coal Ash Effluent
Discharge				19

Introduction.										19

Methods		.........			20

Results								..			22

Discussion										26

5.	Effects of the Coal Ash Effluent on Macroinvertebrate

Community Structure..........					31

Introduction.								31

Methods					32

Results					..			»,	35

Discussion.				40

6.	Evaluation of a Variable Environmental Stress in the

Columbia Coal Ash Drainage System..........		 					46

Introduction...... —					46

Influence of Generating Station Activities		46

Chemical and Biological Indicators......			49

Discussion and Summary										50

References 		.........					51

v

-------
FIGURES

Number	Page

1	Columbia Electric Generating Station				4

2	Laboratory stream chambers					10

3	Plexiglas chambers used in substrate choice experiment.....		13

4	Results of laboratory drift study....			15

5	Daily coal ash effluent discharge Into ash pit drain........		21

6	Changes in conductance, total filterable solids, macroinverte-
brate density and number of taxa upon resumption of coal ash

effluent discharge,,.,.					25

7	Taxonomic composition of artificial substrate samplers..........	36

8	Abundance of selected taxa.						38

9	Shannon-Weaver diversity, evenness, number of taxa, and number

of individuals..............					39

10	Comparison between conductivity, macroinvertebrate abundance

and number of taxa at upstream and downstream sites.............	41

11	Conductivitys effluent discharge rate and frequency between,

1974 and I960					47

vi

-------
1

2

3

4

5

6

7

8

9

10

11

12

TABLES

Page

Typical physical and chemical measurements at site AFD-1,
downstream of the effluent and at the upstream site, APD-3... 5

Concentrations of selected trace elements at the point of

effluent discharge at site APD-1 and at site APD-3 compared
to toxicity thresholds..						 6

Treatment combinations tested in the eight laboratory

streams .........................		 11

Significant main effects and interactions influencing drift

of Asellus vacovibzai, Ganvnams peeudolimnaeue and

Corlxidae in laboratory streams			 16

Median of a series of physical and chemical parameters

measured at sites APD-1, APD-2, and RRC before and after

effluent discharge began.*			— 		 23

Correlation between daily effluent discharge rate and a

series of chemical and physical parameters measured at

sites APD-1 and APD-2									24

Taxonoraic composition of ponar samples collected at the two
experimental sites and at the reference site			 2?

Comparison of two models describing macroinvertebrate

response to resumption of coal ash effluent discharge	 29

Results from multiple comparison test..,,.,.......,.,,		 29

Dates showing intervals for macroinvertebrate colonization
of artificial substrate samplers in the ash pit drain........ 32

Comparison of macroinvertebrate taxa colonizing basket and
modified Dendy samplers at site APD-1			 34

Summary of experimental studies investigating macroinverte-
brate responses to coal ash effluent.......					42

vii

-------
Numbef	Page

13	Summary of observations on conductivity and extent of floe
deposits at site APD-L and a set of generating station
activities potentially influencing ash pit drain habitats.... 48

14	Estimates of severity of enviromental stress at APD-1

based on chemical and biological indicators	 49

viii

-------
ACKNOWLEDGMENTS

Jim Koenig, Paula Kuzmarski, Waiter Gauthier, and Jean Brussock helped
in the field and laboratory. Walter Gauthier built Che laboratory stream
chambers with the help from Glen Lee, who also assisted with the construc-
tion of many laboratory apparati. Stephen Lozano, Dennis Rondorf, and Bill
Horns assisted in experimental design and statistical analysis. William
Hilsenhoff, Cynthia Duszkowski, William Tonn, Thomas Frost, and Joseph
Eilers provided advice and new insights into the data set. Cheryle Hughes
drafted many of the figures.

Financial assistance was provided by the U.S. Environmental Protection
Agency (Grant No. R803971) with additional funds from the University of
Wisconsin-Madison, Madison Gas and Electric Company, the Wisconsin Public
Service Corporation and the Wisconsin Public Service Commission. This
assistance, as well as that provided by the Water Resources Center,
University of Wisconsin, is gratefully acknowledged.

ix

-------
SECTION 1

INTRODUCTION

Discharge of coal ash supernatants from wet storage ponds at coal-fired
power plants can have a significant impact on aquatic ecosystems (Cherry
et al. 1979, Coutant et al. 1978, Grossman et al. 1974). Acute and sub-
lethal impacts on aquatic organisms can occur in response to one or more of
the stresses typically associated with coal ash effluent discharge: (1) the
addition of elements to the ecosystem, including some potentially toxic
heavy metalsf (2) increases in suspended solids, (3) changes in pH and
chemical balance, (4) sedimentation of chemical solids, and (5) absorption
of P by fly ash interfering with nutrient cycling (Guthrie et al. 1982, Roy
et al. 1981).

The Columbia Electric Generating Station near Portage, Wisconsin,
consists of two coal-fired operating units, each capable of producing 52 7
MW/day. Fly ash from Unit I and bottom ash from both units are slurried
into a series of settling basins. The coal ash effluent is discharged into
a stream, the ashpit drain, situated in a floodplain of the Wisconsin
River. This discharge severely modified stream habitats through
precipitation of a chemical floe and increases in suspended and dissolved
materials, including some toxic heavy metals (Magnuson et al. 1980a; Andren
et al. 1977).

Habitats believed to receive severe Impacts included a 5 km stretch of
the ash pit drain (Forbes et al. 1981). The structure of the ash pit
drain's macroinvertebrate community, first monitored prior to generating
station operation, showed dramatic declines in density and species richness
in response to these environmental stresses (Magnuson et al. 1980a).
Hypotheses proposed to explain these responses were tested in a series of
laboratory and field manipulations (Forbes et al. 1981; Forbes and Magnuson
1980; Magnuson et al. 1980 a,b). Direct toxicity and sublethal responses
associated with the uptake and accumulation of toxic metals, reductions in
food quality, and declines in metabolism, were observed.

None of the mechanisms explored in the previously described studies
fully explained the declines in field distributions. One of the purposes of
the research reported here was to Investigate a third hypothesis based on a
behavioral response of habitat avoidance. In one study, the drift responses
of stream invertebrates to coal ash chemicals were measured in laboratory
streams (Section 3) (Webster et al. 1981). A second study (Section 4)
attempted to verify the laboratory drift results in the field by observing

1

-------
changes ta macro invertebrate distributions following resumption of coal ash
discharge after a 15-day pause ta effluent discharge.

The final two sections evaluate the severity of environmental stress at
one ash pit drain site. Section 5 extends the study conducted by Magnuson
et al» (1980a), by comparing their 1974, 1975 and 1977 macroinvertebrate
distributions with the recovery observed in 1980. Community and species
responses are examined and compared with examples from other pollutions!
stresses. Throughout the 7-yr study period (1974-80), observations on
macro invertebrate distributions and chemical measures of habitat modifi-
cation, indicated a variable intensity of environmental stress. At the sane
tine studies were being conducted, effluent discharge characteristics,
methods of coal and effluent treatment, and other generating station
activities were changing. The Impact of these activities on the severity of
environmental stress observed In the coal ash drainage system Is discussed
in Section 6.

Much attention has been focused on other serious environmental problems
associated with the operation of coal-fired power plants, including
emissions of sulfur oxides that contribute to acidic deposition, release of
thermal discharges, entrainroent, and chlorine toxicity. Environmentally
sound disposal of two of the major byproducts of coal consumption — power
plant aggregate (bottom ash and slag) and fly ash — is likely to become an
equally pressing concern as the volume of waste being generated increases.
Although the Clean Water Act of 197 7 prevents newly constructed coal-fired
power plants from using wet storage of ash residues, this method of disposal
was prevalent prior to passage of the act. The results of this research
support the need for stricter control on the amount of effluent discharged
into lotic environments by these older power plants.

2

-------
SECTION 2

SITE DESCRIPTION

COLUMBIA ELECTRIC GENERATING STATION

The Columbia Electric Generating Sttion was constructed in 1972 on 1100
ha of Wisconsin River floodplain near Portage, Wisconsin (Figure 1). Two
operating units each produce 527 MW of electricity, burning 5,000 to 10,000
tons of low S western coal per day. Units I and II began operation in April
1975 and March 1978, respectively.

The low S pulverized coal typically contains 7 to 8% ash. Approxi-
mately 981 of the fly ash residue is collected by high energy electrostatic
precipitators located In the stacks. Fly ash from Unit I is deposited as a
slurry into Sector I of a 30-ha settling basin north of the generating
station. Fly ash from Unit II is disposed of dry. Water from Sector I
drains directly through the dike walls into the final settling pond (Sector
III). Water from Sector II, which receives bottom ash slurried from the
boilers of both units, is circulated back into the plant and pumped into
Sector III when water levels are excessively high, usually only in winter.
Prior to the summer of 1977, the settling basin was divided into a primary
settling pond, which received fly ash and bottom ash, and a secondary
settling pond.

High concentrations of reactive metal oxides raise the pH of water in
the final settling basin to 10 or 11 (Andren et al. 1977 ). To meet
Wisconsin stream water quality standards the effluent is acidified upon
discharge into a shallow ditch that flows along the eastern boundary of the
cooling lake (Figure I). About 2 km downstream, the effluent ditch joins
the ash pit drain, a creek draining a wetland and mint farm east of the
generating station site. The ash pit drain flows through an arcifically
ditched channel along the east and south borders of the cooling lake. In
contrast to ash pit drain reaches upstream from the effluent which are
characterized by unconsolidated organic substrates, substrates in the
ditched area are predominantly sandy with small patches of water grasses and
sedges. Depth Is relatively constant across the width of the stream
(approximately 5 m) due to the ditched stream bed. Riffles are rare, stream
banks are grassed with a few small shrubs. Approximately 5 tan from the
point of discharge, the effluent enters a wetland and floodplain forest
where It joins Rocky Run Creek (RRC), a tributary to the Wisconsin River.
This creek originates in a marsh lake and flows through 20 km of agricul-
tural land before reaching the generating station site (Magnuson et al.
1980a).

3

-------
Figure 1. Columbia Electric Generating Station study site. Sampling
locations marked with closed circles. Arrows denote direc-
tion of flow. The star marks the point of coal ash effluent
discharge from the settling basin into the effluent ditch
(dashed line) which joins with the ash pit drain (solid
line).

4

-------
ALTERATIONS Of STREAM HABITATS

Water Quality

Magnuson et al. (1930a) observed alteration of several water quality
parameters due to coal ash chemicals- Their 1977 results are summarized in
Table 1. In 1977 the downstream site (APD-1) had higher conductivity and
turbidity but lower alkalinity than the upstream site (APD-3). Temperature
and pH increased slightly at APD-1. Concentrations of the trace elements
Cr, Ba, Al, Cd, and Cu in dissolved form and Cr, Al and Ba In samples of
suspended particulates were higher at downstream sites compared to upstream
sites in 1977 (Table 2) (Andren et al, 1977, Helmke et al. 1976a,b).
Concentrations of Cd, Cr, and Cu approached water quality criteria for
aquatic life suggested by the U.S. Environmental Protection Agency (1976,
1980).

TABLE 1. TYPICAL PHYSICAL AND CHEMICAL MEASUREMENTS AT SITE APD-1,

DOWNSTREAM OF THE EFFLUENT AND AT THE UPSTREAM SITE, APD-3. DATA
BASED ON TWO SAMPLES COLLECTED IN SEPTEMBER AND OCTOBER 1977
(MODIFIED FROM MAGNUSON ET AL. (1980a)



Parameter (Units)

Downstream Site
APD-1

Upstream Site
APD-3

Temperature (C)

21.1

18.0

Current speed (cm/sec)

21.4

8.0a

Dissolved oxygen (ppm)

8,6

7.6

Conductivity (uohos 25C)

2515.0

454.0

Alkalinity (ppm)

93.8

250,2

Hardness (ppm)

130.0

244.4

PH

7.35

7.45

Turbidity

27.0

3.8

aTypical value at site APD-3 on other dates#

Conductivity provided an easily monitored measure o£ gradients of
effluent strength (Forbes et al. 1981). A major source of this conductivity
in the coal-ash effluent was the addition of NaCO^ to pulverized coal to
increase efficiency of fly ash collection in the electrostatic

5

-------
TABLE 2. CONCENTRATIONS OF SELECTED TRACE ELEMENTS (PPM) AT THE POINT OF
EFFLUENT DISCHARGE, AT SITE APD-1, AND AT SITE APD-3 COMPARED TO
TOXICITY THRESHOLDS MODIFIED FROM MAGNUSON ET AL, (1980a)

Element:

Water Quality
Criteria

Ashpit
Discharge

APD-1

APD-3

Dissolved'
A1

Ba
Cd
Cr

Cu

50.0
0.012°
0.100c
0.022d

Suspeaded particulates*

Ba
Cr

0.100-11.4
0.730

0.0021-0.0031

0.014-0.07?

0.004-0.045

17,240
134

0.045-0,476

0.0001-0.0012

0.006-0.028
0.002-0.024

1.294
1.225

0.003-0.165

0.0001-0.0002
0.001

<0,0003-0.002
107

aBa concentrations from Helnke eC al, (1976b); other dissolved elements from

November 1976 to April 1977 observations In Andren et al, (1977).

^Suggested as a toxicity level, but not proposed as water quality criteria

by U.S. Environmental Protection Agency (L976) .
e0.S. Environmental Protection Agency (1976).
dU.S. Environmental Protection Agency (1980).

eHelmke et al. (1976a) fall 1975 data; Al was art important constituent of
suspended particulates, but could not be measured by neutron activation
analysis.

6

-------
precipitators. The extreme conductivities measured where the effluent
joined the ash pit drain (e.g., 3000 ymhos let 197?) progressively declined
as the effluent was diluted by water leaking from the cooling lake (Andrews
and Anderson 1976} and mixed with waters of the ash pit drain, sedge meadow
and Rocky Run Creek. The conductivity of Wisconsin River water was not
affected below the entry of Rocky Run Creek (Magnuson et al. 1980a).

Substrate Quality

Substrates in the ash pit drain prior to plant operation were sil ty
with high organic content (Magnuson et al. 1980a), After generating station
operation began, substrates in the artificially ditched reaches of the ash
pit drain became more sandy as silt was washed away (Magnuson et al, 1980a).
Acidification of the effluent caused precipitation of oxides of elements
such as Al, Ba and Cr into a flocculent material (Andrea et al. 1977). This
floe coated stream substrates at the downstream sites in 1977 and 1978 and
was carried by the current into Rocky Run Creek and the Wisconsin River
(Forbes et al. 1981, Magnuson et al. 1980a)» Extensive areas of the ash pit
drain and the slough near Rocky Run Creek were covered by thick deposits of
floe during low flow conditions in the falls of 1977 and 1978. Spring
flooding removed much of the accumulated layer (Forbes et al. 1981). Floe
deposits were not nearly as extensive in the fall of 1979 having disappeared
from much of the ash pit drain. Deposits were only observed in backwater
pools among patches of grases and sedge growing on the otherwise open sandy
substrates in the main channel.

7

-------
SECTION 3

BEHAVIORAL RESPONSES OF STREAM MACROINVERTEBRATES
TO THE COAL ASE EFFLUENT

INTRODUCTION

Examination of effects of pollution on stream macroinvertebrates has
generally focused on acute toxicity. Considerably less attention has been
directed toward investigating behavioral responses to sublethal habitat
modifications by pollutants. While catastrophic downstream movement or
drift accompanied by depletion of standing crop can result from toxic
pollutant stress (Coutant 1964; Waters 1972), behavioral drift has been
recognized as an important mechanism by which macroinvertebrates can avoid
unfavorable environments (Corkum et al. 1977). In the field, increased
drift of stream macroinvertebrates has been induced by stream acidification
(Hall et al. 1980) and sediment addition (Rosenberg and Wiens 1978). In
laboratory streams, unfavorable substrate type, current velocity, and lack
of food supply have increased the drift response (Otto 1976; Corkum et al.
1977; Walton 1978).

Previous studies documented changes in macroinvertebrate distributions
In stream habitats modified by a coal ash effluent from an electric
generating station in Columbia Co., Wisconsin {Forbes et al. 1981; Magnuson
et al. 1980a). After generating station operation began in April 1975 the
abundance of most taxa inhabiting the ash pit drain declined. In 1977 few-
individuals and taxa colonized artificial substrate samplers, indicating an
even more impoverished community. In Rocky Run Creek, downstream of the
junction with the ash pit drain, colonization of artificial substrates was
reduced only when coal ash effluent concentrations , indicated by conduc-
tivity, were high (Forbes et al. 1981). Since declines in abundance and
species diversity could not be adequately explained by direct toxicity,
field distributions might be explained by habitat avoidance through
increased drift response. Macro invertebrate drift responses to a coal ash
modified habitat simulated in laboratory streams were used to test this
hypothesis. The results of this study have previously been published
(Webster et al. 1981).

Water quality degradation has generally been the focus of studies of
the impacts of pollutant discharges into lotic environments. Howevert the
effluent altered both water and substrate quality in the coal ash drainage
system. Substrate characteristics generally determine macroinvertebrate
habitat preferences, while physical and chemical factors determine habitat
tolerance (Cummins 1966, Cummins and Lauff 1969). Decreases In

8

-------
nacroinvertebrate diversity attributed to declines in water quality have
been more closely correlated with substrate parameters (Ruggiero and
Merchant 1979). The experimental design used was developed to allow

evaluation of the relative importance of substrate, water, and food quantity
changes on the macroinvertebrate drift response.

METHODS

Drift Experiment

Macroinvertebrate drift was measured in laboratory streams containing
combinations of substrate and water collected from two sites, one contam-
inated by coal ash effluent (APD-1) and the other an unaffected reference
site in Rocky Run Creek (RRC-1) (Figure 1). A third treatment of food

addition was included to investigate possible food limitation in the coal
ash effluent-modified substrates.

Laboratory Streams—

The design of the plexiglas laboratory streams (Figure 2) was modified
from Gee and Bartnik (1969). A current generated by airstones circulated
water within each laboratory stream over a channel 11 x 52 cm. Nets of
fiberglass screen collected organisms that drifted to the end of each
channel. Two replicate experiments were run In a controlled environment
room. Temperature (5.0 C) and photoperiod (12L:12D) approximated the field
conditions at the tine of the study in March 1979. Laboratory streams were
illuminated by 15-W fluorescent lamps controlled by a timer set to the
ambient photoperiod.

Macroinvertebrates—

Macroinvertebrates were collected from the reference site RRC-1 in
Rocky Run Creek, upstream of its junction with the ash pit drain, 2 weeks
before each experiment, and kept static in aerated buckets at the
experimental temperature and photoperiod. Representatives from five taxa
were selected: Isopoda (A set I us racovitzai), Amphipoda (Gammarue
pseudolirmaeus), and aquatic insects from the orders Hemlptera
(HeBperocorixa sp. and Sigara sp.) , Ephemeroptera (Leptophlebia sp. and
Callibaetis sp.) and Trlehoptera (Pycnopsyohe sp.).

Substrate, Water and Food Addition—

Macroinvertebrate drift was measured in response to two levels of each
of three treatments — substrate, water, and food addition (Table 3).
Substrates were collected from the reference site and from the modified
habitat 2 weeks before the experiments and were frozen to eliminate
macroinvertebrates. Water was collected from the same two sites on the day
before the experiments. The presence or absence of shredded sugar maple

9

-------
Figure 2* Laboratory scream chamber, Arrows denote the direction of flow

along Che 52-cm stream channel located between the upstream screen
and drift net. Shaded areas do not contain water. This design
was modified Erom that of Gee and Bartnik (1969),

10

-------
Acer" sacchartm leaves, incubated In mesh bags In Rocky Run Creek for 4
months, represented the food addition treatment.

TABLE 3* TREATMENT COMBINATIONS TESTED IN THE EIGHT LABORATORY STREAMS3

Stream

Chamber	Substrate	Water	Food

1

H

M

+

2

R

M

+

3

M

R

+

4

R

R

+

5

M

M

0

6

R

M

0

?

M

R

0

8

1

R

0

aFor the factors substrate and water, M is coal ash modified site and R
is reference site; for food addition + is food added and 0 is no food
added.

Experimental Procedure—

Treatment combinations were randomly assigned to laboratory streams.
Thawed substrates were added to the stream channels to a depth of 2 cm*

After circulating distilled water continuously within the streams for 24 h
to remove easily suspended materials from the substrates, the streams were
drained and filled with the appropriate treatment waters to a depth of 9
cm. Shredded leaf material was partially buried at the surface of the
substrates in those streams that received the food treatment. A total of 3?
individuals (10 Aeellus raaovitzai , 10 Gammams peeudol irnnaeuz , 10
Corlxidae, 5 Ephemeroptera, acid 2 Trichoptera) were placed in the upstream
third of each stream channel and allowed to acclimate to exper iroenta1
conditions for 2 h without current. After starting water circulation, the
drift nets at the end of each channel were emptied and drifting organisms
counted after I h and every 3 h thereafter for 48 h. At the conclusion of
each trial, nondrif ting invertebrates were recovered from the sediments.
Missing individuals were counted as mortalities.

n

-------
Hater Analysis—

Temperature and current speed were monitored throughout the experiment
using a mercury thermometer arid a 4-cm drogue, respectively. Dissolved
oxygen levels were measured by the Winkler Method (American Public Health
Association 1976)• At the end of 48 h water samples from each laboratory
stream and from unused water from the reference and. modified sites were
analyzed for methyl orange alkalinity (American Public Health Association
1976), total dissolved residue (American Public Health Association 1976),
conductivity (YSI Model. 33 Conductivity Meter) and pH (Orion Model 339A pH
meter).

Data Analysts™-

The effects of the three treatments (substrate, water and food) at two
levels each on macroinvertebrate drift was measured using a 2 factorial
design (Box et al. 1978). All treatment combinations (Table 3) were tested
simultaneously in the eight laboratory streams; the entire run was
replicated. Such a design allows pairwise comparison of means within each
treatment or main effect (sample size of eight for each treatment level) as
well as evaluation of interactions between treatments (sample size of four
for each of four possible combinations).

Data for Aeellus racovitzai , Gcurma.rus peeudolimnaeue and corixids were
analyzed separately using a three-way analysis of variance (RUMMAGE VER 3,
Madison Academic Computing Center), with the number drifting as the
dependent variable. Since nearly all Ephemeroptera, Trichoptera, and
corixids had drifted by 48 h in all treatments, the data from these taxa did
not meet the assumptions for Ah'OVA and were not analyzed.

Substrate Choice Experiment

In August 1979 the substrate preferences of GammruB pseudci imnasus
offered a choice between sediments from the ash effluent modified site and
from the reference site were investigated. Plexiglas chambers (30 x 16 x LI
cm) contained substrates on either side of a 2.0-cm high longitudinal
divider (Figure 3). Twenty chambers were randomly placed in respect to a
light source (15-W incandescent lamps) in a constant temperature water
bath. Ten of the chambers received only reference substrates (pretreated by
freezing as previously described) and served as controls. The other 10
chambers contained two substrates randomly assigned to the right or left
side. On one side, 2.0 cm of reference substrate was added; on the other,
ash effluent modified substrates were layered over reference substrates.
All chambers contained fresh reference site water and were aerated•
Experimental conditions of temperature (22C) and photoperiod (14L:10D)
approximated ambient field conditions.

Ten Garmarus pseudolirnnaeue were introduced into each chamber, five per
side. At the end of 48 h the chamber halves were separated by a plexiglas
divider and the Garmna.7*it£ settling on each side were recovered. Data were

12

-------
Ficure 3 Plexiglas chamber used in substrate choice experiment. Different substrates
were placed on either side of a divider. The chamber could be divided in
half by a plexiglas sheet placed inside the chamber divider holder.

-------
pooled separately for treatment and for control chambers and analyzed using
Che chi-square method.

RESULTS

Drift Experiment

Factors Influencing Drift after 48 h—

The hypothesis that aquatic macroinvertebrates would drift at higher
rates in laboratory streams with modified conditions was not supported by
the patterns of drift observed for Gammarus and Asellus. The highest drift
occurred in stream chambers containing substrate and water from the
reference site (Figure 4),

Asellus vacovitzai drifted more in response to food addition compared
to no food addition and in reference substrates compared to modified
substrates. The significant response to substrate should be considered as
part of the substrate-water interaction. Drift was significantly higher In
streams where substrate and water were from the reference site compared to
all other stream chambers, while drift in streans with modified substrate
and reference water was significantly lower.

Except for a non-significant effect of food addition, Gammzrus
pseudolinmaeuB exhibited similar responses. Drift increased In the
reference substrate and the substrate-water interaction was significant.
Again, the highest drift occurred in laboratory streams containing substrate
and water from the reference site.

Drift rates during the first hour were low for Asellus vacovitzai (4%)
and Ganmavus pseudolimnaeus (102). Drift activity of the remaining Asellus
vacovitzai occurred predominantly during the night hours (38%) compared to
the day (5%). In contrast, the Gammarids were collected In roughly
equivalent numbers during light (15%) and dark (13%) periods. In general,
drifting invertebrates in streams exhibit a diel periodicity, with higher
rates during the night (Waters 1972). This pattern was not exhibited by
Sammarus pseudolirmaeus In the laboratory streams.

Changes in Significant Factors Influencing Drift over Time—

To determine whether the factors influencing drift changed over time,
similar analyses of drift were made at 12 and 24 h (Table 4). For Asellus,
no main effects or Interactions were significant at 12 h; by 24 h, drift was
significantly higher in chambers with food added. The influence of
substrate did not become important until 48 h. For Gammarus, no factors
were significant at 12 h; by 24 h the substrate-water and substrate-food
interactions were significant.

14

-------
MAIN EFFECTS

INTERACTIONS

SUBSTRATE WATER

FOOD

s w

Islw

a
E

o
¦C

a o

™ OT

c 44

I«

s»
ja

Asellus racovitzat

* *

lu 1 |iiu

Gommarus pseudotimnoeus

10r

n n

a
o
z

u
z

m
m
u.

CJ
UJ

o
o
z

o
z

o
<

a
o
o

o
w
a
a
-------
TABLE 4, SIGNIFICANT MAIS EFFECTS AND INTERACTIONS INFLUENCING DRIFT OF
ASELLUS RACOVIT7.AI, GAMMARUS PSEUDOLIMMBUSt AND CORIXIDAE IN
LABORATORY STREAMS AFTER 12, 24 AND 48 h. SUB|x|WATER AND
SUB | X | FOOD DENOTE TBI SUB STRATI-WATER AND SUBSTRATE-FOOD
INTERACTIONS, RESPECTIVELY

Macroinvertebrate

Taxon

12 h

24 h

48 h

Asellue

None

Food*

Substrate*

rmovitmi

Food*

Sub|xjWater*

Commands

None

Sub

xlWater*

Substrate*

pseudolinmaeue

Sub

XiFood*

Sub[x|Water*

Corixidae

Water**

Food*

(3)

Pood*

*p < 0.05.
**p < 0.001.

aData are non-normal.

Results of the analysis of the corixid drift response at 12 and 24 h

suggested that Initially (12 h) the main effects of water and food were
Important; only the effect of food was significant at 24 h. Drift in those
cases was higher in reference site water and with food addition, ly 48 h
nearly all Corixidae had drifted in all stream chambers.

Analysis of Mortalities—

There was no evidence for differential mortality between stream
chambers or treatments. Analyses of variance using the numbers of Asellus
and Gcvmarus missing or found dead In each stream chamber after 48 h
revealed only one significant effect (the substrate by food interaction for
Gammarus at p < 0,05). Mortality was highest in chambers with modified
substrate and food addition but was low In all other chambers .

Water Quality—

Water quality measurements corresponded to typical field measurements
from earlier studies (Magnuson et al. 1980a) with the exception of ash pit
drain conductivity. As discussed previously, conductivity was lower In the
ash pit drain after January 1979. For laboratory streams containing water
from reference and modified sites, respectively, average alkalinities were
200 and 100 Eg/liter CaCO-j, conductivities ware 430 and 530 wahos/cm and pHs
were 8-3 and 8.0. Those values were comparable to concentrations measured

-------
in unused water from each site. Dissolved oxygen values were near
saturation in all stream chambers. Total dissolved residue values were
erratic, possibly due to inadvertent siphoning of substrate material into
the sample containers. Temperatures ranged from 4.4 to 5.1 C and current
speeds from 5.2 to 5.8 cm/sec.

Substrate Choice Experiment

The hypothesis tested was that Ganrniarus pseudolimnaeus would display no
preference for modified substrates. Chi-square analysis of data from the
control chambers showed that araphipods did not discriminate between chamber
sides (right vs. left or position in respect to light source) after 48 h.
Likewise, no differences were observed between modified and reference
substrates (X = 0.40 for pooled data from 10 chambers).

DISCUSSION

Drift of the beothic invertebrates Gcmmarus pseudolinmaeue and Asellus
vacovitzai in laboratory streams was influenced by substrate and water
quality. Food addition in the form of detritus played a secondary role. In
contrast, the Corixidae, which usually Inhabit the weedy margins of streams,
drifted in high numbers under all conditions at the end of 48 h; the habitat
provided was not their natural one. Water and food were significant
Influences on early drift rates.

The results do not support the hypothesis that macroinvertebrates avoid
habitats modified by the coal ash effluent through Increased drift. To the
contrary, the drift of Gamnarus pseudolimnaeus and Asellus vacovitzai was
uniformly lower in laboratory streams where substrate, water, or both were
from the modified site. This result suggests an effect due to chemical
constituents of the coal ash effluent present in the sediments and water.

The reduced drift response observed could result from a preference for
the modified over the reference conditions or from a reduction in activity
in modified habitats. Preference for the modified habitat seems unlikely.
When presented a choice Ganrniarus pseudolimnaeus did not distinguish between
modified and reference substrates.

Although no evidence was found for behavioral avoidance of coal ash-
modified habitats in the 1979 laboratory drift study, this cannot be totally
discounted as a partial explanation for the lack of organisms present in
1977. Habitat modifications were less severe in 1979 and a fall survey
demonstrated that a community dominated by Asellus racoviisai and yya.il el a
azteca had recolonized the ash pit drain {Section 4). The coal ash-modified
habitat simulated in the laboratory streams may not have represented the
severity of environmental stress that occurred in 1977 or accurately
reproduced the field condition.

Sublethal effects on arthropod physiology following exposure to the
same coal ash effluent have been documented. Crayfish metabolism, as

-------
measured by oxygen consumption, was significantly lower after 10 weeks of
exposure to ash effluent-modified habitats as compared to controls {Magnuson
et al. 1980a).

Changes in Invertebrate behavior have been observed following exposure
to sublethal concentrations of other pollutants. Maciorowski et al. (1980)
observed a decline in locomotor activity levels of the crayfish (Cambarus
acuminatus) when exposd to sublethal Cd concentrations. They suggest that
disruption of crayfish chemoreceptive sensitivity may have caused the
altered behavioral patterns. Similarly, the coal ash effluent which
contains sublethal concentrations of several heavy metals (Table 2) may have
depressed the physiological activity of the stream invertebrate studies.

If behavior can be interpreted as a manifestation of underlying
biochemical and physiological responses to environmental contaminants
(Sherer 1976), the results of this study provide further evidence of
sublethal stress imposed by the coal ash effluent. Sublethal concentrations
of toxic materials may have a more devastating impact on benthic communities
than acute pollutants, affecting not only behavior, but also fecundity,
longevity and reproduction (Lehmkuhl 1979).

-------
SECTION 4

MACROINVERTEBRATE RESPONSES TO
INTERMITTENT COAL ASH EFFLUENT DISCHARGE

INTRODUCTION

The laboratory drift study (Section 3) demonstrated a reduction in
drift activity of macroinvertebrates exposed to coal ash chemicals. This
result did not support the hypothesis of habitat avoidance proposed to
explain the lack of macro invertebrates at severely stressed sites in 1977.
In 1979 when the laboratory drift study was performed, effluent concentra-
tions were below the 1000 ymhos conductivity threshold observed for a series
of acute and sublethal responses measured in 1977 (Forbes et al. 1981) •

Thus, while the drift study may not have been a true test of a mechanism
operating earlier, it did provide evidence for sublethal impacts on activity
at concentrations below the effluent response threshold. Later observations
of macroinvertebrate community structure support this contention: although
populations recovered when effluent concentrations declined, the structure
was characteristic of a disturbed community (Section 5).

In October 1979, an attempt was made to further quantify the influence
of sub-threshold coal ash effluent concentrations on macroinvertebrate
distributions• A plot of daily effluent discharge rates from June-October
1979 (Figure 5) shows that the Columbia Generating Station did not
continuously discharge effluent. After varying periods of time stream water
quality would revert to background conditions. This intermittent discharge
pattern was used to examine the response of the macroinvertebrate community
to renewed effluent exposure. These conditions offered a partial
translation of the laboratory drift study to a field situation; an important
difference was the lack of control over substrate quality in the field.

The objectives of this study were: (1) to document changes in macro-
invertebrate abundance and species composition attributable to renewed
discharge of the coal ash effluent; (2) to determine if intensity of
response varied with effluent concentration; (3) to attempt to verify the
results of the laboratory drift study in the field; and (4) to evaluate the
sublethal stress on the macroinvertebrate community imposed by the inter-
mittent effluent discharge pattern.

-------
methods

Study Design

Benthlc macro invertebrates were sampled on two dates prior to effluent

release (October 3 and 4) and on the first, second, fourth and eighth days
of discharge (October 5, 6, 8 and 12) (Figure 5). Pumping of the effluent
began at 0300 on October 5. The last discharge prior to this date occurred
on Septembr 20, 15 days earlier.

Two experimental sites were chosen for study: APD-1 and APD-2, located
1.7 and 3.4 km, respectively, downstream of the junction of the ash. pit
drain and the effluent ditch (see Figure 1). A reference site (RRC-2) in
Rocky tun Creek 0,5 km upstream of its junction with the ash pit drain was
also sampled to estimate natural population variability. Differences In
stream habitat characteristics preclude use of this site as a strict control
for the two experimental sites.

Macroinvertebrate Sampling

Benthlc macroinvertebrates were collected with a Wildco Petite ponar
(open jaw 15 x 17 cm). Triplicate samples were taken at each site. At
altea APD-1 and APD-2, organisms were collected within patches of sedges
{Eteoahans aoioulansi personal communication Judy Capelli, 1980) and
grasses rather than from the surrounding predominantly sandy substrates.
Substrates at RRC-2 were more homogeneous, consisting mainly of fine silt
and detrital material.

Upon collection, the contents of each ponar grab were emptied Into a
bucket with a fine wire mesh bottom and backwashed with stream water to
remove excess silt. The remaining material was stored In double plastic
bags and preserved with buffered formalin (5%). In the laboratory,
invertebrates collected In each ponar sample were hand sorted and
individuals counted. Only Asellus raaovitsai in APD-1 samples were
subsampled. Identification was to genus except for Hlrudinea, Triciadida
and some aquatic insects which were only Identified to family (e.g.,
Chlronomidae) .

Chemical and Physical Data

Conductivity (Yellow Springs Instrument Model 33 SCT meter), current
speed (Ocean Equipment Model 451), sample depth, and temperature were
measured at each site on each date. Grab water samples were analyzed In the
labortory for methyl orange alkalinity (American Public Health Association
1976), turbidity (Hach Model 2100A), pH (Orlan Model 399A) and total
filterable residue (TFR) (American Public Health Association 1976). The top
8 cm of stream sediments were collected near the site of each ponar sample
with a I-® length of 9.5-mm tygon tubing capped with a pipette bulb (Forbes
et al» 1981). In the laboratory, 7.5 ml of each sediment sample was
thoroughly mixed with an equal volume of distilled water in a graduated

-------
DRILY EFFLUENT 0 ISCHRRGE
JUNE - OCTOBER 1979

25.Of
22 .5

20-0
17-S
15-0
12.5
10.0
7-5
5.0
2.5

1 Li I A. I I I I 111 t t I I I 1 t i ill t I .1 Lli I I

1 I 1 1 » I I I II I M 1 1 1 1 1 1 I I

30 190 200 210 220 230 240' 250 260 270 280 290 300

JULIRN DAY

Figure 5. Daily volume of coal ash effluent discharged Into the ash
pit drain, June to October 1979. The study period is
denoted by the solid bar under the x-axis. Data from
dally discharge records, Wisconsin Power and Light Co*

-------
centrifuge tube. Following a settling period of several days, the amount of

floe layering on top of each sediment sample was measured. No floe was
observed in any reference substrate samples.

Data Analysis

Multiple regression and ANOVA were used to detect significant, changes
in macroinvertebrate abundance related to effluent discharge. Two models
were compared. Model I had independent variables for site (APD-1, APD-2 and
RRC-2) and time (before and after discharge began). Model II included the
above variables plus an independent variable for the site by time inter-
action. The model providing better fit to the dependent variable (measures
of macroinvertebrate abundance arid species richness) was determined by
calculating the F ratio between the regression sum of squares from the ANOVA
of the two models (Draper and Smith 1966). Two alternative hypotheses were
tested: (Hq) The site by time interaction was not significant (i.e., non-
significant F-ratio) and, thus, any changes in the Invertebrate community
following the resumption of effluent discharge could not be distinguished
from natural variability exhibited at the reference site; (Hj_) The site by
time interaction was significant (significant F-ratio) and the changes at
the experimental sites following resumption of effluent discharge were
measurably different from the reference site and could be attributed to
exposure to the effluent.

Where F-tests were significant a multiple comparison test (Box et al.
1978) identified individual sites with significant differences in before vs.
after macroinvertebrate populations. This test provides 95% confidence
intervals for a particular difference in means between two treatments within
an AHOVA,

The dependent variables analyzed In this fashion were total macroinver-
tebrate abundance, Isopoda, Amphipoda and Trichoptera abundances, and the
number of taxa. The taxa from the three replicate ponars were pooled for
each date. Other variables were calculated as the mean from the three
replicate ponars following transformation to natural logs [In (x+1)] (Elliot
1977). Data collected on the first day of effluent discharge (Day 3 -
October 5, 1979) were not included in the analysis because water chemistry
data suggested the effluent may not have reached site APD-2 by the time of
sampling.

RESULTS

Effect on Stream Habitats

Changes in water chemistry at the experimental sites following effluent
discharges paralleled those reported in earlier studies (Magnuson et al-
1980a), Conductivity, turbidity and total filterable solids (TFR)
increased, while pH and alkalinity declined (Table 5). Water temperature,
current speed, stream depth and floe (%) in sediments did not show

-------
TABLE 5. MEDIAN (RANGE) OF A SERIES OF PHYSICAL AND CHEMICAL PARAMETERS MEASURED AT SITES
APD—1, APD-2 AND RRC-2 BEFORE AND AFTER EFFLUENT DISCHARGE BEGAN

Parameter

APD-1

APD-2

RRC-

¦2

Before

After

Before

After

Before

After

Temperature

(C)

(13-16)

(11-16)

(12-14)

(11-13)

(12-13)

(8-12)

Current speed

(cm/sec)

(18-22)

(18-25)

(17-18)

(17-23)

(15-30)

(15-25)

Mean sample

depth (cm)

(20)

(19-27)

(32-33)

(31-32)

(40-48)

(32-42)

Floe in

substrate (Z)

(57-65)

(47-68)

(0-44)

(1-16)

Conductivity

467

610

434

570

539

494

£ mhos @ 25C)

(433-500)

(546-635)

(432-436)

(455-589)

(503-574)

(490-537)

Alkalinity

229

203

105

249

245

(ppm CaCOj)

(226-232)

(7 8-86)

(202-204)

(101-213)

(248-249)

(244-248)

7.6

7.1

7.7

7.4

7.9

(7.6)

(7.1-7.4)

(7.6-7.7)

(7.3-7.5)

(7.9)

(7.9-8.0)

Total filterable

3.8

29.7

4.5

28.8

13,1

11.4

residue

(3.4-4.2)

(21.1-31.1)

(4.3-13.5)

(25.2-31.1)

(12.7-13,5)

(9.4-14.7)

(nig/ liter)

Turbid ity

3.6

6.3

4.3

7.1

3.1

3.3

(JTU)

(3.3-3.8)

(5.0-7.6)

(4.0-4.6)

(6.3-7.5)

(3.0-3.2)

(3.0-3.7)

-------
appreciable changes. At the reference site, stream habitat variables were

relatively stable throughout the study period.

Correlations between water chemistry parameters and effluent discharge
rate suggested differences in response time at the two experimental sites.
Ac site APD-1 all the water chemistry parameters were significantly
correlated with discharge rate (Table 6)• In contrast, at APD-2 located
further downstream, only turbidity and TFR were significantly correlated
with discharge rate. Other parameters (conductivity, alkalinit; and pH)
were more closely related to the discharge rate of the previous a ay (.fable
6). Thus, while organisms at APD-2 may have been exposed to some precursor
of the effluent in the form of increased suspended material on October 5,
there was a lag of 1 day in exposure to the chemical components of the coal
ash effluent. Figure 6 shows daily values for conductivity and total
filterable residue at each site during the study period.

TABLE 6. CORRELATION (R) BETWEEN DAILY EFFLUENT DISCHARGE RATE AND A SERIES
OF CHEMICAL AND PHYSICAL PARAMETERS MEASURED AT SITES APD-1 AND
APD-2. CORRELATION COEFFICIENTS ARE GIVEN FOR THE RELATIONSHIP
BETWEEN EACH STREAM PARAMETER AND EFFLUENT DISCHARGE RATE OF THE
SAME DAY AND FOR EACH STREAM PARAMETER AND EFFLUENT DISCHARGE RATE
OF THE PREVIOUS DAY FOR SITE APD-2

Variable

Correlation Coefficient (r)

Correlated with

APD-L

APD-2

Discharge Rate

Same Day

Same Day Previous Day

Current speed

0.27

0.23

-0.50

-0.82*

-0. 72

-0.61

Conductivity

0.82*

0.69

0.97**

Alkalinity

-0.98**

-0.55

-0.98**

Turbidity

0.90*

0.96**

0. 52

Total filterable

solids

0.97**

1.00**

0. 50

(n)a

^Significant difference at the 0.05 level.

**Highly significant difference at the 0.0L level.
aNumber of days for which observations were compared.

Effect on Mac ro invertebrates

Changes in Abundance--

Beginning October 6, the second day of coal ash effluent discharge,
macroinvertebrate abundance increased at APD-1 (Figure 6)• This was due

-------
ASH PIT DRAIN

3,7 km 5.4 km

REFERENCE
SITE

(fl

JZ
£

X
CT>

to
_i

—5 <
5z

— Q-

m
K
<

z
o

O
z

800

400

40
20

1000
100

20
10

CONDUCTIVITY

_L.

_L_

TOTAL FILTERABLE SOLIDS

—1V_

J I I L

MACRO!NVERTEBRATES

TAXA

_LJ_i

-2 02460-2 02 4 68 -2. 02468

TIME

(Days before and after initiation of
effluent discharge)

Figure 6. Changes in conductance, total filterable solids, macro-
invertebrate density and number of taxa upon resumption
of coal ash effluent discharge. Ash pit drain sites
3.7 km and 5,4 km refer to APD-1 and APD-2, respectively.
The vertical dashed line indicates the first day of dis-
charge. ttacroinvertebrate density (log scale) shows
number of individuals in each of three basket samplers
indicated by dots; the solid line connects the median.
Macro invertebrate taxa represent pooled value for three
samples.

-------
primarily to higher densities of Aeellus vacovitzai and secondarily to
Hyallela azteca and Crangonyx sp. (Table 7). Of the remaining 27 taxa
sampled at the site, four were not collected following effluent discharge
while nine were present only after discharge began. Most of those taxa were
rarely encountered; only Phryganea sp., Oecetis sp. and Nemote I us sp. were
collected in more than three ponar samples.

In contrast to the response at APD-1, total macroinvertebrate abundance
at APD-2 declined following effluent discharge (Figure 6). This decline

began on October 5 when the amount of suspended material increased, but
there was no chemical alteration of water quality. Decreases in densities
of Hyallela azteca and, to a lesser extent, Asellus raaovitzai accounted for
much of the response (Table 7). Of the 12 other taxa collected at this
site, only Cheumatopsyche, Chironomidae and Hydroptilidae were present in
three or more ponar samples• No appreciable differences between before and
after populations for these taxa were apparent.

The macroinvertebrate community at the reference site, RRC-2, remained
relatively stable throughout the study period (Figure 6). As at the two
experimental sites, the dominant macroinvertebratea were Aeellus raaovitzai
and Hyallela azteca. The remaining community members were comprised of 10
less common taxa (Table 7).

Regression Analysis—

Model II provided better fit to the total macroinvertebrate abundance,
Isopoda, Amphipoda and Trichoptera data than did Model I (Table 8). The
site by time interaction explained no additional variation of the
independent variable, the number of taxa.

Upon further analyses with the multiple comparison test, only a few
individual site-specific changes were significant: increases at APD-1 and
decreases at APD-2 in total individuals and Amphipoda abundances following
effluent discharge (Table 9). Although Model II provided better fit for
Isopoda and Trichoptera data, variability within cells apparently masked any
site-specific changes. No significant before vs. after differences were
measured at RRC-2.

DISCUSSION

The response of benthic macroinvertebrates to resumption of coal ash
effluent discharge was not consistent between the two experimental sites.
Macroinvertebrate densities increased at APD-1, while at APD-2 numbers
declined. Explanations for the lack of consistency between responses can
only be speculative. Site APD-1 may provide a more suitable habitat for
macroinvertebrate colonization despite receiving a more concentrated
effluent. Twice as many taxa were collected at this site than at APD-2.
APD-1 is located in an artificially ditched and banked segment of the ash
pit drain approximately 30 m downstream from a culvert. Sedges and water
grasses are recent (1978) colonizers of previously floc-covered sandy

-------
TABLE 7. COMPOSITION OF PONAR SAMPLES COLLECTED AT THE TOO EXPERIMENTAL SITES, APD-1 AND APD-2,
AND AT THE REFERENCE SITE, RRC-1. NUMBERS REFER TO THE MEDIAN (RANGE) ABUNDANCE IN
PONAR SAMPLES COLLECTED BEFORE (n-6) AND AFTER (n-12) DISCHARGE BEGAN, ABUNDANCES ARE
GIVEN FOR GENERA APPEARING IN MORE THAN TB.REE PONAR SAMPLES AT A SITE.

APD-1

APD-

-2

RRC-

-1

Taxon

Before

After

Before

After

Before

After

Isopoda

Aeellue

216(143-342)

555(44-1000)

129(33-239)

25(9-109)

14(0-19)

8(0-45)

Aaphlpoda
Crarigonyx
Comma rue
Hyalella

1.5(1-5)
36(16-56)

5(0-12)

84(8-144)

253(194-357)

49(8-310)

19(0-57)

24(1-137)

Insecta

Trichoptera

Cheumatopsyche

0(0-9)

2(0-9)

0(0-15)

0(0-6)

0(0-1)

Hydroptilidae

0(0-2)

1(0-3)

0(0-1)

Oecetis

0( )

0(0-1)

Phryganea

Polycentropus

0(0-2)

Piiloetomis

0(0-2)

0(0-5)

Ephemeroptera

Bactie
Caenis
Callibaetic
Hexagenia

0(0-1)
0(0-3)

0(0-1)
0(0-4)

Odonata

Coenagrlonidae

Libellula a

Tetrogoneurna

-------
TABLE 1. (Continued)

Taxon

APD-1

APD-2

RRC-l

Before

After

Before After

Coleoptera

Agabuc Xv.

0(0-3)

0(0-1)

Bevocuc lv.

Dineutus ad.

Dubiraphia lv.

Dubiraphia ad.

tialiplus lv.

1(0-2)

2(0-9)

Peltodytes ad.

0(0-1)

Dlptera

Chlronomldae

1.5(0-10)

2(0-14)

5(0-1) 0(0-1)

1(0-4) 1.5(0-6)

Chrysopo

0(0-1)

0(0-2)

Hexatoma

0(0-1)

Nemotelue

0( )

0.5(0-1)

Palpomyia

1(0-8)

3.5(0-8)

Pi la ria

Hemiptera

Beloetoma

Corixidae

0.5(0-1) 0(0-2)

Plea

a a

Megaloptera

Sialis

Lepidoptera

Paraponyx

aPresent in three or fewer ponar samples.

-------
TABLE 8. COMPARISON OF TWO MODELS DESCRIBING MACROINVERTEBRATE RESPONSE TO
RESUMPTION OF COAL ASH EFFLUENT DISCHARGE. GIVEN ARE THE F-
RATIOS, COMPARING THE REGRESSION SUM OF SQUARES FROM THE ANOVA OF
THE TWO MODELS AND THE LEVEL OF SIGNIFICANCE (NS = NOT SIGNIFI-
CANT) OF THESE F-RATIOS FOR FIVE DEPENDENT VARIABLES

Dependent F-Ratio Level of

Variable (2,9)d .f. Significance

Total abundance 27.00 p < 0,001

Number of taxa 1.61 NS

Isopoda 6.53 p < 0.05

Amphlpoda 26.69 p < 0.001

Trichoptera 4.93 p < 0.05

TABLE 9. RESULTS FROM MULTIPLE COMPARISON TEST. PRESENTED ARE THE

DIFFERENCE BETWEEN BEFORE AND AFTER MEANS FOR EACH SITE AND THE
95% CONFIDENCE INTERVAL SURROUNDING THAT DIFFERENCE

Dependent

Sample

Difference

Confidence

Variable

Site

in Means

Interval

Total

APD-1

-1.00*

tO.9!

abundance

APD-2

1.39*

tO. 91

RRC-2

-0.37

tO. 91

Isopoda

APD-1

-1.01

±1.74

APD-2

1.32

±1.74

RRC-2

-0.10

±1.74

Amphipoda

APD-1

-1.10*

±0.98

APD-2

1.41*

±0.98

RRC-2

-0.64

±0.98

Trichoptera

APD-1

-0.66

±0.88

APD-2

0.35

±0.88

RRC-2

0.05

±0.88

*Significant difference at the 0.05 level.

-------
substrates• The habitat offered at site APD-L may provide a refuge for
invertebrates (primarily Aeellus vacovitzai and Hyallela azteca) disturbed
at upstream sites by release of the effluent or entering the coal ash
modified habitat from undisturbed sites. Pennak (1978) notes striking
aggregations of Asellus in streams where velocities are too high to allow
further upstream migrations, yet alow enough to allow then to maintain their
position.

Site APD-2 is situated in the main channel of the ash pit drain as it
flows through a floodplain forest north of the junction with Rocky Run
Creek. The stream is subject to wide seasonal fluctuations in water levels
and streambank erosion as different channels become formed within the
floodplain. As a result, the available stream habitats may provide a
naturally less stable environment for invertebrate colonization.

The timing of response was a second difference between the two sites.
Although macroinvertebrates at APD-L were exposed to the coal ash effluent
on October 5, no increase in density was observed until the following day.
In contrast, at APD-2, decreased abundance occurred on the first day of
discharge (October 5) when suspended material Increased, but no chemical
modifications of water quality — suggested by altered conductivity or
alkalinity — were measured. The increase in suspended material may have
immediately affected the APD-2 community, eliciting a response similar to
habitat avoidance. In contrast, at APD-1 the community was unaffected by
either the suspended load or chemical alterations of water quality.
Population densities did not increase until upstream individuals reached the
site and remained there.

In the laboratory drift study (Section 3), macroinvertebrates were
associated with the substrate and reduced drift activity when exposed to
water and/or sediments contaminated by coal ash effluent. Although the
increase in densities at APD-l may correspond to this response, the lack of
consistency with the results from site APD-2 do not make this probable.

The overall impact of resumption of coal ash effluent discharge on the
established macroinvertebrate community was not dramatic. The organisms
inhabiting the ash pit drain have had to acclimate to periodic disturbances
caused by the intermittent pauses in effluent discharge. The detection of a
slight response provides some additional evidence for the existence of
sublethal stress below the effluent threshold.

-------
SECTION 5

EFFECTS OF THE COAL ASH EFFLUENT
ON MACROINVERTEBRATE COMMUNITY STRUCTURE

INTRODUCTION

The effects of environmental contaminants on aquatic communities range
from immediate acute toxicity, such as results from spills of highly toxic
substances (Crossman et al. 1974), to chronic sublethal impacts which may be
more devesting Co ecosystems in the long term (Lehmkuhl 1979). Sublethal
impacts are difficult to evaluate through monitoring of Individual physical
and chemical parameters, because the concentrations of toxicants that induce
chronic stress are often ill defined. In addition, ecosystems are often
subjected to a variety of stresses operating simultaneously and, perhaps ,
synergistically.

For these and other reasons, monitoring of biological responses to
pollutants has become an integral part of water quality evaluation proce-
dures • Aquatic organisms respond to the environment in an integrated
fashion, making them better indicators of environmental stress than chemical
and physical parameters (Cairns et al. 1973). Aquatic macroinvertebrates
possess several attributes which enhance their use as biological indicators
(Cover and Harrell 1978). These include limited mobility, a relatively long
life span leading to responses which are a reflection of recent conditions,
occupation of several trophic positions as well as being important links in
aquatic food webs, and exhibition of a range of known tolerance limits to
different environmental factors •

Previous reports have documented impacts of the coal ash effluent
discharged by the Columbia Electric Generating Station on the macroin-
vertebrate community inhabiting the receiving stream. The effects were
measured temporally—-through comparisons of pre- and post-operation
community structure—and spatially—through examination of distributions
along a gradient of effluent strength (Magnuson et al. 1980a, Forbes et al.
1981).

This section will describe changes in macro invertebrate community
structure at one highly impacted site within the coal ash drainage system.
An additional year of data (1980) allowed documentation of the following
sequence of events: (1) initial decline of intolerant taxa following
effluent discharge in 1975; (2) disappearance of most taxa in 1977; and (3)
subsequent partial recovery in 1980. Analysis was based on examination of

-------
responses of individual taxa and an analysts of changes in community
structure. Details on this data set can. be found in Magnuson ec al. (1980),

METHODS

Macro invertebrate Samplers

Artificial substrate samplers were used to investigate the macroinver-
terate community inhabiting site APD-1 of the ash pit drain. The advantages
of these samplers are the elimination of substrate differences as effects
between sites and over time, ability to retrieve quantitative data from hard
to sample sites, ease of sample processing, and higher precision than other
methods (Weber 1973). Disadvantages include the length of exposure time,
vulnerability to vandalism and dislodgement, inability to measure effects of
alteration of natural substrates, and development of a community which may
not reflect prior conditions (Weber 1973).

Two types of artificial substrates were used. Basket samplers (Mason
et al. 1970) were used during the summers of 1974, 1975, and 1980 (Table
10). These consisted of closed cylinders 20 x 30 cm constructed of chicken
wire and filled with 4.5 kg of limestone rocks. At the downstream ash pit
drain site, three baskets were suspended approximately 10 cm above the
stream sediments at mid-channel, Following a colonization period of
approximately 4 weeks, the contents of each basket were shaken into a 1-mm
mesh dip net and then preserved in 70% ethanol. Macro invertebrates were

TABLE 10. DATES SHOWING INTERVALS FOE MACROINVERTEBRATE COLONIZATION OF

ARTIFICIAL SUBSTRATE SAMPLERS IN THE ASH PIT DRAIN. TEE NUMBER
OF SAMPLERS RETRIEVED IS SHOWN IN PARENTHESES

Sampler
Type

Year

1974

1975

197?

1980

16 May-13 June

15 May-13 June

(3)

29 May-I July

13 June-9 July

13 June-16 July

(2)

(3)

1 July-30 July

9 July-7 Aug.

16 July-8 Aug.

(1)

(3)

30 July-27 Aug.

7 Aug.-5 Sept.

8 Aug.-10 Sept.

(3)

2 7 Aug.-24 Sept.

5 Sept.-30 0c t.

(3)

Basket

Modified

Dendy

26 May-2 June 2 June-9 June

(3) (?)

1 Sept.-9 Sept.

(3)

-------
later hand-sorted and all individuals counted. If more than 2,5 h were
required to sort the first quarter of a sample, subsamples were analyzed
(Magnuson et al. 1980a). With the exception of a few taxa which were
Identified to family, identification was to the genus level using keys
designed by illseohoff (1975). Identification of some taxa was verified by
Dr. W. Hilsenhoff, University of Wisconsin-Madison -

A modified multiplate Hester-Dandy artificial substrate (Hester and
Dendy 1962) was used to compare macro invertebrate populations upstream ( APD-
3) and downstream (APD-1) of the effluent in the ash pit drain In 1977 and
1980. Site locations are shown in Figure 1- Colonisation periods are
listed in Table 10. The modified Dendy consisted of a Tuffy® scrub ball
held between two 8-ctu masonite plates by an eyebolt. Samplers were
suspended by floats 2.0 cm above streaa sediments and anchored by cement
blocks. Three to seven samplers were placed 1 dm apart longitudinally at
random locations across the width of the stream. After a 1-week coloniza-
tion period, samplers were retrieved, placed in freezer containers, and
preserved in 701 ethanol. For processing, sample contents were washed
through 70 ws mesh to concentrate organisms and to remove fine silt.
Macroinvertebrates were sorted, counted, and identified as previously
described.

Comparison of Samplers

The macro invertebrate taxa colonising basket and modified Dendy artifi-
cial. subtrates in June 1980 were similar (Table 11). Eleven of the 20 taxa
collected colonised both sampler types. These included most of the major
taxa based on abundance. Ranks for individual taxa, based on relative
abundance, were also similar for the two sampler types, the Spearman rank
correlation coefficient (Siegel 1956) comparing ranks by abundance of each
taxon, collected by baskets and Dendy samplers, was 0,626, significant at p
< 0.01.

Three taxa--caddis£lies (Cheumatopeyehe spp. and Hydvopsyche spp.) and
Coenagrionidae daraselflies—colonized only basket samplers in June 1980.
These taxa have previously colonized modified Dendy samplers at other sites
in the ash pit drain (Magnuson et al. 1980a). Modified Dendy samplers were
colonized by six taxa not observed in baskets. These taxa occurred at low
abundances (means of 0.1 to 0.4 individuals per sampler).

Although basket samplers collected more individuals due partly to their
greater surface area available for colonization, taxonomic richness was
similar between sampler types. The medians and range of number of taxa
collected per basket and Dendy were 9 (6 to 11) and 9 (8 to 14),
respectively. Mason et al. (1913) also observed similar abundances and
macroinvertebrate taxa colonizing basket and multiplate samplers, although a
few species showed distinct preference for one sample type.

-------
TABLE H. COMPARISON OF MACROINVERTEBRATE TAXA COLONIZING BASKET AND MODIFIED DENDY SAMPLERS AT
SITE APD-1. COLONIZATION PERIODS WERE MAY 15 TO JUNE 13 (BASKETS) AND JUNE 2 TO 9
(DENDY) IN 1980. SAMPLE SIZES WERE 3 AND 7 FOR BASKETS AND DENDY SAMPLERS,
RESPECTIVELY. RANKS ARE BASED ON ABUNDANCE

Number per Sampler Rank

Macro1nvertebrate Taxon Basket Sampler Modified Dendy Basket Sampler Modified Sampler

Taxa la Both Sampler Types

Aeellue racovitzai 205

Myallela azteca 58

Pel tody tee lv. 26

Crangonyx 5.7

Ganmarue peeudoIimnaeue 3.7

Litrmephi lus 2.0

Hydroptilldae 1.3

Chironomidae 1.3

Haliplua lv. 0.3

Helod idae lv. 0.3

Tropistemus lv. 0,7

Taxa in One Sampler Type

Cheumatopeyche 3.7

CoenagrionIdae 1.0

Hydropsyche 0.3

Asynarchus
Pa Ipomyia
Caenie
Triaenodes

Liodeseus lv.
Cauliodes

127 1 i

51 2 2
3.7 3 5

9.0 4 3

7.6 5.5 4

0.7 7 7

0.6 8.5 8.5

1.0 8.5 6

0.6 13 8.5

0.1 13 14.5

0.1 11 14.5

5.5
10
13

0.4 10.5

0*4 10.5

0.1 14.5

0.I 14.5

0.1 14.5

-------
Data Analysis

Various measures were used to compare macro invertebrate community
structure in the pre-operational year 1974 and post-operational years of
1975, 1977, and 1980, The responses of Individual taxa were measured as
changes in occurrence and abundance throughout the study period. Community
responses were evaluated through a series of Indices Including number of
Individuals, number of taxa, Shannon Heaver diversity (H) and evenness (e)
(Pielou 1969) (Equation 1).

ff - */i i°82 (pi> and e " T^fT'

i"i

where s is the total number of taxa and pj Is the proportion of total
individuals In the ith taxa. All indices were calculated from, mean
abundances determined from the contents of replicate basket samplers
collected on the same date. A Friedman two-way MOM, a measure of the
effect of K treatments upon the ranks of a response, was used to examine
differences in Index values between years (Siegel 1956).

Diversity indices have long been used as biological indicators of water
quality and measures of community response to pollutants (Wilhm and Dorris
1968) - However, Interpretation of pollutant Impacts using diversity indices
alone without understanding the ecology of the community can lead to
misleading conclusions (Hynes 1970, Hocutt 1975). Although environmental
conditions aay vary, diversity Indices may equate pollution tolerant
communities with more typically clean water communities (Godfrey 1978).

RESULTS

Community Composition
Presence/Absence—

The presence of each major macro invertebrate taxon artificial sub-
strates collected at site APD-1 In June and August in 1974, 197 5, 1977 and
1980, is Indicated by solid lines in Figure 7. Data for 1977 were based on
colonization of modified Dendy type samplers; other years are from basket
samplers. In Figure 7 the presence of a particular taxa in months other
than June or August is indicated by a dashed line.

In 1975, only two pre-operatlon taxa did not return (Tabanidae and
Stenonem, sp»). One genus, the amphipod Crangonyx sp., first appeared after
pumping began. The lack of organisms in 1977 is striking. During June,
only a few Individuals of the isopod , Asellus racoviizai , the amphipod
Hualtela azteca, and Chironooldae were found. In August 1977, no organisms
colonized Dendy samplers at site APD-1. By 1980 recovery of most of the

-------
MAJOR MACRO INVERTEBRATE TAXA

19T4 1975 197? 1980

ZOzOzOzO

i——i—r™i [ " I

i sopoda

Asellus racovitzai " "• : .

Amph i poda

Hyalelta ozteca — —" —-

Gammarus pseudolimnaeus

Crangonyx * —

Ephemeroptera

Baetis —

Caen is 1 1 :: : —

Stenacron

Stenonema

Odonata

Coenagrionidae — —

Trichoptera

Cheumatopsyche

Hydropsyche 1 1 — ; —

Hydropt i 11 dae — —

Oecetis .———

Pol ycent ropus

Tr iaenodes —

Coleoptera

Ho!iplus Lv. — —

Tropisternus Lv.

Dipt era

Chironomldae

Tabantdae ——

Cera topogon idoe —

Simu 11 tdae — —

Figure 7. Occurrence of major raacroinvertebrate taxa in artificial sub-
strate sarapLers collected at APD-1 in June and August 1974,
1975, 1977, and 1980. A solid line indicates presence during
the month represented; a dashed line .indicates presence during
a month other than June or August.

-------
major taxa was evident. Only two mayflies, Stenonema sp., and Stenaaron sp.
and two caddiafliea , Oeoefie sp. and Polycentwpue sp. did riot return.

Relative Abundance—

Although no major differences in the presence of taxa In 1974, 1975 and
1980 samples were apparent, changes occurred in the relative and absolute
abundances of several groups (Figure 8). The pre-operation 1974 community
was dominated numerically by Hydropsychid caddis flies, mainly of the genus
Cheumatopsychs , with lesser numbers of Hydmpsyehe • Chironaidae were the
next important group numerically.

In 1975, community composition in June and July was similar to that of
1974 (no data were collected in May 1974). By August, however, several
taxa, notably Ckeumatopeyche spp., Hydropeyche sp. and Chironomidae, had
disappeared. Other taxa not present in August 1975 samples Include GrnmmniS
pseudol iirmaeua and Stenocron . Abundances of Asellus raeovitsai and Hyal lela
azteca increased. The effects of the coal ash effluent were thus not
detected until 4 months after discharge began In April 1975 (Magnuson et al.
1980a).

In 1980, 5 yr after effluent discharge was initiated, the community was
dominated by Asellus raeovitsai and Hyallela azteca, two of the three taxa
colonizing modified Dendy samplers at site APD-1 in 1977. Formerly dominant
taxa Cheumatopeyche spp. and Hydropeyche sp. returned, but at greatly
reduced abundances .

Indices of Community Structure—

Monthly values for four indices of community structure at site APD-1
are shown in Figure 9 for 1974, 1975 and 1980. On a monthly basis, Shannon
Weaver diversity and evenness show general, but not appreciable, declines
from 1974 to 1980. The numerical dominance of the 1980 community by Asellus
raeovitsai (66 to 83£ of total Individuals) probably caused these two
indices to decrease.

Number of taxa was similar for all 3 yr for May, June, July and August.
Although a more substantial difference is apparent In September between 1974
and 1975, no data was collected in September 1980 for additional comparison.

Whlie total abundances from May to July were similar between years,
late summer abundances were dramatically different. Populations exceeded
2,000 individuals per basket sample in August 1974. This late summer
increase in total abundance did not occur in 197 5, when the first effects of
coal ash effluent discharge were observed. A similar pattern of late straraer
increase took place in 1980, but total abundance was < 502 of 1974 values.
None of the differences between years for these indices were significant
based on the Friedman two-way ANOVA.

-------
Total Number
M J J ft

1974

(c>

1975

1980

Aiallus

rocovittei

MJJA

I 1 1—I

°(

Chiranamidas Chtumotopsyctte

MJJA M J J A

I 1 1 1

i J r-

Hyoiella Gammorus

azttca psaudohmnatu* Crangonyt Hydropsycha

MJJA MJJA MJJA MJJA

, ,

(974 |° D=<]

D<]
O—=,

1973

1980

i 1 1 1

> —

I 1 1 1

Cotnoqriomda*

1974'°' |S —
-------
3,0
2.0
1.0
0
to
0.5
0
30
20
10
0

2500
2000
1500
1000

500
0

SHANNON-WEAVER DIVERSITY

EVENESS

NUMBER OF TAXA

J__ l_

J L,

NUMBER OF INDIVIDUALS

I I 1 !

M J J A

MONTH

1974

-— 1975
1980

Figure 9, Shannon-Weaver diversity, evenness, number of
caxa and number of individuals at site APD-1
during the summer months of 1974, 1975, and
1980.

-------
L'pstream-Downstream Comparisons

The results of the June 1980 replication of the 1977 sampling,
comparing colonization of substrates at the upstream (APJ3-3) and downstream
(APD-L) sites in. the ashpit drain, are shorn in Figure 10. While there were
dramatic differences between the two sites In total macroinvertebrate
abundance and number of taxa in 1977, no substantial differences in these
parameters were apparent in 1980.

Evidence for a decline in environmental stress imposed by the coal ash
effluent in 1980 was provided by evaluation of conductivities measured at
the upstream and downstream sites in 1977 and 1980. Forbes et al. (1981)
derived a biological response threshold based on effluent conductivity from
the series of laboratory and field studies summarized in Table 12. Ash pit
drain sites receiving concentrations of coal ash chemicals exceeding 1000

mhos conductance were unsuitable for macro invertebrate colonization (Forbes
et al. 1981). This relationship was not meant to imply elevated conductivity
as a cause of the observed responses; rather, it provided a chemical
Indicator of effluent concentrations harmful to invertebrates.

As shown in Figure 10, effluent concentrations at site AFD-1 clearly
exceeded the habitable limit. Conductivity was nearly 2500 mhos,
significantly higher than the 1000 pmho response threshold - The lack of
invertebrates mirrors the severity of environmental stress. When effluent
concentrations fell below the 1000 umbo threshold in 1980, total abundance
and number of taxa were similar to upstream values. This result supports
the ability of the conductivity threshold to predict the severity of
environmental stress in stream habitats altered by coal ash chemicals
discharged by the Columbia Generating Station.

DISCUSSION

The following sequence of macroinvertebrate responses to habitat
modification by the coal ash effluent was observed: (1) decline in total
abundance and number of taxa 4 months after effluent discharge began; (2)
near total disappearance of all macroinvertebrates in 1977; and (3) recovery
of most taxa in 1980. The macroinvertebrate community in 1980, while
similar in taxonomic composition to the 1974 pre-operation community, had
experienced significant shifts in absolute abundance as well as the relative
abundances of individual taxa. The resulting community was characterized by
lower total abundances overall and domination by Aeellus raeovitzai. The
intensity of community response was related to effluent concentration; when
concentrations exceeded the conductivity response threshold established by
Forbes et al. (1981), macroinvertebrate populations were severely reduced.

Environmental stress can have a selective or non-selective impact on
aquatic communities (Hocuttt 1975). A selective stress, such as organic
pollution generally results In a decline in the number of taxa and increases
in abundance of tolerant forms. A non-selective stress, such as scouring or
sedimentaiton, will reduce the number of individuals but not necessarily the
number of taxa. According to Kocutt (1975), the initial reduction in the

-------
cn —

< j

3 a.

Q 2

- <

> w

Q >

y Q

f: z
u

U. O

O a

• U

O a.

Z -

< s!

X 3E

< <
H «

^ o
O Z

d g

2 DC
UJ

£L

t —

> E

_ o

I- «

O O

3 -c

a £

z x

o w

600

400

200

3000

2000

1000

y MEDIAN

JRANGE

1977

1980

Figure 10. Comparison between conductivity, macroinvertebrate
abundance and number of taxa at upstream APD-3)
and downstream (APD-1) sites in the ash pit drain
in 19?? and 1980.

-------
TABLE 12. SUMMARY OF EXPERIMENTAL STUDIES INVESTIGATING RESPONSES TO COAL ASH EFFLUENT, LISTED ARE
THE RESPONSES INVESTIGATED, WHETHER THESE RESPONSES WERE MEASURED ACROSS AN EFFLUENT
GRADIENT, THE DIRECTION OF RESPONSE, THE CONDUCTIVITY LEVEL AT WHICH THE RESPONSE WAS
DETECTED, AND THE CONDUCTIVITY INTERVAL BRACKETING THE RESPONSE THRESHOLD. STUDIES
CONDUCTED AT ONE EFFLUENT CONCENTRATION OR FOR WHICH NO THRESHOLD INTERVAL COULD BE
DERIVED ARE DESIGNATED BY A N/A

Year(s)

Biological Response
Investigated

Conductivity
Gradient

Response to
Coal Ash

Conductivity

Level

Response
Threshold

Reference

1977

Survival of crayfish caged
on-si te

Yes

None

1800

a,b5c

Metabolic rate of caged
crayfish

Yes

Reduced

1500

730-1500

a.b.c

Body burden toxic elements
in caged crayfish

Yes

Generally
increased

730

n/a

a.b.c

Hatching success and
survival northern pike
eggs and sac fry

Yes

Reduced

1160

520-1160

1977-
1978

Survival of A sell us in

laboratory bioassay

Yes

None

1200

a,b,c

1978

Survival of adult Gaima.ru6
in laboratory bioassay

Yes

None

1900

a,b,c

Survival of juvenile Gamma rue
in laboratory biossay

Yes

Reduced

1900

1100-1900

a ,b ,c

-------
TABLE 12. (Continued)

Biological Response Conductivity Response Co Conductivity Response

Year(s) Investigated Gradient Coal Ash Level Threshold Reference

1978-
1979

Leaf decomposition and
microbial colonization

Reduced

900

n/a

1979

Macroinvertebrate drift

in laboratory streams

Reduced

500

n/a

e,£

^Magnuson et al. (1980a).

Forbes et al• (1981) .
^Harrel1 (1978).

Magnuson et al. (1980b).
®Webatur et al. (1981).
Section I.

-------
numbers of taxa and Individuals due to subii thai or toxic stress tg followed
by a slow adaptation and recovery to a stable, but submarginal environment
dominated by a few highly productive species.

The coal ash effluent discharged from the Columbia Electric Generating
Station elicited changes In macro invertebrate community structure similar to
those described for non-selective and sublethal/toxic stresses. Over the 7-
yr period of observation, the coal ash effluent imposed a variety of chemi-
cal (higher concentrations of dissolved and suspended material, occasionally
harmful concentrations of dissolved metal elements) and physical (habitat
reduction due to smothering of stream sediments by a chemical floe)
stresses• The severity of environmental stress changed during the study
period, reaching a peak In 1977 when the most devastating effects of the
coal ash effluent were documented (see Section 6). Recovery was observed
following reductions in habitat stress, but with a change in community
structure.

This shift from a community dominated by hydropsychld caddlsflies
(1974-75) to one dominated by the isopod Aselluls recavitzai (1980) was one
of the most dramatic responses documented at site APD-1. Hydropyschid
caddlsflies function as collector-filterers, constructing nets which retain
algae, detritus, and animals carried by stream currents (Merritt and Cummins
1978). The accumulation of a chemical floe coating stream substrates
reduced the available habitat for caddlsfly larvae and probably played an
Important role in their disappearance during 1977. Although apparently
tolerant of organic loading (Hllsenhoff 1982), the Hydropsychldae are fairly
Intolerant of toxic pollutants (Roback 1974). Continued exposure to coal
ash chemicals through 1980 nay have been sufficient to prevent total
recovery of these insects even though floe deposits were significantly
reduced.

In contrast, densities of Asellus recavitzai increased at site APD-L.
Numerous individuals were collected in 1979 ponar samples (Section 5) and
1980 basket samplers. The location of site APD-1 below a culvert may
represent a point where stream velocity is favorable for the development of
Isopod aggregations (Pennak 1978). Substrate modification, may not have been
as detrimental to Asellus as to the Hydrops) cMdae. Isopods are scavengers,
consuming dead and injured animals and green and decaying leaves, grass and
aquatic vegetation (Pennak. 1978).

The tolerance of Asellus to organic loadings has been noted (Aston and
Milner 1980, Hllsenhoff 1982), and this genus has been used as an Indicator
of organic pollution in Europe (Hynes 1970). Literature on tolerance to
toxic pollutants is sparse. In a laboratory study, Fraser (1980) increased
Che tolerance of Asellus aquations to Pb by exposing intolerant animals from
an unpolluted site to low levels of Pb. The demonstrated tolerance of
Isopods to other types of pollutants may justify extension to coal ash
chemicals. Asellus racovitzai was one of three invertebrate taxa collected
(in small numbers) at site APD-1 in 1977 when effluent concentrations were
highest and environmental stress most severe.

-------
Although the structure of the macroinvertebrate community inhabiting
the ash pit drain typifies an organically polluted condition (Hilsenhoff
1982), it is unlikely that organic enrichment Is the cause of the observed
responses. Both crayfish and northern pike eggs and sac fry have survived
long-term in situ caging experiments (Magnuson et al. 1980a, b) indicating
diurnal oxygen depletion is not significant. Signs of over-fertilization
were not apparent; visible macrophyte and algal growth were low, although
occasional clumps of floe colonized by periphyton were observed. Host
likely, the ashpit drain macro invertebrate community is a reflection of the
source of colonizers, an enriched agricultural drain.

The actual cause and effect mechanisms by which the coal ash effluent
impacted macroinvertebrate populations were investigated in a series of
studies conducted between 19?? and 1979 (Table 12), As summarized by Forbes
et al. (1981), the only evidence for acute toxicity was from bioassays using
more sensitive early life stages. The major impacts of the coal ash
chemicals may have been through sublethal effects on physiology and behavior
resulting from accumulations of trace metals in body tissues and/or
inadequate food supply (Forbes et al. 1981, Forbes and Magnuson 1980)
(Section 3).

Results from other studies have demonstrated a variety of sublethal
effects of trace elements on aquatic organism, including impacts on
behavior, fecundity, longevity, and reproduction (Lehmkuhl 1979).

Reductions in activity of crayfish were observed following exposure to
sublethal concentrations of Cd (Maciorowski et al. 1977). Chlronomids
exposed in the laboratory to sediments contaminated with Cd, Cr and Zn
exhibited avoidance responses which increased in intensity with increasing
Cd and Zn levels (Wentsel et al. 1977a). In a separate experiment
chironoaids suffered reductions in survival and growth (Wentsel et al.
1977b). Sublethal chronic exposure to Cu reduced the life span and brood
size of Daphnia magna (Winner and Farrell 1976). Sublethal Cu concentra-
tions also reduced brood size in other Daphnia species, completely halting
reproduction in the three most sensitive species.

Effects on physiology, measured as declines in metabolism, can also be
attributable to inadequate food supply. Forbes and Magnuson (1980) docu-
mented low decomposition rates of leaves in the ash pit drain, suggesting
reduced food quality. A visual inspection indicated that food quantity was
also low (Forbes et al. 1981). A decline in oxygen consumption has been
reported for some intertidal invertebrates in response to starvation (Newell
1973).

It was difficult to attribute the observed changes in community
structure to any one of the potential environmental stresses imposed by the
coal ash effluent. Impacts of altered water chemistry could not be
separated from the phyaial effects of habitat loss due to accumulations of a
chemical floe as the Intensity of stress Imposed by these two factors varied
in a similar fashion during the study period (Section 6). In addition,
hypotheses of acute and sublethal impacts of the effluent were supported by
the results of the laboratory and field manipulations.

-------
SECTION 6

EVALUATION Of ENVIRONMENTAL STRESS IN
THE COLOMBIA COAL ASH DRAINAGE SYSTEM

INTRODUCTION

Environmental stress In Che coal ash drainage system of the Columbia
Electric Generating Station was evaluated using a combination of biological
and chemical Indicators. Severely stressed sites were characterized by a
substantially reduced macroInvertebrate fauna and by coal ash effluent
concentrateons exceeding the biological response threshold of 1000 umbo
conductivity (Forbes et al. 1981, Magnuson et al. 1980a). A temporal
variation in the Intensity of environmental stress was detected at site APD-
1 over the 7-yr study period. A partial recovery was documented 3 yr after
a depauperate fauna was observed (Section 5).

Variations in the intensity oE environmental stress can be attributed
to changes In generating station procedures, Including methods of coal and
effluent treatment and rate of effluent discharge. This section will
describe activities influencing effluent concentration (Indicated by
conductivity) and occurrence of floe deposits. The latter qualitative
variable is included as a measure of substrate modification, an important
influence on macro Invertebrate distributions. The close correspondence
between estimates of environmental stress based on effluent conductivity and
substrate modification and the estimates based on biological data will be
discussed.

INFLUENCE OF GENERTING STATION ACTIVITIES
Conductivity

The major sources of conductivity In the effluent were the NaHCOj
(1975-78) and NH^HSO^ (1979-80) added to the pulverized coal to increase the
efficiency of the electrostatic precipitator. The transport of these
compounds with the fly ash through the settling ponds and into the ashpit
drain, elevated conductance values in proportion to the concentration of the
effluent. Thus, the main generating station activities influencing
conductivity values were the frequency and rate of discharge (Table 1.3).

The relationship Is shown in detail in Figure 11, which presents plots
of frequency of discharge, (percentage of days in the year when effluent was
released), the mean and range of discharge rate and the mean and range of

-------
M£Q;j£*C*

' ii?i "i H sim isisTw™

R«IE

ccNOucmin

g TSX-

ina-

iixj-

f "
5 iw-

y 0

1>U It'll Ears ?91T l»TI LIT* i tM

res*

Figure 11. Conductivity, effluent discharge rate and frequency of discharge
between 1974 and 1980. Bars and horizontal lines represent

range and mean for discharge rate and range and median for
conductivity. Number of conductivity observations are shorn
under bars.

-------
TABLE 13. SUMMARY OF OBSERVATIONS ON CONDUCTIVITY AMD EXTENT OF FLOC
DEPOSITS AT SITE APD-I AND GENERATING STATION ACTIVITIES
POTENTIALLY INFLUENCING ASH PIT DRAIN HABITATS. DISCHARGE RATES
ARE MEAN DAILY DISCHARGE FOR YEAR; CONDUCTANCE IS MEDIAN FOR YEAR

Station Operation Indicators

Year

Discharge
(10"liter/day)

Coal
Treatment

PH
Control

Floe
Removal

Conductance
(ymhos)

Floe Deposits

1975

None

h2S04

400

(No observation)

197?

NaHCO-j

h2sq4

1800

Extensive

1979

nh4hso4

h2so4

Yes

600

Diminished

1980

m4hso4

co2

Yes

500

Rate

conductivity observations from 1974 through 1980, There was no systematalc
monitoring of conductivity; the data represent all the available information
for site APD-1. The highest conductivity values (median 1805 mhos) were
measured in 1977 when discharge volume per day and frequency of discharge
were at or near peak. Only five of 25 conductivity observations made during
that year were < the 1000 ymhos biological response threshold. By 1979 and
1980, when the frequency and rate of discharge had declined, conductivities
rarely exceeded the 1000 ymhos threshold.

Floe Deposits

The extent of floe deposits varied in a fashion similar to that of
conductivity throughout the study period. The thick layers of floe which
accumulated on stream substrates in 1977 were substantially diminished by
1979 and were rarely observed in backwater pools in 1980,

The reductions in floe formation by 1979 were also associated with
declines in effluent discharge (Table 13). Removing accumulated floe in the
effluent ditch immediately below the discharge point was an important factor
because it reduced the downstream transport of precipitated materials.

Before 1979, ^SO^ was added to achieve effluent pH control; after 1979 C02
bubbling was used to control pH. This change in methods may also have
reduced precipitation of the dissolved metals which formed the floe (R.
Perry, Personal communication, Wisconsin Power and Light Co., June 1980).

-------
CHEMICAL AND BIOLOGICAL INDICATORS

Macroinvertebrate community structure at APD-i reflected the severity
of environmental stress predicted by the chemical indicators (Table 14). In
197? when conductivities were predominantly above threshold and floe
deposits were extensive, only a few macro invertebrates colonized modified
Dendy samplers in June; no individuals were collected in September (Manguson
et al. 1980a). The chemical and biological indicators emphasized the
severely stressed conditions at APD-L during that year. When conductivities
declined below threshold and floe deposits diminished in 1979-80, many
macro invertebrate taxa recolonized the previously depauperate site. This
recovery was first observed in the fall of 1979 (Section 4) and documented
in 1980 with the same sampling designs used in previous years (Section 5).

TABLE 14. ESTIMATES OF SEVERITY OF ENVIRONMENTAL STRESS AT APD-I IN 1975,
1977 AND 1980 BASED ON CHEMICAL AND BIOLOGICAL INDICATORS.

Chemical Indicators
Environmental Conductance Floe

Year Stress (ymhos) Deposits

Biological Indicator
Macroinvertebrate Community
Structure at APD-1

1975

Low

400

(No observation)

Late summer declines in
intolerant taxa

1977 Severe

1800

Extensive Disappearance of nearly

all individuals

1980

Low

500

Rare

Re-establishment of many
taxa, Asellue now
numerically dominant

Comparisons of macroinvertebrate abundance and species richness at APD-
1 and at the ashpit drain site upstream of the effluent (APD-3) in June 197 7
and 1980 were previously described (Section 5, Figure 10). The dramatic
difference in between-site number of taxa and individuals in 1977 when
conductivites were 2500 pmhos at APD-1, disappeared in 1980 when
conductivities were below threshold and approached background levels.

Although several taxa were able to recolonize the previously severely
disturbed APD-1 site by 1980, community composition differed from the
preoperation (1974) community (Section 5). The tolerant isopod species—
Asellue racovitzai—became numerically dominate in 1980. This species was
present in 1974 prior to effluent discharge but at much lower abundances.
This shift in dominants from Cheumatopsyche spp. to Asellue vaaovitzai
suggests a continued, sub-acute disturbance from exposure to the coal-ash
chemicals below the biological response threshold. Results from three

-------
studies demonstrated possible effects on macroinvertebrates at sub-threshold
effluent concentrations: (1) reduction in food quality through lowered

microbial colonization and decomposition rates of leaves (Forbes and
Magnuson 1980); (2) reduced invertebrate drift activity (Section 3); and {3}
periodic disturbance due to the intermittent nature of effluent discharge
(Section 4),

DISCUSSION AND SUMMARY

The intensity of environmental stress varied substantially over the 7-
yr study period at site APD-1. The cause of this variability was attribut-
able to a series of generating station activities influencing both the
quantity and quality of effluent discharge. Discharge rate and frequency
were closely related to effluent conductivity. Although the extent of floe
deposits was also linked to discharge rate, the removal of accumulated floe
from the upper-most stretch of the drainage ditch probably contributed
greatly towards reducing downstream accumulations.

The importance of efficient coal ash waste handling procedures has been
emphasized in other studies. Cherry et al. (1979) report recovery of most
invertebrate groups in the swamp drainage system of a power plant in Georgia
following construction of an efficient retaining basin system. Invertebrate
densities had been reduced following periodic perturbations of heavy ash
siltatlon, lowered pH, and exposure to toxic coal ash elements. Coutant
et al. (1978) identified seepage of water through ash deposits and
containment dikes at the Bull Run Steam Plant (TVA) as a significant hazard
to biota in the Clinch River. As a result of their assessment, the authors
note containment banks were later enlarged to eliminate the ditch flow
area. Efficient primary and secondary retaining basin design are viewed as
imperative to prevent serious environmental contamination (Coutant et al.
1978) • Accidental release of large quantities of ponded fly ash following
breakdown of dike walls or basin overflow has proven catastrophic to aquatic
biota. In one such case, invertebrates were eliminated from a 5 to 6 km
stretch of the Clinch River and thousands of fish were killed when a caustic
slug of coal ash was released following failure of a dike surrounding a fly
ash retaining pond (Grossman et al. 1974).

Coal ash wastes have been identified as potentially hazardous
pollutants due to their high suspended solids load, toxic element
concentrations and extremely acid or alkaline nature. The impacts on the
aquatic ecosystems to which they are discharged can be considerable. While
this study and those of Cherry et al. (1979) and Crossman et al. (1974)
demonstrate a relatively rapid recovery from a severely stressed condition,
loss of sensitive species and sublethal stress may be inevitable at any
exposure level.

-------
REFERENCES

American Public Health Association. 1976. Standard methods for examination
of water and wastewater. 14th Ed. Washington, D.C. 1193 p.

Andren. A., M. Anderson, N. Loux, and R. Talbot. 1977. Aquatic

chemistry, p. 9-35. In? Documentation of environmental change
related to the Columbia Electric Generating Station. Eleventh Seni-
Annual Report. Report 92. Institute for Environmental Studies,
University of Wisconsin-Madion.

Andrews, C. B., and M. P. Anderson. 1978. Impact of a power plant on the
groundwtaer system of a wetland. Ground Water 16:105-111.

Aston, R. J., and A. G. P. Milner. 1980. A comparison of populations of
the isopod Asellue aquaticua above and below power stations in
organically polluted teaches of the River Trent. Freshwater Biol.
10:1-14.

Box, G. E. P.., W. G. Hunter, and J. S. Hunter. 1978. Statistics for
experimenters. John Wiley and Sons, Inc., New York. 653 p.

Cairns, J., Jr., K, L. Dickson, and G. Lanza. 1973. tepid biological
monitoring system for determining aquatic community structure in

receiving streams, pp. 148-163. In; J. Cairns, Jr. and K. L. Dickson
(eds.) Biological methods for the assessment of water quality. ASTM
STP 528. Amercian Society for Testing and Materials, Philadelphia,
Pennsylvania,

Cherry, D. S., S. R. tartick, R. K. Guthrie, E. M. Davis, and F. F.

Sherberger. 1979. Recovery of invertebrate and vertebrate populations
in a coal ash stressed drainage system. J. Fish. Res. Bd. Can.
36:1089-1096.

Corkum, L. D., P. J. Pointing, and J. J. H. Ciborowski. 1977. The

Influence of current velocity and substrate on the distribution and
drift of two species of mayflies (Ephemeroptera) . Can. J. Zool.
55:1970-1977.

Coutant, C. C. 1964. Insecticide Sevin: Effect of aerial spraying on
drift of stream insects. Science 146:420-421.

-------
Coutamt, C. C., C. S. Wasserman, M. S. Chung, D. B. Rubin, and M. Manning.
1978. Chemistry and biological hazard of a coal ash seepage stream.
J. Water Pollut. Control Fed. 50:747-753.

Cover, E. C., and R. C. Harrel. 1978. Sequences of colonization diversity,
biomass, and produc tivity of macro invertebrates on artificial
substrates in a freshwater canal. Hydrobiologia 59:81-95.

Crossman, J. S., R. L. Kaesler, and J. Cairns, Jr. 1974. The use of

cluster analysis in the assessment of spills of hazardous materials.
Am. Midi. Nat. 92:94-114.

Cummins, K. W. 1966. A review and future problems In benthic ecology. pp.
2-51. In: Organism-substrate relationships in streams• Special Pub.
No. 4 Pymatuning Laboratory of Ecology, University of Pittsburgh.

Cummins, K. W., and G. H. Lauff• 1969. The influence of substrate particle

size on the microdistrIbution of stream macrobenthos. Hydrobiologia
34:145-181.

Draper, N'. R., and H. Smith. 1966. Applied regression analysis. John
Wiley and Sons, New York. 407 p.

Elliot, J M 1977. Some methods for the statistical analysis of samples

of benthic invertebrates. Freshwater Biological Association Scientific

Publication No. 25. 160 p.

Forbes, A. M., and J. J. Magnuson. 1980. Decomposition and microbial
colonization of leaves in a stream modified by coal ash effluent.
Hydrobiologia 76:263-267 .

Forbes, A. M., J. J. Magnuson, and D. M. Harrell. 1981. Effects of habitat
modifications from coal ash effluents on stream macroinvertebrates. p.
241-249. In: L. A. Krumholz (ed.) Wanswater St reams Symposium.
American Fisheries Society, Washington, D.C.

Fraser, J. 1980. Acclimation to lead In the freshwater lsopod Aeellue
aquaticue. Oecologia (Berl.) 45:419-420.

Gee, J. H., and V. G. Bartnlk. 1969 Simple stream tank simulating a
rapids environment. J. Fish Ke 3. Bd. Can. 26:2227-2230.

Godfrey, F. J. 1978. Diversity as a measure of benthic macroInvertebrate
community response to water pollution. Hydrobiologia 57:111-122.

Guthrie, R. K., E. M. Davis, D. S. Cherry, and J. R, Walton. 1982. Impact
of coal ash from electric power production on changes in water
quality. Water Res. Bull. 19:135-138.

Hall, R. J., G. E. Likens, S. B. Fiance, and G. R. Hendrey. 1980.
Experimental acidH & atl3n of a stream in the Hubbard Brook
Experimental Forest New Hampshire. Ecology 61:976—989.

-------
Harrell, D. M. 1978. Effects on crayfish of exposure to coal ash effluent
and to chromium contaminated food. M.S. Thesis. University of
Wisconsin-Madison. Madison, Wisconsin. 134 pp.

Helake, P. A., E. M. Larsen, W. P. Robarge, R. D. Koons, and J. Thresher.

1976a. Trace elements. p. 13-27. In: Documentation of environmental
change related to the Columbia Electric Generating Station. Seventh
Seal-Annual Report. Institute for Enviornmental Studies, University of
Wisconsin-Madison.

Helmke, P. A., E. M. Larsen, W. P. Robarge, R. D. Koons, J. Thresher, and M.
Schoenfield. 1976b. Trace elements, p. 152-160. In: Documentation
of environmental change related to the Columbia Electric Generating
Station. Eighth Semi-Annual Report. Institute for Environmental
Studies, University of Wisconsin-Madison.

Hester, F. E. and J. S. Dendy. 1962. A multiple-plate sampler for aquatic
macro invertebrates. Trans. Am. Fish. Soc. 91:420-421.

Hilsenhoff, W. L. 1975. Aquatic insects of Wisconsin. Tech. Bull. 89.

Wisconsin Department of Natural Resources, Madison, Wisconsin. 52 p.

Hilsenhoff, W. L. 1982. Using a biotic index to evaluate water quality in
streams. Tech. Bull. 132. Wisconsin Department of Natural Resources,
Madison, Wisconsin. 22 p.

Hocutt, C. H. 1975. Assessment of a stressed macro invertebrate
community. Water Res. Bull. 11:820-835.

Hynes, H. B. N. 1970. The ecology of running waters. University of
Toronto Press, Toronto. 555 p.

Lehmkuhl, D. M. 1979. Environmental disturbance and life histories:
principles and examples. J. Fish. Res. Bd. Can. 36:32 9-334.

Maclorowskl, H. D., R. McV. Clarke, and E. Scherer. 1977. The use of

avoidance-preference bloassays with aquatic invertebrates. Proc. 3rd
Aquatic Toxicity Workshop, Halifax, N.S. Rept. No. EPS-5AR-77-1.
Enviornmental Protection Services, Halifax, Canada. p. 49-58.

Maclorowski, A. F., E. F. Benfield, and J. Cairns, Jr. 1980. Preference-
avoidance reactions of crayfish to sublethal concentrations of
cadmium. Hydrobiologia 74:105-112.

Magnuson, J. J., A. M. Forbes, D. M. Harrell, and J. D. Schwarzmeler»
1980a. Responses of stream invertebrates to an ash pit effluent.
Wisconsin Power Plant Impact Study. EPA-600/3-80-081. U.S.
Environmental Protection Agency. Cincinnati, Ohio.

-------
Magnuson, J. J., F. J. Rahel, M. J. Talbot, A. M. Forbes, and P. A.

Medvlck, 1980b. Ecological studies of fish near a coal-fired electric
generating station and related laboratory studies. Wisconsin Power
Plant Impact Study. EPA-600/3-80-078. U.S. Environmental Protection
Agency, Cincinnati, Ohio•

Mason, W. T., J. B. Anderson, R. D. Kreis, and W. C. Johnson. 1970.

Artificial substrate sampling, macroinvertebrates In a polluted reach
of the Klamath River, Oregon. J. Water Follut. Control Fed. 2:315-328.

Mason, W. T., C. I. Weber, P. A. Lewis, and E. C. Julian. 1973. Factors
affecting the performance of basket and raulitplate macrolnvertebrate
samplers. Freshwater Biol. 3:409-436.

Merritt, R. tf., and K. W. Cummins. 1978. An Introduction to the aquatic
insects of North America. Kendall/Hunt Publishing Co., Dubuque,

Iowa. 441 p.

Newell, R. C. 1973. Factors affecting the respiration of lntertldal
invertebrates. Am. Zool. 13:513-528.

Otto, C. 1976. Factors affecting the drift of Potcmophylax aingulatue
(Trichoptera) larvae. Olkos 27:93-100.

Pennak, R. W. 1978. Freshwater Invertebrates of the United States. 2nd
Ed. John Wiley and Sons, New York. 803 p.

Pielou, E. C. 1969. An introduction to mathematical ecology. Wlley-
Intersclence, New York. 286 p.

Roback, S. S. 1974. Insects (Arthropoda:Insecta) . p • 315-376. In: C. W.
Hart and S. L. H. Fuller (eds.) Pollution ecology of freshwater
invertebrates. Academic Press, New York. 387 p.

Rosenberg, D. M., and A. P. Wiens. 1978. Effects of sediment addition on
macrobenthic invertebrates In a northern Canadian River. Water Res.
12:753-763.

Roy, W. R., R. G. Thiery, R. M. Schuller, and J. J. Suloway. 1981. Coal

fly ash: A review of the literature and proposed classification system
with emphasis on environmental Impacts. Environ. Geol. Notes 96.
Illinois Institute of Natural Resources, Chicago, Illinois. 69 p.

Ruggiero, M. A., and H. C. Merchant. 1979. Water qiality, substrate, and
distribution of macro Invertebrates in the Patuxent River, Maryland.
Hydrobiologia 64:183-189.

Scherer, E. 1977. Behavioral assays - principles, results and problems.
Proc. 3rd Aquatic Toxicity Workshop, Halifax, N. S., Tech. Rept. No.
EPS-5AT-77-1. Environmental Protection Service. Halifax, Canada. p.
33-40.

-------
Siegel, S. 1956. Nonparaoetric statistics for the behavioral sciences.
McGraw-Hill, New York. 312 p.

Soholt, L. F.» N- J. Beskid, W. J. Derikson, M. J. Knight, D. D. Ness,
B. Siskind, and W. S. White. 1980. Handling of combustion and
emission-abatement wastes Cor coal-fired power plants: Implications
for fish and wildlife resources. FWS/OBS-80/33- Washington, D.C. U.S.
Fish and Wildlife Service- 184 p.

U.S. Environmental Protection Agency. 1976. Quality criteria foe water,
U.S. Environmental Protection Agency, Washington, D.C. 256 p.

Walton, 0. E. 1978. Substrate attachment by drifting aquatic insect
larvae . Ecology 59: 1023-1030.

Waters, T. F. 1972. The drift of stream insects. Ann. Rev. Entomol.
17:253-272.

Weber, C. I. 1973. Biological field and laboratory methods for measuring
the quality of surface waters and effluents. Environmental monitoring
series. EPA-670/4-73-001. U.S. Environmental Protection Agency,
Cincinnati, Ohio. 125 p•

Webster, K. E., A. M. Forbes, and J. J Hagnuaon. 1981. Behavioral

responses of stream macrolnvertelrates to habitat modification by a
coal ash effluent. p. 408-414. In: L. A. Kruraholz (ed.) Warrawater
Streams Symposium. American Fisheries Society, Washington, D.C.

Wentsel, R., A. Mcintosh, W. P. McCafferty, G. Atchison, and V. Anderson.
1977a. Avoidance response of midge larvae {.Chironomus tetane) to
sediments containing heavy metals). Hydroblologia 55:171-175.

Wentsel, R., A. Mcintosh, and C. Atchison. 1977b. Sublethal effects of
heavy metal contaminated sediment on midge larvae {Chironomus
tetans). Hydroblologia 56:153-156.

Wilho, J. L., and T. C. Dorris. 1968. Biological parameters for water
quality criteria. Bioscience 18:477-481.

Winner, R. W•, and M. P. Farrell. 1976. Acute and chronic toxicity of
copper to four species of Daphnia. J. Fish. Res. Bd. Can. 33:1685-
1691.

-------