Toxic Chemicals and Trace Metals from Urban and Rural Louisiana Lakes: Historical Profiles and Toxicological Significance W . J A M E S CATALLO,*,+ M A T T H E W SCHLENKER,* ROBERT P. GAMBRELL,§ AND BARBARA S. SHANEl Laboratory for Ecological Chemistry and Toxicology (SVM), Institute for Environmental Studies, and Wetland Biogeochemistry Institute, Louisiana State University, Baton Rouge, Louisiana 70803-7511, and Radiation Protection Division, Louisiana Department of Environmental Quality, Baton Rouge, Louisiana
~~~~
Sediment cores collected from lakes in rural and urban/industrial areas of Louisiana were dated using 137Cs,sectioned, and analyzed for a wide range of pollutant chemicals deposited during the period 19501991. Mutagenicity testing also was performed on extracts from the core sections. Statistical and other comparisons of chemical data indicated that annual loadings of polycyclic aromatic hydrocarbons, chlorinated chemicals, and trace metals were not significantly different in the urban lake vs the rural lake over the historical period compared. Mutagenic activity was detected in both lakes, primarily in sediments deposited between 1955 and 1980, with minimal activity before and after that period. This was a time of widespread industrial and agricultural activity in Louisiana, before restrictions on chemical releases to the environment were instituted and enforced.
The effects of environmental pollution on ecological and human health is of concern in Louisiana and elsewhere ( 1-3). Declines in ecosystem function or “health”resulting from chronic pollution occur in concert with other modifications (e.g.,fragmentation of ecosystems) operating over decades or longer (4). Human health effects from exposure to environmental toxicants also may require extended periods to manifest and can reflect prolonged exposure to varying levels of pollution in conjunction with other factors (5). In both cases, the relationships between pollution and ecological and human health can be understood more fully when data on historical levels and types of environmental chemicals are integrated with estimates of exposure and biological effects (4, 6 ) . This is the case especially when areas with different kinds and degrees of industrial and agricultural activities are compared. Records of the distribution and abundances of natural and anthropogenic chemicals can be preserved in sequentially deposited, undisturbed sediment profiles in aquatic systems (7, 8). Geochronological data have been used to document historical changes in levels and compositional complexity of chemicals found in sediment profiles, and these have been related to human and natural processes (9- 13).
The purpose of this work was to examine the historical profiles of toxic chemicals in lake sediments in rural vs industrialized areas of Louisiana over the last 50 years and to evaluate changes in mutagenicity of sediment extracts over this period. Related objectives were (a) to determine any significant differences in historical pollutant loadings between rural and urbadindustrial areas in Louisiana, (b) to identify the sources of chemicals based on analysis of chemical distributions within samples, and (c) to evaluate these approaches for comparative impact and health assessment and epidemiological studies.
Methods Study Sites. Two sites were selected in natural freshwater lakes in industrialized and rural areas of Louisiana (Figure 1). The lakes receive sediments primarily from proximal rather than allochthonous sources (Le., no direct connections to large rivers). There has been no dredging or filling at either site and no direct inputs of point source wastes. Chemical inputs were restricted to the atmosphere and aquatic and terrestrial sources in the watershed. The rural site was Larto Lake (LL), Catahoula Parish, LA, a backwater lake of approximately 1000ha (1.5). The nearest city (Alexandria)is approximately 35 mi west of LL, and as of 1990, there were four hazardous waste sites in Catahoula Parish located ’30 mi away from LL ( 1 6 ) . The urban/ industrial site was Lac des Allemands (LD),located in the Barataria Basin, LA (1 7, 18). The lake is located in a region +
Laboratory for Ecological Chemistry and Toxicology, SVM, LSU.
5
Wetland Biochemistry Institute, LSU. Department of Environmental Quality.
* Institute for Environmental Studies, LSU. 1 Louisiana
1436 1 ENVIRONMENTAL SCIENCE &TECHNOLOGY / VOL. 29, NO. 6,1995
0013-936X/95/0929-1436509.00/0
Q 1995 American Chemical Society
Arkansas I (
FIGURE 1. Location of the study sites in rural and industrial areas of Louisiana.
sometimes referredto as the “industrialcorridor”or “cancer alley”. At least 10 major industrial facilities are located within 10 mi of the shores of LD, and there are some 25 CERCUS waste sites in the same general area (16). Sedimentation rates at five sites in LD have been measured between 0.34 and 0.9 cm/yr (18). Further data on the hydrology,ecology, and historical land use in the lakes can be found elsewhere (15,17, 18). SamplingandAnalytical Procedures. Aluminum core tubes (10 cm x 1.22 m) were cleaned with soap, 4 N acid, organic solvents, and distilled water prior to use. Sampling was performed from an anchored boat within an area of approximately20 x 20mz, in 1.5-2.5 mofwater. Multiple (6-8) samples were collected at each site using a handheld corer with a remote activated closing device, and care was taken to minimize compression during sampling (19). The sediment level in the cores was marked before storage and immediately prior to sectioning. Duplicate cores from each site were examined using X-ray radiography (20). Vertical thin sections (0.5 cm) of the sediment columnwere exposed over photographic film using a Norelco MG-150 X-ray generator. The negatives then were examined for horizontal laminations indicative of superposition of sediments. I3’Cs Analysis. 137Csanalyses were performed on 3 cm sections using a Canberra Series Model 90 Gemini system (low background, 4096 efficiency coaxial germanium detector, 60 000 s live-time count). Interpretation of profiles was performed using methods described elsewhere (21). Other radioisotopes monitored in each sample included (a) r°K, which can be associated with fertilizers, (b)2z6Ra, a naturally occurring radioisotope found in sediments, surfacewaters, oil fieldbrines, and the atmosphere, and (c) 7Be,the presenceofwhichinsurfacesediment scanindicate rapid sedimentation (22). PAHs and Heterocycles. Core sediments (40-60 g dry) were weighed into extraction thimbles, amended with surrogate standards (2-fluorobiphenyl,2-fluoronaphthalene), and Soxhlet extracted for 24 h in dichloromethane (DCM). This was followed by volume reduction, chemical drying, and analysis by gas chromatography-mass spectrometry (GC/MS;HP5890gas chromatographwitha5970
mass selective detector in the selected ion monitoring (SIM) mode). The molecular ion and secondary ions (e.& mlz 126 for the benzofluoranthenes, benzopyrenes, and perylene) of the analytes were monitored. SIM data were referenced to external Calibration files of 16 polycyclic aromatic hydrocarbon (PAH)standards (EPA 610 PAH mixture with 100 pg of perylene added, Supelco Inc. Bellafonte,PA) for quantitation. Sixperdeuterated internal standards (naphthalene, acenaphthene, phenanthrene, chrysene,perylene, and terphenyl)were includedwitheach extract analyzed. Detection limits (-250 pg) were determined using an internal detection standard (5-fluoroindole). Quantitation reproducibility was determined from the results of 5 replicate runs of the calibration standards at concentrations in the middle of the linear range. This was found to be 88%-9596, depending on the compound (heavier analytes showed the highest variability). Analyte retention times were reproducible within 0.3 min. In the results, the concentrations (ng/g)of the PAHs are presented for each section analyzed. Some compounds (e&, fluoranthene and pyrene) eluted within the same SIM window and had approximately equal concentrations in all samples. The summed 1x1 concentrations of these compounds are reported. Total resolved PAHs (TRPAHs) were determined by adding the concentrations of PAHs, alkyl homologues, and sulfur heterocycles (e.g., dibenzothiophenes). AnnualPAHfluxesto eachsectionwereestimated using the I3’Cs sedimentation rates (cm/yr) and the dry weight ofsediment in eachsection. With the sedimentation rate and the dimensions of the core sections known, the annual sediment mass accumulated through the surface area of the core mouth was calculated (g c m P yr). This was multiplied by PAH concentration (ng/gl to give flux (ng c m 2 yr-I). Selected core section extracts also were analyzed in the full scan mode, and mass spectra were identified manually. Chlorinated Hydrocarbons. Screening for PCBs (Aroclors 1242,1248,1254,and 12601,lindane, DDT. DDD, DDE, aldrin, dieldrin, toxaphene, and endrin was performed on all sections from the LL and LD cores. Dried, ground sediment samples (10 g) were Soxhlet extracted (24h in 1:l hexane/acetone), followed by volume reduction, chemical drying, and column chromatography on Florisil. Finished extracts were analyzed using the HP 5890 gas chromatography with a DB-5 capillary column and electron capture detection. Trace Metals Analysis. Potentially bioavailable metals were extracted using hot acid (23) and analyzed using inductively coupled argon plasma emission spectromeuy. Mercurywas analyzed using cold vapor atomic absorption spectrometry (24). Sediment Characterization. Particle size distribution andorganic mattercontent were determined ineachsection using standard hydrometer and combustion methods, respectively (25). Mutagenicity Evaluation. Wet sediments (100 9, were extractedultrasonicallyinan excess ofDCMwithanhydrous NaS04 (Heat Systems XU020 ultrasonicator, 20 min extraction; 3 / 4 in. horn; ice bath), followed by rotary evaporation and solvent exchange with DMSO. Percent solids in each section was determined gravimetrically after drying. Thefinishedextractswereevaluated fortheabilityto revert auxotrophic (his/bio-) mutants ofSalmonellafyphimurium to the wild type in the Ames mutagenicity assay IAMA). VOL. 29, NO. 6.1995 I ENVIRONMENTAL SCIENCE &TECHNOLOGY m (437
LAC DES ALLEMANDS
LARTO LAKE
tl Q
5
z
c
0 ._
0
al
v)
0
80%
40%
20%
60%
100%
CLAY SILT SAND
Composition (%)
0
40% 80% 20% 60% 100% Composition (Oh)
FIGURE 2. Sediment particle size distribution with depth at the study sites.
Strains TA 98 and TA 100 were used in plate incorporation assays with and without liver S-9 preparations. A comprehensive description of the experimental methods and data interpretation protocols for this assay are provided elsewhere (26, 27). The extractionprocedure for the AMAwas differentfrom that used for the PAH analyses. In order to compare the results of both analyses, recovery efficiencieswere determined for each extraction method. Mixtures of naphthalene, benzothiazole,pemuoronaphthalene, l-methyhaphthalene, 2,3-dimethylquinoxaline,1,2-dimethylnaphthalene, dihenzofuran, phenanthrene, benzomquinoline, l-methylphenanthrene, and 1,8-dimethylphenanthrene were extracted from sediments using both procedures and then analyzed by GC with flame ionization detection.
Results and Discussion Comparability of Cores within Sites. Measurement of sediment column length after collection and prior to sectioningindicated that core compaction during sampling/ storage was minimal (
