Printer Friendly

Mobilization of mercury from contaminated soil in to an aquatic food chain.


Fish were collected from a site with mercury (Hg) contaminated soil to determine if Hg was moving from the soil into the aquatic food chain. Fish were collected from a site in Oxford, AL, where Snow Creek, a third-order perennial creek, passes through Hg contaminated soil. Fish were also collected upstream where the soil Hg concentration is at background. Fill deposits used as landscape material in the construction of Oxford Park may have been contaminated with Hg from industrial sources. This land now serves as a non-point source of Hg contaminated soil. Soil samples from the park have a mean Hg content of 1.195 mg/Kg, compared to a background value for the area of 0.039 mg/Kg. Sediment samples taken from Snow Creek as it passes through the park have a mean Hg content of 0.319 mg/Kg. Fish were collected from Snow Creek as it passes through the park and from the upstream site. Fish were analyzed for total Hg using cold vapor atomic absorption. Significant differences were seen in the Hg levels between fish species collected from the park and those collected from the upstream site. The data indicates that Hg from the park is entering the aquatic food chain of Snow Creek.


Mercury (Hg) has been a global environmental pollutant for several decades and is a worldwide health concern. In the Clean Air Act, the United States Environmental Protection Agency (USEPA) identified Hg as the pollutant posing the greatest threat to human health (Gray et al., 2004). Mercury is continuing to contaminate fish in rivers, streams, and lakes across the United States. The metal can enter the aquatic environment from either natural or anthropogenic point sources. Sometimes no direct source of the Hg can be determined. The major Hg contamination source for remote rivers, streams, and lakes is thought to be atmospheric transport and deposition of atmospheric Hg (Lange et al., 1993). Advisories for fish consumption due to elevated Hg contamination are a common phenomenon in many regions of the United States (Balogh et al., 1998a). The contamination of Hg in fish from affected streams remains decades after the use of Hg has ceased, often after remediation efforts were assumed successful (Southworth et al., 2000). Bioaccumulation of Hg in fish can be hazardous to both human and wildlife consumers (Balogh et al., 1998b). Commercial and sport fish, which frequently contain the highest concentrations of Hg, are often used to monitor the bioaccumulation of Hg in the aquatic ecosystems (Peterson et al., 1996).

In the United States, methylmercury is one of the most widespread contaminants of aquatic ecosystems. Methylmercury is a potent neurotoxin to piscivorous wildlife and also poses a serious threat to humans (USGS, 2001). While Hg can be retained in soil, it also can leach into surface and subsurface waters (Paller et al., 2004). Although Hg usually enters the aquatic environment in an inorganic form (Nichols et al., 2002), it is converted to methylmercury by microorganisms before entering the aquatic food chain (Ramlal et al., 1986). Methylmercury is transferred from the sediment of streams, rivers, and lakes into the water by microorganisms and eventually into the biota. Methylmercury is a water soluble compound and is readily absorbed by organisms (Gray et al., 2004).

Any form of Hg can be converted to methylmercury by natural means and then bioaccumulate and biomagnify as it progresses through aquatic food webs. Methylmercury has a strong affinity for sulfur-containing organic compounds and is absorbed by fish from ingestion of dietary sources. Over 90% of Hg found in fish is in the form of methylmercury (Spry and Wiener, 1991). Mercury can bioaccumulate in fish populations as benthic organisms recycle Hg contaminated sediments (Nichols et al., 2002).

Snow Creek flows through highly urbanized areas of Anniston and Oxford, Alabama. The stream is third-order perennial and is relatively calm with alternating pools and riffles. This area includes a shopping mall parking lot and a city park in Oxford, Alabama (Fig. 1).


The city park contains a large lake, baseball fields, playground areas, and tennis courts and is on the floodplain of Snow Creek. Local newspapers have reported that sand from local foundries was used for landscape materials during construction of the city park and the shopping mall parking lot. This area currently serves as a non-point source of Hg contaminated soil. Approximately one mile downstream from the park, Snow Creek empties into the Choccolocco watershed. Choccolocco Creek and the surrounding floodplain have gradually been contaminated with Hg through erosion of contaminated sediment from Oxford City Park (Nichols et al., 2005).

This study was conducted to determine if Hg from contaminated land fill was entering the aquatic food chain in Snow Creek. We previously analyzed 54 bias surface soil and sediment samples from 18 sites in the Snow Creek Watershed (Steffy and Nichols, 2005). Mercury-contaminated soil was found downstream from Quintard Mall and Oxford City Park in both the floodplain and in channel sediment deposits of the creek. Soil from the floodplain in the park had mercury levels at 1.195 [+ or -] 0.713 mg/Kg (mean [+ or -] SD). Channel deposits at the park and downstream have measured mercury levels at 0.319 [+ or -] 0.395 mg/Kg, whereas upstream channel sediment have levels at 0.070 [+ or -] 0.029 mg/Kg. Background soil samples taken outside of the floodplain have measured mercury levels at 0.039 [+ or -] 0.018 mg/Kg.


Fish were collected from Snow Creek in Oxford Park and from an upstream site. The fish were collected between early spring and late fall of 2005 using backpack electrofishing (Model 12 POW Electrofisher). Fish were immediately euthanized by placing on ice. In the lab, fish were sorted by species using dichotomous keys, and wet weight was determined using an electronic balance recorded to the nearest 0.001 g. Once the fish were weighed and measured, they were stored in zip-seal bags at -17 [degrees]C until processing.

Whole fish sampling was conducted for each species. Stonerollers, a minnow size fish, were sampled in groups to obtain adequate tissue for testing. These small fish were divided into three groups by total length (cm). In each sample an average of 25 fish were placed in the 0.0 to 5.5 cm group, an average of 25 fish in the 5.5 to 7.5 cm group, and an average of 20 fish in the 7.5 to 12.5 cm group. The other fishes, all of the family centrarchidae (common name "sunfishes") were processed individually. The fish were thawed and homogenized using 20 mL aliquot of ultrapure water (17 megaohm resistance) and a Waring Laboratory blender. Each sample was homogenized for 20-40 s on low speed. Homogenized samples were frozen at -70 [degrees]C for a minimum of 24 h. Tissue was then freeze-dried (VirTis Freezemobile 12) for 3-4 d. Triplicate freeze-dried samples from the same homogenate (approximately 2.0 g) were weighed using an electronic balance and placed in acid washed 150 mL beakers. The exact weight of each sample of dry tissue was recorded to the nearest 0.001 g using an electronic balance. Dry tissue samples were digested in a mixture of 15 mL of double-distilled nitric acid and 2 mL of 30% hydrogen peroxide at 45 [degrees]C on a hotplate under a fume hood. A reagent control treated in a similar fashion was included with each batch of tissue. The residue from each sample was dissolved in 7% nitric acid solution and filtered (Fisherbrand Q8 filter paper) into an acid-washed 50 mL volumetric flask and diluted to 50 mL with 7% nitric acid. Fish tissue was further digested in sulfuric acid, nitric acid, potassium persulfate, and potassium permanganate solution by heating the mixture to 95 [degrees]C for 2 h. This treatment ensured that all forms of Hg were converted to the mercuric ion. Unreacted potassium permanganate was reduced by adding 1 mL of a 12% hydroxylamine solution to each sample.

The samples were analyzed for total Hg using USEPA Method 245.1, Manual Cold Vapor Technique (USEPA, 1983). Hg analysis was conducted using a CETAC Quick Trace Mercury Analyzer M-1600 cold vapor atomic absorption Hg analyzer with an ASX-400 AutoSampler. During flow injection, a 7% stannous chloride solution reduced all mercuric ions to elemental Hg. 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 0.999 or better), and spiked specimens. Matrix spikes gave 85-90% recovery. Specimen split between two batches had a variation of less than 5%.

InStat, version 3.0 for Windows (Graphpad Software, Inc., San Diego, CA) was used for data analyses which included Kolmogorov-Smirnot test for normality, the unpaired t-test, the Mann-Whitney test, one-way analysis of variance (ANOVA), Fisher's protected least significant difference test, the Krushkal-Wallis test for nonparametric ANOVA, and Dunn's multiple comparisons test.


Among the fish species collected from both sampling locations were longear sunfish (Lepomis megalotis), bluegill sunfish (Lepomis macrochirus), green sunfish (Lepomis cyanellus), redbreast sunfish (Lepomis auritus), and largescale stonerollers (Campostoma oligolepis). Mercury was detected in the tissue of all fish tested (Table 1). Only three species of fish were collected in numbers sufficient to perform statistical analysis between the park and the upstream sites. These were stonerollers, longear sunfish, and bluegill sunfish. Green sunfish and redbreast sunfish were only collected at the upstream location. The green sunfish had a mean Hg tissue concentration of 0.028 [+ or -] 0.006 [micro]g/g (mean [+ or -] SD, dry weight). The redbreast sunfish had a mean Hg tissue concentration of 0.037 [+ or -] 0.011 [micro]g/g (Fig. 2).

Significant differences were seen in the Hg levels between fish collected from the park and fish collected from the upstream site. The largescale stonerollers, Campostoma oligolepis, had the lowest Hg concentrations at both sampling locations. Stonerollers collected upstream from the park had a mean Hg concentration of 0.031 [+ or -] 0.005 [micro]g/g (Fig. 2), whereas those collected from the park had a mean Hg tissue concentration of 0.047 [+ or -] 0.009 [micro]g/g (Fig. 3). An unpaired t-test on these groups resulted in a two-tailed P value of < 0.001.



Bluegill sunfish, Lepomis macrochirus, and longear sunfish, Lepomis megalotis, had Hg tissue concentrations higher than those seen in stonerollers. Bluegill sunfish collected upstream had a mean Hg concentration of 0.032 [+ or -] 0.018 [micro]g/g (Fig. 2). The bluegill sunfish collected from the park had a mean Hg concentration of 0.072 [+ or -] 0.054 [micro]g/g (Fig. 3). An unpaired t-test conducted on these two groups produced a two-tailed P value of 0.012, significant at P < 0.05.

Longear sunfish accumulated the highest amount of Hg among all fish collected in the study. The longear sunfish had a mean Hg tissue concentration of 0.067 [+ or -] 0.015 [micro]g/g at the upstream site (Fig. 2). The mean Hg tissue concentration of longear sunfish from the upstream sampling site was twice that seen in stonerollers and bluegill sunfish. Longear sunfish collected from the park had a mean Hg tissue concentration of 0.121 [+ or -] 0.057 [micro]g/g (Fig. 3), which was three times the Hg tissue concentration found in stonerollers. An unpaired t-test conducted on longear sunfish from the park and upstream resulted in a two-tailed P value of 0.014, significant at P < 0.05.

When Hg concentrations in bluegill and longear sunfish collected from the park and from upstream were compared using ANOVA, the ANOVA F value was 12.016 with a P value of 0.0012. Fisher's protected least significant difference test showed significant difference at P < 0.05 between the park and upstream sunfish. The mean Hg concentration for park sunfish was 0.091 [+ or -] .060 [micro]g/g, and for upstream bluegill and longear sunfish combined, the mean Hg level was 0.045 [+ or -] 0.01.

Mercury concentrations in four different fish species collected at the upstream site were compared using the Kruskal-Wallis test (nonparametric ANOVA). This generated a P value of < 0.0001 and a Kruskal-Wallis statistic KW = 20.260. Dunn's multiple comparisons test showed significant differences between bluegill and longear (P < 0.001), between longear and redbreast (P < 0.05), and between longear and green sunfish (P < 0.01). In each comparison, the longear sunfish had significantly higher Hg levels.

The Mann-Whitney test was used to compare Hg levels between the two different sunfish species collected from Oxford Park. Bluegills collected in the park had a mean Hg level of 0.072 [+ or -] 0.054 [micro]g/g, and longear had a mean of 0.122 [+ or -] 0.057. The Mann-Whitney U-statistic for this comparison was 22.500, which gave a two-tailed P value of 0.008. The longear sunfish accumulated more Hg under both high and low environmental exposures. At the upstream site, there was no significant difference in Hg levels among bluegills, redbreast, and green sunfish.


Mercury was detected in all fish collected in this study. The fish collected were representative of area streams. All the fish collected appeared healthy. In a previous study conducted in northwest Alabama involving catfish collected from rivers, lakes, and ponds detectable levels of Hg were found in all fish. The Hg levels of the catfish collected from the two impoundments of the Tennessee River ranged from 0.05 [micro]g/g to 3.28 [micro]g/g (dry weight) in the liver tissue and 0.01 [micro]g/g to 0.21 [micro]g/g in the muscle tissue. The Hg levels of the catfish collected from the neighboring ponds ranged from 0.03 [micro]g/g to 4.79 [micro]g/g in the liver tissue and 0.01 [micro]g/g to 0.54 [micro]g/g in the muscle tissue. The Hg levels in the fish were attributed to atmospheric depositions from coal-burning power plants (Nichols et al., 2002). Mercury emissions from electric utilities are the largest anthropogenic source of Hg in the atmosphere. Approximately 51 tons of Hg are emitted nationwide each year from coal-burning power plants (USEPA, 1998). Atmosphere deposition from area coal-burning power plants could account for the Hg detected in fish collected at the upstream site.

A significant difference was seen in the Hg levels between fish collected upstream and those collected in Oxford Park for each of the three species examined. Stonerollers had the lowest Hg concentrations at both locations. This was expected, as they are a primary consumer in the aquatic food chain (Boschung and Mayden, 2004; Mettee et al., 1996). Bluegill and longear sunfish are opportunistic carnivores and, as such, are higher up the aquatic food chain than stonerollers. Both species had Hg tissue concentrations higher than those seen in stonerollers. Longear sunfish accumulated more Hg than the other species at both the low concentration and high concentration sites.

Snow Creek is a perennial stream. During the time specimens were collected, pools in the creek ranged from 2 to 3 feet in depth. As the sunfish increase in size, they tend to migrate to deeper water downstream, in this case to an impoundment lake (Logan-Martin) on the Coosa River. Sunfish are popular for local sports fishing (especially for children as the fish are easily caught from the bank) and eating. Also, as sunfish grow and migrate into deeper waters of the Coosa River, they become prey for larger fish. This can result in biomagnification as Hg moves up the food chain. While the Hg levels found in this study are below those considered hazardous for human consumption (1.0 [micro]g/g, wet weight), it is disturbing that detectable levels of the metal are present at all. For each species tested, Hg levels were significantly higher in fish obtained from Oxford Park than those found in fish sampled from upstream. We have previously presented data showing that Hg levels above background can be found in creek sediments as far as 14 miles downstream from Oxford Park (Steffy and Nichols, 2005). We conclude that Hg contained in contaminated land fill has not been immobilized by interactions with the soil. On the contrary, our data indicates that Hg from Oxford Park is slowly leaching into Snow Creek where it is entering the aquatic food chain as Snow Creek drains into Lake Logan Martin and the Coosa River system.


The authors would like to thank Jacksonville State University for providing funding for this research.


Balogh, S. J., M. L. Meyer, and D. K. Johnson. 1998a. Transport of mercury in three contrasting river basins. Environmental Science and Technology. 32: 456-462.

Balogh, S., M. Meyer, and K. Johnson. 1998b. Diffuse and point source mercury inputs to the Mississippi, Minnesota, and St. Croix Rivers. The Science of the Total Environment. 213: 109-113.

Boschung, H. T., and R. L. Mayden. 2004. Fishes of Alabama. Smithsonian Books, Washington D. C., USA.

Gray, J. E., M. E. Hines, P. L. Higueras, I. Adatto, and B. K. Lasorsa. 2004. Mercury speciation and microbial transformations in mine wastes, stream sediments, and surface waters at the Almaden mining district, Spain. Environmental Science and Technology. 38: 4285-4292.

Lange, T. R., H. E. Royals, and L. L. Connor. 1993. Influence of water chemistry on mercury concentration in largemouth bass from Florida lakes. Transactions of the American Fisheries Society. 122: 74-84.

Mettee, M. F., P. E. O'Neil, and J. M. Pierson. 1996. Fishes of Alabama and the Mobile Basin. Oxmoor House, Birmingham, Alabama, USA.

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

Nichols, A. C., D. A. Steffy, and S. Al-Hamdani. 2005. Mercury distribution in sediments and uptake by plants along Snow and Choccolocco Creeks in Northeast Alabama. The Toxicologist. 84: 324.

Paller, M. H., C. H. Jagoe, H. Bennett, H. A. Brant, and J. A. Bowers. 2004. Influence of methylmercury from tributary streams on mercury levels in Savannah River Asiatic clams. Science of the Total Environment. 325: 209-219.

Peterson, M. J., G. R. Southworth, and W. D. Crumby. 1996. Monitoring mercury in fish in a stream system receiving multiple industrial inputs. Environmental Monitoring and Assessment. 40: 91-105.

Ramlal, P. S., J. W. M. Rudd, and R. E. Hecky. 1986. Methods for measuring specific rates of mercury methylation and degradation and their use in determining factors controlling net rates of mercury methylation. Applied and Environmental Microbiology. 51: 110-114.

Southworth, G. R., R. R. Turner, M. J. Peterson, M. A. Bogle, and M. G. Ryon. 2000. Response of mercury contamination in fish to decreased aqueous concentrations and loading of inorganic mercury in a small stream. Environmental Monitoring and Assessment. 63: 481-494.

Spry, D. J., and J. G. Wiener. 1991. Metal bioavailability and toxicity to fish in low alkalinity lakes: a critical review. Environmental Pollution. 71: 243-304.

Steffy, D.A., and A. C. Nichols. 2005. Mercury occurrence in the Middle Choccolocco Creek Watershed, Calhoun County, Alabama, presented at the Southeast Regional Meeting for the Society of Environmental Toxicologists and Chemists, Dauphin Island, AL.

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. 1998. Mercury emissions and electric utilities. U. S. Environmental Protection Agency.

USGS. 2001. A national pilot study of mercury contamination of aquatic ecosystems along multiple gradients: bioaccumulation in fish. U. S. Geological Survey.

Candice R (1). Kohute, Alfred C. Nichols (2), David A. Steffy (2), and Mark E. Meade (1)

Department of Biology (1) and Department of Physical and Earth Sciences (2)

700 Pelham Road North, JSU Box 7990

Jacksonville State University

Jacksonville, Alabama 36265

Correspondence: Kohute, Candice (
Table 1. Hg concentrations in fish collected from Oxford Park and
upstream sites along Snow Creek in Anniston and Oxford, Alabama

 Hg Concentration
Species Location ([micro]g/g, dry weight)

Stonerollers Oxford Park 0.047 [+ or -] 0.009
 Upstream 0.031 [+ or -] 0.005
Longear sunfish Oxford Park 0.121 [+ or -] 0.057
 Upstream 0.067 [+ or -] 0.015
Bluegill sunfish Oxford Park 0.072 [+ or -] 0.054
 Upstream 0.032 [+ or -] 0.018
Redbreast sunfish Upstream 0.037 [+ or -] 0.011
Green sunfish Upstream 0.028 [+ or -] 0.006
COPYRIGHT 2006 Alabama Academy of Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kohute, Candice R.; Nichols, Alfred C.; Steffy, David A.; Meade, Mark E.
Publication:Journal of the Alabama Academy of Science
Geographic Code:1U6AL
Date:Jul 1, 2006
Previous Article:Plasmid profile of antibiotic resistant Escherichia coli isolated from chicken intestines.
Next Article:Applications of laser induced breakdown spectroscopy.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters