NOAA Technical Memorandum NMFS NE 157
Contaminant
levels in muscle
of four species of recreational fish
from the New York
Bight apex
by Ashok
Deshpande1, Andrew F.J. Draxler1, Vincent S.
Zdanowicz1,2, Mary E. Schrock3, Anthony J.
Paulson1,
Tom W. Finneran1, Beth L. Sharack1, Kathy Corbo1,
Linda Arlen1, Elizabeth A. Leimburg1,
Bruce W. Dockum1, Robert A. Pikanowski1, Brian May4,
and Lisa B. Rosman4,5
1National Marine
Fisheries Serv., 74 Magruder Rd., Highlands, NJ 07732
2Current address: U.S. Customs Serv., Research Lab., 7501
Boston Blvd., Ste. 113, Springfield, VA 22153
3Battelle Memorial Inst., 505 King Ave., Columbus, OH 43201-2693
4U.S. Army Corps of Engineers, 26 Federal Plaza, New York,
NY 10278-0090
5Current Address: National Ocean Serv., 290 Broadway, Rm.
1831, New York, NY 10007
Print
publication date June 2000;
web version posted June 13, 2001
Citation: Deshpande A, Draxler AFJ, Zdanowicz VS, Schrock ME, Paulson AJ, Finneran TW, Sharack BL, Corbo K, Arlen L, Leimburg EA, Dockum BW, Pikanowski RA, May B, Rosman LB. 2000. Contaminant
levels in muscle
of four species of recreational fish
from the New York
Bight apex. NOAA Tech Memo NMFS NE 157; 99 p.
Download complete PDF/print version
A survey was conducted to establish a benchmark for concentrations of
selected trace metals and organic contaminants in the edible flesh of
four species of fish important to the recreational fishery of the New
York Bight Apex. Bluefish (Pomatomus saltatrix), summer flounder
(Paralichthys dentatus), black sea bass (Centropristes striatus),
and tautog (Tautoga onitis) were caught by rod and reel during
September-December 1993 at 15 sites in the New York Bight Apex. Fourteen
composite samples of muscle tissue from each fish species were analyzed
for 9 trace metals, 25 polychlorinated biphenyl (PCB) congeners, 17 organochlorine
pesticides, 24 polycyclic aromatic hydrocarbons (PAHs), seven 2,3,7,8-substituted
polychlorinated dibenzo[p]dioxins (PCDDs), and ten 2,3,7,8-substituted
polychlorinated dibenzofurans (PCDFs).
Concentrations of trace metals were low and within the range of values
normally found in muscle tissues of finfish from relatively pristine
ecosystems. Total mercury levels in all fish composites were <0.11 µg/g
(ppm) wet weight, which is an order of magnitude below the U.S. Food
and Drug Administration (FDA) action level of 1.0 µ="3">g/g (ppm)
wet weight for methylmercury.
PCB and organochlorine pesticide concentrations were relatively low
and were related to the lipid content of the muscle tissue. The Aroclor-based estimates
(see Glossary... for definition)
for all composite samples were below the FDA tolerance level of 2.0 µg/g
(ppm) wet weight for PCBs. Average sums of 23 PCB congeners were 0.37 µg/g
for bluefish, <0.05 µg/g (i.e., below the detection limit)
for summer flounder, 0.08 µg/g for black sea bass, and 0.06 µg/g
for tautog.
Average sums of DDTs and their metabolites for all composite samples
were well below the FDA action level of 5.0 µg/g (ppm) wet weight.
Average sums of DDTs and their metabolites were 0.16 µg/g for bluefish, <0.009 µg/g
(i.e., below the detection limit) for summer flounder, 0.02 µg/g
for black sea bass, and 0.014 µg/g for tautog.
Average sums of chlordanes for each species, which ranged from 0.04
to 0.08 µg/g, were below the FDA action level of 0.3 µg/g
(ppm) wet weight.
With few exceptions, PAHs were undetected.
Concentrations of 2,3,7,8-tetrachlorodibenzo[p]dioxin (TCDD) in all
composite samples were below the FDA advisory level of 25 pg/g (pptr)
wet weight for limited consumption. Concentrations of 2,3,7,8-TCDD were
below the method detection limit of 1.63 pg/g in all summer flounder
and black sea bass composites, 10 of 14 tautog composites, and 4 of 14
bluefish composites. The concentrations of 2,3,7,8-TCDD were near the
detection limit in the 4 remaining tautog composites, and in 9 of 10
remaining bluefish composites. The remaining bluefish composite contained
the highest concentrations of PCBs (0.57 µg/g), DDTs (0.27 µg/g),
chlordanes (0.062 µg/g), and 2,3,7,8-TCDD (7.27 pg/g). This bluefish
composite had the highest average composite weight, included the heaviest
individual specimen, and had the highest lipid content.
INTRODUCTION
A survey was conducted to establish a benchmark for concentrations of selected
trace metals and organic contaminants in the edible flesh of fish species
important to the recreational fishery of the New York Bight Apex (i.e.,
the area bounded by the coasts of New Jersey and Long Island, 73°30W
longitude, and 40°15N latitude; Bowman and Wunderlich 1976; Figure
1). Four species were targeted based on their importance to
the recreational fishery, their life habits, and the regional ecology:
bluefish, Pomatomus saltatrix (pelagic habitat); summer flounder, Paralichthys
dentatus (demersal habitat); black sea bass, Centropristis striata (reef
habitat); and tautog, Tautoga onitis (reef habitat). Refer to Figure
2 for species illustrations and synoptic descriptions of range,
habitat use, spawning, stock structure, migratory behavior, predation,
and management.
The survey collected and analyzed fish caught by local recreational
fishermen during the fall when fish physiological condition and lipid
(i.e., fat) levels would likely be highest. To the extent that
any metal or contaminant concentration is positively associated with
lipid levels, the timing of the sampling would be most useful from a
public health standpoint. Measured concentrations were compared with
the U.S. Food and Drug Administrations guidelines for human consumption.
METHODS
SAMPLE
COLLECTION, DISSECTION, AND COMPOSITING
Bluefish, summer flounder, black sea bass, and tautog were caught by
rod and reel during September-December 1993 at 15 sites in the New York
Bight Apex selected on the basis of their popularity with fishermen (Figure
1; Appendix Table A1). Site locations
were determined by LORAN-C.
Guidelines of NOAA-FDA-EPA (1986) were used for handling the fish. Whole
fish were returned to the laboratory on ice, and dissected within 48
hr. In the laboratory, each whole specimen was weighed to the nearest
gram and measured to the nearest millimeter. Total length was measured
for summer flounder, black sea bass, and tautog; fork length was measured
for bluefish. Individual specimens were also examined for gross abnormalities
and sex. In keeping with local consumption practices, fillets (i.e.,
boneless muscle tissue) of summer flounder and tautog were prepared with
the skin and scales removed. Bluefish and black sea bass fillets included
the skin, with the scales removed.
Dissections were performed in the laboratory under a high-efficiency
particle air (HEPA) laminar-flow hood. Dissecting implements and containers
were cleaned in a manner appropriate for the specific analyses. Implements
were cleaned with ultrapure 10% nitric acid, double-deionized (DDI) water,
methanol, and methylene chloride from a commercial supplier. Plastic
containers for trace metals samples were washed in dilute Micro liquid
laboratory cleaner, rinsed in tap water, washed in 10% nitric acid, triple
rinsed in DDI water, and dried under a HEPA clean-air hood. For trace
metal analyses, three adjoining pieces (approximately 2 cm3 each)
of white muscle were excised from the anterior dorsal portion of each
fillet and stored in acid-cleaned plastic vials at -20°C (Figure
3). For organic contaminant analyses, the remainder of each dorsal
fillet was homogenized in a stainless steel blender, and stored in precleaned
glass jars at -80°C. Both dorsal and ventral fillets from each summer
flounder, and small samples from other species, were homogenized to obtain
an adequate sample size.
Allocation of muscle tissue from individual specimens to composites was
based on fish length. Outlier specimens for each species at each site were
identified using the Dixon outlier test (Sokal and Rohlf 1981); those specimens
were excluded from further consideration. The number of composites for
each species for each station was based on the number of normally distributed
specimens and the need for three specimens per composite. A random number
was then assigned to each of the normally distributed specimens. Given
that N specimens were to be composited for a specific station, we selected
the N specimens with the lowest random number from the available specimen
pool. For example, if five composites were to be prepared for a particular
station, specimens with the lowest 15 random numbers were selected. The
selected specimens were sorted by length and grouped in sets of three to
form the five composites (Appendix Tables A2-A5). The
specimens identified as outliers and those not randomly selected for the composite
preparations are listed in Appendix Tables A6-A9.
SAMPLE ANALYSIS
Trace Metals
Analyses of nine trace metals (Appendix Table B1)
in the muscle composite samples were performed in two separate batches following
the procedures of Zdanowicz et al. (1993). Each batch included 28 muscle
composites (i.e., 14 for each of two species), three replicates of dogfish
liver standard reference material (SRM; DOLT-1, National Research Council of
Canada), three method blanks, and one composite in duplicate for each of the
two fish species. Quality assurance (QA) and quality control (QC) procedures
included participation in the annual NOAA-NRC [National Research Council Canada]
intercomparison exercise (Willie and Berman 1995).
Approximately 0.5 g of muscle from each of three individual specimens constituting
a composite were placed in acid-cleaned teflon vials and dried overnight at
60-65°C. Five milliliters of ultrapure, concentrated nitric acid were added
to each vial, and the vials were allowed to stand at room temperature for 2-4
hr. Vials were then placed inside teflon-lined bombs, and the tissue was digested
overnight at 120°C. After cooling, the bombs were vented, the vials removed,
and the digests allowed to degas at room temperature overnight. The digests
were then quantitatively transferred to 25-ml glass graduated cylinders, brought
to volume using double-deionized water, and analyzed for arsenic (As), cadmium
(Cd), chromium (Cr), copper (Cu), lead (Pb), mercury (Hg), nickel (Ni), silver
(Ag), and zinc (Zn) using atomic absorption spectrophotometry or inductively
coupled plasma mass spectrometry. Wet weight and dry weight for each muscle
composite were used in the percent water determinations.
PCBs, Organochlorine Pesticides,
and PAHs
Analyses of 25 PCB congeners (Appendix Table B2),
17 organochlorine pesticides (Appendix Table B3),
and 24 PAHs (Appendix Table B4) in muscle composites
were performed in six separate batches following the guidelines of NOAA (Krahn et
al. 1988; Sloan et al. 1993) and the EMAP [Environmental Monitoring
and Assessment Program] procedures of the U.S. Environmental Protection Agency
(EPA 1993a). Each batch of 24 extractions included one method blank, one matrix
spike, one mussel tissue SRM (mussel tissue V, QA93TIS5 - SRM 1974a, NIST [National
Institute of Standards and Technology]), and one muscle composite in triplicate.
One batch included seven spiked replicates of a summer flounder muscle composite
for the method detection limit (MDL) determination. The remaining soxhlet extraction
setups in this latter batch were allocated to the analyses of other QA samples
and the muscle composite samples. QA and QC procedures followed EMAP protocols
(Valente et al. 1992), and included participation in the annual NIST/NOAA/NS&T/EPA
EMAP intercomparison exercise (Parris 1995). A separate sample of each composite
was dried overnight in an oven at 120°C, and reweighed to determine the
percent wet weight (Appendix Tables A2-A5). Although
all values in this report are wet weight concentrations, this measurement allows
the reader to convert concentrations to a dry-weight basis. In addition, the
lipid content of each composite was determined gravimetrically.
Approximately 4 g of muscle from each of the three individual specimens constituting
a composite were placed in a mortar and mixed, using a pestle, with anhydrous
sodium sulfate until the composite was dry. The mixture was soxhlet extracted
with methylene chloride following NIST protocols (Wise et al. 1991).
Twenty percent of the methylene chloride extract was evaporated to dryness
for lipid determination. Silica gel/alumina/florisil column chromatography
was used to remove the bulk biogenic and other polar interferences from the
remaining extract. The cleaned fraction was further purified using size-exclusion,
high-performance liquid chromatography (HPLC).
PCBs and chlorinated pesticides were analyzed by capillary gas chromatography
with electron-capture detection (GC/ECD; EPA 1993a). Nomenclature for PCB compounds
follows that of Ballschmiter and Zell (1980). Specific PCB congeners were not
verified by gas chromatography - mass spectrometry (GC-MS), and the apparent
concentrations of specific congeners may be affected by contribution(s) from
the coeluting compound(s) (Appendix Table B2). PAHs
were analyzed by capillary GC-MS in selected ion monitoring mode (EPA 1993a).
2,3,7,8-Substituted PCDD and
PCDF Congeners
Dibenzo[p]dioxin and dibenzofuran are the base structures for two sets of
compounds in which chlorine atoms are added to form PCDDs and PCDFs. There
are 75 PCDD and 135 PCDF congeners. Those congeners with chlorine atoms in
the 2,3,7, and 8 positions (of which there are 7 PCDDs and 10 PCDFs; Appendix Table
B5) are considered toxic, with 2,3,7,8-TCDD being the most toxic of all
PCDD and PCDF congeners (EPA 1989).
Muscle composites were analyzed for the seven 2,3,7,8-substituted PCDD congeners
and ten 2,3,7,8-substituted PCDF congeners in three separate batches, using
EPA Method 8290 with selected modifications from EPA Method 1613 (EPA 1993b,
1994; Battelle 1996). Each batch of 18-20 samples included one method blank,
one matrix spike, one fish SRM (EDF-2526, Cambridge Isotope Laboratory), and
one muscle composite in triplicate. One batch included four replicates of a
summer flounder muscle composite for the MDL verification. Internal standards
used in identification and quantification of PCDD and PCDF congeners were 13C-labeled
analogs of each dioxin congener except for 1,2,3,7,8,9-hexachlorodibenzo[p]dioxin,
and of each dibenzofuran except for octachlorodibenzofuran. Isomers of each
homolog series were resolved on a DB-5 column and analyzed by high-resolution
MS. Second-column confirmation of 2,3,7,8-TCDF levels above 1 pptr were performed
on a DB-Dioxin column.
STATISTICAL TESTS
The nonparametric Kruskal-Wallis test (SAS 1989) was performed to examine
the statistically significant differences among stations for those analytes
in which the mean concentration value from at least one station was three times
the method detection limit. For composites for which a specific congener was
not detected, one-half perform the test of interstation differences.
The Spearman rank order correlation test was used to determine associations
among PCBs, DDTs, lipid content, and average composite length. Nondetectable
values were not used in the Spearman rank order correlation test.
QUALITY ASSURANCE
Precision and accuracy are the measures of QA determined in this study. For
trace metals, QA for precision included analyses of MDLs and relative percent
differences (RPDs), while for accuracy, it included comparative analyses with
SRM. For organic contaminants, QA for precision included analyses of laboratory
methods blanks, MDLs, and laboratory triplicates, while for accuracy, it included
analyses of internal surrogate standards, matrix spike analytes, and SRM. The
data quality objectives (DQOs) for organic analyses are listed in Appendix Table
B6.
Trace Metals
The MDLs for each metal for each batch were computed as three times the standard
deviation of six method blank measurements. The blank values were low and the
MDLs were below 1 µg/g dry weight for all metals (Appendix Table
B7). The percent recoveries of the nine metals in the DOLT-1 SRM varied
between 92 and 104%. The relative standard deviations (RSD) of the dry weight
measurements were 10% or less for all metals, except for Cr which had an RSD
of 17%.
Based upon the percent water determination in muscles of different fish species
(Appendix Tables A2-A5), a nominal water content
of 75% was used for muscle composites for the purpose of determination of MDLs
on a wet weight basis (Appendix Table B8). For duplicate
fish tissue samples, the RPD was computed as the range divided by the mean
and then multiplied by 100. Greater than 75% of the duplicate samples exhibited
an RPD <20% based on wet weight, with much of the variation attributed to
differences in the percent solids between duplicate fillets.
PCBs
GC/ECD analyses were performed on seven replicate solutions of each PCB congener
(approximately 40 pg/µL of each congener, 1 µL injected). The instrumental
detection limit (IDL) for each PCB congener was determined by multiplying the
standard deviation for seven replicate measurements by a Students t value
of 3.143 (EPA 1984b). None of the PCB replicate determinations exceeded the
DQO criterion for IDL (Appendix Table B9). The estimated
method detection limit (EMDL) for each PCB congener was calculated using the
IDL and the nominal values of 10 g wet weight for sample size, 50% for extraction
and cleanup recovery efficiency, 250 µL for final extraction volume,
and 1 µL for GC injection volumes. The MDL for each PCB congener was
determined on seven spiked replicates of summer flounder muscle composites
following the procedure outlined in EPA (1984b). The RSD values were <10%
for most replicate measurements in the MDL determination of PCB congeners (Appendix Table
B10). The fact that the MDL is greater than the EMDL indicates that the
method is limited by random variation in the recovery from samples at low concentrations
rather than by the sensitivity of the instrument. None of the laboratory method
blank values exceeded the DQO criterion (Appendix Table
B6). Approximately 90% of laboratory triplicate values met the DQO criterion
(Appendix Tables B11-B12).
Consistent recoveries (i.e., 85.3-94.9%) were found for the relatively
nonvolatile BZ #198 (Appendix Tables B13-B16),
while recoveries for the surrogate 4-4'-dibromooctafluorobiphenyl ranged between
46 and 70%. The higher apparent recoveries (i.e., 167%) of HPLC surrogate
1,2,3-trichlorobenzene (TCB) in bluefish may be due to the coelution of unknown
interfering compound(s) with the TCB peak. Approximately 69% of PCB congeners
(Appendix Table B17) met the matrix spike DQO criterion.
For PCB congeners, 99% of analyses met the DQO criterion for analysis of accuracy
based on reference material (Appendix Table B18).
Organochlorine Pesticides
GC/ECD analyses were performed on seven replicate solutions of organochlorine
pesticides (approximately 40 pg/µL of each pesticide, 1 µL injected).
The IDL for each pesticide analyte was determined by multiplying the standard
deviation for seven replicate measurements by a Students t value of 3.143
(EPA 1984b). One of 19 pesticide replicate determinations exceeded the DQO criterion
for IDL (Appendix Table B19), although this value
(5.52%) was near the DQO target of 5%. The MDLs were determined on seven spiked
replicates of summer flounder muscle composites following the procedure outlined
in EPA (1984b), and were greater than the EMDLs. The RSD values were <10%
for most replicate measurements in the MDL determination of pesticide analytes
(Appendix Table B20). Approximately 65% of pesticide
analytes met the DQO criterion for analysis of laboratory triplicate samples
(Appendix Tables B21-B22).
Approximately 82% of recovery values for internal pesticide surrogate standards
met the DQO criterion (Appendix Tables B13-B16).
Approximately 79% of pesticide analytes met the matrix spike DQO criterion
(Appendix Table B23). For pesticide analytes, 100%
of the determinations met the DQO criterion for analysis of accuracy based
on reference material (Appendix Table B24).
PAHs
GC-MS analyses were performed on seven replicate solutions of PAH analytes (approximately
200 pg/µL of each congener, 1 µL injected). The IDL for each PAH
analyte was determined by multiplying the standard deviation for seven replicate
measurements
by a Students t value of 3.143 (EPA 1984b). Seven of 24 PAH replicate determinations
exceeded the DQO criterion for IDL (Appendix Table B25a,b),
although the highest RSD value was only 8.3%. The MDLs for PAH analytes were
determined on seven spiked replicates of summer flounder muscle composites following
the procedure outlined in the EPA (1984b). The overspiking of PAHs resulted in
high MDL values. The reported detection limits for PAHs were computed from the
replicate
analyses of mussel tissue SRM (Appendix Table B26a,b).
For four PAH compounds for which peaks were not found in the chromatograms, the
detection limits were estimated to be 10 ppb wet weight based on the following
assumptions: 1) 10 g wet weight of muscle tissue, 2) 50% efficiency in sample
extraction and cleanup steps, 3) 250 µL as the final sample volume, and
4) an IDL (i.e., GC-MS) of 200 pg/µL. Of the three samples that
exhibited one PAH value above the MDL, all analyses met the DQO for analysis
of laboratory
triplicates (Appendix Tables B27a,b, B28a,b).
Approximately 75% of recovery values for internal surrogate standards met
the DQO criterion (Appendix Tables B13-16). The
low recoveries (i.e., 23%) for deuterium-labeled naphthalene are not
unexpected considering the volatility of this compound. Only 37.5% of PAH determinations
(Appendix Table B29a,b) met the matrix spike DQO
criterion. The matrix spike recovery data for PAHs are apparently skewed by
the low-molecular-weight PAHs. These compounds are somewhat more volatile than
their high-molecular-weight counterparts, and thus, are prone to evaporative
losses during sample preparation. There seems to be no apparent explanation
for the lower recoveries of perylene. The analysis of NIST SRM 1974a (intercomparison
sample QA93TIS5) with each analytical batch indicates good precision from batch
to batch, although this material contained low concentrations of contaminants
(Appendix Table B30).
2,3,7,8-Substituted PCDD and PCDF Congeners
The MDLs for the PCDD and PCDF congeners were calculated using only three
replicates because one of the four replicates used in the MDL verification
exercise was lost during preparation (Appendix Table B31a,b). None of the laboratory
triplicate analyses exceeded the DQO criterion of #25% RSD for analytes that
had concentrations greater than 10 times the MDL (Appendix Tables
B32a,b, B33a,b).
Approximately 98% of internal surrogates and 96% of cleanup surrogates for
dioxin analyses met the DQO criterion (Appendix >Tables
B34a,b-37a,b). Recoveries of internal surrogate standards varied from 7
to 116%, with an average of 73% (RSD = 21%). Recoveries of a cleanup standard,
in which all four chlorines of 2,3,7,8-TCDD were labeled with 37Cl,
ranged from 16 to 385%, with an average of 104% (RSD = 41%). All internal standard
DQOs were exceeded for one tautog composite (i.e., composite #155).
For tautog composite #155, the average recovery of congeners in which all 12
carbons were labeled with 13C was 12.1% (range of 7-17%), and the
recovery of the cleanup standard, 37Cl-labeled 2,3,7,8-TCDD, was
16%. The highest recovery for this composite is below the lowest recovery in
any other composite (i.e., 29%). Therefore, values for tautog composite
#155 should be used with caution.
Approximately 94% of the matrix spike analytes met the DQO criterion (Appendix Table
B38a,b). None of the PCDD and PCDF analyses of the accuracy-based SRM
exceeded the DQO criterion (Appendix Table B39a,b).
Quality Assurance Summary
The MDLs for metals were low, and the accuracy and precision of the SRM measurements
were generally within 10%. Duplicate analysis of fish tissue samples resulted
in RPD measurements which exceeded 20% in 25% of the samples, partially due
to differences in tissue density.
For organic contaminants, some of the quality assurance goals were not met,
but the overall quality assurance compliance is judged typical of organic analytical
data. The potential exists that organic data can be affected by interfering
compounds coeluting with contaminant analyte peaks. Because each sample matrix
is different, methods do not exist for adjustment of the data for these variations.
The number of replicates per site, the number of fish per composite, and the
agreement with other laboratories participating in the NIST/NOAA/NS&T/EPA
EMAP intercomparison exercise do, however, provide confidence in the final
estimates. As a practical matter, the general conclusions derived from this
survey are not limited by the quality assurance data, but data should be interpreted
with caution for specific samples in which quality assurance goals were not
met.
RESULTS
AND DISCUSSION
GENERAL CHARACTERISTICS
OF FISH COMPOSITES
Bluefish had significantly higher lipid content (i.e., 6.80%)
than the other three species (Figure 4A).
Tautog and black sea bass had similar lipid content (i.e., 2.03
and 2.09%, respectively), while the lipid content of summer flounder
was 0.56%. For all four species, there were no significant differences
in lipid content among stations (Table 3, Appendix Tables
A2-A5).
The bluefish caught at Station BL1 were shorter, lighter, and less dense
(i.e., more water) than the bluefish caught at Stations BL2 and
BL3. The black sea bass caught at Station SB1 had significantly higher
water content than those caught at Stations SB2 and SB3. Within the four
species, there was a significant correlation between average length of
the composite and average weight (Table 1),
in part because randomly selected specimens were sorted into composite
samples by length.
A correlation between percent lipids and average length (or weight)
was observed only for tautog (Table 1).
CONTAMINANT CONCENTRATIONS
IN FISH COMPOSITES
Trace Metals
Complete listings of analytical results for metals in each composite
are given in Appendix Tables C1-C4. Metal
concentrations in all muscle composites were above the detection limits.
These concentrations were generally low (Table
2) and well within the ranges expected for metals in muscle of fish
from relatively uncontaminated locations. Values on a wet weight basis
were <0.05 µg/g (ppm) for Ag; 0.1-0.5 µg/g for Cd, Cr,
Cu, Ni, and Pb; 3.5-13.8 µg/g for Zn; and 0.4-3.8 µg/g for
As. The Hg analyses in this study determined total Hg content, a measurement
which encompasses all species of mercury present, including methyl mercury.
The highest value of total mercury in this study was 0.11 µg/g
which was an order of magnitude lower than the FDA action level of 1.0 µg/g
wet weight for methyl mercury in fish or shellfish for human consumption
(Kennedy 1979).
Statistically significant differences in metal concentrations among
locations were not detected for any metal (i.e., P values > 0.05
for all metals for all species; Appendix Tables
C1-C4). Therefore, data from all sites were pooled by species, and
mean species concentrations (Table 2) were
compared (Figure 5). The following significant
differences in mean metal levels were found: Cr: bluefish = black sea
bass > tautog = summer flounder; Zn: bluefish > tautog = summer
flounder = black sea bass; As: black sea bass > summer flounder > tautog > bluefish;
and Hg: tautog = bluefish > summer flounder = black sea bass.
These differences are probably related to the unique nature, diet, and
behavior of each species. It is not uncommon to find differences in contaminant
levels among species (Zdanowicz et al. 1992). The differences,
however, do not appear to be related to habitat type.
PCBs
A significant number of composite-congener combinations had PCB concentrations
that were below the MDL (i.e., 20% for bluefish, 91% for summer
flounder, 57% for black sea bass, and 63% for tautog). Two PCB congeners
were not detected in any composite (Appendix Table
C5), and were not included in subsequent calculations. Complete
listings of analytical results for PCB concentrations for each composite
are given in Appendix Tables C6-C9.
Of the 16 PCB congeners which were found in bluefish and which were
statistically analyzed (Appendix Table C6),
two low-concentration congeners (i.e., BZ #8 and BZ #128) and
one high-concentration congener (i.e., BZ #153) were found to
be significantly higher at Station BL3 (P < 0.05). Differences
among the stations were not observed for the remaining 13 congeners,
nor for the sum of 23 PCB congeners (P = 0.37). No interstation
difference was assumed, and therefore, the bluefish means for the 14
composites are given in Table 3. Bluefish muscle
composites contained the highest mean PCB concentrations of the four
species. The sum of 23 PCB congeners (i.e., PCBs)
ranged from 0.21 to 0.57 µg/g (ppm), with a mean of 0.37 ppm (Appendix Table
C6). The PCB congener composition in bluefish composites was dominated
by seven congeners: BZ #1 (monochloro), BZ #66 (tetrachloro), BZ #101
(pentachloro), BZ #118 (pentachloro), BZ #153 (hexachloro), BZ #138 (hexachloro),
and BZ #180 (heptachloro). The maximum concentration of 0.57 ppm for PCBs
was found in a bluefish composite (i.e., #113; Appendix Table
C6). The sum of concentrations of 18 specific PCB congeners was multiplied
by 2 to generate an approximation of Aroclor-based total
PCB data for comparison with the historical total PCB data; the approximations
are shown in the second data column of Table 3 and
in the last column of Appendix Tables C6-C9 (NOAA
1989; ACE-EPA 1992). The highest of these estimates was 0.9 ppm, which
is below the FDA tolerance level of 2 ppm wet weight (FDA 1991).
No PCB congener had a concentration greater than 3 times the MDL in
summer flounder composites. Of the six congeners (i.e., BZ #1,
BZ #66, BZ #101, BZ #118, BZ #153, and BZ #138) detected in summer flounder
composites, only one (i.e., BZ #1) was found at all stations.
The PCB levels in black sea bass composites were considerably lower
than those found in bluefish composites. The PCBs
in black sea bass composites ranged from 0.043 to 0.14 ppm, with a mean
of 0.079 ppm (Appendix Table C8). The PCB
congener composition in black sea bass composites was dominated by five
congeners: BZ #66, BZ #101, BZ #118, BZ #153, and BZ #138. Of the six
congeners (i.e., the aforementioned five plus BZ #1) in black
sea bass composites for which statistics were performed (i.e.,
those where the station mean concentrations were greater than 3 times
the MDL), interstation differences were found for four congeners (i.e.,
BZ #66, BZ #101, BZ #118, and BZ #138; Appendix Table
C8). Since higher PCB concentrations were found at Station SB3 for
the majority of congeners and for PCBs
(P = 0.02), we consider black sea bass at the entrance to Ambrose Channel
(Station SB3; Figure 1) to have higher PCB
concentrations than black sea bass at the two stations farther south
(Table 3).
Mean PCB levels in tautog composites were similar to those in black
sea bass composites. The PCBs
in tautog composites ranged from 0.038 to 0.12 ppm, with a mean of 0.059
ppm (Appendix Table C9). The PCB congener
composition in tautog composites was dominated by three congeners: BZ
#1, BZ #101, and BZ #153. No interstation differences were detected for
PCB levels in tautog composites.
The trend in PCB concentrations among species (Figure
4B) follows the trend in lipid content (Figure
4A). Among composite samples of bluefish, black sea bass, and tautog,
PCB concentrations increased with lipid content in each species (Figure
6A). Correlation of PCBs and lipids in the pelagic bluefish (r =
0.724) was similar to the correlation for the two reef fish, black
sea bass (r = 0.763) and tautog (r = 0.814) (Table
1). PCB concentrations in tautog composites were also correlated
with tautog length (r = 0.695; Table 1; Figure
6B), probably because tautog length is correlated with the muscle
lipid content (r = 0.559).
Organochlorine Pesticides
A significant number of station means for chlorinated pesticides were
below the MDL (i.e., 29% for bluefish, 96% for summer flounder,
73% for black sea bass, and 80% for tautog). Four chlorinated pesticides
were not detected in any of the 56 muscle composites (Appendix Table
C5). The complete listing of the remaining 13 chlorinated pesticides
is given in Appendix Tables C10-C13, and
summarized in Table 3.
The pesticide composition in bluefish and black sea bass composites
was dominated by total chlordanes and total DDTs. No pesticide analyte
had a concentration greater than 3 times the MDL in summer flounder composites.
The pesticide composition in tautog was dominated by total DDTs. No interstation
differences were detected for total chlordanes or total DDTs in any fish
species.
When total DDT concentrations were compared among species, the observed
trend was similar to the pattern for lipids and PCBs
(i.e., bluefish > black sea bass = tautog > summer flounder; Figure
4). The DDTs were correlated with lipid content in black sea bass
(r = 0.918) and tautog (r = 0.757) composites, and to a
lesser degree in bluefish composites (r = 0.512) (Figure
6C). The DDTs were correlated
with tautog composite length (Figure 6B),
which covaried with the lipid content (<Table 1).
The DDTs correlated with PCBs
(Figure 6D) in bluefish, black sea bass,
and tautog muscle composites. This relationship was stronger for black
sea bass and tautog composites than for bluefish composites (Table
1).
The trend for total chlordane concentrations among species followed
the same order as PCBs, DDTs,
and lipids for bluefish, black sea bass, and summer flounder composites
(Figure 4D).
Average sums of DDTs and metabolites for all composite samples (i.e.,
0.014-0.16 µg/g) were below the FDA action level of 5.0 µg/g
(ppm) wet weight, and average sums of chlordanes for all composite samples
(i.e., 0.04-0.08 µg/g) were below the FDA action level of
0.3 µg/g (ppm) wet weight (FDA 1986, 1987).
PAHs
Nineteen of the 24 PAHs were undetected in any fish muscle composite
(Appendix Table C5). The detected PAHs were
not consistently found among fish species. The concentrations of the
five PAHs that were detected in at least one composite are listed in
Appendix Table C14. Acenaphthene was found
only in bluefish composites, while benz[a]anthracene was detected only
in summer flounder, black sea bass, and tautog composites. The PAH 2-methylnaphthalene
was detected only in summer flounder and tautog. Summer flounder was
the only species with measurable concentrations of naphthalene and 1-methylnaphthalene.
The station mean for acenaphthene in bluefish was above the MDL for all
three stations (Table 3; Appendix Table
C14). The station mean for benz[a]anthracene was above the MDL for
one black sea bass station and two tautog stations.
No apparent explanation is evident for the differing presence of these
PAH compounds among the species. The absence or low concentration levels
of PAHs are expected as these compounds are extensively metabolized by
the fish hepatic microsomal enzymes, and the metabolites are temporarily
stored in the bile until their excretion (Deshpande 1989; Varanasi et
al. 1989).
2,3,7,8-Substituted PCDD
and PCDF Congeners
Thirteen of the seventeen dioxin and furan congeners were not detected
above the MDL in any muscle composite samples analyzed for this study
(Appendix Table C5), and were not included
in subsequent calculations. Complete listings of the analytical results
for the four detected dioxins and furans in each composite are given
in Appendix Table C15, and summarized in Table
3.
The dominant PCDD and PCDF congeners were 2,3,7,8-TCDD, octachlorodibenzo[p]dioxin,
2,3,7,8-TCDF, and 1,2,3,4,6,7,8-hepatochlorodibenzo[p]dioxin. The concentrations
of 2,3,7,8-TCDD, considered the most toxic dioxin congener (EPA 1989),
were below the MDL of 1.63 pg/g wet weight (pptr) in all summer flounder
and black sea bass composites. Concentrations of 2,3,7,8-TCDD were below
the detection limit in 10 of 14 tautog composites, and near the detection
limit in the other four tautog composites (i.e., ranging from
2.36 to 3.39 pg/g wet weight). The overall tautog species mean for 2,3,7,8-TCDD
was less than the MDL (Table 3).
The concentrations of 2,3,7,8-TCDD in 4 of 14 bluefish composites were
below the MDL. The concentrations ranged from 1.82 to 3.76 pg/g in 9
of 10 remaining bluefish composites. The remaining bluefish composite
(i.e., #113; Station BL3) contained 7.27 pg/g of 2,3,7,8-TCDD.
This composite also contained the highest concentrations of PCBs
(i.e., 566 ng/g), DDTs
(i.e., 268 ng/g), chlordanes
(i.e., 62.4 ng/g), and approximately twice (i.e., 13.2%)
the average bluefish muscle lipid content of all composites analyzed
in this survey. This composite exhibited the highest average composite
weight (i.e., 3.6 kg; Appendix Table A2),
and included the heaviest individual (i.e., 4.0 kg; Appendix Table
A2) of all bluefish collected. The concentration of 2,3,7,8-TCDD
in the other three bluefish muscle composites from Station BL3 were below
the MDL of 1.63 pg/g. The station means and the species mean for bluefish
muscle composites were only about 50% greater than the MDL of 1.63 pg/g
(Table 3). All values for 2,3,7,8-TCDD for
all composites are well below the FDA guidance level of 25 pg/g for limited
consumption (Cordle 1981; Green 1981; Niemann 1986).
The furan congener 2,3,7,8-TCDF is about one-tenth as toxic as 2,3,7,8-TCDD
(EPA 1989). No summer flounder composites had concentrations of 2,3,7,8-TCDF
that were above the MDL. Two station means for black sea bass (i.e.,
Stations SB2 and SB3) were only slightly higher than the MDL, no 2,3,7,8-TCDF
was found at the third (i.e., Station SB1), and the species mean
was below the MDL. For bluefish and tautog, the three station means were
above the MDL, with two station means for bluefish and one for tautog
being greater than 3 times the MDL. For tautog, spatial differences were
not significant (P = 0.13). In contrast, the mean concentration
of 2,3,7,8-TCDF at bluefish Station BL3 (i.e., 7.26 pg/g) was
statistically higher than those at Stations BL1 and BL2 (P = 0.02).
The higher 2,3,7,8-TCDF concentrations at Station BL3 can partially be
explained by the longer, heavier, and more dense bluefish at Station
BL3 (Appendix Table A2).
The sum of 2,3,7,8-TCDD toxic equivalent (2,3,7,8-TCDD TE; EPA 1994)
values for the three TCDD congeners and one TCDF congener detected in
at least one composite ranged from a baseline of 0.90 pg/g (using one-half
MDL when all four compounds were below the MDL) to 8.3 pg/g. The 13 congeners
that were not detected in any of the 56 composites had a 2,3,7,8-TCDD
TE value of 6.2 pg/g wet weight, at concentrations of one-half MDL.
CONCLUSIONS
- Total mercury levels in all fish composites were <0.11 µg/g
wet weight, which is an order of magnitude below the FDA action level
of 1.0 µg/g wet weight.
- PCB concentrations in black sea bass were higher at Station SB3
(i.e., entrance to Ambrose Channel) than at stations farther
south along the New Jersey coast.
- PCB and organochlorine pesticide concentrations were relatively
low, and were correlated with the lipid content of the muscle tissue.
Bluefish, with its higher lipid content, had both the highest mean
PCB and pesticide concentrations. The individual bluefish composite
with the highest lipid content also exhibited the highest PCB and pesticide
concentrations.
- The Aroclor-based estimate maximum of 0.9 µg/g wet weight
was below the FDA tolerance level of 2.0 µg/g (ppm) for PCBs.
- Average sums of DDTs and metabolites for all composite samples (i.e.,
0.014-0.16 µg/g) were below the FDA action level of 5.0 µg/g (ppm)
wet weight.
- Average sums of chlordanes for all composite samples (i.e.,
0.04-0.08 µg/g) were below the FDA action level of 0.3 g/g (ppm) wet
weight.
- Consistent with findings in the scientific literature, PAHs were
largely undetected.
- All but one of the 56 fish muscle composites analyzed in this study
had concentrations <4 pg/g for 2,3,7,8-TCDD, the most toxic dioxin
congener.
- Concentrations of 2,3,7,8-TCDD in all composite samples were below
the FDA guidance level of 25 pg/g (pptr) wet weight for limited consumption.
- The bluefish composite with the highest PCB, organochlorine pesticide,
and lipid content also had the highest 2,3,7,8-TCDD concentration (i.e.,
7.3 pg/g), which was still well below the FDA guidance level of 25
pg/g (pptr) wet weight for limited consumption.
ACKNOWLEDGMENTS
The
authors acknowledge the assistance of Donald Macmillan and Stuart Wilk
(NMFS James J. Howard Marine Sciences Laboratory) in fish collection,
and of
Jeffrey Cross (NMFS James J. Howard Marine Sciences Laboratory) and Joel OConnor
and Douglas Pabst (U.S. Environmental Protection Agency - Region II) in technical
review of the manuscript. This research received financial support, in part,
from the U.S. Environmental Protection Agency - Region II, and the U.S. Army
Corps of Engineers - New York District.
REFERENCES
CITED
ACE [U.S. Army Corps of Engineers]-EPA
[U.S. Environmental Protection Agency]. 1992. Guidance for performing
tests on dredged material proposed for ocean disposal.
Draft report. New York, NY: ACE - New York District, and, EPA - Region
II; 81 p.
Ballschmiter, K.; Zell, M. 1980. Analysis of polychlorinated
biphenyls by capillary gas chromatography. Fresenius Z. Anal.
Chem. 302:20-31.
Battelle. 1996. Dioxin levels in flesh of four species of recreational
fish from the New York Bight Apex. Final report. EPA Contract
No. 68-C2-0134. New York, NY: U.S. Environmental Protection Agency
- Region II; 103 p.
Bowman, M.J.; Wunderlich, L.D. 1976. Distribution of physical
properties in the New York Bight Apex. Am. Soc. Limnol. Oceanogr.
Spec. Symp. 2:58-68.
Cordle, F. 1981. The use of epidemiology in the regulation of
dioxins in the food supply. Regul. Toxicol. Pharmacol.
1:379-387.
Deshpande, A.D. 1989. High performance liquid chromatographic
separation of fish biliary polynuclear aromatic hydrocarbon metabolites. Arch.
Environ. Contam. Toxicol. 18:900-907.
EPA [U.S. Environmental Protection Agency]. 1984a. Analytical
reference standards and supplemental data: the pesticides and
industrial chemicals repository. EPA Doc. 600/4-84/082;
207 p. Available from: EPA Environmental Monitoring Systems
Laboratory, Las Vegas, NV.
EPA [U.S. Environmental Protection Agency]. 1984b. Rules and
regulations. Appendix B to Part 136definition and procedure
for the determination of the method detection limit. Revision
1.11. Fed. Regist. 49(209).
EPA [U.S. Environmental Protection Agency]. 1989. Interim measures
for estimating risks associated with exposures to mixtures of
chlorinated dibenzo-p-dioxins and dibenzofurans (CDDs and CDFs)
and 1989 update. EPA Doc. 625/3-89/016. Available from:
EPA Office of Research and Development, Cincinnati, OH.
EPA [U.S. Environmental Protection Agency]. 1993a. Laboratory
methods manual estuaries. EPA Doc. 600/4-91/024; 289 p.
Available from: EPA Office of Research and Development,
Cincinnati, OH.
EPA [U.S. Environmental Protection Agency]. 1993b. Method 1613,
Rev. A: tetra- through octa-chlorinated dioxins and furans by
isotope dilution HRGC/HRMS. Appendix: modifications to Method
1613 for use in the analysis of 2,3,7,8-TCDD and 2,3,7,8-TCDF
only. In: Analytical methods for the determination of
pollutants in pulp and paper industry wastewater. EPA Doc. 821-R-93-017:1-60.
Available from: EPA Water, Engineering, and Analysis Division,
Washington, DC.
EPA [U.S. Environmental Protection Agency]. 1994. Method 8290:
polychlorinated dibenzodioxins (PCDDs) and polychlorinated dibenzofurans
(PCDFs) by high-resolution gas chromatography/high-resolution
mass spectrometry (HRGC/HRMS). Revision O. In: Test methods for
evaluating solid waste, physical/chemical methods. EPA Doc.
SW-846; 71 p. Available from: EPA Office of Solid Waste,
Arlington, VA.
Erickson, M.D. 1992. Analytical chemistry of PCBs. Chelsea,
MI: CRC Press; 508 p.
FDA [U.S. Food and Drug Administration]. 1986. FDA action levels
for unavoidable pesticide residues in food and feed commodities. FDA
Policy Compliance Guide 7141.01-B. Rockville, MD: Public
Health Service.
FDA [U.S. Food and Drug Administration]. 1987. FDA action levels
for unavoidable pesticide residues in food and feed commodities. FDA
Policy Compliance Guide 7141.01-B. Rockville, MD: Public
Health Service.
FDA [U.S. Food and Drug Administration]. 1991. Tolerances of
polychlorinated biphenyls (PCBs). 21 [U.S.] Code Fed. Regul.
ch. 1 (Apr. 1, 1991), sect. 109.30(a)(7), p. 96-98.
Green, J. 1981. Dioxin in fish. FDA Talk. Pap. T81-32:1-2.
Available from: U.S. Public Health Service, Rockville,
MD.
Harvey, R.G. 1997. Polycyclic aromatic hydrocarbons. New York,
NY: Wiley-VCH; 667 p.
Hites, R.A. 1985. Handbook of mass spectra of environmental
contaminants. Boca Raton, FL: CRC Press; 433 p.
IUPAC [International Union of Pure and Applied Chemistry]. 1993.
The international harmonized protocol for the proficiency testing
of (chemical) analytical laboratories. Pure Appl. Chem.
65(9):2123-2144.
Jandel Corporation. 1995. SigmaStat 2.0 for Windows. Version
2.0. Available from: Jandel Corporation, San Rafael, CA.
Kennedy, D. 1979. Action level for mercury in fish, shellfish,
crustaceans, and other aquatic animals. Fed Regist. 44(14):3990-3993.
Krahn, M.M.; Wigren, C.A.; Pearce, R.W.; Moore, L.K.; Bogar,
R.G.; MacLeod, W.D., Jr.; Chan, S.-L.; Brown, D.W. 1988. Standard
analytical procedures of the NOAA National Analytical Facility,
1988: new HPLC cleanup and revised extraction procedures for
organic contaminants. NOAA Tech Memo. NMFS FNWC-153; 52
p.
Niemann, R.A. 1986. Surrogate-assisted determination of 2,3,7,8-tetrachlorodibenzo-p-dioxin
in fish by electron capture capillary gas chromatography. J.
Assoc. Off. Anal. Chem. 69:976-980.
NIST [National Institute of Standards and Technology]. 1993.
NIST/NOAA/NS&T/EPA EMAP Intercomparison Exercise Program
for Organic Contaminants in the Marine Environment. Exercise
description and results: exercise material mussel tissue V (QA93TIS5).
Gaithersburg, MD: National Institute of Standards and Technology;
62 p.
NOAA [National Oceanic and Atmospheric Administration]. 1989.
A summary of data on tissue contamination from the first three
years (1986-1988) of the Mussel Watch Program. NOAA Tech.
Memo. NOS OMA 49; 154 p.
NOAA [National Oceanic and Atmospheric Administration]-FDA [U.S.
Food and Drug Administration]-EPA [U.S. Environmental Protection
Agency]. 1986. Report of 1984-86 federal survey of PCBs in Atlantic
Coast bluefish - data report. Washington, DC: NOAA, FDA, & EPA;
179 p. [NTIS Access. No. PB86-218070/AS].
Parris, R.M. 1995. Report for the 1994 NIST/NOAA NS&T/EPA
EMAP organics intercomparison exercises (mussel VI and sediment
IV). Gaithersburg, MD: National Institute of Standards and Technology;
15 p.
Sander, L.C.; Wise, S.A. 1997. Polycyclic aromatic hydrocarbon
structure index. NIST Spec. Publ. 922; 105 p.
SAS [SAS Institute, Inc]. 1989. SAS/STAT users guide.
Version 6. 4th ed. Vol. 2. Cary, NC: SAS Institute Inc.; 846
p.
Sloan, C.A.; Adams, N.G.; Pearce, R.W.; Brown, D.W.; Chan, S.-L.
1993. Northwest Fisheries Science Center organic analytical procedures. In:
Lauenstein, G.G.; Cantillo, A.Y., eds. Sampling and analytical
methods of the National Status and Trends Program, National Benthic
Surveillance and Mussel Watch Projects, 1984-1992. Vol. 4. Comprehensive
descriptions of trace organic analytical methods. NOAA Tech.
Memo. NOS ORCA 71; 182 p.
Sokal, R.R.; Rohlf, F.J. 1981. Biometry. 2nd ed. New York, NY:
W.H. Freeman; 859 p.
Stemmler, E.A.; Hites, R.A. 1988. Electron capture negative
ion mass spectra of environmental contaminants and related compounds.
New York, NY: VCH Publishers; 390 p.
Valente, R.; Strobel, C.J.; Schimmel, S.C. 1992. Environmental
Monitoring and Assessment Program, EMAP-estuaries Virginian Province.
1992 quality assurance project plan. Revision O. Narragansett,
RI: EPA Environmental Research Laboratory; 83 p.
Varanasi, U.; Stein, J.E.; Nishimoto, M. 1989. Biotransformation
and disposition of polycyclic aromatic hydrocarbons (PAH) in
fish. In: Varanasi, U., ed. Metabolism of polycyclic aromatic
hydrocarbons in the aquatic environment. Boca Raton, FL: CRC
Press; p. 93-150.
Willie, S.; Berman, S. 1995. NOAA National Status and Trends
Program. Eighth round intercomparison exercise results for trace
metals in marine sediments and biological tissues. NOAA Tech.
Memo. NOS ORCA 83; 50 p.
Wise, S.A.; Benner, B.A., Jr.; Christensen, R.G.; Koster, B.J.;
Kurz, J.; Schantz, M.M.; Zeisler, R. 1991. Preparation and analysis
of a frozen mussel tissue reference material for the determination
of trace organic constituents. Environ. Sci. Technol.
25(10):1695-1704.
Zdanowicz, V.S.; Finneran, T.W.; Kothe, R. 1993. Digestion of
fish tissue and atomic absorption analysis of trace elements. In:
Lauenstein, G.G.; Cantillo, A.Y., eds. Sampling and analytical
methods of the National Status and Trends Program, National Benthic
Surveillance and Mussel Watch Projects, 1984-1992. Vol. 3. Comprehensive
descriptions of elemental analytical methods. NOAA Tech. Memo.
NOS ORCA 71; 219 p.
Zdanowicz, V.S.; McKinley, B.; Finneran, T.[W.]; Leimburg, E.
1992. Trace metals in midwater fish. In: Second annual
report on monitoring the biological effects of sludge dumping
at the 106-Mile Dumpsite. Report prepared for: National
Ocean Service, Coastal Monitoring and Bioeffects Assessment Division,
Rockville, MD; 157 p.