xvEPA
United States
Environmental Protection
Agency
TECHNOLOGY TRANSFER
Biomonitoring to
Achieve Control
of Toxic Effluents
EPA/625/8-87/013
*^-*>sS?5i*r-:, ••.'•sA&s&i-ti.. fe
-------
-------
EPA/625/8-37/01 3
Biomonitoring to Achieve
Control of Toxic Effluents
September 1987
U.S. Environmental Protection Agency
Office of Water, Permits Division
Washington, DC 20460
Office of Research and Development
Environmental Research Laboratory
Duluth, MN 55804
Printed on Recycled Paper
-------
This document was prepared by Alexis Steen of EA Engineer-
ing, Science, and Technology, Inc., subcontractor to JACA
Corp. of Fort Washington, PA. Assistance was provided by Rick
Brandes and Thomas Wall of EPA Office of Water Enforcement
and Permits; Orville Macomber of EPA Center for Environmen-
tal Research Information; and Donald Mount of EPA Environ-
mental Research Laboratory in Duluth, Minnesota.
The principal investigator was Donald I. Mount. Assistance
was provided by Daniel Dudley of Ohio EPA in Columbus and
Robert Wysenski of Ohio EPA's Northeast District Office. Mr.
Wysenski's co-workers, Robert Davic and David Stroud, com-
piled detailed information on the plant flows and bypassing
volumes. Sample collection and testing was conducted by
Teresa Norberg-King and Floyd Boettcher at U.S. EPA's Envi-
ronmental Research Laboratory-Duluth. On-site assistance and
some sample collection was provided by personnel of the City
of Akron: Randall Monteith, Frank Ertle, and their staff. The au-
thors wish to acknowledge the assistance of D.F. Bishop, Cor-
nelius Weber, Timothy Neiheisel, and William Horning of U.S.
EPA Office of Research and Development in Cincinnati, Ohio,
and James Giattina of U.S. EPA Region V in Chicago, Illinois.
Technical review of the document was provided by William
Peltier, III, EPA Region IV; Charles Webster, Ohio EPA; Linda
Anderson-Carnahan of EPA Region V; Fred Bishop of EPA Of-
fice of Research and Development; Daniel Dudley of Ohio EPA;
and Randall Monteith of the City of Akron. Editorial and techni-
cal assistance was also provided by Camilla Greene and
Sheree Romanoff of JACA Corp., Fort Washington, PA.
This report has been reviewed by the U.S. Environmental Pro-
tection Agency and approved for publication. The process al-
ternatives, trade names, or commercial products are only ex-
amples and are not endorsed or recommended by the U.S.
Environmental Protection Agency. Other alternatives may ex-
ist or may be developed. In addition, the information in this
document does not necessarily reflect the policy of the Agen-
cy, and no official endorsement should be inferred.
This guidance was published by
U.S. Environmental Protection Agency
Center for Environmental Research Information
Office of Research Program Management
Office of Research and Development
Cincinnati, OH 45268
-------
Foreword
When the Cuyahoga River caught fire at Cleveland, Ohio
on June 22,1969 the river immediately became a symbol
of the degraded state of the nation's surface waters.
Since then, significant efforts have been made to im-
prove the river's water quality, and the Cuyahoga is a |
much cleaner river than it was 18 years ago.
Downstream from Akron, Ohio, the Cuyahoga flows
through rural surroundings and the Cuyahoga Valley
National Recreation Area before reaching Cleveland,
where it discharges into Lake Erie. It is a shallow, swift-'
flowing river that has the potential for being an excellent
warm water habitat, which is the river's designated use.
The physical habitat necessary for supporting a healthy
fishery is present and the oxygen levels in the free-flowj
ing portions of the river are consistently high. In 1984,
however, the few fish that were found in a 16-km stretch
of the river downstream of Akron were juveniles of pol-
lution-tolerant species (white suckers and creek chubs), j
most of which had lesions, eroded fins, or skeletal defor-
mities. Few of the fish were expected to live to adult-
hood.
The Akron Publicly Owned Treatment Works (Akron
POTW) discharges its effluent at the point in the river
where a severely depleted fish population was first ob-;
served. The Akron POTW is a secondary treatment facili-
ty for 2.2 to 4.4 m3/s (50 to 100 mgd)*of industrial and do-
mestic waste. At certain times during the year, the
wastewater comprises up to 60 percent of the total flow^
of the river.
This report provides a case study of how water quality-,
based toxicity control procedures can be combined with
chemical analyses and biological stream surveys to
achieve more effective water pollution control. It de-
scribes how regulatory agencies used laboratory toxicity
testing and biological stream surveys to confirm that the
Akron POTW causes the Cuyahoga's water quality prob--
lem downstream from Akron, and that effluent toxicity,
rather than conventional or nonconventional pollutants,
causes the effects observed. It describes the toxicity testr
ing and chemical analysis procedures that researchers
used to search for toxicants responsible for the effluent
toxicity. Finally, the report presents sample permit limit
derivation for the Akron POTW. This derivation is pre- i
sented for educational purposes only and does not re- ;
present an official EPA regulatory action. Derivation cal1
culations are followed sequentially through to ',
recommendations which outline the toxicity tests that ;
would be most valuable to monitor the toxicity of the Ak-
ron POTW's effluent.
in
-------
Contents
1. Introduction 1
1.1 Regulatory Background 1
1.2 The Cuyahoga River Study Area 1
2. Site Selection Process 3
2.1 Identification of the Cuyahoga River as Having a Water '
Quality Problem 3
2.2 1984 and 1985 Fish Survey Data 3
2.3 1984 Benthic Invertebrate Survey Data 5
2.4 Evaluation of Stream Surveys 6 '•
2.5 Previous Effluent Toxicity Surveys 7
2.6 Post-Study, 1986 Stream Survey Data 8
3. Site Description and Reconnaissance Data 9
3.1 Site Description 9
3.2 Initial Sample Collection During Site Reconnaissance 12
3.3 Site Reconnaissance Toxicity Test Data 12
4. Development of the Study Plan for Toxicity Assessment 15
4.1 Introduction 15
4.2 Effluent Sampling and Toxicity Analysis Plan 15
4.3 Ambient Sampling and Toxicity Analysis Plan 16
4.4 Ambient River Station Descriptions 16
4.5 Toxicity Testing Procedures 18
5. Toxicity Testing Results and Data Interpretation 20
5.1 Introduction 20
5.2 Akron POTW Effluent Toxicity Results 20
5.3 Ambient River Toxicity Results 25
5.4 Summary 27
6. Toxicity Source Investigation 31
6.1 Introduction 31
6.2 General Procedures for the Toxicity Source Investiga-
tion 31
6.3 Phase I Test Procedures 32
6.4 Phase II Test Procedures 33 j
6.5 Phase I Data and Results 34
6.6 Phase II Data and Results 35 ;
6.7 Wastewater Bypassing Concerns 36
6.8 Summary 37
7. Conclusions and Application of Case Study Data to Setting
Permit Limits 38
7.1 Introduction 38
7.2 Toxicity Testing Needs to Reduce Uncertainty Levels ; 38
7.3 Derivation of Sample Permit Limits 39
7.4 Conclusions 43 ;
8. References 45
8.1 Complex Effluent Toxicity Testing Program Reports i46
Glossary 47
IV
-------
1. Introduction
This case study describes the water quality-based toxic-
ity control procedures that researchers used to address
a serious water quality problem on the Cuyahoga River
in northeastern Ohio. The objective was to provide Na- <
tional Pollutant Discharge Elimination System (NPDES) ,
program managers, NPDES permit writers, and water ;
quality specialists with an example of how toxicity test-,
ing can be used to address pre-identified toxic water '
quality problems.
1.1 Regulatory Background
During the 1960s and 1970s the major focus of pollution:
control was conventional pollutants — oxygen-demand-
ing materials, heat, and suspended solids — which were
causing severe degradation of rivers, lakes, and
streams. Industries installed technologies to control the!
discharge of these types of pollutants and billions of
dollars were invested in publicly owned treatment
works (POTWs). Progress was so dramatic that by the |
1980s, point source contributions to conventional pollu-i
tion problems were considered largely under control.
The focus of point source abatement programs then
shifted to the control of toxic chemicals.
i
In regulating toxic discharges, water pollution control
agencies have usually used a "chemical-specific" ap-
proach. Typically, a regulatory authority would require a!
large industrial discharger to monitor and submit data
on the concentrations at which the priority pollutants'
list toxicants occurred in its wastewater discharge. The :
concentrations reported were often based on one efflu- |
ent sample per week or month. To keep laboratory test- !
ing costs low, regulators would allow multipollutant
scans with detection levels for individual compounds of
about 50 to 100 parts per billion — levels higher than the ;
toxic effect concentrations of some priority pollutants. :
For many dischargers, regulatory authorities set con-
centration limits on compounds based on state water :
quality standards. Many such standards limit only the
concentration of 10 or fewer metals and other inorgan- •
ics, and five or fewer organic pollutants. This occurs
when states have adopted numerical criteria based on
EPA's Red Book values from 1976 (1), although narrative >
water quality standards are now available.
Nationwide, certain categories of industrial dischargers
are required to meet minimum treatment levels (e.g.,
best available technolgy [BAT], best possible technology i
[BPT]) for a list of conventional and priority pollutants
that have been found in the effluent discharged by most.
facilities in that category. These treatment standards do
not necessarily relate to acceptable concentrations to ;
prevent toxicity to aquatic life. Only POTWs that receive
wastes from many industrial users, or users in certain
industrial categories, must analyze their effluent for toxi-
cants. Using this chemical-specific methodology, regu-
lators have reduced the discharge of the priority list
toxicants and of other toxicants with similar physical
and chemical properties by millions of pounds annually.
For many dischargers this level of control has reduced
effluent toxicity to acceptable levels. Limited toxicity
testing by the U.S. Environmental Protection Agency •
(EPA), some states, and the regulated community, how-
ever, has shown that controlling conventional pollutants
and the 126 priority pollutants does not reduce or elimi-
nate the toxic effects of all discharges. In particular,
large, multiprocess industrial facilities and POTWs that
treat industrial plant wastewaters may have effluents
laden with nonpriority list toxicants. These toxicants
may significantly contribute to the degradation of some
surface waters. In less complex situations a discharger's
effluent may contain a few nonpriority list toxicants in
concentrations that adversely affect the instream biota
and exceed water quality standards.
A chemical-by-chemical approach to controlling non-
priority toxic pollutants can be inadequate for several
reasons. Many of these compounds cannot be detected
by gas chromatography/mass spectroscopy (GC/MS) or
other widely available detection methods. Of the subset
of toxicants that can be detected by GC/MS analysis,
complex wastewaters often contain many more than
can be identified using the most comprehensive mass
spectroscopy libraries. For many of the compounds that
can be detected and identified, little or no definitive
aquatic toxicology data are available. These difficulties
point up the need for a more direct and cost-effective
method for toxicity assessment and control.
1.2 The Cuyahoga River Study Area
The Cuyahoga River's water quality problems down-
stream from the Akron POTW represent a typical case of
the need for advanced water quality-based pollution
controls for certain dischargers on certain receiving wa-
ters. The Akron POTW is a well-operated municipal
wastewater treatment facility that achieves significant
reductions in conventional and toxic pollutant levels
through sophisticated treatment processes (see Sec-
tions 2.5 and 3.1.1). Yet the condition of the Cuyahoga
River downstream of the Akron POTW as a fishery is
poor and its recreational and aesthetic value is lowered.
At the time of this study, several factors indicated that
the Akron POTW was causing the water quality problem,
and that toxic rather than conventional or nonconven-
-------
INTRODUCTION
tional pollutants were producing the adverse effects on
water quality.These factors were:
» Industrial wastes comprised about 14 percent of the
influent to the Akron POTW. These wastes may have
contained many toxic pollutants which exhibited
inhibitory effects to the biota at very low concentra-
tions.
• GC/MS scans of the effluent showed several organ-
ic pollutants in the effluent. The pollutants caus-
ing the toxicity may not even have been seen in
these scans because their low concentrations were
undetectable or they could not be extracted for
analysis.
" The Akron POTW effluent constituted up to 60 per-
cent of the river's flow at certain times of the year.
• The fewfish found in the river exhibited characteris-
tics such as lesions and other deformities associat-
ed with exposure to toxicants.
The Akron POTW was selected for the case study as a re-
sult of cooperative efforts between various Federal and
state agencies. The Ohio EPA, U.S. EPA Office of Re-
search and Development (ORD), Cincinnati, and U.S.
EPA Region V identified the toxicity problem as de-
scribed in Chapter 2. A reconnaissance trip was con-
ducted by U.S. EPA Environmental Research Laboratory
(ERL), Duluth, and others to confirm the site selection
and to refine the study plan for an on-site test program.
The results of this field reconnaissance trip are present-
ed in Chapter 3. The evolution of the study plan (Chapter
4) is presented in view of the site selection data, recon-
naissance data, and site description. Chapter 5 contains
the results of the on-site testing of effluents and ambient
river water.The toxicity source investigation for identify-
ing toxic components and to suggest appropriate treat-
ment technologies is presented and used to make sam-
ple NPDES permit recommendations for the Akron
POTW (Chapter 7). Technical methods are not included
in this document but are referenced as appropriate.
-------
2. Site Selection Process
2.1 Identification of the Cuyahoga River as
Having a Water Quality Problem
Ohio environmental regulatory agencies (e.g., Departmerjt
of Health and Ohio EPA) and other cooperating groups '
have recognized problems and have monitored conditions
in the Cuyahoga River at intervals over the last 40 years.
During that time, the discharge of municipal and industrial
wastes has been substantially reduced and there have
been documented improvements in the chemical water .
quality of the river (2,3,4,5,6,7). In 1960, dissolved oxygen
(DO) concentrations measured in the Cuyahoga down-
stream from Akron showed that a 27-km segment of the
river between Akron and the Canal Diversion Dam (near
river kilometer [RK] 32) was anoxic (see Glossary) (4). ;
Since that time, the DO profile has steadily improved and
1984 to 1985 results show that DO is no longer a limiting!
factor to the biota (8,9). Likewise, the instream average :
ammonia concentrations downstream from the Akron !
POTW have declined from >5 mg/L in 1969 to <1 mg/L in
recent years (8,9).
The segment of the Cuyahoga River downstream from the
City of Akron was termed grossly polluted in the 1960s (4).
Benthic macroinvertebrate sampling was conducted in i
1973,1974, and 1980 by researchers at the University of ;
Akron, with results indicating that the macroinvertebrate <
communities in the river between Akron and Indepen- i
dence, Ohio were predominantly pollution-tolerant organ-
isms, primarily pulmonate snails (Physidae), oligochaetes,
and midges (Chironomidae) (see Figure 2-1) (10, 11). Con-i
ditions were blamed primarily on pollutants originating in
the Akron area of the river.
Historical data on the fish communities in the Cuyahoga '.
River downstream from Akron are rather sparse. Fish
were not observed or collected at any of 12 sampling sites
between the Akron POTW (RK 62) and Independence, I
Ohio (RK 23), in the summer of 1967 survey (4). Intensive
sampling in the early 1970s between the Akron POTW and
Peninsula, Ohio (RK47) yielded no fish, while downstream
from the dam in Peninsula, a fish community composed
only of 19 species representing three families (Cyprinidae,
Catastonidae, Centrarchidae) was found (12, 13).
U.S. EPA Region V data analyses from 1981 indicated a
water quality problem in the Cuyahoga due to the Akron ,
POTW. Concentrations of phthalate esters, phenols, and !
ammonia in ambient water samples were greater than ;
state standards at the three ambient stations nearest to, |
and downstream of, the Akron POTW (the Ohio phenol
standard is set to prevent taste and odor problems, not !
toxicity). A number of actions were initiated to bring the I
Akron POTW into compliance with its NPDES permit lim-
its. These included improvements in sludge handling, pri-
mary settling, aeration, final settling, and phosphorus re-
moval. As a result, the Akron POTW made a number of
structural and process changes to improve the efficiency
of operation and treatment.
In 1984 Ohio EPA conducted a biological impact assess-
ment on the main stem of the Cuyahoga. The results of
this study indicated a 16-km reach where fish collections
(species and individuals) were much reduced. This reach
traversed a rural area, including a portion of a national
park and recreation area. The Akron POTW is located at
the upstream end of this 16-km reach.
2.2 1984 and 1985 Fish Survey Data
Water quality data collected in 1984 confirmed that the
chemical/physical condition of the Cuyahoga River
downstream from Akron had greatly improved (2).
Along with improved DO and lower ammonia concen-
trations, the data also revealed that levels of other con-
ventional pollutants (e.g., biochemical oxygen demand
[BOD], suspended solids [SS]) and chlorine and heavy
metals met or nearly met (less than 10 percent frequen-
cy of exceedance) Ohio EPA water quality standards for
warm water aquatic life habitat. ;
Physical habitat conditions were excellent and judged to
be fully capable of supporting a warm water fish com-
munity (2). The biological data, however, revealed that
the river was nearly devoid offish for 27 km down-
stream of the Akron POTW. The fish survey was con-
ducted along a 104-km segment of the river in 1984 and
along a 72-km segment in 1985. Two ecological indices
were used to characterize the overall health of the fish
community and the river system in 1984. The profile of
the fish community using the Index of Well-Being illus-
trated a rapid decline downstream from the Akron
POTW (Figure 2-2) (14). The loss of nearly all fish just
downstream of the Akron POTW and the failure of the
fish community to recover to upstream levels indicated
a persistent toxic influence. A similar pattern and sever-
ity of impact was also apparent using the Invertebrate
Community Index (ICI) (See Section 2.3) (13). Some
component of the Akron POTW discharge was limiting
the fish fauna, although the conditions measured in
1985 were slightly improved (Figure 2-2). The predomi-
nant species were juvenile white suckers and creek
chubs (13). Observations of fin erosion, lesions, and ex-
ternal deformities on the fish collected in 1985 added
more evidence of a serious environmental stress in the
Cuyahoga River downstream from Akron (9).
-------
SITE SELECTION PROCESS
• Akron POTW
•••— River-basin boundary
• USGS gauge and water-quality monitor
Kilometers
0 2 4 6 8 10
FIguro 2-1. Tha Cuyahoga River Basin (19)
-------
SITE SELECTION PROCESS
Lake
Rockwell
Little
Cuyahoga
River
Peninsula
Dam
Tinkers
Creek
Cleveland
Harbor
Monroe
Falls Dam
Akron
POTW
Brandywine
Creek
Ship
Channel
I2r
_ 10
m
CD
tt °
OQ
_j
_1
111
National Recreational Area
Q
Z
Q
Q
o
Poor
' Very Poor
100
90
80
70
60
50
40
30
20
10
RIVER KILOMETER
Figure 2-2. Profile of the Cuyahoga River Fish Community Using a Modified Index of Well-Being (17)
2.3 1984 Benthic Invertebrate Survey Data
Invertebrate samples were collected in the middle and
lower reaches of the river by Ohio EPA in 1984 (8). Elev-
en stations were located at positions upstream of the i
Little Cuyahoga River, upstream from the Akron POTW,:
and at numerous locations downstream from the Akron
POTW for approximately 48 km. Longitudinal profiles of
taxa richness, Shannon-Wiener diversity index values,
and taxonomic composition are shown in Figures 2-3
and 2-4.
In addition, Ohio EPA developed the Invertebrate Com-i
munity Index (ICI) to measure aquatic invertebrate comj
munity health. This index is composed of eight mea-
sures of taxa and density, and results in ICI values are
designed to reflect stream water quality. Invertebrate
data from 1984 yielded ICI values that reflected good ,
water quality upstream of the Akron POTW at the con-
fluence with the Little Cuyahoga River. Reduced ICI val-
ues reflected poor to fair water quality downstream of
that confluence and for 44 km downstream of the Akron
POTW (Figure 2-5). This decline was first observed at a
station between the confluence and the Akron POTW, in-
dicating that another discharger may have influenced
the water quality, although toxicity data do not show
any effect upstream of the Akron POTW (see Sections,
2.5 and 3.2). An intermediate area near the confluence
-------
SITE SELECTION PROCESS
Lake
Rockwell
Little
Cuyahoga
River
Peninsula
Dam
Tinkers
Creek
Cleveland
Harbor
Monroe
Falls Dam
Akron
POTW
Brandywine
Creek
Ship
Channel
U
60r
SO
40
National Recreational Area
ffl 30
u.
O
s
1 20
z
10
-16.0
5.0
g
m
4.0 c/>
3-0
2.0
1.0
110
100
90
80
70
60
50
40
30
20
10
RIVER KILOMETER
Flguro 2-3. Profile of the Number of Benthic Taxa and Benthic Diversity on the Cuyahoga River
with Tinkers Creek had better ICI values, indicating some
recovery, but that dissipated within 8 km.
A good invertebrate community predominated by pollu-
tion-intolerant and pollution-intermediate mayflies, cad-
disflies, and midges was collected at the station up-
stream from the Little Cuyahoga River. Substantial
compositional changes in the community (from insect
to noninsect dominated) initially occurred downstream
from the Little Cuyahoga followed by further changes
downsteam from the Akron POTW. Less dramatic de-
clines also occurred in taxa richness, density, and diver-
sity over this area of the Cuyahoga River. Predominant
organisms at nearly all the sites downstream of the Ak-
ron POTW were the pulmonate snail genus Physella, the
mayfly species Bactis intercalaris, and the midges Crico-
topus bicinctus, Polypedilum (Polypedilum) fallax
group, and Conchapelopia. This association of inverte-
brates has not been typically encountered in other Ohio
rivers with heavy organic loadings from POTWs. In addi-
tion, the typical community response to organic con-
tamination (i.e., degradation followed by various stages
of downstream recovery) did not occur downstream of
th Akron POTW. Relatively little community change was
detected for nearly 32 km downstream of the Akron
POTW,
2.4 Evaluation of Stream Surveys
Recent sampling has indicated much improved water
quality conditions in the Cuyahoga River. Yet fish sur-
veys have rarely caught fish downstream of the Akron
POTW (Section 2.2), and benthic invertebrate surveys of
this same area have indicated a community change
(Section 2.3). Treatment improvements had succeeded
in reducing levels of conventional pollutants (e.g., SS)
-------
SITE SELECTION PROCESS
Little
Cuyahoga
River
110
100
90
^ 80
d^
« 70
^
w 60
u 50
DC
0 40
30
20
10
Peninsula
Dam
Akron
POTW
Tinkers
Creek
2 Non-Insect
Other Insect
Midge
Caddisfly
Mayfly
68.5 64.7 59.8 56.8 53.4 46.5 33.5 27.8
RIVER KILOMETER
25.1
21.1
15.3
Figure 2-4. Benthic Invertebrate Community Composition From Selected Stations on the Cuyahoga River (7)
discharged to the river; however, a corresponding grad-
ual improvement in aquatic organism populations or
community composition compared to upstream condi-
tions was not observed. In 1985 Ohio EPA began to ex-
amine the possibility of other types of pollutants such as
toxicants.
2.5 Previous Effluent Toxicity Surveys
A separate survey of six Ohio POTWs was conducted by
U.S. EPAORD Cincinnati in 1984 and 1985. The results of
this survey indicated that of the six POTWs, the Akron
plant was receiving the most toxic raw wastewater for
treatment (15). Results of toxicity tests on fathead min-
nows (Pimephalespromelas) and a cladoceran (Cerio-
daphnia dubia) also showed reduced toxicity with treat-:
ment, but no observed effect concentration (NOEC)
values were still 10 to 100 percent. These data indicated
that the treated effluent discharged by the Akron POTW
could still be toxic. The data did show variable toxicity,
but since the effluent often constituted a large portion of
the Cuyahoga River flow, the effluent could have been a
cause of the observed stream effects. The State of Ohio
water quality standards limit acute toxicity within the
mixing zone to less than rapidly lethal conditions, re-
gardless of the quantity of water available for dilution.
The ORD-Cincinnati chemical analyses indicated that
trace metal concentrations were reduced in the effluent
and that conventional pollutant removal requirements
were met (16). This indicated that the Akron POTW was
treating its influent and achieving large reductions in
concentrations of metals and extractable organics, and
that plant performance was not an obvious cause of ef-
fluent toxicity or adverse biological effects. Investigative
efforts were then concentrated on determining the vari-
ability of the effluent toxicity, the magnitude of the tox-
icity, ambient toxicity, and eventually the identification
of toxicants in the effluent.
Since the effluent discharge composed a large portion
of the Cuyahoga River flow (from 20 to 60 percent),
there was probably insufficient initial dilution to remove
toxicity. In addition, examination of plant operating re- ;
cords showed that when influent volumes were too ,;
great, the plant would bypass raw or partially treated
-------
SITE SELECTION PROCESS
Little
Cuyahoga
River
Peninsula
Dam
Tinkers
Creek
Cleveland
Harbor
Monroe
Falls Dam
Akron
POTW
Brandywine
Creek
Ship
Channel
o
I
i
I
60
50
40
30
20
National Recreation Area
J8
S 10
Exceptional
Good
Fair
- Poor
_ Vary Poor
1984
1986
110
100
90 80
70
60
50
40
30
20
10
RIVER KILOMETERS
Fljjuro 2-5. Profile of the Cuyahoga Macroinvertebrate Community Using an Invertebrate Community Index (17)
wastewater into the Cuyahoga River. Bypassing after a
rain may have produced a diluted but possibly toxic dis-
charge. As a result of the observed instream effects on
fish and benthic organisms as well as effluent toxicity,
the Akron POTW was selected for a detailed toxicologi-
cal evaluation.
2.6 Post-Study, 1986 Stream Survey Data
Ohio EPA repeated their previous sampling for inverte-
brate and fish data during the summer of 1986. ICI val-
ues demonstrated a similar pattern to that found in
1984, showing lower water quality downstream of the
Akron POTW. The decline in ICI values downstream of
the Akron POTW, however, occurred over a much
shorter distance, and the maximum values up- and
downstream of the Akron POTW, were greater in 1986
than in 1984 (Figure 2-5).
Fish sampling showed increased numbers and in-
creased diversity in the Cuyahoga in 1986. In addition,
Ohio EPA noted better use of available habitat by the
fish (14). In contrast to 1985 data, adult white suckers
were found in 1986 and in general, fish abundance in-
creased as distance downstream from the Akron POTW
increased (14). Fish community recovery can also be ob-
served over time and with distance from Akron POTW
using the Index of Weil-Being (Figure 2-2).
The Ohio EPA noted that these improvements were in-
cremental and the Cuyahoga River biological communi-
ties were poor and still reflected toxic effects (14). Efflu-
ent toxicity tests during the summer of 1986 resulted in
LC50 values >100 percent effluent forCeriodaphnia (Ta-
ble 6-1). This absence of effluent toxicity should aid in
the continuation of improvements to the Cuyahoga Riv-
er. The Ohio EPA will sample again during the summer
of 1987 for water quality, invertebrate community
health, and fish community health to document any fur-
ther improvements to the Cuyahoga River and to verify
; the absence of acute toxicity noted in 1986.
-------
3. Site Description and Reconnaissance Data
3.1 Site Description
3.1.1 The Akron POTW i
The Akron POTW is on the eastern bank of the Cuyahoga
River located northwest of Akron, Ohio at approximately
RK 62. The service area is approximately 194 km2 and con-
tains an estimated population of 382,000. Wastewater flow
to the Akron POTW is 35 percent residential, 14 percent inj-
dustrial, and 51 percent infiltration plus stormwater and
miscellaneous flows (18). The area is served by a 1,100 km
separate and combined stormwater and sewage collec- '.
tion system.
Wastewater settling basin (chlorine contact chamber and EPA
mobile laboratory in background). :
The original plant was constructed in 1928. Since then,
major revisions have been made so that the treatment
processes now consist of grit removal, screening, preaera-
tion, chemical treatment for phosphorus removal, primary
settling, aeration (secondary treatment), final settling, and j
chlorination. The plant is capable of bypassing influent i
(raw wastewater) prior to entering the plant. In addition,
flows in excess of 6.6 m3/s (150 mgd) may be shunted after'.
grit removal to the equalization basin and then to the chlo-
rine contact basin before discharge. The process schemat-
ic diagram illustrates the waste treatment and indicates :
the locations of bypassing (Figure 3-1). The partially treat-'
ed wastewater is chlorinated with the final effluent only
during the seasonal chlorination period of May 1 to Octo-
ber 31. The effluent has its greatest effect from June to Oc- •
tober when it often constitutes more than 20 percent of the
river flow.
Influent (raw) and partially treated wastewater bypasses
occur irregularly. The Akron POTW Monthly Operating Re-
ports indicate that 1980 to 1985 total raw bypass volumes
varied from 1,134,000 m3/yr to 11,340,000 m3/yr (300 to
3,000 mil gal annually) (19). Annual volumes of partially .
treated wastewater bypasses are probably similar, but to-
tals are not available since the volumes are not reported
during the seasonal chlorination period of May 1 to Octo-
ber 31. The bypass volumes alone:exceed 10 percent of
the Cuyahoga River flows (Chapter 5).
3.1.2 The Cuyahoga River
The Cuyahoga River has a designated use as a warm wa-
ter habitat, capable of supporting warm water species.
The river flows north into Lake Erie from headwaters in
Geauga County to the Akron area and through rural Ohio
to Cleveland (Figure 2-1). The river runs through the Cuya-
hoga Valley National Recreation Area for much of the
length of the study area. The river is shallow and the water
is sufficiently clear to reveal that silty sediments have coat-
ed the river bottom in some areas. Such sediments may,
remove some toxicants from the water column, placing
them in greater proximity to benthic organisms and per-
haps at higher concentrations than in the water column.
The study area encompassed portions of the middle and
lower reaches of the Cuyahoga. The middle reach ex-
tends from Lake Rockwell to the Akron POTW, and two
of the major tributaries are in the study area (Little Cuya-
Cuyahoga River downstream of Station 10.
-------
SITE DESCRIPTION/RECONNAISSANCE
Overflow
Supernatant to
Wasto Liquor
Equalization or
Pre-aaratlon
Afh
to
Landfill
• Wastewatsr
Chemical Treatment
for Phosphorus
Removal (Sodium-
Alum or Others)
Pre-aeratibn
, Tanks (Designed
for 30 Min <§>
24 Flow Retention
Basins 10.4 MG
j Volume 60 MG Cap.
Chemical
Treatment for
Phosphorus &S.S.
Removal (Sodium-
Alum & Polymer)
Scum to
Waste Liquor
Stabilization
(Lime)
Pre-aeration
Screening Grit
to Landfill
n
83MGD
'.
00
00
,
'
Influ
83
.(Ai
3 Course
Bar
Screens
Tanks
Filtrate
Plant
" Influent
Raw Influent
By-Pass
^Primary
[Thickener
Sludge 5-9% Solids
Sludge Mixing El-
Blending Tanks i
Overflow From
D.A.F. to W.L.E.
Waste Activated Sludge (W.A.S.) ^
Chlorine Contact Tanks
L (Designed for 30 Min @ 86 MGD)
Effluent to
Cuyahoga River
1-2% Solids Disso|VBd Air
Flotation
Thickeners
.(D.A.F.)
.Zimpra Ash Slurry
Sludge Slucf9e M{*ln®
Blending Tanks
4-Belt Filters
Chemical
Conditioning
2-Vao Filters
4-lncinerators
— — — — Bypass/Overflow
SOURCE: City of Akron Water Pollution Control Division.
Ash to
Landfill
To Future
Compost Facility
Figure 3-1. Schematic Diagram of the Akron POTW Processes
hoga River and Mud Brook). Water quality is considered
to be degraded in the middle reach (20). The lower reach
extends from the Akron POTW to Lake Erie and includes
the Cuyahoga Valley National Recreation Area. The por-
tion of the study area in the lower reach has five tributar-
ies: Yellow Creek, Furnace Run, Brandywine Creek,
Chippewa Creek, and Tinkers Creek (Figures 2-1 and 3-
2). A 1980 Ohio EPA report considered the waters from
the Akron POTW to Furnace Run to be grossly polluted
(21),
Near the Akron POTW the river is approximately 7 m
wide and has had a mean flow at Old Portage of 12.1
m3/s over the 59-year period 1925 to 1984 (7). During the
low flow periods in the summer, the Akron POTW dis-
charge returns much of the water removed from the
Lake Rockwell water supply reservoir for use in the Ak-
ron area and this contributes as much as 60 percent of
the river flow (19). Minimum flows of 0.7 m3/s have been
recorded twice at Old Portage in the last 59 years (7),
where the 7Q10 value for September to November is
1.27 m3/s based on 39 years of data (22). Downstream at
the Independence, Ohio gauging station the mean flow
has been 23.3 m3/s, nearly twice that measured at Old
Portage. The 7Q10 value for the period of September to
November is 1.92 m3/s, based on 44 years of data at the
Independence gauge (22).
The Cuyahoga River and its tributaries receive point
source discharges from various industrial and commer-
cial operations. On the Cuyahoga River proper, Ohio
EPA has issued NPDES permits to 32 dischargers (one of
which is the Akron POTW). There are point source dis-
charges to Mud Brook, Yellow Creek, Furnace Run,
Brandywine Creek, and Tinkers Creek, as well as to other
tributaries. Field stations were located by each of these
four tributaries during the full-scale evaluation. In addi-
tion, a landfill adjacent to the Akron POTW has a small
intermittent discharge to the Cuyahoga River.
10
-------
SITE DESCRIPTION/RECONNAISSANCE
Sampling Stations
O Reconnaissance Sampling
Stations
• POTW
A USGS Gauge
Brandywine Creek
Station 8
Peninsula Road
Furnace
Run J
Cuyahoga River
Figure 3-2. Location of Sampling Stations and Tributaries in Study Area on the Cuyahoga River, Ohio
11
-------
SITE DESCRIPTION/RECONNAISSANCE
3.2 Initial Sample Collection During Site
Reconnaissance
After the Akron POTW site was identified as having a
water quality problem probably due to toxics, an on-site
visit was planned for ERL-Duluth to conduct initial toxic-
ity tests. By the time the reconnaissance visit was ar-
ranged, the following was known:
« The Akron POTW effluent had been shown to be
toxic to fathead minnows and Ceriodaphnia (U.S.
EPA Cincinnati data).
» Ambient water collected just upstream of the Akron
POTW had been shown to have no toxic effects on
fathead minnows (15) and concentrations of met-
als up- and downstream of the outfall were similar
(ORD-Cincinnati data).
m The effluent contains measurable levels of extract-
able organics and metals (ORD-Cincinnati data).
« The effluent had been cited as causing a water qual-
ity problem due to ammonia and phenol levels
which exceeded Ohio state standards (23). Howev-
er, recent water quality data on ammonia and DO
concentrations had shown improved conditions,
but had not indicated any relationship to the ob-
served biological conditions downstream of the Ak-
ron POTW since these had not improved commen-
surately (Ohio EPA data).
• The discharge volume of the Akron POTW is large
compared to the Cuyahoga River flow. At typical
flows of 2.9 m3/s (67 mgd), the Akron POTW dis-
charge contributes more than 30 percent of the
mean river discharge. During low river flows the ef-
fluent may comprise up to 60 percent of the flow.
This typical flow level also represents over twice
the flow of the Cuyahoga River under 7Q10 condi-
tions at Old Portage gauge.
• The Little Cuyahoga River joins the Cuyahoga River
several kilometers upstream of the Akron POTW
and may supply toxic materials.
A reconnaissance trip was conducted from July 15 to 17,
1985, to familiarize the ERL-Duluth staff with the site and
to collect effluent and ambient water samples. Samples
of river water were collected from four areas near the
eventual location of Stations 3,4,6, and 9 during the re-
connaissance site visit (Figure 3-2). These particular sta-
tions were selected to obtain samples from just up-
stream and downstream of the outfall (Stations 3 and 4),
several kilometers downstream at the confluence with
Furnace Run (Station 6), and farther downstream at the
confluence with Brandywine Creek (Station 9). These
ambient stations were selected to include the 16-km
area that Ohio EPA reported as having a markedly re-
duced fish community. Grab samples were collected on
July 16,1985 and shipped to ERL-Duluth to determine
toxicity to Ceriodaphnia (Table 3-1). A 24-hour compos-
ite Akron POTW effluent sample was collected on July
17 and shipped to ERL-Duluth to determine toxicity to
fathead minnow larvae and Ceriodaphnia (Tables 3-2
and 3-3). The ambient and effluent toxicity tests were
conducted in accordance with accepted chronic toxicity
testing protocols. This general testing information is
available from U.S. EPA (24). Specific test methods for
Ceriodaphnia are described in Mount and Norberg (25)
and for fathead minnows in Norberg and Mount (26).
Statistical analyses were conducted (27) with a Dun-
nett's two-tailed t-test for the effluent data and Tukey's
Honestly Significant Difference Test (28) for the ambient
data. The fathead minnow data were statistically ana-
lyzed using MINITAB © (Pennsylvania State University,
1982) to determine a t-statistic for comparison with Dun-
nett's two-tailed t-test.
Table 3-1. Reconnaissance Sample Analysis for Ambient
Toxicity Measured by Ceriodaphnia Exposed to Cuyahoga
River Water, July 16,1985
Station
3
4
6
9
Control13
Mean
Young/Female
15.4
16.6
14.4a
7.5a
17.3
95%
Confidence Interval
13.9-16.9
15.5-17.6
12.8-16.0
4.6-10.5
15.9-18.7
7-Day
Survival
100
90
20a
70
100
;aVa!ue is significantly lower than the control at P <0.05.
bDilution water from Lester River, Duluth, MM.
3.3 Site Reconnaissance Toxicity Test Data
Tests using Ceriodaphnia and fathead minnows were
conducted off-site (at ERL-Duluth) using the static re-
newal method on the samples shipped from the recon-
naissance trip.The composite samples were less than 24
hours old including travel time, when testing was be-
gun. Sampling locations are described in Section 3.2
and in Figure 3-2. Routine water chemistry data were
collected during the effluent toxicity tests. Dissolved
oxygen ranged from 4.3 to 8.9 mg/L.The range of pH val-
ues was 6.9 to 8.4.
3.3.1 Data Analysis and Interpretation
A seven-day Ceriodaphnia reproductive potential test
was conducted on the four Cuyahoga River samples (Ta-
ble 3-1). Highest young production occurred in the con-
12
-------
SITE DESCRIPTION/RECONNAISSANCE
Table 3-2. Effluent Variability Test Results for Ceriodaphnia
Exposed to Akron POTW Effluent, July-August 1985
Table 3-3. Effluent Variability Test Results for Fathead
Minnows Exposed to Akron POTW Effluent, July-August 1985
Exposure 95%
Concen- Mean Confidence
tration Young/Female Interval
Control3
1
3
10
30
100
Control3
1
3
10
30
100
Control3
1
3
10
30
100
17.3
16.7
15.5
16.4
14.9
16.5
20.2
18.8
19.1
19.1
15.5
Ob
16.5
15.9
17.7
18.3
14.7
Ob
July 17
15.9-18.7
15.3-18.1
14.0-17.0
15.6-17.3
13.7-16.1
15.4-17.6
NOEL> 100% effluent
July 26
18.6-21.8
17.1-20.5
17.4-20.7
18.1-20.2
b 13.8-17.2
—
NOEL = 10% effluent
August 16
14.3-18.7
12.1-19.5
14.8-20.6
15.9-20.7
11.2-18.2
NOEL = 30% effluent
7-Day
Survival
(%) !
100 :
100 i
100
90
80
100 ;
100
100
90
100
100 :
ob
100
90 :
100
100 ,
100
ob ;
"Dilution water from Lester River, Duluth, MM.
bValue is significantly lower than the control at P :
; 0.05.
trol and at Station 4, which was the closest downstream
station to the Akron POTW outfall. Significantly lower (P
<0.05) young production was observed in the two far-
ther downstream stations. Survival followed the same
pattern, with lowest values at Station 6 (significantly
lower than the control at P <0.05) and at Station 9. These
stations are located 7.8 km and 22 km downstream of
the Akron discharge. In contrast, survival near the Akron;
POTW outfall at Station 4 was 90 percent. In addition,
the following tests were conducted on the composite
Akron effluent samples (Tables 3-2 and 3-3): >'
• Seven-day Ceriodaphnia reproductive potential test:
(The NOEL was >100 percent effluent).
• Seven-day fathead minnow larval growth test (The i
NOEL was >100 percent effluent). I
These three sets of results did not, by themselves, show!
that the Akron POTW effluent was toxic.
Exposure
Concen-
tration
Control13
1
3
10
30
100
Control13
1
3
10
30
100
Survival
(%)
100
100
100
90
70
80
80
100
90
90
100
100
NOEL
90
70
90
90
100
90
90
80
78C
100
100
80
NOEL
Mean
Survival ,
(%)
July 17
, 100
95
75
90
90
100
Dry
0.537
0.364
0.527
0.360
0.531
0.496
0.364
0.572
0.448
0.523
0.409
0.487
Mean
Weight3
(mg)
0.451
0.444
0.514
0.468
0.486
0.448
> 100% effluent
August 15 :
80
90 .
95 ;
85 ;
89
90
0.437
0.599
0.539
0.340
0.319
0.337
0.348
0.470
0.302
0.455
0.365
0.354
0.518
0.440
0.328
0.409
0.379
0.360
> 100% effluent
"Initial fry weight: 0.078 mg, July 18,1985; 0.086 mg, August 15,1985.
bDilution water from Lester River, Duluth, MN.
°n = 9
The second effluent toxicity test using Ceriodaphnia did
show toxicity at 30 and 100 percent concentrations (Ta-
ble 3-2). The NOEL was 10 percent. A third Ceriodaphnia
test resulted in young productioniof 14.7 to 18.3 per fe-
male and high survival until the 100 percent exposure
concentration, where there was no young production or
survival (Table 3-2). The NOEL was 30 percent.
The second effluent toxicity test using fathead minnow
larvae showed lowest mean survival in the control (Ta-
ble 3-3). Consequently, the NOEL was >100 percent.
Results from this second series of preliminary tests dif-
fered from the first test series results. Ceriodaphnia
13
-------
SITE DESCRIPTION/RECONNAISSANCE
young production and survival were significantly low-
ered at the 30 and 100 percent effluent concentrations.
In contrast, fathead minnow larvae exhibited little or no
change in survival or weight with increasing exposure
concentrations.
The combined results of all the preliminary effluent tests
indicated that: 1) there did appear to be differences in
the toxicity of the effluent with time; 2) the effluent was
sometimes toxic to Ceriodaphnia at concentrations of 30
percent and greater; and 3) Ceriodaphnia were equally
or slightly more sensitive to the toxic agents in the efflu-
ent than fathead minnow larvae. Evidence suggested
the need for a more complete toxicity assessment.
Chapters 4 and 5 illustrate the development, execution,
and results of that assessment.
14
-------
4. Development Of The Study Plan For Toxicity
Assessment
4.1 Introduction
Having identified the Akron POTW as a probable source of
water quality impact to the Cuyahoga River, the next task
was to quantify the toxicity. This section discusses the de-
velopment of the procedures used to identify and quantify
the source of toxicity. ;
As discussed in Chapters 2 and 3, it appeared likely that
toxicants were damaging the fish and invertebrate popula-
tions in the river below Akron. The available field data and
toxicity screening data suggested that the Akron POTW
was a source of this toxicity despite good treatment of the
influent to reduce concentrations of metals and extract- ;
able organics (Section 2.5).
A regulatory authority can use two approaches to assess^
ing and controlling effluent toxicity. The traditional ap-
proach is to identify the pollutants being discharged, com-
pare their concentrations instream to existing state and i
Federal water quality standards, and set limits for each of
the pollutants exceeding the standard. The other approach
is to analyze the toxicity of the whole effluent and set lim-
its either on the parameter toxicity (e.g., the effluent shall
not exhibit a NOEL of less than 25 percent) or require the
identification of the causative agents of the measured tox-
icity and limit these toxicants individually. This approach is
accomplished by using toxicity testing to find the chemi-:
cals causing the effect. As it was possible that other
sources such as leaching landfills, combined sewer over-
flows, bypasses of the Akron POTW, or other tributaries ,
were causing or contributing to the problem, it was necest
sary to analyze the toxicity of the Akron POTW effluent ;
and analyze the pattern of ambient toxicity in the receiving
water. It was recognized that direct analysis of other po- ;
tential toxicant sources might also be appropriate.
The whole effluent toxicity testing approach was the best
assessment technique for analyzing the Akron POTW's ef-
fluent because:
1. Available data showed that the Akron effluent is a
highly complex mixture of compounds that varies
over time. Identifying each of the compounds that
could cause toxicity would be time-consuming and ,
costly. :
2. Even if all the specific toxicants could be identified, :
there was likely to be little information on the toxicity
of most of them. Few state or Federal standards exist
for many pollutants. ;
3. The impact of effluent toxicity depends on many fac-
tors, particularly the interaction in the water column
of the toxicants and such other water quality param-
eters as suspended solids, pH, alkalinity, and hard-
ness. Most individual pollutants limits do not take
into account such interactions (except ammonia,
which considers pH, and some metals). Limiting tox-
icity accounts for these water quality parameters and
permit limits are set to protect aquatic life.
4.2 Effluent Sampling And Toxicity Analysis
Plan
The effluent toxicity testing procedure, was designed
considering that:
1. Effluent toxicity data were needed for comparison
with instream effluent concentrations (IWC) to de-
termine if the effluent caused instream toxicity. The
toxic effect from continuous discharge of an efflu-
ent must be determined through toxicity testing.
2. Because the Akron POTW effluent dominates the
Cuyahoga's flow, aquatic organisms are exposed
to relatively high concentrations of effluent over
long periods (during low flow, high exposure may
last for days). Chronic toxicity tests, which measure
toxic effects over a species' lifespan, had to be per-
formed.These tests simulate the exposure to which
the organisms are subjected in the stream.
3. Testing one organism, which may not be sensitive
to the toxicants in the effluent, could lead to an er-
roneous conclusion that the effluent is not toxic.
Therefore, to avoid a false negative conclusion two
to three aquatic organisms representing fish, inver-
tebrates, and algae should be tested (24). The range
of sensitivity of test species is thus measured.
Finding a sensitive test species is important in the
permitting process (24). Oncfe identified, the regu-
latory authority can use the most sensitive of the
test species in a Toxicity Source Investigation (TSI)
and for permit compliance monitoring.
4. To determine the effects on aquatic biota, the toxic-
ity tests should simulate effluent/receiving water
interaction at the point of discharge. Therefore, di-
lution water for the tests should be taken directly
upstream of the discharge point. To measure the
inherent toxicity of an effluent to avoid interactions
with natural waters or upstream sources of conta-
minants, a standard laboratory dilution water
should be used. This is especially important in situ-
ations with multiple discharges.
5. All potential sources of toxicity associated with the
Akron POTW had to be analyzed. In this case, pre-
and postchlorination, bypass/overflow, and landfill
leachate wastewaters.
15
-------
DEVELOPMENT OF STUDY PLAN
6. Wastewaters should be sampled to simulate expo-
sure. For the continuously discharged effluent,
composite sampling simulates the chronic expo-
sure conditions experienced by the resident biota.
For bypass situations, the acute toxicity associated
with intermittent flows of short duration are best
simulated using multiple grab samples collected
over the duration. The effect of the landfill leachate
was also assessed using grab samples. Grab sam-
ples for pre- and postchlorination are sufficient be-
cause chlorine is reactive in water and should ex-
hibit toxicity only for a short (acute) period of time.
7. Knowledge of the variability of the effluent's toxic-
ity is needed to assess toxic impact. Since the Ak-
ron POTW effluent was known to be variable, toxic-
ity testing was to be repeated as often as
practicable from samples collected over a period of
time, to observe potential changes in effluent com-
position and thus in effluent toxicity.
4.3 Ambient Sampling And Toxicity Analysis
Plan
It is often helpful for a regulatory authority to perform
ambient toxicity tests to develop a toxicity profile of the
stream segment under study. Ambient toxicity tests:
• Will demonstrate whether instream toxicity occurs
• May locate the source(s) of toxicity
• May determine the fate/persistence of toxicity.
For regulatory purposes, a sampling station should be
placed at the perimeter or downstream edge of the mix-
ing zone to determine chronic effects. Since this case
study was conducted for research purposes, such a sta-
tion was not precisely located, but was placed beyond
the mixing zone edge at the nearest access point. A dye
dilution study should be conducted to properly locate
stations with respect to the effluent plume and to mea-
sure the concentration of effluent instream. An anlysis
of this type was not within the scope of work for this
project.
In ambient toxicity testing the river does the mixing for
the dilutions. The regulatory authority only needs to
know the concentration of the effluent instream and lo-
cations of other sources of dilution or potential toxicity.
Sampling should be conducted when possible during
critical low flow periods so that dilution levels will match
those specified in the water quality and toxicity stan-
dards. Otherwise, toxicity test results will reflect condi-
tions of higher effluent dilution and ambient samples
may be so dilute as to have no toxic effects.
Ambient sampling stations should be established at the
following locations:
1. A control station upstream of the point of dis-
charge and, if possible, any other potential sources
of toxicity.
2. A station just upstream of the discharger and
downstream of other potential sources of toxicity
to determine upstream toxicity levels.
3. A station as close as possible to the point of com-
plete mixing where the concentration of effluent is
greatest after mixing, but before the effluent toxic-
ity may begin to decay.
4. Several stations at points of increasing distance
downstream of the point of discharge (in this case
the Akron POTW) to measure the decay in toxicity
of the sampled effluent, the effect of other sources
'. of toxicity, and the dilution or additional pollution
that occurs as tributaries flow into the stream un-
der study. These stations are generally placed im-
mediately above points where other sources of wa-
ter enter the river to identify and separate any
tributary influences on water quality. Where tribu-
taries are known to receive discharges of toxicants
should be stations located up- and downstream of
their confluences.
The same biological species used in the effluent toxicity
tests should be used in the ambient toxicity tests. Sam-
pling should be conducted to best simulate the expo-
sure received by the resident biota. As discussed, com-
posite samples provide the best duplication of chronic
exposure.
4.4 Ambient River Station Descriptions
The Cuyahoga River banks are generally steeply sloped
and wooded. The ambient stations were placed either
along these wooded banks or on bridge foundations,
with one exception (Table 4-1). At Station 6, because of
the difficult access to the bridge foundation, surface
samples were collected below the bridge using a poly-
ethylene bucket hung from a rope.
Station locations were selected based on the principles
described in Section 4.2, with some modifications be-
cause of nearby discharges, topography, and accessibil-
ity. Three stations were located upstream of the Akron
POTW: one immediately upstream of the Akron POTW,
and two others each upstream of a tributary confluence
with the Cuyahoga. The furthest upstream location is
the control and reflects all water quality influences up-
; stream of the study area. Samples from the next two
16
-------
"DEVELOPMENT OF STUDY PLAN
Table 4-1. Station Locations and Descriptions
Station
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
River
Kilometer
68.1
64.6
61.1
60.3
56.8
53.2
46.8
42.6
38.8
33.5
27.8
25.1
21.2
Description
Upstream of Little Cuyahoga River confluence with Cuyahoga River on the Cuyahoga River
control station
Upstream of confluence of Mud Run with Cuyahoga River, 50 m upstream of USGS
gauging station at Old Portage
At POTW, 100 m upstream of outfall
Upstream of Bath Road bridge and landfill drainage confluence with Cuyahoga River, 100 m
downstream of POTW outfall
Upstream of Ira Road bridge
Furnace Run confluence with Cuyahoga River
Upstream of Peninsula Road bridge
Downstream of Boston Road bridge
Upstream of Brandywine Creek confluence with Cuyahoga River at Vaughn Road bridge
Upstream of Route 82 bridge at Station Road ' !
Upstream of Tinkers Creek confluence with the Cuyahoga River at Fitzwater Road
Downstream of Tinkers Creek confluence with Cuyahoga River at Hillside Road
Downstream of Old Rockside Road
stations downstream reflect all water quality influences,
including tributaries downstream of the control, yet still
upstream of the Akron POTW (see Figure 3-2.) \
Ten stations were located downstream of the Akron
POTW, one just downstream of the outfall and the other
nine covering the 16-km area identified by Ohio EPA. ;
These nine stations were approximately equidistant and
located to identify any downstream tributary effects.
Sampling Stations 11 and 12 were set just upstream and
downstream of the Tinkers Creek confluence with the
Cuyahoga River, respectively. This was done because
Tinkers Creek was known to receive discharges from
four small POTWs.
There were some complications with the placement of
the composite water samplers. At Station 2, an intermit-
tent discharger was discovered at the desired station lo-
cation so the sampler was moved about 50 m upstream
to the opposite river bank. At Station 5, because there
was an old culvert on the river bank, the composite sam-i
pier was repositioned upstream (—145 m) as a precau-
tionary measure. At Station 7, the composite sample
was moved about 12m upstream from a rusted drain-
pipe with a very low volume outflow which smelled of [
sewage.
In general, ambient samples were collected as 24-hour
composites to further examine any day-to-day variabil-
ity in toxicity. However, due to access difficulties and
lack of equipment, grab samples were collected at Sta-
tions 6 and 13.
Drainage pipe at Station 7. Water sampler was located
upstream.
17
-------
DEVELOPMENT OF STUDY PLAN
4.5 Toxicity Testing Procedures
The mobile toxicity testing laboratory and staff mem-
bers from ERL-Duluth began effluent and ambient toxic-
ity testing on September 26 and ended on October 3,
1985."
The mobile laboratory was parked at the Akron POTW
so that there was sufficient electricity and water, securi-
ty, 24:hour access, and proximity to POTW operation
and management personnel. This location provided ac-
cess to plant operations and allowed bypass sample
testing and observation of effluent variations.
Three test organisms were selected for the toxicity test-
ing procedures—the fathead minnow, Ceriodaphnia,
and a nonoperculate snail, Aptexia sp.The test method-
ology iorAplexia is still under development. The meth-
ods of Mount and Norberg (25) and Norberg and Mount
(26) were used for the other test species. Each of these
tests is considered a short-term chronic toxicity test.
4.5.1 Effluent Sample Tests
Toxicity data collected previously by ORD-Cincinnati
had indicated that the chronic toxicity effect concentra-
tion (NOEL) of the effluent was similar to its projected in-
Jlw EPA mobile laboratory for ambient and effluent toxicity
testing.
"Toxicity testing was performed on-site, although all data could have
boon generated off-site had samples been provided. This would have
significantly lowered costs.
Inside EPA's mobile laboratory. Ceriodaphnia and fathead
minnow tests are being set up.
stream waste concentration at low flow periods (Section
2.3). Therefore, concentrations of 50 and 75 percent
were added to the usual dilution series (100, 30, 10, 3,
and 1%) to bracket the expected toxicity concentration of
the wastewater. For the effluent tests, dilution water was
collected upstream of the point of discharge at Station 3.
Since ambient testing was also being conducted, any
toxicity contribution from upstream could be accounted
for in the analysis of the toxicity results. Further, based
on tjie State of Ohio's fish survey, the fish population
upstream of the Akron POTW was in very good condi-
tion. Upstream toxicity, therefore, was expected to be
negligible.
Both effluent and ambient samples were collected as 24-
hour composite samples using commercial, battery-
powered samplers (except at Stations 6 and 13). Sam-
pling was conducted during low flow, but not 7Q10
conditions (Section 3.1.2 and Table 5-4). Each day the or-
ganisms were transferred into a new sample so that
over the period of seven days they were exposed to sev-
en different samples.
Several different types of bypasses occur at the Akron
POTW. One is a straight bypass of raw sewage to the riv-
er when rainfall precludes treatment of the entire vol-
ume of wastewater entering the Akron POTW. One by-
pass event of this type occurred during the course of this
study. The overflow basin was allowed to discharge for
several hours before a grab sample was taken so that a
18
-------
DEVELOPMENT OF STUDY PLAN
Making effluent dilutions for use in toxicity tests.
sample representative of bypassed wastewater after the
first flush would be obtained. This sample was tested for
acute toxicity since it was from a short-term event (ap-
proximately four hours) resulting in short-term expo- :
sure. No further bypassing samples were tested for tox-
icity. Another type of bypass occurs when influent from
the primary settling basins is diverted around the secon-
dary treatment system and into the chlorine contact ba--
sin. This partially treated wastewater is mixed with the '
final effluent, so effluent toxicity testing during this time
was considered to account for this type of bypass. I
Variability of the effluent's toxicity was assessed in two
ways. First, the tests were conducted over a seven day
period to measure variation in toxicity. Second, effluent
toxicity tests were repeated six times over six months,
beginning with a sample collected in July 1985 and end-
ing with a sample collected in December 1985. These ;
samples were 24-hour composite samples collected on-
site and shipped to the ERL-Duluth. The toxicity test con-
ducted on the bypass of raw sewage also assessed efflu-
ent variability since this potential source of increased
toxics contribution of the Akron POTW was measured. !
fluent occurred at each sampling station, plus other am-
bient conditions that existed in the river.
Sampling locations are described in Table 4-1. The area
of interest along the Cuyahoga River was originally de-
fined from Ohio EPA's data which indicated a 16-km
stretch of river from the Akron POTW downstream
where fish and benthic populations were greatly re-
duced. An upstream area was used as a control and an
area farther downstream was also included. Ambient
toxicity analysis was terminated at Station 13, below
which several large-volume tributaries and effluents en-
tered the river.
Due to the high volume flow of the effluent into the
Cuyahoga, instantaneous, complete mixing was as-
sumed to occur at the outfall. This assumption was con-
firmed when a green dye spill from an indirect industrial
discharger was observed in the plant effluent. Acting as
an unintentional dye dilution study, this green effluent
mixed at the point of discharge and colored the entire
river near the Akron POTW for approximately 24 hours.
The green effluent was observed to cross the Cuyahoga
almost perpendicular to the direction of flow and reflect
back toward midstream where an eddy formed. Down-
stream of this eddy, the green color was greatly dimin-
ished.
4.5.2 Ambient Toxicity Tests
The ambient toxicity tests were conducted using seven
daily samples of 100 percent of the sample (i.e., tests
were renewed daily with new sample water). Test organ-
isms were exposed to whatever concentration of the ef-
19
-------
5. Toxicity Testing Results and Data
Interpretation
5.1 Introduction
Cuyahoga River and Akron POTW effluent samples were
collected, following the study plan described in Chapter 4..
Fathead minnows, Ceriodaphnia, and Aplexia were the
three organisms used as toxicity detectors in the effluent
and ambient toxicity tests. Statistical analyses were con-
ducted according to procedures described in Section 3.2
and were typically used in the Complex Effluent Toxicity
Testing Program report series, unless specified otherwise.
Adding effluent dilutions to fathead minnow test chambers for
sample renewal.
5.2 Akron POTW Effluent Toxicity Results
Fathead minnows were not as sensitive as Ceriodaphnia
to the Akron POTW effluent. Exposure to 100 percent efflu-
ent did not affect fathead minnow survival, although
mean weight was significantly lower (P <0.05) than the
control.The NOEL is 75 percent effluent for the fathead
minnow (Table 5-1) and the Acceptable Effluent Concen-
tration (AEC) is 86.6 percent effluent.
Effects on survival and mean young production were ob-
served for Ceriodaphnia (Table 5-2). The NOEL is 10 per-
cent effluent and the AEC is 17.3 percent.
A trial test using Aplexia indicated no effluent effects on
survival, although mean larval snail weight decreased
with increasing exposure concentration (Table 5-3).
The Akron POTW effluent (in all concentrations) produced
extreme convulsive behavior in Ceriodaphnia for the first
three to four days of exposure before mortality began. The
Ceriodaphnia were so convulsive that they were difficult
to capture when being transferred during the sample re-
newal process. In almost every case, adult mortality oc-
curred when a brood of young was produced (day 4 of the
test). As adults molt at the time of brood release and are
known to be more sensitive at that time, this behavior is
Table 5-1. Effluent Toxicity Test Results for Akron POTW
Using Fathead Minnows, September 26-October 3,1985
Exposure
Concentration
(% Effluent)
Control8
: 1
3
10
30
50C
75C
100
Mean Weight
(mg)
0.397
0.303
0.395
0.365
0.303
0.337
0.314
0.291b
NOEL = 75%
AEC = 86%
Standard
Error ±
0.021
0.022
0.021
0.022
0.021
0.020
0.021
0.020
effluent
effluent
Mean
Survival
(%)
90
85
100
85
95
95
90
100
aStation 3 water.
bSignificantly lower (P < 0.05) than the control.
Additional dilutions to typical series to ensure better
resolution of toxic level.
Adding dilutions to test cups for effluent toxicity test using
Ceriodaphnia.
20
-------
TOX1CITY TESTING RESULTS
not unusual. The young appeared normal and were alive ,
when the daily observation was made, but they were not
kept to see how long they would have survived. ;
It is important to record how and when test organisms are
affected in the tests. Symptoms enable biologists to better
interpret toxic impact. For example, knowing thatCer/o- I
daphnia adults died in Akron effluent after brood release
Table 5-2. Effluent Toxicity Test Results for Akron POTW
Using Ceriodaphnia, September 26-October 3,1985
Exposure
Concentration
(% Effluent)
Control3
1
3
10
30
50C
75°
100
Mean Survival Mean
(%)
100
90
100
100
40b
Ob
Ob
Ob
NOEL =
AEC =
Young/Female
34.6
35.8
28.1
30.5
31.7
b
b
b
10% effluent
17.3% effluent
95% ,
Confidence
Limits
28.4-40.8 I
29.4-42.4
22.9-33.3
20.2-40.8
22.8-40.2
— :
— '••
— • ;
1
aStation 3 water.
bSignificantly lower (P < 0.05) than the control.
"Additional dilutions to typical series to ensure better
resolution of toxic level.
Table 5-3. Effluent Toxicity Test Results Using the
Snail Aplexia "
Effluent
Exposure
Concentration
(% Effluent)
Control13
1
3
10
30
50d
75d
100
Mean Weight
(mg)
0.296
NA
NA
0.264
0.251
0.237
0.215
0.208°
NOEL> 100% effluent
AEC not calculated
i
Mean Survival
(%) i
100
NA i
NA j
100 I
90 i
100
100
100
:
aTest methods are being developed by U.S. EPA
Environmental Protection Research Laboratory-Duluth.
bStation 3 water.
"Significantly lower (P < 0.05) than the control.
Additional dilutions to typical series to ensure better
resolution of toxic level.
explained why mortality of adults, not reduced reproduc-
tion, was the endpoint of importance in these chronic toxr
icity tests.
Routine water chemistry data for the effluent tests were
collected for fathead minnows and Ceriodaphnia. Dis-
solved oxygen ranged from 4.6 to 9.5 mg/L. The range of
pH was 7.3 to 7.8. Temperature was controlled and varied
from 24.8 to 25.0°C. Conductivity varied from 774 to 929
(jimhos. These values represent all effluent dilution and
control tests.
5.2.1 Test Species Sensitivity
Three aquatic organisms were tested as surrogates for the
range of biological community response in the Cuyahoga
River to the Akron POTW effluent. Of the three, Ceriodaph-
nia was the most sensitive. Fathead minnows were less
sensitive and Aplexia were unaffected by exposure to the
effluent. In this case, the fish population in the Cuyahoga
River has been severely affected by an effluent that was
not toxic to fathead minnow test organisms/This demonr
strates the relation between test species and biological
community sensitivity that was well documented in the
Complex Effluent Toxicity Testing Program. A test organ-
ism should represent the sensitivity of a community under
study, but which component of the community is not
known a priori. "Careful analysis of our knowledge of toxi-
cology, effluent decay, and relative sensitivity tells us that
we cannot expect: 1) Ceriodaphnia toxicity to always re-
semble toxicity to benthic invertebrates; 2) fathead min-
now toxicity to always resemble toxicity to fish; 3) fathead
minnows and other fish to display the same relative sensi-
tivity to different effluents" (29).
Ceriodaphnia, the most sensitive test organism, repre-
sents the sensitive part of the Cuyahoga's resident com-
munity, not necessarily the invertebrates or organisms
closely associated with it physiologically. Further, there
may be residents of the Cuyahoga River which are more
sensitive than Ceriodaphnia to the Akron POTW effluent.
When the effluent effect concentration of this representa-
tive sensitive species is exceeded an entire group of simi-
larly sensitive or more sensitive species will be affected.
Such an occurrence may affect an entire biological com-
munity, fish included.
5.2.2 Instream Effects Due To Effluent Toxicilty
The results of effluent toxicity tests may be used to as-
certain whether the Akron POTW discharge causes in-
stream effects. To determine the effect of an effluent, the
test NOEL concentrations are compared to instream ef-
21
-------
TOXICITY TESTING RESULTS
fluent concentrations. The potential for adverse environ-
mental impact is determined by (1):
IWC £ NOEL (Equation 5-1)
where
IWC = concentration of wastewaterinstream at a criti-
cal flow period (Instream Waste Concentration)
NOEL = the concentration of wastewater at which no
toxic effect is observed (No Observed Effect
Level)
If the NOEL is equal to or greater than the IWC, then no
toxic effect is expected to occur instream. This relation-
ship has been demonstrated at a number of NPDES dis-
charge sites (see the Complex Effluent Toxicity Testing
Program report series in the References).
Using Akron POTW discharge data and USGS flow data
for the Cuyahoga River, the proportion of the river flow
contributed by the effluent (IWC) was calculated (Table
5-4). The contribution was calculated for the river near
the Akron POTW using available data on the mean flow
from the Old Portage gauging station. Daily information
on effluent discharge and river flow rates indicated that
the Akron POTW effluent represented 12 to 48 percent of
the Cuyahoga in 1984 and 6 to 56 percent in 1985 (19).
The IWC values for the September 26 to October 3 test
period are 38 to 54 percent (19). The NOEL value is 10
percent for the most sensitive species, Ceriodaphnia
(Table 5-7). Therefore, the NOEL < IWC and a toxic im-
pact would be expected in the river from the Akron
POTW discharge.
Table 5-4. Akron POTW Effluent Contribution to
Cuyahoga River Flow
Table 5-5. Acute Toxicity to Ceriodaphnia of
Bypassed Wastewater From the Akron POTW3
Date
Sep 26 85
Sep 27 85
Sep 28 85
Sep 29 85
Sep 30 85
Oct 1 85
Oct 285
Mean
Akron POTW
Effluent Flow8
(m3/s)
3.22
2.61
2.59
2.54
3.17
2.78
2.66
2.79
River Flow
Contribution13
(% Effluent)
46
49
54
53
52
38
38
47
"From Monthly Operating Reports submitted by the City of
Akron to Ohio EPA.
bCalculated using the river flow data from the USGS gauging
station at Old Portage, Ohio (near Station 2 and upstream of
the POTW discharge).
Note that the 7Q10 value at Old Portage gauge for this time
period is 1.27 m3/s.
Wastewater
Concentration
(%)
Control"
3
6.2
12.5
25
50
Survival
(%)
90
100
90
100
0
0
NOEL = 12.5% wastewater
LC50 = 17.7% wastewater
aSample collected September 26, 1985.
bControl and dilution water from Station 3.
Since complete mixing at the point of discharge is as-
sumed to occur (Section 4.4.2), the NOEL of Ceriodaph-
nia is continuously exceeded instream from the point of
discharge down to Station 13, a distance of about 40 km.
Thus, a chronic toxic effect would be expected over this
entire segment when there is no toxicity decay or dilu-
tion. This is confirmed by the fish and invertebrate popu-
lation data (Chapter 2) and in the ambient toxicity data
(Section 5.2).
5.2.3 Wastewater Bypassing Toxicity Results
Regulatory authorities have been concerned about the
impact of wastewater bypassing at the Akron POTW.
The influent was also known to be highly toxic (more
toxic than the effluent) and thus could cause severe im-
pact to the river. The volumes of total raw wastewater
bypassed annually vary from 1.13 mil m3to 113 mil m3
(300 to 3,000 mil gal) (Section 3.1.1). To estimate the tox-
icity of bypassed influent (raw wastewater), a bypass
sample was collected during an event which occurred
during the on-site testing. The sample was tested for
acute toxicity due to the short release times typical of
bypass events. Since Ceriodaphnia had a greater sensi-
tivity to the Akron POTW effluent, they were used to test
the toxicity of the bypass sample (Table 5-5). The LC50
of the raw sewage was 17.7 percent (using straight line
interpolation). Complete mortality occurred after 24
hours in the 25 and 50 percent bypassed wastewater
concentrations. There was no mortality in the 12.5 per-
cent concentration of bypassed wastewater in 48 hours.
In terms of the volume discharged, the contribution of
bypassing to the river flow and instream toxicity seems
quite large. However, the actual contribution to instream
22
-------
TOXICITY TESTING RESULTS
i
o
cc
100
90
80
70
60
50
40
30
20
10
100
90
80-
70
60
50
40
30
20
10
* = Likely higher bypass volume, although,
unreported during chlorination season.
| = Bypass
_Chlorination period, partially treated ^,
wastewater bypasses not reported.
January February March
ii i i i i i i[ i i i i i
April May June
July
ri i i i i—q~i—n—i—i—ITT—i—i—n—q—i—i—rr
August September October November1 December'
On Site
1985
fT"!—I I II I I I I I I I i I I I II I I I I I II ' I I I ' II I I—I I I || I I I I I l| I I I 1 I l| I I I I I l| I I I I I l| I I I I I l| I I I I I l|
January February March April May June July August September October November December
EFFLUENT DISCHARGE
Figure 5-1. Percent Contribution of the Akron POTW Effluent and Bypass Flow to the Cuyahoga River
toxicity may be limited for several reasons. The duration
of these events is often short, less than 12 hours (Figure
5-1) (19). In addition, bypassing has usually occurred ;
during rainfall events when the volume of wastewater is
increased beyond the capacity of the Akron POTW.
Since the Cuyahoga River is a relatively swift-flowing
river, bypassed wastewaters move quickly downstream,
particularly during such rainfall events. It is estimated :
that the effluent has a flow time of less than 24 hours
from the discharge point through the segment of river :
under study. It is important to note that the bypass sam-
ple did notaffecttheCer/odap/7/7/a until after 24 hours of
exposure. Therefore, impacts of the bypassed
wastewaters may be minimized by their quick passage
and dilution downstream. ;
The type of injury to biota is not consistent with the ex-
pected effects from bypassing episodes. While instrearri
organisms are exposed to pulses of very toxic raw
wastewater for brief periods, these periods are probably
not long enough to cause the toxic effects observed on
the fish collected during the Ohio EPA biosurveys. Those
fish exhibited a highjncidence of lesions, tumors, and
other abnormalities which are characteristic of long-
term exposures. Also, this evaluation did not consider
sediment toxicity from an accumulation of toxicants.
The instream effects from bypassed wastewater may be
reduced by dilution with rainwater. If the toxicant source
flow is not proportional to the flow increases associated
with rainfall (e.g., an industrial vs. storm drain source for
toxicants), the toxicant concentration will be reduced by
dilution from the addition of stormwater regardless of
whether the Akron POTW treatment is effective. This di-
lution assumes a continuous passive discharge in con-
trast to the potential for deliberate releases during rain-
23
-------
TOXIC1TY TESTING RESULTS
fall (which has not been observed at Akron POTW).
Further dilution of the toxicant will occur as the effluent
mixes with the Cuyahoga River.
The role of bypassed wastewaters in causing a toxic ef-
fect on the Cuyahoga River is uncertain and deserves
further study. On the basis of results from this study it
appears that raw or partially treated sewage probably
does contribute to instream impact. However, the rela-
tive importance of bypassed wastewater and the con-
tinuous, high-volume discharge of toxicants in the Ak-
ron POTW's effluent is uncertain. Steps are currently
being taken at the Akron POTW to minimize bypassing
by making process improvements.
5.2.4 Effluent Toxicity Variability
The Akron POTW's effluent was tested for chronic toxic-
ity in August 1984 and May 1985 by ORD-Cincinnati and
over a six-month period in 1985 by ERL-Duluth. ORD-
Cincinnati tested three forms of the effluent (unchlorin-
ated, chlorinated, and dechlorinated—achieved by hold-
ing time) twice. ERL-Duluth tested the chlorinated
effluent before the site visit, while on-site, and twice fol-
lowing the site visit (Tables 3-2,3-3, 5-1, 5-2, and 5-6).
Table 5-6. Effluent Variability Test Results for Ceriodaphnia
Exposed to Akron POTW Effluent, December 1985
Exposure
Concentration
95%
Mean Confidence
Young/Female Limits
7-Day
Survival
(%)
December 2
Control*
1
3
10
30
100
16.5
19.7
20.1
19.3
23.1
18.1
NOEL = 30%
9.9-23.1
15.4-24.0
16.7-23.5
17.6-21.0
20.9-25.3
14.6-21.6
effluent
100
100
100
100
100
50b
December 9
Control*
1
3
10
30
100
13.7
15.6
15.3
18.3
21.5
17.1
NOEL = 30%
10.2-17.2
9.4-21 .8
11.2-19.4
13.3-23.2
17.8-25.2
13.7-20.3
effluent
100
70
100
90
100
Ob
The results of these tests were used to initially assess
the variability of the effluent's toxicity before the toxicity
source investigation testing, described in Chapter 6, be-
gan.
Toxicity varied by one order of magnitude, from a NOEL
of 10 percent to greater than 100 percent (essentially no
measurable toxicity) (Table 5-7). Based on previous ex-
perience with effluent toxicity, this is not an unusually
variable effluent. The cause of the variability is likely a
combination of factors, including variable contributions
by industrial dischargers and varying treatment efficien-
cy.
Collecting an ambient water sample from the Cuyahoga River.
Table 5-7. Effluent Variability Characterized by No Observed
Effect Levels (NOELs) From Toxicity Tests Using Fathead
Minnows and Ceriodaphnia
Fathead Minnows Ceriodaphnia
Survival Growth Survival Reproduction
"Dilution water from Lester River, Duluth, MM.
'Value is significantly lower than the control at P :
; 0.05.
EPA-Cincinnati
Aug 1984
May 1985
'ERL-Duluth
Jul 1985
Jul 1985
Aug 1985
Sep 1985
Dec 1985
Dec 1985
30
30
>100
NA
>100
>100
—
>100
>100
>100
NA
>100
75
—
30
>100
>100
30
30
10
30
30
>100
NA
>100
10
30
30
>100
>100
24
-------
TOXICITY TESTING RESULTS
The variability data would be very important in a waste-
load allocation and permit issuance. For example, the
procedure for deriving water quality-based permit limits
accounts for effluent variability. Procedures for such cal-
culations are found in References 24 and 30. Data on the
variability of effluent toxicity is also of importance in try-
ing to determine the causative agents of the measured
effluent toxicity.
5.3 Ambient River Toxicity Results
Ambient toxicity tests were conducted using fathead i
minnows, Ceriodaphnia, and Aplexia exposed to com- ,
posite Cuyahoga River samples from 13 stations. The
stations were located on a 44-km length of the Cuya- j
hoga River from above the confluence with the Little
Cuyahoga to downstream of Tinkers Creek (Figure 3-3).
Station selection and locations are described in Section
4.3. ;
Routine water chemistry data were collected during the
ambient tests for the fathead minnows and Ceriodaph-,
nia. Dissolved oxygen ranged from 4.2 to 9.3 mg/L (final
values below 5.5 mg/L were consistently obtained in the
fathead minnow test chambers). The range of pH was
6.7 to 7.9. Temperature was controlled and varied from
24.9 to 25.4°C. Conductivity varied from 789 to 871
5.3.1 Aplexia Test Results
The trial chronic toxicity test using Aplexia did not indi-
cate ambient toxicity (Table 5-10). Larval snail weight
varied 0.261 to 0.400 and the higher weights were from
stations downstream of the Akron POTW. Survival was
90 to 100 percent, except at Station 10 where it was 80
percent.
5.3.2 Fathead Minnow Test Results
The fathead minnows did not show any effects from am-
bient toxicity on their weight. Only at Station 8 was sur-
vival significantly lower (P < 0.05) than upstream of the
Akron POTW at Station 1 (Table 5-8). Mean weights var-
ied 0.297 to 0.413 mg. Mean survival varied 50 to 100
percent; however, with the exception of Station 8, sur-
vival was >70 percent. i
5.3.3 Ceriodaphnia Test Results
Survival and young production of Ceriodaphnia were re-
duced at stations downstream of the Akron POTW (Ta-
ble 5-9). Survival of adult Ceriodaphnia declined from
Fathead minnow larvae are being added to test chambers at
initiation of an effluent toxicity test.
Table 5-8. Ambient Toxicity Test Results for the
Cuyahoga River Using Fathead Minnows
Station
1
2
3
POTW
4
5
6-
7
8
9
10
11
12
13
Mean Weight
(mg)
0.391
0.355
0.403
0.383
0.319
0.374
0.369
0.413
0.381
0.297
0.329
0.317
0.329
Standard
Error ±
0.029
0.031
0.030
0.029
0.030
0.031
0.032
0.030
0.034
0.032
0.032
0.033
0.030
Mean
Survival
(%)
100
90
95
100
95
90
85
50a
70
85
80
80
90
Significantly lower (P < 0.05) than Station 3.
100 percent at Stations 1 through 3, located upstream of
the effluent discharge, to 44 percent or less at stations
located downstream of the discharge.
Upstream of the point of discharge, no instream toxicity
was observed. The Akron POTW discharge enters the
river between Stations 3 and 4. At Station 4, the first sign
of instream toxicity was observed when the mean sur-
25
-------
TOXICITY TESTING RESULTS
yival was only 40 percent (remember that adult mortality
is the parameter of concern). At Station 5, there was no
survival. The reasons for the continued decline in Cerio-
daphnfa mortality between Stations 4 and 5 are not
known. Possible explanations are that some aspect of
the river's water chemistry briefly changed the equilibri-
um ofthetoxicant(s) responsible for the effluenttoxicity
or possibly the presence of nonpoint source dischargers
or the influence of Yellow Creek. See Chapter 6 for
information on the causes of toxicity.
The pattern of reduction in survival oiCeriodaphnia con-
tinues downstream with some fluctuations. This pattern
is not unexpected because the flow of the Cuyahoga Riv-
er is swift and the time of travel for the effluent between
Stations 4 and 13 is less than 24 hours. Toxicity of the ef-
fluent is persistent at least for that period of time and
distance on the Cuyahoga River.
The number of young produced by Ceriodaphnia ex-
posed to Cuyahoga River water did not indicate any con-
sistent pattern at downstream stations. There was no
young production at Stations 5 and 13 due to adult Cer-
iodaphnia mortality, and low young production at Sta-
tions 1,6, and 10 of from 13 to 16 per female. For the re-
maining stations, young production was >20 per
female.
From the effluent toxicity test, the effect observed was
adult mortality after a brood was produced, so the lack
of a large effect on young production is not surprising. In
Table 5-9. Ambient Toxicity Test Results for the
Cuyahoga River Using Ceriodaphnia
Table 5-10. Ambient Toxicity Test Results for the
Cuyahoga River Using Aplexia"
Station
1
2
3
POTW
4
5
6
7
8
9
10
11
12
13
Mean Survival
(%)
100
100
100
40
Oa
10a
10a
33s
10a
44
12.5a
33a
Oa
Mean
Young/Female
16.1
20.1
21.5
24.6
a
15.1a
39.4
20.8
21.0
13.6
27.3
23.3
a
95%
Confidence
Limits
9.4-22.8
13.8-26.5
17.0-26.0
14.9-34.1
—
0.66-25.2
23.7-51.3
5.4-35.8
11.0-32.9
10.9-16.4
25.6-29.5
15.3-31.2
Station
1
2
3
POTW
4
: 5
6
7
; 8
9
10
; 11
12
13
Mean Weight
(mg)
0.261
0.276
0.296
0.317
0.309
0.300
0.316
0.278
0.267
0.400
0.313
0.300
0.311
Mean Survival
(%)
100
100
100
100
100
100
100
100
100
80
90
100
90
'Significantly lower as 0.05) than Station 3.
"Statistical analysis of these data was not conducted.
contrast, survival was low at all stations downstream of
the Akron POTW, and was significantly lower (P <0.05)
than the control at Stations 5, 6, 7, 8, 9,11, and 13. In the
statistical analyses program used (27), mortalities occur-
ring on day 7 do not affect statistical differences as much
as early mortality. Most of the mortality occurred on
days 5 and 6 in the effluent and ambient tests. The rou-
tine water chemistry data did not indicate changes in
DO, pH, temperature, or conductivity. The test organ-
isms that did survive at Stations 6 through 12 were ex-
tremely convulsive and would probably have rapidly
died with additional test time (31).
5.3.4 Landfill Drainage Test
An acute toxicity test was conducted on a sample from
the landfill drainage stream located immediately down-
stream of the Akron POTW. When Ceriodaphnia were
exposed to 100 percent sample over a 24-hour period,
survival was 100 percent. Since there was no acute tox-
icity and the flow volume was intermittent and too small
to cause chronic toxicity instream (far below 1% of the
Cuyahoga River flow), further testing was not per-
formed.
5.3.5 Comparison of Ambient Toxicity to Ceriodaphnia
with River Survey Data
Examination oiCeriodaphnia survival data in conjunc-
tion with the biosurvey data from Ohio EPA (Section 2.2)
Indicates a similar pattern near the Akron POTW. The
26
-------
TOXICITY TESTING RESULTS
number of benthic taxa and benthic diversity values de-;
clinesfrom maximum levels of about 50 upstream of the '-•
Akron POTW to minimum levels of about 20 at RK 60
through 40 in 1984 (Figure 5-2). The Invertebrate Com-
munity Index data illustrate the same trend of a decline
from good water quality beginning downstream of the :
confluence with the Little Cuyahoga (upstream of the '-
Akron POTW) to poor quality near RK 55 downstream of.
the Akron POTW (Figure 5-3). Recovery toward up-
stream levels was better in the 1986 survey than in the '
1984 survey. More noticeably, the fish community Index|
of Well-Being declines rapidly from good to very poor
status just downstream of the Akron POTW (Figure 5-3).
Data from 1984 to 1986 show this decline, although the
magnitude is smaller in 1985 and 1986. The fish index |
never improves to the level found upstream of the Akron
POTW.
5.4 Summary
The principal conclusion from this data is that the Akron
POTW's effluent is a major cause of the toxic impact to
the Cuyahoga River. Of the three organisms tested,Cer-
iodaphnia is most sensitive to the effluent and ambient
samples.TheCeriodaphnia effluent toxicity data and riv-
er flow data indicate that the concentrations of effluent
in the river are sufficient to cause an adverse biological
effect.
No evidence was obtained to show that bypassing
events greatly influence or markedly changed the pat-
tern of ambient toxicity observed; In studies of effluent
and ambient toxicity at other sites, wastewater bypass-
ing did affect ambient toxicity (32). The good survival of
adu\tCeriodaphnia at Stations 1, 2, and 3 suggests that
upstream of the Akron POTW the water quality is good.
Lake
Rockwell
Little
Cuyahoga
River
Peninsula
! Dam
Tinkers
Creek
Cleveland
Harbor
Monroe
Falls Dam
Akron
POTW
Brandywine
Creek
Ship
Channel
X
£
o
X
UJ
m
DC
UJ
QQ
eor
50
40
30
20
10
National Recreational Area
uj
Q
Z
55
oc
UJ
5
100
90
80
70
60
50
40
30
20
10
RIVER KILOMETER
100-
80-
60-
40-
20-
-6.0
-5.0
-4.0
-3.0
i
1
CO
c:
30
-2.0 ^
-1.0 -
0--0
Figure 5-2. Profile of the Number of Benthic Taxa, Benthic Diversity, and Ceriodaphnia Survival in the Cuyahoga River, Ohio
27
-------
TOXICITY TESTING RESULTS
Little
Cuyahoga
River
Peninsula
Dam
Tinkers
Creek
Cleveland
Harbor
O
UJ
i
I National Recreation Area
§
O
§
1
C/3
30
I
'- 20
90
80
70
60 50 40 30
RIVER KILOMETERS
20
10
Flguro 5-3. Profile of the Cuyahoga Macroipvertebrate Community (17) and Ceriodaphnia Survival
This is supported by Ohio EPA's fish and invertebrate
surveys data. Although there are other dischargers to
the Cuyahoga River and several tributaries, the Cerio-
daphnia survival and young production data did not
show a pattern which would indicate other important
sources of toxicity (Figure 5-5).
Analysis of the various types of information yields the
following conclusions:
1. The Akron POTW effluent causes the impact to the
biota observed in the Ohio EPA surveys. This is
clearly seen in Figures 5-3 and 5-4, where the ambi-
ent toxicity data is plotted against the fish and in-
vertebrate biosurvey data.
2. Whatever the role of bypassing in causing an in-
stream toxic effect, the toxicity of the continuously
discharged effluent is sufficient to cause an in-
stream effect.
3. No other important sources of toxicity were ob-
served within the study area. There are point-
source discharges on tributaries up- and down-
stream of the Akron POTW, as well as nonpoint
source dischargers. However, the Akron POTW ef-
fluent dominates the Cuyahoga in the reach where
adverse biological community impacts were ob-
served.
4. The Akron POTW must reduce the toxicity of its ef-
fluent before improvement in the Cuyahoga's resi-
dent biota will occur.
5. Investigation into the source of the effluent's toxic-
ity should take place to account for any observed
variability in the toxicity.
28
-------
TOXICITY TESTING RESULTS
HI
00
UJ
u.
O
X
UJ
Q
Q
UJ
Q
O
Lake
Rockwell
Little
Cuyahoga
River
Peninsula
Dam
Tinkers
Creek
Cleveland
Harbor
Monroe
Falls Dam
Akron
POTW
Brandywine
Creek
Ship
Channel
12r
National Recreational Area
10
Exceptional
¥¥TV—
100
80
60
40
20
100 90
80
70 60
50 40
30
20 10
RIVER KILOMETER
§
1
1
-------
TOXICITY TESTING RESULTS
110T
100
80
§ 60
w
40
20
• Aplexia
• Fathead Minnow
- Ceriodaphnia
Little !
Cuyahoga
River POTW .Yellow Creek
Mud
Brook
H
Furnace
Run
Brandywine
Creek
Tinkers
Creek
I I
100
90
80
70
60 ; 50
40
30
20
RIVER KILOMETERS
10
Figure 5-5. Survival of Fathead Minnows, Ceriodaphnia, and Aplexia Exposed to Ambient Water From the Cuyahoga River
30
-------
6. Toxicity Source Investigation
6.1 Introduction
From the results of analyses presented in this document,
the discharge from the Akron POTW was identified as a
source of toxicants to the Cuyahoga River. The next tier in:
pollution control is to reduce the level of toxicity. U.S. EPA;
recommends that sources which discharge unacceptable
levels of toxicity be required to perform a Toxicity Reduc- ,
tion Evaluation (TRE). The initial steps in a TRE are pre- •
sented in Mount and Anderson-Carnahan (31) and include
determining what component(s) of the effluent are con-
tributing to the toxicity. Chemical and toxicological infor- .
mation on the effluent is needed to answer these ques-
tions:
• What is the variability associated with the com- :
pound(s) causing the toxicity in the POTW effluent?
• What compound(s) are causing that toxicity? ;
• Is the same chemical always responsible for the toxic-
ity? i
This information is used to determine what measures are
needed to control the effluent toxicity to acceptable levels
by helping to select treatment methods for toxicant re-
moval or to identify toxicants which can be prevented
from entering the wastestream. '•
In practice, a POTW discharging a toxic effluent may first ;
desire to investigate toxicant origins within their collection
system. That investigation may be accomplished by test- ;
ing wastewater from specific locations in the collection ;
system to identify important contributors of toxicants.
Toxic inputs may be controlled through implementation of
a pretreatment program. In this case, the Akron POTW ef-
fluent was tested for toxicity and fractionated to chemical-
ly isolate toxicants in an attempt to identify important ;
compounds and to determine appropriate treatments ;
which would reduce effluent toxicity.
In the first phase of a TRE, investigators conduct toxicity ,
tests to screen and characterize the chemical and physical
properties of the effluent and suspected toxicants. "A par-
allel series of relatively simple, low cost analyses" and
sample treatment are conducted, each of which is de-
signed "to remove or render biologically unavailable a ;
specific group of toxicants such as oxidants, metals, non-
polar organics and metal chelates" (33). After these sam-
ple treatments (for example, aeration or filtration) toxicity
tests are conducted to determine their effectivenes in re-
ducing toxicity.
The second phase of the TRE process is the Toxicity !
Source Investigation (TSI) which may be conducted in two
ways: using bench-scale treatability studies to further ex-
amine the success of various treatments on the toxicity
without identifying the toxicant; or identifying the toxicant
through the chemical and toxicological examination of se-
quential effluent fractions intended to isolate toxicants.
This second option was selected for the Akron POTW. This
type of analysis is still being developed and detailed pro-
cedures are to be published in 1988 by U.S EPA in a man-
ual titled "Phase II - Toxicant Identification/Source Investi-
gation and Control Manual."
Information concerning the waste treatment processes
and the effluent composition at the Akron POTW was ac-
cumulated during this study. The effluent toxicity test re-
sults showed variable toxicity (Table 5-7), although this is
not unusual for a POTW. Hence the TSI process paid par-
ticular attention to changes in the effluent characteristics
over time. The ambient test results demonstrated that tox-
ic effects occurred in the river, and that survival decreased
with distance downstream from the discharger (Tables 5-8
and 5-9). The lowest survival of fathead minnows occurred
when exposed to Station 8 and Station 9 waters. Signifi-
cantly lower survival oiCeriodaphnia occurred when ex-
posed to water from Station 5 and further downstream,
with the exception of Stations 10 and 12. This persistent
effect indicated that the toxicant(s) were not rapidly de- ;
graded, and was consistent with the observed impacts to
the biological community of the Cuyahoga River down-
stream of the Akron POTW.
6.2 General Procedures for the Toxicity Source
Investigation
6.2.1 Sample Type and Sampling Frequency
Samples were initially collected by compositing effluent
over a 24-hour period. As effluent toxicity declined, the
later samples, primarily those collected on and after
April 22, 1986, were grab samples to capture more toxic
effluent by avoiding the averaging effect of composite
samples. The samples were placed on ice and shipped
to ERL-Duluth via overnight carrier for testing. This sam-
ple collection effort began in January 1986 and contin*
ued through September 1986. The sampling schedule,
was initially set to obtain three samples each month on
consecutive days. The schedule was modified slightly
when grab samples were collected so that from two to
four samples monthly were tested. Aliquots from each
sample were taken for the various test procedures.
6.2.2 Test Species
Laboratory toxicity tests use aquatic species as detec-
tors of toxicity. Consequently, it is crucial for a sensitive
species to be used as the detector. Toxicity testing data
31
-------
TOXICITY SOURCE
presented in Chapters 3 and 5 indicated thatCeriodaph-
nla was the most sensitive of the three aquatic organ-
isms tested. Further, Ceriodaphnia are adaptable to frac-
tionation testing. ERL-Duluth maintains cultures of
Ceriodaphnia for testing. Use of this species would pro-
vide continuity with previous test results using Cerio-
daphnia.
6.3 Phase I Test Procedures
The testing procedures were developed during this ef-
fort and have since been modified. The most advanced
description of these procedures is in Mount and Ander-
son-Carnahan (33).
6.3.1 Physical Treatment
The three samples collected in January 1986 were sub-
jected to air stripping for removal of volatile compounds
and then filtered to remove suspended particulate mate-
rial. Procedures for these sample preparations are de-
scribed in Mount and Anderson-Carnahan (33). The air
stripping procedures included adjusting the pH so that
acidic, neutral, and basic conditions were created. The
altered pH levels were maintained during the aeration
process. After aeration, the pH was returned to original
levels before performing toxicity testing.
Test results were examined to determine whether toxic-
ity was reduced and if so, by how much in each of the
three pH conditions. If the toxicity is reduced similarly in
the three aerated samples, but not removed, this would
indicate that the toxicity is partially due to volatile com-
pounds which are unaffected by pH. If toxicity is reduced
and dissimilar between the aerated samples, then the
general type of compound(s) contributing to the toxicity
is revealed (e.g., greater toxicity reduction in the basic
aerated sample indicates that a basic volatile compound
was contributing more to the toxicity).
Filtering was the second physical treatment for the Ak-
ron POTW effluent. Samples were filtered with glass-fi-
ber filters so that adsorption of dissolved organics
would be reduced. Toxic compounds may be adsorbed
on suspended material. The filtered samples were used
in toxicity tests to determine the effect of the removal of
suspended particulates (i.e., filterable solids). If toxicity
is removed by filtering, then filtration may be used as an
effluent treatment to reduce toxicity.
6.3.2 EDTA-Chelation Test
The ethylenediaminetetraacetate ligand (EDTA) is used
to observe whether EDTA-chelatable cations (possibly
metals) are causing toxicity. Direct analyses for such
metals are less useful since much of the measured con-
centration may be biologically unavailable in the efflu-
ent and does not contribute to the observed toxicity. In
the EDTA-chelation procedure, the results of toxicity
tests on whole effluent samples spiked with EDTA are
compared to results of tests on samples without the ad-
dition of EDTA If toxicity is removed with the addition of
EDTA then toxicity may be caused by a metal.
6.3.3 Oxidant Reduction Test
Sodium thiosulfate is used as a reducing agent to deter-
mine whether oxidants are contributing to the toxicity of
the effluent. In sufficient concentration, sodium thiosul-
fate is toxic to Ceriodaphnia and this toxicity is a func-
tion of the concentration of electrophiles in the sample.
if sufficient electrophiles are present in the effluent to
consume the thiosulfate, then toxic effects will not be
observed. Therefore, varying amounts of sodium thio-
sulfate are added to determine the toxic concentration
and, once known, the predetermined concentrations are
added to the effluent. The results of toxicity tests on
whole effluent samples with sodium thiosulfate added
are compared to results of tests without the addition.
Chlorine is a typical oxidant that is extensively used for
disinfecting wastewaters. The information from this test
is also used to aid in the search for the toxicant and to
initially identify an effective treatment for the removal of
toxicity.
6.3.4 Whole Effluent Acute Toxicity Test Procedures
Typical effluent static acute toxicity tests were conduct-
ed to obtain 48-hour LC50s using a concentration series
that progressed at 0.5 concentration intervals: 0, 6, 12,
25,50, and 100 percent effluent. If greater resolution was
desired for the higher concentration range, then a 75
percent concentration was substituted for the six per-
cent concentration. The testing methods for Ceriodaph-
nia are generally those from Peltier and Weber (34). The
LC50 values were calculated using graphical interpola-
tion.
The results of these static toxicity tests were used to
judge the toxicity of the effluent in comparison to the
various solid phase extraction fractions and to the re-
sults of the EDTA-chelation and/or oxidant reduction
tests.
32
-------
TOXICITY SOURCE
6.4 Phase II Test Procedures
On receipt of the whole effluent sample, an aliquot of
the sample was immediately tested for acute toxicity. In
general, aliquots of whole effluent samples were sub- ;
jected to the solid phase extraction (SPE) test to investi-l
gate toxicity due to nonpolar organics. These extracted
samples were sometimes subjected to the EDTA-Chela-
tion test to investigate toxicity due to cations, and/or the
Oxidant Reduction test to investigate toxicity due to oxi-
dants.
6.4.1 Solid Phase Extraction (SPE) Test Procedures
The purpose of the SPE procedure is to simplify the
identification of nonpolar organic toxicants. The effluent;
is separated into fractions containing smaller numbers
of compounds than in the whole effluent. This proce- !
dure was capable of removing toxicity and therefore
was the predominant procedure used after the initial
January 1986 samples. Effluent sample material parti-
tioned on a C-18 (octadecylsilane bonded silica gel solid
phase extraction column [J.T. Baker Chemical Co., Phil-
lipsburg. New Jersey]) was sequentially extracted from ;
the column using 25, 50, 75, 80, 85, 90, 95, and 100 per-!
cent methanol solutions in water for the SPE test. i
Methanol concentrations are increased in successive
elution volumes so that the polarity decreases for each
fraction of effluent material that is eluted with each vol-
ume of methanol and water.
Effluent fractions from the SPE test, as performed here,
represent concentrations of the effluent which may be
up to five times more concentrated than the original
sample (whole effluent) depending on the efficiency of
sorption onto the C-18 column and elution into the ;
methanol-water fraction. Consequently, if toxicity is not
observed from exposure to such concentrated samples,
then the components in that fraction are assumed to not
cause toxicity in the whole effluent. If toxicity is ob- ;
served from exposure to such concentrated samples,
then toxicity may contribute to the toxicity of the whole
effluent. However, the observed toxicity may not be of
sufficient strength to cause a toxic effect in the whole ef-
fluent (a dilution of up to five times the concentration in
the SPE fraction). i
6.4.2 Acute Toxicity Fractionation Tests
Acute toxicity tests were conducted using effluent sam-
ple material that had been eluted from the C-18 column i
during the pass-through of each methanol-water frac-
tion. Toxic effects were determined to have occurred
when the mortalities oiCeriodaphnia exposed to the
methanol fractions were greater than in the effluent con-
trol.
To determine whether a toxic effect in one of the con-
centrated SPE test fractions causes toxicity in the whole
effluent, definitive toxicity tests should be conducted on
the effluent fraction(s) of interest. Concurrent chemical
analysis should be performed to determine the actual
concentrations of a particular toxicant in the fraction of
concern. To determine the toxicant concentration and
exposure concentration, the compound(s) believed to be
causing the toxicity must be identified and the recovery
efficiency of that compound from the C-18 column into
the methanol fraction must be known.
In this effort at the Akron POTW, the toxic compound(s)
was never identified. Consequently, definitive toxicity
tests and associated chemical analyses were not con-
ducted.
6.4.3 EDTA-Chelation of SPE Fractions
After June 1985 the EDTA-chelation test was conducted
on some of the methanol fractions from the SPE test.
This was done to ensure that the metals in the effluent
were not causing toxicity in the methanol fractions.
When EDTA was added to specific SPE fractions, toxicity
tests were again performed, but only to indicate the
presence of a toxic effect (i.e., greater Ceriodaphnia
mortalities than in the effluent controls).
6.4.4 Chemical Analysis of Metals
Effluent samples from January, April, July, and August
were analyzed for the presence of metals. The concen-
trations of cadmium, chromium, copper and zinc were'
determined using standard U.S. EPA methods for atom-
ic absorption spectroscopy.
6.4.5 Chemical Analysis of Organics
Extracts of methanol fractions of whole effluent from the
four August 1985 samples were injected into a gas chro-
matograph with an Ion Trap Mass Detector to examine
the classes of compounds in the SPE fractions and to
search for suspect toxicants. The methanol fractions
were first diluted with water to reduce the concentration
of methanol, extracted with hexane, concentrated
(about 10 times), and then injected into the gas chroma-
tograph. The analytical methods met the performance
criteria for quality control listed in Section 8.2 of U.S.
EPA Method 625.
33
-------
TOXICITY SOURCE
The chromatograms from each sample were examined
to locate peaks either common to the toxic fractions or
of sufficient concentration to yield valuable spectra.
Spectra identification information from the U.S.
EPA/NIH Mass Spectral Database was used to identify
common peaks.
6.5 Phase I Data and Results
6.5.1 Physical Treatment
Whole effluent samples with different pH values were
subjected to air stripping. The samples were returned to
their original pH and then tested for toxicity to Cerio-
daphnia. This treatment was used to remove volatile
compounds. No effect on the toxicity of the Akron POTW
effluent was found as a result of air stripping.
Whole effluent samples were filtered to remove sus-
pended particulates and, similarly this treatment did not
affect the toxicity to Ceriodaphnia when compared to
untreated effluent.
Results of these two physical treatments indicate that
the toxicants were nonvolatile compounds and they
were not associated with solids in the effluent.
6.5.2 Whole Effluent Toxicity
Twenty-nine effluent samples (composite and grab)
were collected over the course of this study and static
acute toxicity tests were conducted using Ceriodaphnia
to determine 48-hour LC50 values. The LC50s varied
from 38 to >100 percent effluent for samples collected
from Janaury 13 to September 26,1986 (Table 6-1). Over
Table 6-1. Whole Effluent Acute Toxicity Test Results for Ceriodaphnia Exposed to Akron POTW Effluent, 1986
Sample
Date/Type8
January 13, C
January 14, C
January 15, C
February 7, C
February 8, C
February 9, C
March 28, C
March 29, C
March 30, C
April 22, 1G
2G
3G
4G
1Gb
June 16, C
June 26, 1G
2G
June 27, 1G
2G
JulyS, 1G
2G
July 10, 1G
2G
August 6, C
August 14, C
August 18, C
August 26, C
September 18, C
September 22, C
48-hour
LC50
(% Effluent)
82
>100
>100
>100
79
38
77
81
67
50
82
100
67
<25
>100
>100
>100
>100
>100
>100
>100
>100
72
>100
>100
>100
>100
>100
>100
Zinc
Concentration
(M-g/L)
136
274
146
58
66
57
26
EDTA-Chelation
Toxicity Test
Result
Toxicity removed
No effect
No effect
No effect
No effect
No effect
No effect
No effect
No effect
(c)
No effect
No effect
No effect
No effect
No effect
No effect
Oxidant
Reduction
Test Result
No effect
No effect
No effect
8C * composite, G = grab sample
"The first grab sample for April 22 was retested after 30 days. ;
°Toxicity reduced in 80 and 95 percent, but unaffected in 85 percent methanol fraction.
34
-------
TOXIC1TY SOURCE
half of the LC50 values were >100 percent effluent (17 of;
the 29 samples). The effluent grab samples collected in ,
June and July had LC50 values >100 percent effluent, |
except one of the July 10 samples which had an LC50 of;
72 percent effluent.
The majority of the LC50 values for the winter and spring1
samples collected through April 22, 1986 indicated that ;
the effluent was acutely toxic. In contrast, whole effluent
samples collected after that date had no acutely toxic ef-
fects (except one). :
6.5.3 EDTA-Chelation !
The EDTA-chelation test was performed on 14 whole ef-
fluent samples from June through July. None of these
effluents had shown acutely toxic effects on Ceriodaph-
nia, and the addition of EDTA had no effect on the ob-
served toxicity, with two exceptions (Table 6-1). The re-.
tested sample from April 22 was acutely toxic at <25 !
percent effluent, and the addition of EDTA removed the
toxicity. The whole effluent grab sample from July 10
was acutely toxic at an LC50 of 72 percent and the addi-:
tion of EDTA reduced the effluent toxicity. This result in-
dicates that metals were causing the toxic effects on this
particular sample. \
6.5.4 Oxidant Reduction
The oxidant reduction test was performed on three sam-
ples collected in February 1986. There was no reduction
of toxicity from the addition of sodium thiosulfate to the
effluent (Table 6-1). The lack of toxicity reduction indi-
cates that oxidants are not the cause of toxicity in the ef-
fluent samples. A typical oxidant present is chlorine
used for disinfection. The chlorination period at the Ak-
ron POTW is from May 1 to October 31, so that the sam-
ple tested did not contain chlorine.
6.6 Phase II Data and Results
6.6.1 SPE Fractionation
For each of the samples obtained from the Akron POTW,
the SPE toxicity tests were conducted in conjunction
with effluent toxicity tests to determine 48-hour LCBOs of
the whole effluent. The results of the SPE tests indicated
that, even in cases where the LCBOs were >100 percent
effluent, there was toxicity in at least one of the eight
methanol fractions (Tables 6-1 arid 6-2). And conversely,
while the March 29 sample had an LC50 of 81 percent ef-
fluent, the SPE fractions did not indicate toxicity. This
sample was not tested for the EDTA-chelation or Oxi-
Table 6-2. Solid Phase Extraction Toxicity Test Results3
Date/Type13
January 13, C
January 14, C
January 15, C
February 8, C
February 9, C
March 28, C
March 29, C
March 30, C
April 22, 1G
2G
3G
4G
July 10, 2G
August 6, G
August 14, C
August 1 8, C
August 26, C
% Methanol
25 50 75 80 85 90
T°
}
' Tc
Tc
i T°
: Tc
.
Tc
i Tc
; T°
Tc
i Tc
Tc Tc
; Tc
TC
, -p.
Tc
95 100
Tc
-re
I
T°
Tc
Tc
Tc
Tc
T°
a - denotes not tested
bC = composite, G = grab sample
Toxicity found in that fraction
35
-------
TOXICITY SOURCE
dant Reduction effects to further determine if the whole
effluent toxicity was due to either metals or oxidants, re-
spectively.
The SPE fraction which predominantly contained toxic
components was the 85 percent methanol fraction. This
result indicated that it was the more nonpolar organic
compounds which caused toxicity and that such com-
pounds were consistently present in the effluent. Fur-
ther, the 100 percent methanol fraction, containing the
most nonpolar compounds, resulted in toxic effects to
Ceriodaphnia in four of the samples. Toxic effects in the
10 percent methanol fraction were only observed when
the toxic effects in the 85 percent methanol fraction were
great (when mortality was ensuing rapidly). This sug-
gests that there is more than one toxic, nonpolar com-
pound in the effluent, and that while their polarities dif-
fer they may have a common industrial source.
6.6.2 EDTA-Chelation
Further examination of the July 10,1986 sample indicat-
ed that toxic effects were observed in the 80, 85, and 95
percent methanol fractions. The EDTA-chelation test re-
sults for the 80 and 95 percent fractions indicated that
the toxicity was again reduced with the use of EDTA. So,
in contrast to the 85 and 100 percent methanol fractions,
the toxic effects exhibited in the 80 and 95 percent frac-
tions were contributed by metal compounds. Further,
the EDTA test on the 85 percent fraction indicated that
toxicity was not reduced when EDTA was added.
These results illustrate that there are at least two types
of toxicants present in the Akron POTW effluent—non-
polar organics and metals. The nonpolar organic toxicity
appears to be due to a compound that is eluted off the C-
18 columns using 85 and 100 percent methanol solu-
tions. The 85 percent methanol fraction demonstrated
the greatest and most consistent toxicity, while the 100
percent methanol fraction was less consistent. In addi-
tion, the loss of toxic effects by the addition of EDTA
demonstrated that at least one of the toxic components
was chelatable.
6.6.3 Organic Chemicals
From the examination of the GC/MS chromatogram of
the April 22 grab samples, six compounds were found in
relatively high concentrations. Three of those were iden-
tified using computerized compound libraries. The com-
pounds were 1-methyl naphthalene, 2-methyl naphtha-
lene, and 3,5-bis (1,1-dimethylethyl) phenol. Original
concentrations in whole effluent for these three com-
pounds in the SPE fractions were estimated to be in the
mg/L range. Based on quantitative structure activity rela-
tionship estimates and toxicity data, these concentration
levels are too low to be suspected of causing toxicity
(35,36). Plans to purify other toxic SPE fractions, espe-
cially the 85 percent methanol fraction, to further exam-
ine the predominant source of toxicity could not be com-
pleted since effluent samples after April 22 were not
acutely toxic to Ceriodaphnia.
6.6.4 Metals
Zinc had been identified earlier as a potential toxicant
because of the very high concentrations in the influent
|(over 600 mg/L). The high influent concentrations par-
tially arose from the use of zinc orthophosphate in the
municipal water treatment plant at concentrations of 200
to 300 mg/L and from metal plating facilities discharging
to the Akron POTW (37). Therefore, zinc concentrations
were monitored in some of the samples for the whole ef-
fluent acute toxicity tests. Concentrations of zinc ranged
from 136 to 274 mg/L for three samples collected in Jan-
uary, April, and July. These concentrations are similar to
the effluent concentrations of approximately 200 mg/L
zinc (37). Lower effluent zinc concentrations were ob-
tained in the four August samples, 26 to 66 mg/L of zinc,
but do not correspond to the high influent levels during
August (mean concentration of 846 mg/L zinc). Appar-
ently, treatment was effective in removing a large por-
tion of the zinc present in the influent. Concentrations of
zinc found to be acutely toxic range from 32 to 76 mg/L
(38, 25), so that even the low concentrations of zinc in
the effluent may have been acutely toxic. However,
since the bioavailability of zinc and the bioavailable por-
tion in the effluent was not known, the actual toxic con-
centration is unknown.
The concentrations of cadmium, chromium, and copper
were also measured from effluent samples collected in
April, July, and August 1986. The concentrations were at
or below the water quality criteria for protection of
aquatic life (hardnesses of up to 200 mg/L CaC03). These
metals were judged not to be contributors to the ob-
served effluent toxicity in April. During July and August
the effluent was not toxic.
6.7 Wastewater Bypassing Concerns
Wastewater bypassing had been identified as a potential
source of the toxicants causing the adverse impacts ob-
served in the Cuyahoga River when this case study be-
gan. The frequency and duration of bypassing was ex-
amined and indicated that most events were of less than
12 hours duration and were often the consequence of
rainfall events when the Akron POTW capacity was ex-
36
-------
TOXICITY SOURCE
ceeded (19). Travel time through the study area was esti-;
mated to be less than 24 hours, but acute toxicity toCer-
iodaphnia did not occur until after 24 hours of exposure ,-
in the one sample of bypassed wastewater tested. The |
information gained in this effort does not confirm initial'
suspicions that bypassing of untreated or partially treat-.
ed wastewaters is the cause of adverse conditions in the
Cuyahoga River. Wastewater bypassing, however, could'
be a contributing factor.
6.8 Summary ;
The tools used to conduct the TSI were refined and ex-
panded during this study, and are the basis for the pro-'
cedures described in the U.S. EPA manual for TREs by '
Mount and Anderson-Carnahan (31). In this case study, I
effluent toxicity decreased markedly before toxicity
characterization procedures were able to identify causa-:
live toxicants.
37
-------
7. Conclusions and Application of Case Study
Data to Setting Permit Limits
7.1 Introduction
The Akron POTW was studied through a cooperative ef-
fort between various Federal and state agencies. The pur-
pose of the work was to determine if the Akron POTW dis-
charge was a source of contaminants/toxicants to the
Cuyahoga River and to document the process of discov-
ery and research using water quality-based procedures for
the control of toxics. These conclusions and recommenda-
tions are based on the research conducted at the Akron
POTW only and are not meant to influence the pollution
control and abatement activities of Ohio EPA or the City of
Akron. The results of the toxicity testing are summarized
in Table 7-1 and then used in the development of sample
permit limits for the Akron POTW.
7.2 Toxicity Testing Needs to Reduce
Uncertainty Levels
U.S. EPA (24) provided guidance for the degree of whole
effluent toxicity testing needed to determine whether to 1)
continue toxicity data generation, 2) begin permit limit
Table 7-1. Toxicity Testing Summary for the Akron POTW
derivation, or 3) place the discharger in a low priority per-
mitting category. This guidance includes approximating
the uncertainty associated with toxicity test data due to ef-
fluent variablity, test species sensitivity, and acute-to-
chronic ratios (ACR). U.S. EPA (24) provides uncertainty
factor values for each, from 1 to 1,000 and the more infor-
mation available the lower the factor value. The goal is to
obtain sufficient information to lower the level of uncer-
tainty so that
Toxic Response
> level of uncertainty (Equation 7-1)
where
Toxic response is measured either by LC50s for acute
toxicity or by NOELs for chronic toxicity.
IWC = instream waste concentration;
Q effluent/ Q river + Q effluent
Level of Uncertainty (UF) = product of the three uncer-
tainty factors
Test Dates
Aug 1984
Aug 1984
May 1985
May 1985
Jul 1985
Jul-Aug 1985
Jul-Aug 1985
Sep-Oct 1985
Sep-0ct1985
Sep-Oct 1985
Sep-Oct 1985
Sep-Oct 1985
Sep-Oct 1985
Dec 1985
Jan 1986
Feb 1986
Mar 1986
Apr 1986
Jun 1986
Jul 1986
Aug 1986
Sep 1986
Investigators
EPA-Cincinnatib
EPA-Cincinnatib
EPA-Cincinnatib
EPA-Cincinnatib
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
ERL-Duluth
Test Species
Fathead minnow .
Ceriodaphnia
Fathead minnow
Ceriodaphnia
Ceriodaphnia
Ceriodaphnia
Fathead minnow
Ceriodaphnia
Fathead minnow
Aplexia
Ceriodaphnia
Fathead minnow
Ceriodaphnia
Ceriodaphnia
Ceriodaphnia
Ceriodaphnia
Ceriodaphnia '•
Ceriodaphnia
Ceriodaphnia
Ceriodaphnia
Ceriodaphnia
Ceriodaphnia
Toxicity
Range3
(% Effluent)
30->100 NOEC
30 NOEC
30 NOEC
>100 NOEC
20-100
30->100 NOEL
>100 NOEL
10 NOEL
75 NOEL
>100 NOEL
0-100
50-100
80-100
30 NOEL
82->100LC50
38->100LC50
67-81 LC50
50-100LC50
>100LC50
>100LC50
>100LC50
>100LC50
Sample Type
Effluent
Effluent
Effluent
Effluent
Ambient
Effluent
Effluent
Effluent
Effluent
Effluent
Ambient
Ambient
Ambient
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Number
of Tests
2
2
1
1
4
3
2
1
1
1
13
13
13
2
3
3
3
40
5
4
4
2
"Survival data only.
'Tests were conducted on unchlorinated, chlorinated, and dechlorinated effluent. Data used were those from the latter two cate-
gories.
T0sf data from the 30-day toxicity decay test conducted on the April 22 sample are not included.
38
-------
CONCLUSIONS
Using the lowest LC50 value of 38 percent effluent and
flows corresponding to the 7Q10 and mean Akron POTW
flow (Table 5-4), Equation 7-1 was solved for the necessary
level of uncertainty. The left side of the equation has a val-
ue of 0.55, and for this to be greater than the level of un-
certainty would require that each of the uncertainty fac-
tors have a value less than 1, which is not possible.
0.38
2.79 m3/s
, = 0.55
1.27m3/s + 2.79 m3/s
Initial values for the uncertainty factors are: 10to100for
effluent variability, 10 for species sensitivity, and 10 for
the acute-to-chronic ratio (2). While short-term effluent :
variability was briefly examined in Chapter 6 (Table 6-1) ,-
by using grab samples in acute toxicity tests, U.S EPA
recommended that such testing be conducted monthly
for at least one year. Similarly, long-term effluent vari-
ability tests were not conducted in sufficient number to
address this source of uncertainty. The uncertainty fac- '
tor for effluent variability when the variability in effluent,
toxicity is less than an order of magnitude is 10 (24)
(Table 7-1).
Three species were used for chronic toxicity testing dur-
ing the on-site visit (Chapter 5) and Ceriodaphnia was !
identified as the most sensitive species. While monthly ,
effluent samples were not tested at the frequency rec-
ommended by U.S. EPA (24), it may be argued that suffi-
. cient testing was conducted before the reconnaissance ,
trip, during the reconnaissance trip, and during the site '.
visit to have adequately addressed species sensitivity. ;
Therefore, the uncertainty factor for species sensitivity is
1. Chronic toxicity tests are to be conducted using three,
species for comparison to the acute toxicity test data
generated for the species sensitivity factor. The frequen-
cy of recommended testing for the determination is
from 144 to 216 tests per year. This level of testing was ,
not met; however, ambient toxicity testing was conduct-
ed using three species to measure actual instream toxic-
ity so that an ACR is not needed (24). Consequently, the
uncertainty factor for ACR is 1. ;
The minimum level of uncertainty based on the above
discussion is equal to 10 X 1 X 1, or 10. This value is ;
greater than 0.55 from Equation 7-1. In such cases, fur-
ther testing cannot reduce the level of uncertainty to sat-
isfy Equation 7-1, so the data generated from definitive ;
toxicity testing can be used to set permit limits (24).
When effluent dilution in the receiving water at critical
low flows (or design flows) is < 100 to 1, then definitive
data generation will always be recommended. Note that
the above discussion was presented for illustrative pur-
poses as part of the permit derivation process since tox-
icity data are not necessary to set permit limits.
7.3 Derivation of Sample Permit Limits
The concepts developed and presented in the Technical
Support Document for Water Quality-based Toxics Con-
trol (24) and the Permit Writer's Guide (30) will be dis-
cussed as appropriate to the work conducted at the Ak-
ron POTW for the derivation of sample permit limits.The
derivation of sample limits is presented for educational
purposes only and does not represent an official U.S.
EPA regulatory action. The most sensitive species to the
Akron POTW effluent and ambient waters was Cerio-
daphnia and results from those toxicity tests will be
used for calculating the sample limits.
7.3.1 Exposure Criteria
Exposure criteria were developed for acute and chronic
exposures as the next step in the permit limit derivation
process. For these two exposures, the recommended
criteria were defined as: criterion maximum concentraj
tion (CMC) for acute exposures, and a criterion continu-
ous concentration (CCC) for chronic exposures (24, 30).
These exposure criteria are for completely mixed dis-
charges when measuring whole effluent toxic effects. A
completely mixed discharge is one that mixes rapidly
with the entire receiving water flow. The CMC and CCC
are recommended targets to limit effluent toxicity and
are defined below.
CMC< 0.3 TUa, and (Equation 7-2)
CCC < 1.0 TUC (Equation 7-3)
where
TU =
100
LC50 or NOEL
7.3.2 Wasteload Allocation
The purpose of the wasteload allocation (WLA) is to set
values for acceptable effluent concentrations. U.S. EPA
(30) defines wasteload allocation as "the portion of a re-
ceiving water's total maximum daily pollutant load that
is allocated to one of its existing or future point sources
of pollution." State definitions of mixing zones and de-
sign flow specifications will greatly influence the WLA
process through model selection and will determine
where the permit limits are to be'met.
Several types of models have been developed for the
WLA process and selection is dependent upon the avail-
able data and the type of discharge. A steady-state
39
-------
CONCLUSIONS
wasteload allocation model is acceptable to determine
exposure limits and effluent quality levels which will
meet discharge criteria. U.S. EPA (24) regards this model
type as appropriate for effluent-dominated streams like
the Cuyahoga River. The discharge from the Akron
POTW is sufficiently large relative to the flow of the
Cuyahoga so that the effluent often composes more
than 20 percent of the river flow. This procedure also as-
sumes that there are no upstream concentrations of pol-
lutants or toxicants (30). This is in agreement with the
ambient toxicity data for the Cuyahoga upstream of the
Akron POTW discharge, where Ceriodaphnia survival
was 100 percent.
Since the Akron POTW is being treated as a single dis-
charger, the allocation process is quite simple. There are
no other dischargers for which allowable toxicant con-
centrations must be apportioned.
For situations in which the effluent is rapidly and com-
pletely mixed, U.S. EPA (24) states that two wasteload
allocations should be conducted to meet the CMC and
CCC at specific exposure durations and frequency. From
visual observations of dye dilution at the Akron POTW
discharge (Chapter 4), the effluent was seen to be rapid-
ly mixed and for this discussion will be regarded as com-
pletely mixed. However, U.S. EPA (24) states that when
mixing zones have been specified, the wasteload alloca-
tion cannot be based on the assumption that an effluent
is completely mixed at the point of discharge. The State
of Ohio standards allow the Ohio EPA Director to define
the size of a mixing zone for a receiving watercourse up
to one-third of the cross-sectional area of that water-
course with a length up to five times the width of the
river at the point of discharge (Ohio Water Quality Stan-
dards. Chapter 3745-1 of Ohio Annotated Code). Within
this mixing zone acutely toxic conditions are to be
prevented (24).
Receiving water flow information is needed in the form
of design flows for calculating the acceptable effluent
concentrations (24). If design flows are not specified by
the regulatory agency for calculation of the WLA, then
hydrologically-based design flows can be used to calcu-
late acceptable effluent loads. It is the policy of Ohio EPA
to allow use of the whole design flow in WLA calcula-
tions, without accounting for the mixing zone, where the
receiving water and effluent mix rapidly (e.g., for the
Cuyahoga River) (39). Such design flows for stressed
systems are recommended to be the 1Q10 for calculat-
ing the CMC and the 7Q10 for calculating the CCC (30).
Note that the Ohio EPA does not use 1Q10 design flows
for permitting; however, for consistency with U.S. EPA
general recommendations (24,30) 1Q10 design flows
will be used in this example. U.S. EPA (30) states that
since the "acceptable effluent toxicity is a function of di-
lution and the ambient criteria, the available dilution will
drive the WLA and thus the permit limit. The key to this
procedure is to determine the dilution factor." Accept-
able instream toxicity values to meet the ambient crite-
ria are determined by U.S. EPA (24,30) as
: dilution factor X CMC or CCC (Equation 7-4)
; WLA
where
• the dilution factor = Qe + Qs / Qe and
Qs = design flow for the receiving water
! (Cuyahoga River)
; Qe = design flow for the effluent (Akron POTW)
The equation above is appropriate for situations where
the source water and receiving water are not the same.
The source of the Akron drinking water and much of the
.subsequent Akron POTW influent is the Lake Rockwell
reservoir, located upstream of the Akron POTW.
The 7Q10 value is 1.27 m3/s and the 1Q10 value is 0.87
m3/s for the Cuyahoga River (21). The "effluent flow
which could cause the greatest impact" is recommend-
ed to be used by U.S. EPA (30). The Akron POTW has its
NPDES permit limits for conventional pollutants based
on an effluent flow of 3.94 m3/s. In contrast to the worst-
case conditions which are examined using the design
flows, the mean effluent discharge was 2.79 m3/s during
the site visit when chronic toxicity tests were conducted
(Table 5-7). In addition, acute toxicity data were generat-
ed for the TSI during 1986 when the mean discharge was
3.22 m3/s (Table 6-1). When not accounting for the mix-
ing zone, a dilution factor of 1.22 for the CMC and a dilu-
tion factor of 1.32 for the CCC would be used for the cal-
culations of acceptable effluent toxicity. With the mixing
zone, the available receiving water for dilution is re-
duced and the dilutions are 1.07 for the CMC and 1.11 for
the CCC. Further calculations are based on the dilutions
without the mixing zone following Ohio EPA policy, and
solutions have been rounded.
Dilution With No Mixing Zone
3.94 m3/s + 0.87 m3/s
CMC:
CCC:
3.94 m-Vs
3.94 m3/s + 1.27m3/s
3.94 m3/s
= 1.22
Dilution With the Ohio Mixing Zone
CMC.
3.94 m3/s
1/3 (0.87 m3/s)
u/
3.94 m3/s
40
-------
CONCLUSIONS
3.94 m3/s + 1/3(1.27m3/s) _ 1 „
3.94 m3/s ;
The solution for Equation 7-4 for WLAa is <0.366 TUa ,-
andforWLAcis<1.32TUc. :
WLAa: 1.22 x 0.3 TUa = 0.366
WLAC: 1.32x1.0TUc =1.32 !
The WLAa is converted to a chronic measure by multi-
plying by the ACR value. Using a discharger-specific
ACR is the first opportunity to use toxicity data in the cal-
culation of permit limits. :
7.3.3 Permit Limit Calculations
The chronic and acute wasteload calculations corre-
spond to two criteria which are used as the basis for de-:
termining long-term averages (LTAs). An LTA is a mean
effluent value that will result in an acceptable record of :
compliance with water quality standards and it is based,
in part, on past effluent performance information such
as variablity. U.S. EPA has set probability estimates for
these levels at 99 percent which means that statistically,:
there will be one time in a hundred that the level will be,
exceeded. The one-hour LTA corresponds to acute expo-
sures and the four-day LTA corresponds to chronic ex-
posures.
1-hour LTA = e(lJ- + °'5 ff2) (Equation 7-5);
where \
(JL = ln(WLAa)-z + za> (Equation 7-8)
where
z = 1 .645 for the 95th percentile occurrence probability
(JL = In LTA-0.50-2
o-2 = ln(CV2 + 1)
Average Monthly Limit = e'1*" + ZCTn) Equation 7-9)
where
. (o-2-o-n2)
(An = (J, + - ~-
41
-------
CONCLUSIONS
n = number of monthly effluent samples
The maximum daily limit result using the LTA derived
from the U.S. EPA values for CV and ACR indicated that
the maximum daily limit is 1.25 TUC. The average
monthly limit using the same LTA indicated that, for the
typical case of four effluent samples per month, the
chronic exposure limit is 0.91 TUC.
Maximum Daily Limit = e'"0'510 + 1-645x°-554> = 1.49 TU0
where
p. = In 0.70 - 0.5 x 0.307 = -0.510
Average Monthly Limit = e1"0'400 + ^x0-293' = 1.08TUC
where . .
H = -0.510 +(0-307-0.086), .Q ,
Translated into percent effluent values, these results re-
quire that the mean daily limit of toxicity is >67 percent
effluent (NOEL) and that the mean monthly limit of toxic-
ity is >92 percent effluent (NOEL). Permit compliance
should ensure that acute toxicity does not occur within
the mixing zone and that chronic toxicity does not occur
beyond the mixing zone. During the on-site chronic tox-
icity testing, ambient water from Station 4 located
downstream of the mixing zone resulted m Ceriodaph-
nia survival of 40 percent (Table 5-9). The ambient toxic-
ity data indicates that the Akron POTW is not in compli-
ance with these sample permit limits especially since the
ambient sample was not collected at the design low
flow, so that there was more available dilution than at
7Q10or1Q10flows.
Permit limit calculations can be performed without
using assumed values for the CV or ACR when sufficient
data has been collected. The CV of the acute LC50 data is
0.2 and the CV of the chronic NOEL data is 0.7 forCerio-
daphnia survival (Tables 5-7 and 6-1). The chronic CV
was determined using the Ceriodaphnia survival data
only. For simplicity of calculation, and since neither the
sampling for the acute and chronic testing was conduct-
ed simultaneously nor was the purpose of the testing for
permitting, a mean CV of 0.45 was used. Similarly, for
the determination of the ACR, the ratio of mean values
for the acute (LC50s) and chronic (NOELs for survival)
data was used. The ACR is 4 using this procedure. With
these two new values Equations 7-4 et seq. can be reeva-
luated. The acute LTA is 0.59 TUC and the chronic LTA is
0,86 TUC.
1-hour LTA = e(-°'618 + °'5*0-1844' = 0.59 TUC
where
ix = In (0.366 x 4) - 2^326 x 0.4294 = -0.618
, a =Vln[(0.45)2 + 1] = 0.4294
i 4-day LTA = e'"0'239 + °'5 x °'1844) = 0.86 TUC
where
= In (1.32)-2.326 \/In 1
01844
= -0.239
This four-day LTA value is converted using Equation 7-7
;to obtain a daily value for comparison to the acute LTA.
When the chronic LTA is converted, the 0.86 TUC be-
comes 0.81 TUC.
LTAC
where
= e
(-0.306 + 0.5x0.1844)
= 0.81
r / vi
i |x = -0.239-(0.5x0.1844) + 0.5 x In 1 +/eai844-1\
= -0.306
.The limiting LTA of 0.59 TUC was derived from the acute
LTA using site-specific data, in contrast to the previous
limit calculations which were controlled by the chronic
LTA. When the limiting site-specific LTA value was used
in the limit derivation process, the maximum daily limit
is 1.09 TUC and the average monthly limit is 0.83 TUC.
, Maximum daily limit = e(-°-619 + 1-645x0-4294' = 1.09 TUC
where
M, = In 0.59 - 0.5 x 0.1844 = -0.619
Average monthly limit = e''0'551 + 1-645xa222)=0.83 TUC
where
= .Q_619 +(0.1844-0.0494\ _Q^
0.1844
= -0.0494
Translated into percent effluent values, these results re-
quire that the mean daily limit of toxicity is >91 percent
effluent (NOEL) and that the mean monthly limit of toxic-
ity is >100 percent effluent (NOEL). As mentioned pre-
viously, Ceriodaphnia survival was 40 percent at Station
'4 which is located downstream of the discharge and be-
yond the boundary of the mixing zone. Therefore, the
42
-------
CONCLUSIONS
Akron POTW is also not in compliance with the maxi-
mum daily or average monthly limit calculated with
Akron POTW-specific data.
The abrupt loss of acute toxicity in June 1986 was quite
welcome, but did prevent further investigation of the
source and identification of the toxicant(s). The last four
months of acute toxicity testing for the TSI resulted in
Ceriodaphnia LC50 values of >100 percent effluent. If :
these tests had been conducted for monitoring pur-
poses, the results would have been used to determine
compliance. However, dC50 values are not directly com-
parable with permit limits which are expressed in NOELs
and must be converted |io a chronic measure using the
ACR. The converted LC50 represents a NOEL of 10 per-
cent effluent using the iLl.S. EPA value of 10 for the ACR
and a NOEL of 25 percent effluent using the Akron-spe-
cific ACR of 4. |
i
Comparing these two NpEL values to the permit limits
indicates that the Akron1 POTW is not in compliance
since the limits require NOELs >67 percent effluent.
However, an LC50 value above 100 percent cannot be
determined with present test procedures which use
whole effluent. In orderto meet the sample permit lim-
its, the LC50 values would need to be >670 to 1,000 per-
cent effluent. Procedures to concentrate complex ef-
fluents, which would nqt introduce artifacts into the
original effluent, are now being developed. U.S. EPA
(30) has acknowledged that the detection limit is curent-
ly 100 percent effluent, and has recommended monitor-
ing tests (acute or chronic be selected so that results
would be within detection limits). Because of the low di-
lution available to the Akron POTW effluent, the sample
maximum daily limit and average monthly limit values
require NOEL values which are beyond the level of de-
tection using acute tests. Chronic toxicity tests, which di-
rectly yield a NOEL, would be appropriate for compli- !
ance monitoring.
7.4 Conclusions
Water quality-based control procedures were used at
the Akron POTW to indicate the presence of toxicants in
the effluent, determine the ambient toxicity, seek meth-
ods to reduce toxicity levels, and then to calculate sam-
ple permit limits. Examination of the results of this case
study illustrates that:
1. Tier 1 screening indicated that the Akron POTW
was a priority candidate for toxicity testing.
2. Tier 2 definitive data generation indicated that
there was ambient toxicity and variable whole ef-
fluent toxicity downstream of the Akron POTW
which abruptly ended in the late spring of 1986.
3. Uncertainty factor calculations indicated that it was
not possible for the Akron POTW to satisfy Equa-
tion 7-1, so that further testing to reduce the uncer-
tainty factors was not necessary and calculation of
permit limits may be made based on available data.
4. Ceriodaphnia was the most sensitive test organ-
ism. !
5. Sample permit limits were derived for two cases:
one with values for CV and ACR set by U.S. EPA,
and one with site-specific data from the Akron
POTW for these parameters. Site-specific values
are more appropriate, but can only be used when
data are available (in this case research data was '
used). The limits derived using U.S. EPA's values
were based on chronic LTA performance levels
while the limits derived using the Akron-specific
data were based on acute LTA performance levels.
6. Calculated sample permit limits require NOELs >67
percent effluent. In order for the Akron POTW to
comply with these limits, their effluent cannot
show any measurable acute toxicity.
The Akron POTW has improved its treatment facilities
and processes. New sludge handling facilities have been
constructed and have begun operation for the compost-
ing of this wastewater treatment byproduct. This re-
duces the need for sludge storage at the Akron POTW
and the volume of material which may contain toxicants
for recirculation through the clarifiers.The frequency of
bypassing untreated or partially treated wastewater has
also been reduced as a result of new treatment systems.
In conjunction with the toxicity testing conducted by U.S.
EPA, biological surveys have been conducted annually
by Ohio EPA since 1984. Results of those and other sur-
veys indicate that fish and invertebrate communities in
the Cuyahoga River have been gradually improving, but
an impacted zone still remains downstream of the Akron
POTW even after effluent acute toxicity disappeared.
The disappearance of the effluent acute toxicity is not
known to be permanent since toxicity testing of the ef-
fluent was not performed after September 1986. Results
from future effluent toxicity testing and surveys of the
biota in the Cuyahoga River will detemine whether the
toxicity disappearance has been sustained and the effect
of its absence on the biota. The control of effluent toxic-
ity does not eliminate the potential for other sources of
43
-------
CONCLUSIONS
toxicants to cause adverse biological effects. These
areas were beyond the scope of this study. Traditional
measures of water quality indicate that the river should
be capable of supporting healthy and abundant popula-
tions representative of'a warm water river. Ohio EPA is
continuing the chemical and biological monitoring sur-
veys in the Cuyahoga River to document any improve-
ments. U.S. EPA will also continue monitoring thetoxic-
ity of the effluent.
44
-------
8. References
(1) U.S. EPA. 1976. Quality Criteria for Water. EPA-440/9-76-
023. Washington, D.C.
(2) Ohio EPA. 1984. Biannual 305(b) Report to the U.S. Con-
gress. State of Ohio. Vol. 4. Subbasin Report on the Status
of the Nation's Waters. Submitted to EPA Region V.
(3) Department of Health. 1960. Report of Water Pollution
Study of Cuyahoga River Basin. State of Ohio, Division of
Engineering. :
(4) Simpson, G.D., LW. Curtis, and H.K. Moukle. 1969. The
Cuyahoga River, Lake Rockwell to Lake Erie. In the Cuya- ;
hoga River watershed pp. 87-120. Proceedings of a Sym- i
posium held at Kent State Unversity, Kent, Ohio. Novem- '
ber 1,1968. Department of Biological Sciences, Kent State
University, Kent, Ohio. '
(5) U.S. Geological Survey (USGS). 1971. Water Resources ,
Data-Ohio Water Year 1971. USGS Water Data-Report '.
71-2. ;
(6) USGS. 1980. Water Resources Data-Ohio Water Year !
1980. USGS Water Data Report OH-80-2.
(7) USGS. 1985a. Water Resources Data-Ohio Water Year |
1984. Vol. 2. St. Lawrence River Basin. USGS Water Data ;
Report OH-84-1.
(8) Ohio EPA. 1984. Unpublished survey data from Cuyahoga
River. ;
(9) Ohio EPA. 1985. Unpublished survey data from Cuyahoga '
River.
(10) Olive, J.H. 1976. Chemical-physical and biological assess- ;
ment of water quality in the Cuyahoga River (1973-1974). :
OhioJ.Sci. 76:5-15.
(11) Trauben, B.K. and J.H. Olive. 1983. Benthic macroinverte-
brate assessment of water quality in the Cuyahoga River,
Ohio—An Update. Ohio J. Sci. 83:209-212.
(12) White, A. 1985. Letter to D. Dudley of Ohio EPA. John Car- -
roll University, Ohio. Personal communication. August 9.
(13) Dudley, D. 1986. Ohio EPA, Division of Water Quality. Per-;
sonal communication.
(14) Ohio EPA. 1986. Status Report: Chemical and Biological <
Quality of the Cuyahoga River. Division of Water Quality
Monitoring and Assessment. December 30.
(15) Neiheisel,T. 1985. EPA Office of Research and Develop- ;
ment. Newtown Laboratory. Newtown, OH. Personal
communication. ,
(16) Ehreth, D.J. and D.F. Bishop. 1985. Toxicity reduction
evaluation in municipal wastewater treatment. Presented
at Joint US/USSR Work Group for Prevention of Water '
Pollution from Industrial and Municipal Sources. :
(17) Ohio EPA. 1987. Unpublished data from Ray Beaumier of
the Division of Water Quality Monitoring and Assess-
ment.
(18) Monteith, R.A. 1986. Manager, Water Pollution Control Di-
vision, City of Akron, Ohio. Personal communication.
(19) Ohio EPA. 1985. Unpublished data from Robert Wysenski
of the Northeast District Office.
(20) USGS. 1984. Literature Review and Need for Additional
Study of Surface-Water Quality in the Cuyahoga Valley
National Recreation Area, Ohio. Open-File Report 84-619.
Prepared in cooperation with the National Park Service.
(21) Ohio Environmental Protection Agency. 1980. Cuyahoga
and Chagrin River basin, subbasin summary: 305(b) re- ;
port. Vol. IV. pp. 43-54.
(22) USGS. 1981. Low-Flow Characteristics of Ohio Streams.
Open-File Report 81-1195. Prepared in cooperation with '.
the Ohio Environmental Protection Agency.
(23) U.S. Environmental Protection Agency (U.S. EPA). 1981.:
Compliance Sampling Inspection Field Report—Akron
WWTP, Akron, Ohio. EPA Region V Environmental Ser-
vices Division, Eastern District Office, Westlake, Ohio. Oc-
tober 20.
(24) U.S. EPA. 1985. Technical Support Document for Water
Quality- Based Toxics Control. Office of Water. Washing-
ton, D.C.
(25) Mount, D.I. and T.J. Norberg. 1984. A seven-day life-cycle
cladoceran toxicity test. Environ. Toxicol. Chem. 3:425-
434. :
(26) Norberg, T.J. and Mount D.I. 1985. A new fathead min- ,
now (Pimephales promelas) sub-chronic toxicity test. En-
viron, toxicol. Chem. 4(5).
(27) Hamilton, M.A. 1984. Statistical Analysis of the Seven- '
Day Ceriodaphnia reticulata Reproductivity Toxicity Test.
EPA Contract J 3905NASX-1. '
(28) Sokal, R.R. and FJ. Rohlf. 1981. Biometry. W.H. Freeman
and Company, New York.
(29) Mount, D.I., T.J. Norberg-King, and A.E. Steen. 1986. Va-
lidity of Effluent and Ambient Toxicity Tests for Predicting
Biological Impact, Naugatuck River, Waterbury, Connecti-
cut. EPA-600/8-86/005. U.S. Environmental Protection
Agency.
(30) U.S. EPA. 1986. Permit Writer's Guide to Water Quality-
Based Permitting for Toxic Pollutants. Office of Water.
Washingtn, D.C
(31) Mount, D.I. 1985. EPA Environmental Research Laborato-
ry, Duluth, Minnesota. Personal communication.
(32) Mount, D., N. Thomas, M. Barbour, T. Norberg, T. Roush,
and R. Brandes. 1984. Effluent and Ambient Toxicity Test-
ing and Instream Community Response on the Ottawa
River, Lima, Ohio. EPA-600/2-84-044. Permits Division,
Washington, D.C., Office of Research and Development,
Duluth, Minnesota.
45
-------
(33) Mount, D.I. and L Anderson-Carnahan. Draft 1987. Meth-
ods for Toxicity Reduction Evaluations. Phase I -Toxicity
Characterization Procedures. January. EPA/600. U.S. EPA,
Washington. June.
(34) Peltier, W. and C.I. Weber. 1985. Methods for Measuring
the Acute Toxicity of Effluents to Aquatic Organisms.
Third edition. EPA-600/4-85-013. Office of Research and
Development, Cincinnati, Ohio.
(35) Mount, D.I. 1987. EPA Environmental Research Laborato-
ry, Duluth, Minnesota. Personal communication.
(36) Verschuren, Karel. 1983. Handbook of Environmental
Data on Organic Chemicals. Second Edition. Van Nos-
trand Reinhold Company, New York.
(37) Mosser, J. 1987. City of Akron Wastewater Treatment
Plant. Personal communication.
(38) Carlson, A.R. and T.H. Roush. 1985. Site-specific water
quality studies of the Straight River, Minnesota: Complex
effluent toxicity, zinc toxicity, and biological survey rela-
tionships. EPA-600/3-85-005. National Technical Informa-
tion Service, Springfield, Virginia.
(39) Dudley, D. 1987. Ohio EPA, Division of Water Quality. Per-
sonal communication.
8.1 Complex Effluent Toxicity Testing Program
Reports
Mount, D.I., N.Thomas, M. Barbour,T. Norberg, T. Roush, and
R. Brandes 1984. Effluent and Ambient Toxicity Testing and In-
stream Community Response on the Ottawa River, Lima, Ohio.
EPA-600/2-84-044. Permits Division, Washington, D.C., Office
of Research and Development, Duluth, Minnesota.
Mount, D.I., and T.J. Norberg-King, eds. 1985. Validity of
Effluent and Ambient Toxicity Tests for edicting Biologi-
cal Impact, Scippo Creek, Circleville, Ohio. EPA-600/3-85-
044. U.S. Environmental Protection Agency.
Mount, D.I., T.J. Norberg-King, and A.E. Steen, eds.
1985. Validity of Effluent and Ambient Toxicity Tests for
Predicting Biological Impact, Five Mile Creek, Birming-
ham, Alabama. EPA-600/8-85-015. U.S. Environmental
Protection Agency.
Mount, D.I., A.E. Steen, and T.J. Norberg-King, eds.
1986. Validity of Effluent and Ambient Toxicity Tests for
Predicting Biological Impact, Ohio River, Wheeling, West
Virginia. EPA-600/3-85/071. U.S. Environmental Protec-
tion Agency.
Mount, D.I., A.E. Steen, and T.J. Norberg-King, eds.
1986. Validity of Effluent and Ambient Toxicity Tests for
Predicting Biological Impact, Back River, Baltimore Har-
bor, Maryland. EPA-600/7-86/001. U.S. Environmental
Protection Agency.
Mount, D.I., A.E. Steen, and T.J. Norberg-King, eds.
1986. Validity of Effluent and Ambient Toxicity Tests for
Predicting Biological Impact, Naugatuck River, Water-
bury, Connecticut. EPA-600/8-86/005. U.S. Environmen-
tal Protection Agency.
Mount, D.I. and T.J. Norberg-King, eds. 1986. Validity of
Effluent and Ambient Toxicity Tests for Predicting Bio-
logical Impact, Kanawha River, Charleston, West Virgin-
ia. EPA-600/3-86/006. U.S. Environmental Protection
i Agency.
Norberg-King, T.J. and D.I. Mount, eds. 1986. Validity of
Effluent and Ambient Toxicity Tests for Predicting Bio-
logical Impact, Skeleton Creek, Enid, Oklahoma. EPA-
600/8-86/002. U.S. Environmental Protection Agency.
46
-------
Glossary
acute - involving a stimulus severe enough to rapidly in-
duce a response; in toxicity tests, a response observed
in 96 hours or less typically is considered an acute one.
An acute effect is not always measured in terms of letha-
lity; it can measure a variety of effects. Note that acute
means short, not mortality.
acute-chronic ratio (ACR) - the ratio of the acute toxicity
(expressed as an LC50) of an effluent or a toxicant to its
chronic toxicity (expressed as an NOEL). Used as a fac-
tor for estimating chronic toxicity on the basis of acute
toxicity data.
additivity - the characteristic property of a mixture of
toxicants which exhibits a cumulative toxic effect equal
to the arithmetic sum of the effects of the individual toxi-
cants.
ambient toxicity - toxicity manifested by a sample col-
lected from an aquatic receiving system.
anoxic - without oxygen.
antagonism - the characteristic property of a mixture of
toxicants which exhibits a less than additive cumulative
toxic effect.
bioavailability - the property of a toxicant that governs
its effect on exposed organisms. A reduced bioavailabi-
lity would have a reduced toxic effect.
chronic - involving a stimulus that lingers or continues
for a relatively long period of time, often 1/10 the life
span or more. Chronic should be considered a relative
term depending on the life span of an organism. A
chronic effect can be lethality, growth, reduced repro-
duction, etc. Chronic means long.
conservative pollutant - a pollutant that is persistent and
not subject to decay or transformation.
continuous stimulation model - a fate and transport
model that uses timeseries input data to predict receiv-
ing water quality concentrations in the same chronologi-
cal order as the input variables.
Criteria Continuous Concentration (CCC) - the U.S. EPA
national water quality criteria recommendation for the
highest instream concentration of a toxicant or an efflu-
ent to which organisms can be exposed indefinitely
without causing unacceptable effect.
Criteria Maximum Concentration (CMC) - the U.S. EPA
national water quality criteria recommendation for the
highest instream concentration of a toxicant or an efflu-
ent to which organisms can be exposed for a brief peri-
od of time without causing mortality.
critical life stage - the period of time in an organism's life
span when it is the most susceptible to adverse effect ,
caused by exposure to toxicants, usually during early
development (egg, embryo, larvae). Chronic toxicity
tests are often run on critical life stages to replace long
] duration, life cycle tests since the toxic effect occur dur-
ing the critical life stage.
' design flow - the critical flow used for steady-state was-
teload allocation modeling.
diversity - the number and abundances of species in a
specified location.
i duration - the period of time over which the instream
! concentration is averaged for comparison with criteria
1 concentrations. This specification limits the duration of
concentrations above the criteria.
effluent biomonitoring -the measurement of the bio-
' logical effects of effluents (such as toxicity, biostimula-
tion, and bioaccumulation).
', LC50- the toxicant concentration killing 50 percent of ex-
posed organisms at a specific time of observation.
• lognormal probabilistic dilution model - a dilution mod-
el that calculate the probability distribution of receiving
: water quality concentrations from the lognormal prob-
ability distributions of the input Variables.
, magnitude - how much of a pollutant (or pollutant pa-
rameter such as toxicity), expressed as a concentration
or toxic unit, is allowable.
No Observed Effect Level (NOEL) - the highest measured
continuous concentration of an effluent or a toxicant
j which causes no observed effect on a test organism.
1Q10 - the discharge at the 10-year recurrence interval
! taken from the frequency curve at annual values of the
i lowest mean daily discharge.
, permit averaging period - the duration of time over
. which a permit limit is calculated—day(s), week, or
month.
47
-------
persistence - that property of a toxicant or an effluent
which is a measurement of the duration of its effect. A
persistent toxicant or toxicity maintains effect after mix-
ing, degrading slowly. A nonpersistent toxicant or toxic-
ity may have a quickly reduced effect after mixing, as
degradation processes such as volatilization, photolysis,
etc, transform the chemical.
lug flow sampling - a monitoring procedure that fol-
ows the same slug of wastewater throughout its trans-
port in the receiving water. Water quality samples are
collected at receiving water stations, tributary inflows,
and point source discharges only when a dye slug or
tracer passes that point.
probability- a number expressing the likelihood of oc-
currence of a specific event, such as the ratio of the num-
ber of outcomes that will produce a given event to the
total number of possible outcomes.
7Q10 - the discharge at the 10-year recurrence interval
taken from a frequency curve of annual values of the
lowest mean discharge for seven consecutive days.
steady-state model - a fate and transport model that
uses constant value of input variables to predict con-
stant values of receiving water quality concentrations.
sublethal - involving stimulus below the level that
causes death.
synergism - the characteristic property of a mixture of
toxicants which exhibits a greater than additive cumula-
tive toxic effect.
total maximum daily load (TMDL) - the total allowable
pollutant load to a receiving water such that any addi-
tional loading will produce a violation of water quality
standards.
toxic unit acute (TUa) - the reciprocal of the effluent dilu-
tion that causes 50 percent of the test organisms to die
by the end of the acute exposure period.
toxic unit chronic (TU3) -the reciprocal of the effluent di-
lution that causes no unacceptable effect on the test or-
ganisms by the end of the chronic exposure period.
uncertainty factors - factors used in the adjustment of
toxicity data to account for unknown variations. Where
toxicity is measured on only one test species, other spe-
cies may exhibit more sensitivity to that effluent. An un-
certainty factor would adjust measured toxicity upward
and downward to cover the sensitivity range of other,
potentially more or less sensitive species.
wasteload allocation (WLA) - the portion of a receiving
water's total maximum daily pollutant load that is allo-
cated to one of its existing or future point sources of pol-
. lution.
whole effluent toxicity - the aggregate toxic effect of an
effluent measured dirctly with a toxicity test.
48
------- |