Printer Friendly

Movement of mercury from a contaminated city park in Oxford into the Coosa River, Alabama.


Soil samples from an Oxford, AL, city park and from sites within the same watershed but outside the park area were collected and analyzed for mercury (Hg) content using cold vapor atomic absorption. Soil from the park had a mean Hg concentration of 3.27 [+ or -] 0.59 [micro]g/g (mean [+ or -] SEM., n = 12). Mean Hg concentration for soil collected outside of the park was 0.04 [+ or -] 0.01 [micro]g/g (n = 12). Sediment samples were collected from Snow Creek as it flows through the Oxford park and from Snow Creek upstream of the park. When analyzed for Hg, sediment from the park had a mean concentration of 1.21 [+ or -] 0.13 [micro]g/g (n = 19), while sediment collected upstream of the park had a concentration of 0.12 [+ or -] 0.03 [micro]g/g (n =19). A profile of Hg concentrations in sediments collected from Snow Creek and Choccolocco Creek downstream of the park showed areas with higher Hg concentrations extending 36 miles to the Coosa River. Earthworms (Lumbricus terrestris) grown in soil from the Oxford park had significantly higher tissue levels of Hg than earthworms grown in background soil. The data support the hypothesis that Hg from Oxford park soil is being transported downstream from Snow Creek to the Coosa River.


We have identified a city park, Oxford Lake Park, in Oxford, AL, where the concentration of Hg in the soil is significantly higher than that found in soil samples collected from areas surrounding the park (Figure. 1). We propose that fill deposits used as landscape materials in the park were contaminated with mercury (Hg) from an industrial source. It is not clear from public records when this landscaping occurred, although it was prior to 2003. A likely source for this fill material was an industrial chlor-alkali facility where a mercury cell process had been used to produce chlorine gas for the manufacture of chlorinated biphenyls. The mercury cell was operational from 1952 until 1969 (Bluemilk, 2001). Unlike organic contaminants which generally undergo decomposition over time due to natural oxidation or microbial metabolism, heavy metals are in the elemental state and cannot be further broken down. Once released into the environment, heavy metals such as Hg can constitute a long-lasting hazard. Although these metals are not degraded, they can be changed from one oxidation state to another depending on local conditions (Schwedt, 2001). Such changes affect both the mobility and the bioavailability of metals. Mercury in the environment can occur as elemental mercury ([Hg.sup.0]), ionic mercury ([Hg.sub.2.sup.2+], [Hg.sup.2+]) or alkylated mercury ([CH.sub.3][Hg.sup.+] or [([CH.sub.3]).sub.2]Hg). These forms can coexist and often interchange as a result of interactions with organic material (Si and Ariya, 2008), bacteria (Lefebvre et al., 2007), or flora (Leonard et al., 1998). Such changes can alter not only the water solubility and bioavailability of Hg, but also the volatility.


The oxidation state of Hg has a pronounced effect on its toxicity in humans. Inhaled elemental Hg vapor readily penetrates the lungs, and can distribute from the blood into the central nervous system (CNS) (Hamada and Osame, 1996). Mercury accumulation in the CNS produces a syndrome characterized by tremors, excitability, insomnia, and depression. Ionic Hg is toxic to the kidney, causing a form of glomerulonephritis (Liu et al., 2008). Most human exposure to methyl Hg results from the consumption offish. Almost all of the Hg in most fish species occurs as the alkylated form, [CH.sub.3][Hg.sup.+]. Methyl Hg readily crosses lipid barriers in the body, including the blood-brain barrier and the placenta. In the CNS, methyl Hg can cause cerebral edema, gliosis, and cerebral atrophy. In children exposed prenatally, methyl mercury can produce a form of cerebral palsy (Liu et al., 2008).

Snow Creek, a third order perennial stream, flows through Oxford Lake Park. It is relatively calm with alternating pools and riffles. Snow Creek joins Choccolocco Creek approximately one mile downstream from the park. Choccolocco Creek empties into Logan-Martin Reservoir on the Coosa River. We previously reported that fish collected from Snow Creek as it passes through Oxford Lake Park had significantly higher tissue levels of Hg than fish collected from Snow Creek upstream from the park (Kohute et al., 2006). The July 26, 2008, edition of the Anniston Star published a fish advisory for consumption of spotted bass from Choccolocco Creek due to high levels of Hg (Faulk, 2008). This advisory applies to a reach of Choccolocco Creek downstream from the confluence with Snow Creek. In the present study, data is presented that supports the hypothesis that soil from Oxford Lake Park is a source of the Hg found in the Choccolocco Creek fish.


Soil and Sediment Preparation

Twenty-eight grab samples of the Snow and Choccolocco Creek's bottom sediments were taken at convenient points averaging no more than 1.5-mile apart along a 36-mile longitudinal traverse from its headwater to its confluence with Logan Martin Reservoir (Figure 1). These grab sediment samples from the creek bottom were retrieved by a non-metal trowel and placed in a sterile 120 ml glass jar with a Teflon-lined cap.

Between 4.65 and 7.57 miles along the transverse, a spike in Hg level was detected (Figure 4). At the location of this spike in Hg level, biased soil samples were grabbed from the floodplain of the creek. These samples were taken less than 10 feet from the creek bank. Three grab samples of soil, which were from the upper 12 cm of soil, were mixed thoroughly and placed in a sterile 120 ml clear glass jar with a Teflon-lined cap. To avoid cross contamination, the Teflon spade used to extract soil was brushed clean after each sample. Any grass or rock was separated from each field sample.


Vertical profiles of soil in the same location as the biased samples and in the adjacent bottom of Snow Creek were extracted by a 30 cm long AMS Slotted Soil Recovery Probe with a 2.2 cm diameter butyrate liner. The recovered profile was sealed with plastic end caps and wrapped in aluminum foil for transportation and storage.

For determination of total concentration of Hg in the soil, field samples were divided into triplicate analytical samples, each weighing 2.000 g ([+ or -] 0.001 g). Each two gram analytical sample was placed in an acid washed 500 mL BOD bottle and partially dissolved in 50 mL ultra-pure water (17 megaohm resistance). Mercury in the samples was oxidized to [Hg.sup.2+] using 0.5 mL trace metal grade sulfuric acid, 0.25 mL trace metal grade nitric acid, 1.5 mL potassium permanganate solution (5% wt/wt), and 0.8 mL of potassium persulfate solution (5% wt/wt). Samples were then heated at 95 [degrees]C for two hours. Samples were allowed to cool to room temperature. Immediately before analysis, 1.0 mL of hydroxyiamine solution (12% wt/wt) was added to each sample to reduce unreacted potassium permanganate. The solution phase from each BOD bottle was transferred to an analyzer tube by syringe with a sediment filter attachment. Sediment samples were collected and processed in a similar manner.

Earthworm Treatment and Preparation

Earthworms, Lumbricus terrestris, were obtained from DMF Bait Company (Waterford, MI). Due to a two-year severe local drought during the time of this research, we were notable to collect earthworms from local soils. Worms were housed in 1.38 x [10.sup.-2] [m.sup.3] styrofoam boxes filled with either contaminated soil collected from the Oxford park or background soil collected from a local field. The background soil had a Hg concentration of 0.03 [+ or -] 0.02 [micro]g/g (mean [+ or -] S.D., n = 3). Worms were added to the boxes on day one. After two weeks, ten worms were taken from each box for Hg analysis. Worms ranged in size from 10-20 cm at the time of harvesting. Worms were also harvested for analysis at four-week and six-week time intervals. After harvesting, worms were measured and euthanized by cooling. Worms were frozen at -70 [degrees]C for a minimum of 24 h and then freeze-dried (VirTis Freezemobile 12) for 3-4 days. Freeze-dried tissue was pulverized with a mortar and pestle. Triplicate 2.00 g samples were taken from each batch of worms and digested with a mixture of 15 mL of trace metal grade nitric acid and 2.0 ml of 30% hydrogen peroxide (Fisher OPTIMA grade). Digested tissue was dissolved in 7% trace metal grade nitric acid and filtered (Fisherbrand Q8 filter paper) into 50 ml volumetric flasks. The mixture was diluted to 50 mL with 7% nitric acid. Prior to analysis, samples were further digested in sulfuric acid, nitric acid, potassium persulfate, and potassium permanganate by heating to 95 [degrees]C for one h. This treatment insured that all forms of Hg were converted to [Hg.sup.2+]. Immediately before analysis, 1 mL of 12% hydroxylamine (wt/wt) was added to each sample.

To determine the fraction of Hg contained in the worms that could be attributed to soil in the worm's gastrointestinal (GI) tract, five worms were randomly selected and carefully dissected to remove the intact GI tract. For each worm, the GI tract (and the soil it contained) was separated from the remaining tissue, and both were freeze-dried. Masses of the resulting dried materials were determined to three decimal places.

Mercury Analysis

All reagents used in sample preparation and Hg analysis were trace metal grade. Ultra-pure water (17 megaohm resistance) was used to prepare all solutions. Standard Hg solutions were prepared from a Fisher Certified Mercury Reference Solution, Lot No. 055614-24. All glassware was acid washed prior to each assay. Samples were analyzed for total Hg using USEPA Method 245.1, Manual Cold Vapor Technique (USEPA, 1983). Mercury analysis was conducted using a CETAC Quick Trace Mercury Analyzer M-6100 cold vapor atomic absorption Hg analyzer with an ASX-400 AutoSampler. All specimens were run in batches that included blanks (reagent and instrument), a five point standard calibration curve (standards of 0.0, 0.5, 1.0,2.5, 5.0, and 10.0 [micro]g/L with a linear correlation of 0.995 or better), and spiked specimens. Matrix spikes gave 85-90% recovery. Specimen split between two batches had a variation of less than 5%. The calculated method detection limit (MDL) for Hg in the liquid phase was 0.1 [micro]g/L. NIST Standard Reference Material 2976, mussel tissue, was used to validate the analytical procedure. The concentration of Hg in the reference material was reported as 0.061 [micro]g/g. The experimental value determined in this study was 0.058 [micro]g/g.

Statistical Analyses

GraphPad Prism, version 5.01 (GraphPad Software, Inc., San Diego, CA), was used for data analyses. The unpaired t test was performed to identify differences between contaminated groups and background groups. A one-way A NOVA was used to compare Hg uptake over time for earthworms. Results were considered significant at an alpha level of .05. Means are reported [+ or-] SEM.


In this study, evidence is presented that soil from Oxford Lake Park has significantly higher Hg levels than soil collected from sites within the same watershed but outside the contaminated park area. In addition, sediment collected from Snow Creek as it flows through the park is shown to have significantly higher Hg levels than sediment collected from Snow Creek upstream of the park. A profile of Hg concentrations in Snow Creek sediments extending from the park downstream into Choccolocco Creek, and continuing downstream for a distance of 36 mile to the Coosa River is presented. Earthworms, Lumbricus terrestris, grown in soil from the park are shown to accumulate significantly higher tissue concentrations of Hg compared to earthworms grown in soil having the area background concentration of Hg. Earthworms have been reported to accumulate Hg when grown in Hg contaminated soils (Ernst and Frey, 2007). As earthworms are a food source for many animals, they could serve as a mechanism by which Hg from soil enters the terrestrial food chain.

Mercury in Park Soil

We analyzed 12 grab samples collected from the upper 12 cm of soil at 7 sites locations in the floodplain of Oxford Lake Park located adjacent to the area of Snow Creek that produced the spike observed in the bottom sediment (Figure 4). These data were compared to 12 grab soil samples analyzed from 12 sites from the Snow Creek Watershed outside of the contaminated area of the park. Soil samples collected from the park had a mean Hg concentration of 3.27 [+ or-] 0,59 [micro]g/g whereas background soil samples had a mean of 0.04 [+ or -] 0.01 [micro]g/g (Figure 2). Soil samples collected from the park had significantly higher mean Hg levels than soil collected outside of the park but in the same flood plain (t = 5.448, P = <0.0001).


Mercury in Creek Sediment

Nineteen bottom sediment grab samples were collected from Snow Creek as it flows through Oxford Park and nineteen grab sediment samples were collected from Snow Creek upstream of the park. Park bottom sediment samples had a mean Hg concentration of 1.21 [+ or -]0.13 [micro]g/g. Bottom sediment samples collected from Snow Creek upstream of the park had a mean Hg concentration of 0.12 [+ or -] 0.03 [micro]g/g (Figure 3). Mercury levels were significantly higher in sediment from the park compared to upstream (t= 8.048, P = < 0.0001).


Figure 4 depicts the longitudinal profile of Hg levels in the bottom sediments of Snow Creek continuing downstream into Choccolocco Creek. Each point represents the mean of three replicate samples. These are biased samples collected where mud had collected on the stream bed. The profile starts (0 mile) at an unnamed tributary of Snow Creek upstream of the Oxford park and ends 36 miles downstream at the outflow point of Choccolocco Creek into Logan Martin Reservoir on the Coosa River (Figure 1-darkened stream segments). A spike in Hg concentrations occurs between 4.65 and 7.57 miles along the profile. This corresponds to the area where Snow Creek passes through the Oxford Lake Park and an adjacent shopping mall. Downstream from the park, Snow Creek feeds into Choccolocco Creek which outflows into Logan Martin Reservoir on the Coosa River.

Vertical Profiles of Hg Distribution

Vertical profiles were determined for Hg distribution in both the contaminated soil of the park and the contaminated creek bottom sediment in the park (Figure 5). Each point represents the mean of three replicate samples. Both the park soil and the creek sediment show some level of vertical variation of Hg levels. These profiles suggest that the eroded contaminated soil becomes more homogenized when it is redistributed in the stream bottom sediment load. The measurement of the vertical extent of the contamination both in the soil and sediment was restricted by the corer used to extract the soil/sediment profiles.


Mercury Uptake by Earthworms

In an earlier study (McLaughlin et al., 2007), we presented evidence that earthworms were bioaccumulating Hg from the contaminated soil in the park. This bioaccumulation could be an important mechanism in the vertical and horizontal re-distribution of Hg in soil. Due to the extended drought in the study area, naturally occurring earthworms were hard to sample. Consequently, a laboratory study of earthworm bioaccumulation was undertaken. Earthworms grown in a sample of background soil with a Hg level of 0.03 [micro]g/g had a total Hg tissue concentration of 0.05 [+ or -] 0.004 [micro]g/g (n = 9) dry weight (Figure 6). Earthworms grown in a sample of contaminated park soil having a Hg concentration of 0.52 [micro]g/g had a total Hg tissue concentration of 0.27 [+ or -] 0.07 (n = 9). Total tissue Hg levels were significantly higher in worms grown in park soil compared to worms grown in background soil (t = 3.049, P = 0.0077). While small increases in the Hg levels in worms were seen with increased exposure times, these differences were not found to be significant based on one-way ANOVA analysis.


Data presented in Table 1 was used to determine the fraction of Hg detected in the worm that could be attributed to soil contained the worm's GI tract. The first column of Table 1 gives the mean dry mass of five randomly selected worms as 0.732 [+ or -] 0.047 g (mean [+ or -] SEM). The second column shows the mean dry mass of the five GI tracts plus the soil contained therein to be 0.112 [+ or -] 0.007 g. Based on this data, the GI tract plus soil accounted for approximately 16% of the total worm mass (column 3). Worms grown in park soil had a mean Hg tissue concentration of 0.27 [micro]g/g. Using this tissue concentration to calculate the total body Hg concentration, the five worms in this study contain a mean of 0.195 [+ or -] 0.012 [micro]g of Hg per worm (Table 1). The park soil had a Hg concentration of 0.52 [micro]g/g. Assuming that the entire mass of the GI tract in each worm was soil, this calculates to a mean of 0.058 [+ or -] 0.004 [micro]g of Hg per GI tract. Thus, the Hg in the GI tract of a worm should only account for 30-35% of the Hg total in a worm.
Table 1. Hg concentration in worm tissue and GI tract.

Worm  Total Dry  GI Tract  Tissue Mass  Total Hg  GI Tract  Hg Mass %
      Mass (g)   Dry Mass  % (GI Tract    (mg)     Hg (mg)  (GI Tract
                   (g)      Mass/Total                       Hg/Total
                              Mass)                             Hg)

  1    0.639      0.129       20.19      0.173     0.067      38.73
  2    0.782      0.122       15.60      0.211     0.063      29.86
  3    0.660      0.100       15.15      0.178     0.052      29.21
  4    0.687      0.093       13.54      0.185     0.048      25.95
  5    0.892      0.118       13.23      0.241     0.061      25.31
Mean   0.732      0.112       15.54      0.195     0.058      29.81


The association of environmental Hg contamination with chlor-alkali facilities is well documented (Santschi et al., 1999; Zagury et al., 2006; Neculita et al., 2005). An article in the Anniston newspaper (Bluemink, 2001) chronicles Hg release into the local environment by a chlor-alkali facility that operated from 1952 until 1969. This article reports a 1970 advisory for Hg in fish from Choccolocco Creek downstream from this facility. While there are public accounts of Hg Choccolocco Creek downstream from this facility. While there are public accounts of Hg releases from the former chlor-alkali facility, there does not seem to be any public information explaining the use of soil contaminated with Hg as landfill at Oxford Lake Park. However, there is a public record indicating that the company that operated the chlor-alkali facility was involved in the development of the park. There are also public accounts that both Oxford Lake Park and Snow Creek as it flows through the park had been contaminated with chlorinated biphenyls (PCBs) (Dougan, 2000; USEPA, 2001). The Dougan report attributes PCB contamination of park soil to past flooding of Snow Creek. We propose that the Hg contamination and probably the PCB contamination (although we have yet to investigated this) are the result of contaminated soil use as fill deposits in the park. Our data show significantly higher Hg levels in soil collected from Oxford Lake Park compared to soil samples collected from the same drainage system outside of the contaminated park area. of Hg in the park soil resulted from the flooding of Snow Creek, it would be expected that flood prone areas upstream from the park would also have higher than background Hg levels. This was not found to be the case. Mercury levels in the sediment of Snow Creek as it flows through the park were significantly higher than Hg levels in sediment collected from Snow Creek upstream of the park. This argues against the downstream movement of Hg as the source of the current Hg contamination in park soil. Additional support of the premise that Hg is entering Snow Creek from the park soil and not from an upstream source comes from a previous study in which we found that fish collected from the park area of Snow Creek had significantly higher tissue Hg levels than fish of the same species collected from Snow Creek upstream of the park (Kohute, et al., 2006).

Our data support the hypothesis that Hg from the park soil is moving into Snow Creek and being transported to Choccolocco Creek and downstream into the Coosa River. While it is not clear from the public records how the soil in Oxford Lake Park became contaminated with Hg, our data indicate that the Hg is not remaining sequestered in the park soil. Mercury is moving into Snow Creek where it has entered the aquatic food chain. Mercury from park soil may be entering the terrestrial food chain due to uptake by earthworms. Earthworms grown in contaminated soil from the park had Hg tissue levels over five times higher than earthworms grown in background soil. Atmospheric deposition of Hg from coal burning power plants is a source of Hg contamination across northern Alabama (Nichols et al., 2002), and could account for the background level of Hg seen in our study. However, atmospheric deposition does not explain the high Hg level found solely in the park.


Bluemilk, E. 2001. Monsanto contamination now gets scrutiny after 30 years. In The Anniston Star. July 20. Anniston, AL, USA.

Dougan, K.R. 2000. PCBs found at Oxford's ball fields. In The Anniston Star. July 26. Anniston, AL, USA.

Ernst, G. and B. Frey. 2007. The effect of feeding behavior on Hg accumulation in the ecophysiologically different earthworms Lumbricus terrestris and Octolaseon cyaneum: a microcosm experiment. Soil Biology & Biochemistry. 39: 386-390.

Faulk, M. 2008. Officials advise against eating certain area fish. In The Anniston Star. July 26. Anniston, AL, USA.

Hamada, R. and M. Osame. 1996. Minamata disease and other mercury syndromes. In Toxicology of Metals. CRC Press, Boca Raton, FL, USA. 337-351.

Kohute, C.R., A. C. Nichols, D.A. Steffy, and M.E. Meade. 2006. Mobilization of mercury from contaminated soil into an aquatic food chain. Journal of the Alabama Academy of Science. 77: 160-167.

Lefebvre, D.D., D. Kelly, and K. Budd. 2007. Biotransformation of Hg (II) by cyanobacteria. Applied and Environmental Microbiology. 73: 243-249.

Leonard, T.L., G.E. Taylor, M.S. Gustin, and G.C.J. Fernandez. 1998. Mercury and plants in contaminated soils: 1. Uptake, partitioning, and emission to the atmosphere. Environmental Toxicology and Chemistry. 17: 2063-2071.

Liu, J., R.A. Goyer, and M.P. Waalkes. 2008. Toxic effects of metals. In Casarett and Doull's Toxicology: The Basic Science of Poisons, 7th ed. McGraw-Hill, New York, NY, USA. 931-950.

McLaughlin, A., A. Nichols, and D. Steffy, 2007. Bioaccumulation of mercury in the earthworm Lumbricus terrestris (abstract). American Chemical Society Southeast Regional Meeting, Greenville, October 24-27.

Neculita, C.M., G.J. Zagury, and L. Deschenes. 2005. Mercury speciation in highly contaminated soils from chlor-alkali plants using chemical extractions. Journal of Environmental Quality. 34: 255-262.

Nichols, A.C., T.P. Murray, and T.D. Richardson. 2002. Mercury accumulation in catfish (Ictalurus furcatus and I. punctatus) from the southwestern Tennessee River Valley. Southeastern Naturalist. 1:159-168.

Santschi, P.H., M.A. Allison, S. Asbill, A.B. Perlet, S. Cappellino, C. Dobbs, and L. McShea. 1999. Sediment transport and Hg recovery in Lavaca Bay, as evaluated from radionuclide and Hg distributions. Environmental Science and Technology. 33:378-391.

Si, L. and P.A. Ariya. 2008. Reduction of oxidized mercury species by dicarboxylic acids ([C.sub.2]-[C.sub.4]): kinetic and product studies. Environmental Science and Technology. 42: 5150-5155.

USEPA. 1983. EPA methods for chemical analysis of water and wastes. Mercury, Method 245.1 Manual Cold Vapor Technique. U.S. Environmental Protection Agency.

USEPA. 2001. Superfund Fact Sheet: amended agreement for removal action, Anniston PCB site. U.S. Environmental Protection Agency.

USEPA. 2007. Treatment Technologies for Mercury in Soil, Waste, and Water. U.S. Environmental Protection Agency.

Zagury, G.J., CM. Neculita, C. Bastien, and L. Deschenes. 2006. Mercury fractionation, bioavailability, and ecotoxicity in highly contaminated soils from chlor-alkali plants. Environmental Toxicology and Chemistry. 25: 1138-1147.

Alfred C. Nichols and David A. Steffy

Department of Physical and Earth Sciences, Jacksonville State University, 700 Pelham Rd. North, Jacksonville, AL 36265-1602

Correspondence: David A.
COPYRIGHT 2009 Alabama Academy of Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Nichols, Alfred C.; Steffy, David A.
Publication:Journal of the Alabama Academy of Science
Article Type:Report
Geographic Code:1USA
Date:Jul 1, 2009
Previous Article:Comparison of methods for the extraction of genomic DNA from aerobic digest.
Next Article:Safeguarding revenue-generating assets in austere economic environments: a case study of Dadeville, Alabama.

Related Articles
Mercury rising: our seafood is increasingly contaminated with toxins.
Mobilization of mercury from contaminated soil in to an aquatic food chain.
Mercury found in mining areas.
Spiders boost mercury levels: birds eating arachnids get high dose of toxic metal.
Mercury detected in birds' food supply.
N.H. in suit to reduce mercury.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters