Executive
Summary
The U.S. EPA Great Lakes National Program Office funded Wright State
University to assess the sediment quality in the lower Black River in collaboration with
the Ohio EPA and Dr. Paul Bauman (USGS). The one year study was designed to assess the
effectiveness of the previous remedial dredging activity which was aimed at removing
sediments contaminated with polycyclic aromatic hydrocarbons. Dredging occurred in the
late 1989 and early 1990 below the Kobe Steel outfalls at river miles (RM) 2.83 to 3.55.
The U.S. Army Corps of Engineers (COE) also routinely does maintenance dredging from the
mouth of the river up to RM 2.5 in the Turning Basin. This dredging occurs only in the
main channel of the river. Wright State University focused on measuring the toxicity of
the sediments and overlying waters in the lower 5 miles of the river and comparing those
findings to upstream reference stations. Depositional sediments were sampled near the
river banks and outside of COE dredging areas. Surficial and deeper buried sediments were
analyzed to determine contaminant gradients exist in the river. Toxicity testing included
both laboratory and in situ (field) exposures of four aquatic species: Pimephales
promelas (fathead minnow), Ceriodaphnia dubia (water flea), Hyalella
azteca (amphipod), and Chironomus tentans (midge).
The Fall 1997 a survey was conducted during base flow conditions.
The survey revealed a wide range of toxicity existing in the sediments and/or overlying
waters. The highest levels of sediment toxicity noted in laboratory exposures occurred at
River Miles (RM) 15.0, 2.9, 2.5, 0.5 and 0.3. The East Branch of the Black River RM 18.9,
and the lower Black River at RM 11.6, 5.2, 4.6, and 2.4 showed little to no mortality.
Growth of the amphipod and midge in the upper reference site was good. For the amphipod
growth was lowest at RM 0.3, 2.4, and 5.2; while for the midge, growth was lowest at RM
0.3, 2.5, and 9.8. Surficial sediments tended to be less toxic (survival and growth
(amphipod only)) than deeper, more historical sediments in most cases. In situ
toxicity testing allowed for more realistic exposures to both sediments and overlying
waters and showed better survival of organisms overall. However, high mortality of some
test organisms was observed at RM 2.4 and 0.3. Initial findings indicated that
photo-induced toxicity from polycyclic aromatic hydrocarbons may be a factor at some of
these sites, as there was lower survival in near-surface water exposures.
A survey of the indigenous snail, Physella gyrina, in the
study area showed genetic patterns indicative of stress at 2 locations. When the DNA
pattern of individuals with a population becomes similar, then it suggests the population
has been adversely impacted and is less diverse. This loss of genetic diversity can equate
to greater susceptibility to stress and general population decline. At RM 5.2 above Kobe
and French Creek and in Kline Ditch (a tributary of French Creek) a high degree of genetic
similarity was observed. The Kline Ditch area has had water quality problems attributed to
nearby fly ash disposal sites (Ohio EPA, personal communication). In addition, there was a
significant relationship in the Black River study area between declining fish health (IBI
scores) and snail genetic patterns.
In the Spring of 1998, in situ exposures of test organisms during a
high flow event showed little to no acute toxicity existing at most test sites. This
survey did not focus on sediments, rather organisms were only exposed to near surface
waters during very turbid conditions. Photo-induced toxicity due to PAHs is not a factor
when turbidity is high (Ireland et al. 1996). Therefore, the greater toxicity observed at
base flow conditions, when turbidity is lower, appears to be a PAH effect. This effect can
occur at PAH concentrations at the low to sub microgram per liter level.
The water column acute toxicity at base flow and the sediment acute
toxicity (survival and/or growth) observed in the downstream areas suggest the PAHs may
still be a primary stressor in the lower Black River. Sediments that were several
centimeters deep tended to be more toxic than surficial sediments and may be exposed
during resuspension events (e.g., storms, boat traffic, dredging). However, chemical
analyses of sediments did not show elevated levels of PAHs. Total sediment metal
concentrations were elevated and tended to be higher downstream. Acute toxicity during
high flow suggests that the impacts of nonpoint source runoff and stormwater inputs are
less severe. However, since only acute toxicity was measured, it is unknown whether
chronic toxicity may exist due to nonpoint source runoff.
Introduction
The objectives of the 1992 Great Lakes 5 Year Strategy are unlikely
to be met in the Black River Area of Concern (AOC) unless there is a thorough evaluation
of what role contaminated sediments are having on the system. The Black River Remedial
Action Plan (RAP) Strategy Plan (1997 - 2001) and 1997 Draft Annual Plan state that
contaminated sediments and related toxicity are a significant issue in the system, yet no
studies have been developed for assessing sediment toxicity. The Black River’s
infamous history of severe sediment contamination from polycyclic aromatic hydrocarbon
(PAH) below USS/Kobe Steel and associated impacts on the fish community, the land use and
hydrologic characteristics which result in extensive sediment loading and deposition, and
past and present pollution sources clearly suggest sediments may, in fact, be the dominant
stressor in the system. Controls are being implemented (through the RAP process) to reduce
sediment and contaminant loading. However, the effectiveness of any controls or past
dredging, the contribution of existing sediment contamination to use impairment, or
documentation for use impairment delisiting, cannot be accurately assessed without a
comprehensive assessment of sediment toxicity.
This project was designed to be an integral component of the RAP
Strategic Plan, and as such, effectively promotes U.S. EPA project criteria of helping the
multiple stakeholders involved in the clean-up of this complex AOC. A comprehensive
assessment of contaminated sediments will assist the RAP process in identification of
significant pollution sources, effective remedial designs, and education and outreach
program directions. The current project integrates with the 5 year goals of the RAP in the
4 primary areas of Partnership, Stewardship, Resource, and Habitat.
The Black River basin in north central Ohio has multiple land uses,
all of which have components that are causing beneficial use impairments. Significant
amounts of pollution continue to enter the Black River and its major tributaries from
numerous sources, including: urban and suburban storm water runoff; construction site,
river bank, feedlot and agricultural runoff; septic systems; combined and separate sewer
overflows; and inadequate point source treatment (Ohio EPA 1994). In addition, historical
landfills, municipal wastewater treatment plants, and other unknown sources appear to be
contributing toxicants and responsible for non-attainment in some stream segments. All
these pollution sources have contributed, in varying degrees, to seven beneficial use
impairments. Sediments which were severely contaminated with PAHs were dredged in the
lower mainstem in 1990. Recent studies by Dr. Paul Baumann (U.S. Geological Survey/Ohio
St. Univ. (USGS/OSU)) show that the incidence of fish tumors may be improving; however,
fish are still adversely affected. Further documentation of improvement is required for
cancellation of the fish advisory and delisting of the use impairment. However, it has
been noted that significant PAH loadings are continuing from a variety of urban sources.
This on-going contamination may prevent any further improvements (and subsequent
delisting), as new surficial sediments will continue to be contaminated.
Our understanding of sediment contamination and how to properly
assess its ecological significance has greatly improved in the past few years. Half of
beneficial uses which are impaired in the Black River AOC are likely linked to sediment
contamination. For example, the contamination of fish and associated advisories, the
impaired benthic biological communities, and impacts on fishing are strongly associated
with contact and bioaccumulation of sediment contaminants. In order to delist the
beneficial use impairments, it will be necessary to use a suite of environmental
indicators/criteria to track improvement in the system contamination. This study sought to
assess the apparent role of contaminated sediments on beneficial use impairment. The
assessment considered the following: the severity and spatial extent of sediment
contamination, its ecological and human health significance, and the sources of and their
relative contribution to the sediment contamination.
Project Description
The project utilized a suite of environmental indicators to define
the degree and spatial extent of sediment and runoff contamination in the mainstem of the
Black River. Since ecologically significant chemical concentrations in waters and
sediments have often been found to be site-specific, and unrelated to existing criteria
guidelines, an integrated assessment was conducted in a "weight-of-evidence"
approach. The environmental indicators used in this integrated approach will include: 1)
water (low flow and first flush) and sediment contaminant concentrations (using criteria
or guidelines) (USGS/OSU & OEPA); 2) resident benthic macroinvertebrate populations
(using multiple indices) (OEPA); 3) fish tissue residue analysis of target contaminants
(using advisory levels) (WSU, OEPA, USGS/OSU); 4) presence of fish tumors or other
deformities (OEPA and USGS/OSU); 4) sediment toxicity testing (using U.S. EPA standard
methods for Hyalella azteca and Chironomus tentans) (WSU); 5) In
situ toxicity testing (using H. azteca, C. tentans, Ceriodaphnia
dubia, and Pimephales promelas) during dry and wet weather conditions (to
determine the temporal impacts from runoff) (WSU); and 6) genotoxicity effects in benthic
macroinvertebrate and fish species (using the RAPD assay) (WSU). In situ toxicity testing
allowed differentiation of PAH-related photo-induced toxicity. The following report
documents those portions of the study conducted by WSU.
Methods
Sampling site locations focused on establishing whether the Black
River mouth has improved in quality since the dredging which took place below the Kobe
outfalls and determine if improvements are occurring for beneficial use delisting.
Sediment contamination and toxicity was characterized as recent (surficial sediments) or
historical (deep sediments) to provide crucial information for remedial design strategies,
(e.g., dredging scenarios, management of on-going CSO and/or urban runoff, and risk from
resuspension of historical contamination. The test area extended from the lower end of the
Black River to the confluence with the East and West Branches with upstream reference
stations (Table 1,
Figure
1). Sediment and water sampling was conducted in coordination with the OEPA Intensive
Survey (1997 Study Plan for the Black River Basin, Draft, June 6, 1997), the RAP Process,
and Dr. Paul Baumann (USGS/OSU).
Sediment cores were collected from a minimum of 12 locations within
the study area by WSU, Dr. Baumann, and the OEPA (Table 1).
All sediment collection for WSU, Dr. Baumann, and the OEPA was conducted simultaneously by
OEPA, U.S. EPA and WSU using U.S. EPA draft protocols (1997) and ASTM (1994) protocols.
For this proposal, toxicity testing was conducted on upper (recent) and lower (historical)
layers of sediment. The upper 2 centimeters of the sediment core will be removed and
analyzed for PAHs and the remaining core frozen for possible future analysis. Two core
depths were tested (0 - 2 and 8 - 10 centimeters) by placing into 300 mL beakers for 10
day laboratory toxicity exposures (H. azteca and C. tentans
simultaneously following modified U.S. EPA 1994).
Sediments for toxicity testing were collected with a Ponar dredge
from the same location and at the same time as core samples were collected. The Ponar
collects from 2 to 10 cm of the upper sediment, depending on the nature of the sediment.
These sediments were homogenized (by hand mixing), placed in high density polyethylene
bottles and stored on ice until refrigeration. Sediment sampling occurred on one occasion
during the October 1997 test period.
Field (in situ) toxicity studies followed previous
published protocols of the PI (Chappie and Burton 1996, Ireland et al 1996, Burton et al.
1996) testing H. azteca, C. tentans, C. dubia, and P.
promelas at base and high flow conditions to evaluate the relative contribution of
toxicity from continuous (e.g., bedded sediments, WWTP effluents) vs. non-continuous
(e.g., stormwater runoff) sources. Briefly, triplicate chambers (polyethylene core tubes
with 74 micron nylon mesh windows) were deployed containing the early life stages of the
test organisms (10 organisms/chamber) and exposed for 2 to 7 day periods. Toxicity was
partitioned into overlying water only exposures and also sediment exposures by altering
the design and placement of the exposure chambers. In addition, the effect of PAH
contamination was assessed by focusing on photo-induced toxicity (which results from PAH
interaction with ultraviolet light) using light and dark exposure chambers. Exposures were
conducted in the Fall of 1997 and the late Spring of 1998 to assess seasonal variation.
The project and all procedures followed accepted quality assurance
and quality control guidelines. These are fully described in a separate quality assurance
project plan (QAPP).
Laboratory
Toxicity Test Methods: Fall 1997
Sediment toxicity was measured using the U.S. Environmental
Protection Agency methods (USEPA 1994). The 10 day growth and survival bioassays using the
amphipod Hyalella azteca and the midge Chironomus tentans were conducted
on 18 field sediments.
For the Fall 1997 test period, sediment samples were collected on
September 30 and October 1, 1997. The Grafton Road sediment (upstream reference) was
collected on October 2, 1997. Sediment samples were collected using an Ekman dredge and
placed into a plastic pan. The upper 0 to 2 cm layer was gently scrapped using a spatula
and placed into a labeled sample bottle (high density polyethylene). The lower 8 to 10 cm
layer was also subsampled and placed into a separate labeled sample bottle. These same
sediment samples were also subsampled for chemical analyses by Dr. Paul Bauman, USGS,
Columbus, Ohio. Sample bottles were immediately placed into ice-filled coolers and
transported back to the laboratory for refrigeration.
Toxicity testing was initiated on October 31, 1997 using 7 to 14 day
old H. azteca and 8 to 12 day old C. tentans. At test initiation,
sediment samples were thoroughly homogenized with any overlying water that had separated
out during storage. Sediments were distributed (100 mL) to 300 mL glass beakers (4
replicates). Overlying water (175 mL) of culture water (hardness ~130 mg/L CaCO3) was
gently added to each beaker containing sediment and allowed to settle overnight prior to
test organism addition. Laboratory controls consisted of culture water with synthetic mesh
or sand for the amphipods and midge, respectively. An additional laboratory control
consisted of culture water with Florissant soil. After a 24 hr settling period, the
overlying water was sampled and tested for dissolved oxygen, pH, temperature,
conductivity, alkalinity, hardness, and ammonia. The overlying water was renewed and then
10 organisms were randomly added to each test beaker below the water surface. Each test
chamber was provided the appropriate food ratio then placed into a Zumwalt dilutor system
for daily water renewals. Daily monitoring included water renewal, observance of test
organism behavior, feeding, and measurement of water quality parameters. At test
termination, overlying waters were collected for physicochemical analyses and individual
test sediments sieved with a 45 micron standard sieve for organism enumeration and
collection. Surviving organisms were placed into labeled, preweighed, aluminum weigh boats
and dried at 100 C for 24 hr prior to weighing.
In situ
Toxicity Testing: Fall 1997
The Black River in situ study sites were chosen on the
basis of bottom sediment consistencies as well as proximity to known point source runoff
areas of concern. A total of eleven field locations (including a possible control site)
and a laboratory control were evaluated for potential toxic response. An initial
reconnaissance visit and site evaluation preceded in situ field testing.
Due to variable organism sensitivities to the myriad of contaminants
believed to exist at the host of test sites, four surrogate test species were chosen for in
situ evaluation and included: the fathead minnow Pimephales promelas (24 hours post
hatch), the daphnid, Ceriodaphnia dubia (24 hours old), the midge Chironomus
tentans (8-12 days post hatch) and the amphipod, Hyalella azteca (7-14 days
old). Organisms were transported from cultures maintained at Wright State University,
Dayton, Ohio (traceable to USEPA stocks) to the test sites on the Black River in
Cleveland, Ohio.
The in situ chambers used for this study were constructed
of clear core sampling tubes (cellulose acetate butyrate) cut to a length of approximately
15 cm. Polyethylene closures capped each end. Two rectangular windows (~85% of the core
surface area) covered with 74 micron Nitex mesh were incorporated into the core tube,
opposite each other.
In situ chambers were deployed at all field locations on
the afternoon of 29 September, 1997 and collected in the afternoon of 2 October, 1997
after 72 hours of exposure. A laboratory water control was maintained at the field (hotel)
laboratory for standard quality control purposes. Prior to chamber deployment, ten of each
organism was gently added to 50 ml test tubes of culture water for ease of transport to
field locations (one test tube contained one species only). Transportation of organisms to
field sites by this method has proven to minimize handling and travel related stressors.
In the field, site water temperatures were measured and organisms were slowly acclimated
to the lower temperature field conditions. Upon acclimation, in situ chambers
capped on one end were immersed into the river and test organisms were slowly delivered
from the test tubes to the chambers then capped. Before placement into mesh dip bags,
chambers were held below the water surface and all internal air was expelled. At each test
site, in situ chambers were placed just below the waters surface, on the sediment
bottom or both for 72 hours of exposure. After 72 hours of exposure, in situ
chambers were gently lifted out of the river in the intact mesh bags and returned to the
field laboratory in coolers of site water. Upon arrival to the lab, chambers were
individually emptied into crystallizing dishes and the survivors of each species
enumerated and logged.
In situ
Toxicity Testing: Spring 1998
The spring/high flow Black River in situ study revisited
the same field study sites characterized in the fall/low flow in situ study,
however, chambers were located in the upper water column only as the primary aim of this
study was to measure water column toxicity as a result of storm water runoff loadings but
not sediment toxicity. Original field study sites were chosen based upon potential point
and non-point source influences. A total of six field locations (including a reference
site) and a laboratory control were evaluated for potential toxic response.
The same test species were used as in the Fall sampling period,
including: the fathead minnow P. promelas (24 hours post hatch), the daphnid, C.
dubia (<24 hours old), the midge C. tentans (8-21 days post hatch) and
the amphipod, H. azteca (7-14 days old). Organisms were transported from cultures
maintained at Wright State University, Dayton, Ohio (traceable to USEPA stocks) to the
test sites on the Black River in Cleveland, Ohio.
The in situ chambers used for this study were constructed
of clear core sampling tubes (cellulose acetate butyrate) cut to a length of approximately
15 cm. Polyethylene closures capped each end. Two rectangular windows (~85% of the core
surface area) covered with 74 micron Nitex mesh were incorporated into the core tube,
opposite each other. To decrease the stress of swift flow through chambers and assure
stabilization of in situ chambers at field locations during periods of runoff and
increased flow, chambers were place in flow traps and that were weighted down by bricks
for this study. Flow traps were constructed of poly snap top boxes containing holes for
water exchange. Four chambers fit in one flow trap box.
In situ chambers were deployed at all field locations on
the afternoon of 11 June, 1998 and collected in the afternoon of 13 June, 1998 after 48
hours of exposure. A laboratory water control was maintained at the hotel laboratory for
standard quality control purposes.
Prior to chamber deployment, ten of each organism were gently added
to 50 ml test tubes of culture water for ease of transport to field locations (each test
tube contained one species only). Prior to transportation of organisms to the field sites,
test tube confined organisms were slowly acclimated to the temperature of each field site
as determined during the previous days field measurements. Acclimation was one degree or
less per hour and took a total of 3-5 hours. Organisms were separated and transported to
field sites in coolers of culture water, additional acclimation took place in the field
when necessary. Transportation of organisms to field sites by this method has proven to
minimize handling and travel related stressors. Upon acclimation to field temperatures
within one degree, in situ chambers capped on one end were immersed into the
river and test organisms were slowly delivered from the test tubes into the open end then
capped. Capped chambers were then held below the water surface and all internal air was
gently expelled before being placed in flow traps. Flow traps were sealed with poly twine,
located just below the waters surface then secured in place by tying to stabilized objects
(trees, rocks, etc.). Chambers remained in place for the entire 48 hour period and were
not removed before that time. After 48 hours of exposure, in situ chambers were
gently lifted out of the river, removed from the flow traps and transported to the hotel
laboratory in coolers of site water. Upon arrival to the lab, chambers were checked for
damage then individually emptied into crystallizing dishes and the survivors of each
species enumerated and logged.
Water quality samples were collected upon test initiation (June 11,
1998), twice during the first 24 hour period (June 12, 1998) and then again at test
termination (June 23, 1998). Typical physico-chemical parameters conducted included;
temperature, dissolved oxygen, pH, hardness, alkalinity, turbidity, conductivity and
ammonia. All field water quality monitoring equipment was calibrated prior to each use
according to EPA and or instrument specifications.
During the Spring sampling period a high flow event occurred. Rain
had not occurred during the preceding couple of weeks. On June 12, 1.23 inches (+ 0.28) of
rain fell and on June 13, 0.62 inches (+ 0.08) fell in the lower Black River area. The
river was noticeably higher with increased turbidity.
DNA Fingerprinting
Populations (N=24) of native snails (Physella gyrina) were collected
from 6 test sites on the Black River. Genomic DNA was isolated with QIAquick PCR
purification kits (Qiagen). The resulting DNA pellets were washed with 70% ethanol, dried
and resuspended in 50 uL of TE [10mM TRIS (pH 8.3), 1mM EDTA]. The quantities of DNA
isolated for each sample were estimated by electrophoresis on 1% agarose yield gels. Upon
dilution to make DNA concentrations consistent between samples, all isolates were either
used immediately or stored at –20 C.
RAPD-PCR prfiles were generated from total genomic DNA as described
by Williams et al. (1990). Final reaction volumes were 10mL and contained 2mL of diluted
genomic DNA, 1.5 units of KT1 KlenTaq (Wayne Barnes, Washington University, St. Louis,
MO), 20mM Tris, pH 8.0, 2.5 mM MgCl2, 16 mM (NH4)2SO4, 150 mg/mL bovine serum albumin, 0.2
mM of a single primer (either B-01: 5’-CAGGCCCTTC-3’ or B-02:
5’TGATCCCTGG-3’) and 60 mM dNTP. MJ Research thermocyclers (PTC-100 and
Mini-cycler models) were used for the amplifications for 45 cycles consisting of the
following steps: 92 C for 1 minute, 36 C for 1 minute, 68 C for 2 minutes. An additional
extension at 68 C for 5 minutes followed the last round of amplification. These conditions
were chosen due to their ability to generate DNA profiles from individual snal in a highly
reproducible fashion (each profile was generated on at least two separate occasions and
scored independently by at least two observers). All samples were held at 4 C until
RAPD-PCR products could be resolved by gel electrophoresis. RAPD products were
electrophoresed in 2% agarose gels in TBE buffer (10 mM Tris, pH 8.3, 10 mM boric acid,
1mM EDTA) at 4 C. Gels were stained in ethidium bromide for one half –hour and
destained in water for one hour. Bands were visualized with a UV lamp and documented using
a Gel Print 1000I imaging system. Each RAPD-DNA profile was scored twice, independently,
and the RAPD-PCR amplification of a sample was repeated in cases where scorings were not
in complete agreement. A measure of the genetic similarity of individual crayfish to
others collected at the same site was obtained by determining the fraction (f) of markers
it shared with other snails from the same site using the following equation:
where mxy is the number of bands any two samples
share and mx and my are the number of bands amplified in each organism.
The Jonckheere test for ordered alternatives was performed using a
custom-written FORTRAN program. A probability value was determined based on an analysis of
5,000 random permutations of the data set with the null hypothesis stating that there is
no difference among the mean pairwise genetic similarities from the 8 sites used in this
study (Ho:u1=u2=u3=…=un) and the alternative hypothesis that there is a monotonic
trend based on a priori information (HA: u1>u2>u3>…>un, where at least
one of the inequalities is a strict inequality).
Results of the genetic similarity analyses were compared with the
Ohio EPA Index of Biotic Integrity (IBI) (a measure of fish community health) and the Ohio
EPA Invertebrate Community Index (ICI) (a measure of benthic macroinvertebrate community
health). Both of the Ohio EPA indexes are based on multiple metrics of species diversity,
presence/absence, functional groups, abundance, and group ratios. These data were obtained
from Ohio EPA (1994). A regression analysis was conducted comparing the average value of
all pairwise similarities for each snail vs. all others with either the IBI or ICI values
from the same sampling locations.
Results and
Discussion
Site water quality 1997
During the 1997 sampling event the Black River was at base flow conditions.
No rain occurred during the sampling; however strong winds were blowing from Lake Erie
upstream. The river was relatively turbid, as exists during base flow conditions.
Fall 1997 Laboratory
Testing
Whole sediment toxicity was measured following standard U.S. EPA protocols
for H. azteca and C. tentans in 10 day laboratory exposures. Survival
and growth (dry weight) were measured in both organisms at test termination (Figs. 2-5). Survival was greater than 80% in the
laboratory sediment control (Florissant) for both species (Table
3, Figs. 2 and
4). The water control showed poor survival for both
organisms, suggesting a water related stress was ameliorated in the presence of sediment.
The upstream reference sites showed the East Branch (RM 18.9) to have the best survival
and growth for both species, while only H. azteca survived well (>80%) at RM
15 and C. tentans survived poorly (30%). Poor survival occurred at the most
downstream site in the river mouth at RM 0.3 for C. tentans with 47.5% survival
in deep sediments and 65% survival in near surface sediments. Similar poor survival
occurred at RM 4.6 with survivals of 52.5 and 60% in deep and surficial sediments,
respectively. The amphipod also had poor survival downstream with lowest survival at RM
2.5 (15 and 30%, lower and upper sediments), and poor survival also in the upper sediments
from the harbor (RM 0.2). The deep sediments at RM 2.4 were also toxic to the amphipod
with a survival of 60%.
A comparison of upper (0-2 cm) and deeper (8-10 cm) sediments showed
a tendency for greater toxicity in the more historic sediments. Five of 7 of the amphipod
exposures to deep sediments were more toxic than the surficial sediment. The two sites
near the Lake (RM 0.5 and 0.3) had an opposite response with greater toxicity in the
surficial sediments. The midge showed a similar pattern with 4 of 5 deeper sediments being
more toxic; however, in contrast to the amphipod surficial sediments at RM 0.3 were not as
toxic as deeper ones.
Growth of the amphipod and midge in the upper reference site was
good (Figs. 3 and
5).
For the amphipod growth was lowest at RM 0.3, 2.4, and 5.2; while for the midge, growth
was lowest at RM 0.3, 2.5, and 9.8. Surficial sediments tended to be less toxic (survival
and growth (amphipod only)) than deeper, more historical sediments in most cases.
Fall 1997 In situ
Exposures
The field laboratory control survival for all organisms ranged from
92.5 to 100% in the Sept. 29 – Oct. 2, 1997 field exposures. Survival at the upstream
reference sites (East Branch RM 18.9 and Mainstem RM 15.0) ranged from 82.5 to 100% (Table 2,
Figs. 6-9).
Spurious results were obtained with the C. dubia reference samples and were
deleted. Conditions in the river during the exposures were at base flow. No rain events
occurred during the test period.
Survival of the 4 test species varied at the downstream test sites
from a low of 40% P. promelas survival at RM 0.5 (Erie St. Bridge, upper exposure) to a
high of 97.5% P. promelas survival at RM 2.4 (upper exposure) (Fig. 8). Three of the four test organisms had lowest
survival at RM 0.5. Hyalella azteca exhibited lowest survival at RM 0.3 and second lowest
at RM 2.4 (Fig. 6). The midge C. tentans survived
better than the other 3 organisms overall, ranging from 72.5 to 95% at the downstream test
sites (Fig. 7).
At four test sites, organisms were exposed near the water surface
(upper) and in chambers placed on the sediment surface where there was minimal contact
possible through the chamber mesh. C. dubia was the only organism not exposed in
the upper water column. There was a strong trend towards lower survival of organisms in
the upper water column exposures, than on the sediment surface. Of the four test sites and
3 organisms, survival was lower in 9 of 12 responses in the upper water column. The
differences were not extreme, but suggest greater toxicity exists in the near surface area
of the water column.
Fall 1997 DNA
Fingerprinting
When the DNA pattern of individuals with a population becomes
similar, then it suggests the population has been adversely impacted and is less diverse.
This loss of genetic diversity can equate to greater susceptibility to stress and general
population decline. A survey of the DNA fingerprints of the indigenous snail Physella
gyrina was conducted at 6 sites along the Black River. The test sites and genetic
similarity measures are shown in Table 4. Genetic
similarities ranged from 0.80 to 0.88. The high similarity index indicates that the
population is being stressed, which was observed at the French Creek tributary (Kline
Ditch). The Kline Ditch area has had water quality problems attributed to nearby fly ash
disposal sites (Ohio EPA, personal communication). The other site with a high value was
from snails collected from River Mile 5.2 with a genetic similarity of 0.85. This site is
above Kobe and French Creek. A regression analyses between the mean genetic similarity at
5 sites with the IBI revealed a highly significant correlation of 0.92 (p<0.01) (Fig. 10). This suggests that the DNA fingerprinting
approach may be a useful surrogate indicator of overall aquatic ecosystem health. The
regression analyses with the ICI was not statistically significant.
Spring 1998 In
situ Exposures
In the Spring of 1998, in situ exposures of test organisms
during a high flow event showed little to no acute toxicity existing at most test sites (Table 5,
Figs. 11-14).
This survey did not focus on sediments, rather organisms were only exposed to near surface
waters during very turbid conditions. For the majority of test sites, all organisms met or
exceeded EPA survival performance criteria exhibiting greater than 80% survival. The only
deviation was C. tentans at the Cascade site (RM 15.0) in which the average
survival was 70%. Upon test termination, greater than 90% of dead organisms were accounted
for.
Two field references (river mile 15.0 (Cascade) and 18.9 (East
Branch - Grafton)) and one laboratory control were run concurrent with the test. Since a
field reference for in situ storm water runoff assay is preferential, two
potential reference sites were chosen based upon USEPA historical data. A laboratory
control is standard protocol as back up in the event of chamber loss due to high flow or
vandalism at any of the field reference locations and to verify organism health. The 18.9
Mile field location is the farthest site upstream on the Black River and proved to provide
the best reference conditions. Both field reference conditions, however, exhibited
acceptable in situ test survival criteria.
Hyalella azteca exhibited greater than 90% survival at all
of the field locations and 97.5% survival at the 18.9 Mile field reference site (Fig. 11). Ceriodaphnia dubia exhibited greater than
82.5% at all of the field locations and 92.5% survival at the field reference site (Fig. 14). The cladoceran survival appeared to be greater
than 100% (107.5%) at the 15.0 Mile location but was attributed to inclusion of indigenous
cladoceran during the loading of organisms into the chambers at this particular site. It
was impossible to distinguish which organisms originated from the laboratory and which
were indigenous. Poor survival of the field control of C. dubia was unexplained;
however, the good survival in situ verified organism health. Chironomus
tentans also exhibited greater than 80% survival at all of the field locations with
the exception of the 15.0 Mile site in which survival was 70% (Fig. 12). All dead organisms were located eliminating
the possibility of escape. Midge control survival was 92.5%. The fathead minnow, P.
promelas, exhibited the most variability between sites, yet had greater than 77%
survival on average at all the sites (Fig. 13). The
lowest survival was at the 4.8 Mile site (French Creek mouth) where an oil smell and
droplets emanating from the sediment was noted; however, survival was still 77.5%. One
chamber at each of the 2.3 Mile and 5.2 Mile field locations had zero and one organism,
respectively, which were determined to be outliers and not included in the summary
statistics. P. promelas reference survival was 100% at the 18.9 Mile site.
Photo-induced toxicity due to PAHs is not a factor when turbidity is
high (Ireland et al. 1996). Therefore, the greater toxicity observed at base flow
conditions, when turbidity is lower, appears to be a PAH effect. This effect can occur at
PAH concentrations at the low to sub microgram per liter level.
The water column acute toxicity at base flow and the sediment acute
toxicity (survival and/or growth) observed in the downstream areas suggest the PAHs may
still be a primary stressor in the lower Black River. Sediments that were several
centimeters deep tended to be more toxic than surficial sediments and may be exposed
during resuspension events (e.g., storms, boat traffic, dredging). However, chemical
analyses of sediments did not show elevated levels of PAHs (Table
5). It appears the remedial dredging which took place from 1989-1990 was effective in
reducing PAH concentrations and likely associated toxicity. Total sediment metal
concentrations were elevated and tended to be higher downstream (Table 6). The lack of acute toxicity during high flow
suggests that the impacts of nonpoint source runoff and stormwater inputs are less severe.
However, since only acute toxicity was measured, it is unknown whether chronic toxicity
may exist due to nonpoint source runoff.
Acknowledgements
This project was coordinated with Dr. Paul Baumann (USGS). We
appreciate the valuable assistance of Paul Anderson (Ohio EPA) who provided extensive
logistical support.
Literature Cited
Baumann, P.C. and J.C. Harshbarger. 1995. Bullhead catfish after
coking plant closes and environmental PAHs plummet. Environ. Health Perspec. 103: 168.
Burton, G.A., Jr., C. Hickey, T. DeWitt, D. Morrison, D. Roper, and
M. Nipper. 1996. In situ toxicity testing: Teasing out the environmental
stressors. SETAC NEWS 16(5):20-22.
Chappie, D. J. and G.A. Burton, Jr., 1997. Optimization of in
situ bioassays with Hyalella azteca and Chironomus tentans. Environ. Toxicol. Chem.
16:559-564.
Ireland, D.S., G.A. Burton, Jr., and G.G. Hess. 1996. In Situ
toxicity evaluations of turbidity and photoinduction of polycyclic aromatic hydrocarbons.
Environ. Toxicol.
Chem. 15:574-581.
Ohio EPA . 1994. Biological and Water Quality Study of the Black
River (with selected tributaries) and Beaver Creek. Ecological Assessment Section,
Division of Surface Water. Columbus, OH.
USEPA. 1994. Methods for Measuring the Toxicity and Bioaccumulation
of Sediment-associated Contaminants with Freshwater Invertebrates. EPA 600/R-94-024.
Office of Research and Development. Duluth, MN.
Williams, J.G.K., A.R. Kubelik, K.J. Livak, J.A. Rafalski, and S.A
Tingey. 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic
markers. Nucleic Acids Res. 18:6531-6535.
