1994-05 – Research paper – Mirecki and Parks – Leachate Geochemistry at a Municipal Landfill, Memphis, Tennessee


Leave a Reply

Your email address will not be published. Required fields are marked *

Leachate Geochemistry at a Municipal Landfill,
Memphis, Tennessee

by June E. Mireckia,b and WilliamS. Parksa
A leachate plume emanating from the Shelby County Landfill located at Memphis, Tennessee is defined using selected major and trace inorganic constituents. High specific conductance values and elevated concentrations ofammonia, barium, boron, calcium, chloride, dissolved solids, potassium, sodium, and strontium were measured in water samples from wells screened in the alluvial and Memphis aquifers down gradient from or proximal to the landfill. Ofthese constituents, barium, boron, chloride, and strontium were used in the geochemical model code PHREEQE to estimate the percent leachate component in selected Memphis aquifer samples. Estimates of the leachate component in samples from affected Memphis aquifer wells range from 5 to 37 percent for barium and strontium, the most reliable tracers found during this study.
Leachate plumes from municipal solid-waste landfills are potential sources of contamination to ground-water supplies. Evaluation of the spatial extent and magnitude of leachate contamination is important because elevated con.centrations ofinorganic and organic compounds in leachate can degrade ground-water quality and contribute to higher treatment costs or abandonment of water-supply wells. Although leachate compositions vary, large specific con.ductance values, and elevated concentrations of ammonia, dissolved solids, total organic carbon, chloride, iron, and most major cations commonly are detected (for example, Baedecker and Back, 1979). Although leachate plumes may not always have high concentrations of priority pollutants, occasionally federal maximum contaminant levels in ground water are exceeded with respect to volatile organic com.pounds and trace elements (Bolton and Evans, 1991; Borden and Yanoschak, 1990; Reinhard et al., 1984; Dewalle and Chian, 1981).
This paper describes the geochemistry of a leachate plume emanating from the Shelby County landfill at Memphis, Tennessee based on inorganic constituent con.centration data from ground-water samples. These data were used to define the path of the plume from its source at the landfill into the deeper Memphis aquifer.
Defining the geochemical characteristics ofthe leachate plume in this manner may be applicable to other municipal landfills built on clastic sediment. This work indicates that relatively inexpensive analytes (selected major and trace constituents) can be used to detect leachate contamination
“u.s. Geological Survey, Water Resources Division, 7777 Walnut Grove Road, Memphis, Tennessee 38120. bAlso at Memphis State University, Department of Geologi.cal Sciences, Memphis, Tennessee 38!52. Received February !993, revised August 1993, accepted September 1993.
in downgradient wells, as required by U.S. Environmental Protection Agency (199la; 199lb; 199lc) ground-water monitoring regulations.
This study is an extension of an investigation (Parks and Mirecki, 1992) ofthe Shelby County landfill conducted from 1989 to 1991 by the U.S. Geological Survey in coopera.tion with the Shelby County Department of Public Works.
Site History and Description
The Shelby County landfill was operated as a waste.disposal facility from 1968 to 1988, during which unregu.lated disposal occurred from 1968 to 1972. After 1972, dis.posal was regulated by the Shelby County Department of Public Works (SCDPW) and only domestic and municipal wastes were accepted. The 90-acre landfill is located on the alluvial plain of the Wolf River, south of Walnut Grove Road (Figure l). At present, its capped surface is 40 to 45 feet above the alluvial plain, which consists of a forested levee to the southwest, and agricultural land to the southeast and north. The Shelby County landfill was closed in 1988.
Expansion ofthe landfill north ofWalnut Grove Road was considered, which prompted a geologic and hydrologic investigation by the SCDPW in 1986 to evaluate whether expansion of the existing facility was appropriate. During that investigation, water-level measurements in auger holes and observation wells indicated that the potentiometric sur.face of the alluvial aquifer north of the landfill was depressed below the low-flow altitude of the Wolf River (Bradley, 1991). This anomalous water-level condition was the initial indication that downward leakage of water from the near-surface alluvial aquifer into the underlying Memphis aquifer was taking place.
Hydrogeologic Setting
Near-surface lithologic units beneath the Shelby County landfill include alluvium ofQuaternary age and the Memphis Sand of Tertiary age. These units constitute the alluvial and
Vol. 32, No. 3-GROUND WATER-May-June 1994
Memphis aquifers, respectively. In the upper part of the Memphis Sand, a confining unit consisting of interlensed clay, silt, and fine sand separates the alluvial aquifer from the Memphis aquifer. This confining unit was thought to be laterally extensive near the landfill, thus preventing contami.nation of the Memphis aquifer, which is the principal source of drinking water for the city of Memphis. However, geo.physical log data from test holes and wells indicated that these lenses ofclay, silt, and fine sand in the upper part of the Memphis Sand are not laterally continuous throughout the Memphis area (Parks, 1990). In particular, lithologic data from some test holes indicated that the confining unit is discontinuous adjacent to the Shelby County landfill (test hole MS-8 and well MS-12, Figure 1; Parks and Mirecki, 1992).
Identification of a discontinuity in the confining unit ;illowed an evaluation of ground-water flow direction between the alluvial and Memphis aquifers. Water levels measured in wells screened in the alluvial aquifer near the Wolf River approximated stages of the river. However, a depression in the potentiometric surface of the unconfined alluvial aquifer (as much as 14 feet below the low-flow altitude of the Wolf River) had been defined northeast of the landfill (Figure 2; Bradley, 1991). In the alluvial aquifer, water flows northeast from the Wolf River toward the depression. In the Memphis aquifer, water flows generally to the west (Figure 3), but an anomalous high in the potentio.metric surface existed where the confining unit is thin or absent. This anomaly indicates that water level in the Memphis aquifer was elevated as a result of downward leakage from the alluvial aquifer.
Definition of ground-water flow direction enabled interpretation ofleachate plume geochemical characteristics as the plume moved downgradient from the landfill source. In the alluvial aquifer, background, upgradient, and down.

500 1,000 WET(RS

Fig. 1. Location of the Shelby County landfill and geologic section A-A’ (modified from Parks and Mirecki, 1992).

-230–POTENTIOMETRIC CON’I’OUR·–Show.s altitude at wntch water level
would heve stood in tightly cased wells. Dashed where approximately
located. Hechures lndieate depression Contour interval 5 feet.
Datum is sea level
Fig. 2. Altitude of the water· table in the alluvial aquifer and location of wells used to trace the leachate plume.
gradient wells were designated relative to the flow path through the landfill (Figure 2). The distinction among back.ground, upgradient, and downgradient wells was clear because the geochemical composition of samples from downgradient wells 26, 27, 31, 38, and 39 differs from that of all other alluvial aquifer wells (Table 1). In the Memphis aquifer, anomalies in the altitude of the potentiometric sur.face near the landfill resulted from discontinuities in the confining unit, so designation of upgradient and down.gradient wells was not clear. Therefore, for the Memphis

~2.30-·-POTENTIOMETRIC CONTOUR–Shows alhlude at which water level
would have stood m hghlly cased wells. Dashed where approximately
located. Contour mtervel 5 feet. Datum is sea level


Fig. 3. Altitude of the potentiometric surface of the Memphis aquifer and location of wells used to trace the leachate plume.
WLLFOIA4312-001 -0023331
Table 1. Mean Values of Dissolved Constituents and Properties of Water for Samples Collected from Wells Screened in the Alluvial Aquifer at the Shelby County Landfill, Memphis, Tennessee
Background wells
Constituent Mean+/.or property Well4 We/17 standard deviation
Ammonia 0.23 mg/1 O.ll mgjl 0.17 +I. 0.06 mg/1
Bromide 0.03 mg/1 0.09 mgj! 0.06 +/. 0.03 mgj I
Calcium 19 mgfl 34 mg/1 26.5 +j. 7.5 mgfl
Chloride 2.1 mg/1 9.8 mgfl 6.0 +I. 7.5 mgfl
Dissolved solids 132 mgfl 266 mgf! 199 +/-67 mg/1
Fluoride <0.10 mgfl 0.25 mg/1 0.18 +/. 0.08 mgfl Iodide 0.005 mg/1 0.014 mg/1 0.010 +/. 0.005 mgfl Iron 27.5 mg/1 27.5 mgfl Magnesium 6.4 mgfl 13 mg/1 9.7 +/. 3.3 mgfl Manganese 0.185 mgjl 0.185 mg/1 Nitrate plus nitrate as N <0.100 mgfl 0.65 mg/1 0.51 +j. 0.34 mgfl Phosphate as P04 0.21 mg/1 0.21 mgfl Phosphorus as P 0.08 mg/1 0.01 mgfl 0.05 +1. 0.04 mg/ I Potassium 1.9 mgfl 1.9 mg/1 1.9 +j. 0.0 mg/1 Silica 20 mg/1 21.5 mgfl 21 +/. 0.8 mg/1 Sodium 4.6 mg/1 21 mg/1 13 +/. 8.0 mg/1 Sulfate 72 mg/1 140 mg/1 106 +/-34 mg/1 Total organic carbon 2 mg/1 1.1 mg/1 1.7 +/. 0.6 mgfl Barium Boron 60 <10 llgf I llgfl 83 2


< 10 -s "' ~ i :J j 0 2 10-7 0 10 20 30 40 50 PERCENT LEACHATE WATER EXPLANATION • CHLORIDE y = 3.4 X 10-4 + 6.7 X 10 -B (X) 0 STRONTIUM y = 5.8 X 10-? + 3.8 X 10 -B (X) 0 BORON 10-6 y = 1.3 X + 3.3 X 10 -? (X) e BARIUM Y = 4.2 X 10-7 + 2.9 X 10 -B (X) Fig. 4. Dilution curves prepared from mixture data generated by the geochemical model code PHREEQE. 396 Table 3. Estimated Percentage of Leachate Plume Water in Affected Memphis Aquifer Samples (Wells MS-7, MS-11, and MS-12) Barium equation defining dilution curve: Y = 4.2 X l0-7 + 2.9 X 10-8 (X) Observed concentration %Leachate Memphis aquifer sample J.J.g/1 molal plume water MS-7 81 5.9 X 10-7 5.9 MS-11 86 6.2 X 10-7 6.9 MS-12 77 5.6 X 10 7 4.8 Boron equation defining dilution curve: Y = 1.3 X l0-6 + 3.3 X 10-7 (X) Observed concentration %Leachate Memphis aquifer sample J.Lg/1 molal plume water MS-7 25 2.4 X 10-7 0 MS-11 65 6.0 X 10-7 0 MS-12 25 2.4 X 10-7 0 Chloride equation defining dilution curve: Y = 3.4 X 10-4 + 6.7 X l0-6 (X) Observed concentration %Leachate Memphis aquifer sample mg/1 molal plume water MS-7 10.3 2.9 X 10-4 0 MS-11 61 1.7 X 10-3 greater than 100 MS-12 8.4 2.4 x w-4 0 Strontium equation defining dilution curve: Y = 5.8 X 10-7 + 3.8 X l0-8 (X) Observed concentration %Leachate Memphis aquifer sample J.J.g/1 molal plume water MS-7 82 9.5 X 10-7 9.7 MS-ll 130 1.5 X 10-6 24 MS-12 171 2.0 X 10-6 37 [The percentage ofcontaminated alluvial aquifer water was calcu.lated using measured concentrations of barium, boron, chloride, and strontium from affected samples in equations from model.generated dilution curves (Figure 4). In the dilution curve equa.tions, the X variable is% leachate plume water, and theY variable is observed concentration (molal).] not identical, indicating that some constituents react (that is, are semiconservative) as they move away from the landfill (Figure 4). Saturation indices (SI: log [Ion Activity Product/ Ksp T0 C] where T°C is temperature in degrees Celsius ofthe ground-water samples) listed in the PHREEQE output showed which constituents were involved in pre.cipitation-dissolution reactions (see Appendix). Positive values for the SI suggested that barite (BaS04) should pre.cipitate in mixtures composed of 25 to 100 percent contami.nated alluvial aquifer water. Negative Sl values were obtained for celestite (SrS04) and strontianite (SrC03) in all cases, indicating that these waters were undersaturated with respect to celestite and strontianite (see Appendix). Chloride salts did not precipitate in these low ionic strength conditions. Some boundary conditions were estimated in the PHREEQE model, specifically regarding the redox state (pE: -log{electron}) of the leachate plume. Dissolved fer.rous (Fe2+) and ferric (Fe3+) iron and dissolved hydrogen sulfide (H2S) concentrations in ground waters were not measured. However, iron oxide or hydroxide (presumably goethite, o:-FeOOH, at this pH and Fe concentration range; Stumm and Morgan, 1981) was identified in auger-and test-hole cuttings from the alluvium, downgradient from the landfill (Appendix B, Parks and Mirecki, 1992), as was the odor of hydrogen sulfide during drilling. Iron oxide precipi.tation and sulfate reduction suggested apt range ofO to-2, given that pH measurements ranged from 5.2 to 6.9 in both alluvial and Memphis aquifers (Parks and Mirecki, 1992). All model mixtures were run with a pt: = 0. Precise evalua.tion of the redox environment is not critical for evaluation of dissolved boron, chloride, and strontium mobility because these constituents are not overly sensitive to changes in pE. However, the pE conditions of leachate can affect the distri.bution of sulfate and, consequently, the solubility of cor.responding sulfate mineral phases such as barite and celestite. The redox conditions of the leachate plume near the landfill were assumed to be reducing based on the odor of hydrogen sulfide, and generally low sulfate concentra.tions in samples from downgradient alluvial aquifer wells (Table 1). As the plume became more dilute away from the land.fill, redox conditions became more oxidizing, as suggested by higher sulfate concentrations in samples from distal Memphis aquifer wells (Table 2). An oxidizing environment favors precipitation of sulfate minerals, especially barite. However, barium and strontium concentrations diminish by dilution so that water samples from the distal region of the flow path are undersaturated with respect to barite and celestite (see Appendix). Because the geochemical environ.ment is defined by changing redox conditions and mineral solubilities, barite seems to precipitate only in a limited region along the leachate flow path. Barite precipitation will result in an underestimation of the percentage of contami.nated alluvial aquifer water in some ground-water samples proximal to the landfill. It was assumed that the contribution of barium, boron, chloride, and strontium from desorption or dissolution of quartz sand, gravel, silt, and clay was insignificant. Adsorp.tion by clays would reduce cation concentrations in proxi.mal ground-water samples. Because the Memphis Sand is quartz sand, adsorption and desorption reactions were assumed to be negligible. The effect of the leachate plume on water quality of selected Memphis aquifer samples can be estimated by fit.ting concentration data from affected samples to the model.generated dilution curves (Table 3; Figure 4). The leachate component calculated in samples from Memphis aquifer wells MS-7, MS-ll, and MS-12 ranged from 5 to 7 percent using barium as a tracer, 0 percent using boron, 0 to 100 percent using chloride, and 10 to 37 percent using strontium (Table 3). Of these four constituents, barium and strontium probably serve as the most reliable tracers of the leachate plume for several reasons. First, strontium likely behaves conservatively, as indicated by negative saturation indices for celestite and strontianite (see Appendix). Second, barium and strontium probably behave similarly in the leachate plume, as indicated by similar slopes ofthe model-generated dilution curves (Figure 4). Ideally, chloride and boron should serve as conserva.tive tracers because they are nonreactive in dilute solutions. Chloride anions should not precipitate as mineral phases or adsorb extensively onto clay surfaces. However, due to extensive variations in chloride and boron concentrations (coefficients of variation ranged between 25 and 100 percent in downgradient alluvial aquifer and Memphis aquifer samples), estimates of the leachate component ranged from 0 to 100 percent for these constituents. Summary and Conclusions Concentrations of selected inorganic constituents, specifically barium, boron, chloride, and strontium, were shown to identify the leachate plume emanating from Shelby County landfill at Memphis, Tennessee. In the alluvial aquifer, barium, boron, chloride, and strontium concentrations in samples from downgradient wells were elevated at a statistically significant level above those con.centrations measured in upgradient and background samples. Elevated concentrations of barium, boron, chloride, and strontium persisted as the leachate plume moved away from the landfill toward a depression in the potentiometric sur.face of the alluvial aquifer. Most likely, ground water did not seep continuously through the landfill, or flow at the same velocity in the alluvial aquifer toward the depression in the potentiometric surface. Instead, ground-water movement through the land.fill was affected by recharge from precipitation, and in the alluvial aquifer by the stage of the Wolf River. These rela.tions suggest that leachate probably flows away from the landfill in the alluvial aquifer as discrete "pulses" along preferential flow paths rather than as a continuous plume. These physical factors may explain the variability observed in this geochemical data set. Using the geochemical model code PHREEQE, ground-water mixtures were generated using different pro.portions of leachate plume water (mean concentrations of downgradient wells in the alluvial aquifer) mixed with uncontaminated Memphis aquifer water (mean concentra.tions from background samples). Comparison of these model-generated mixtures with measured concentrations from contaminated Memphis aquifer samples allowed esti.mation of the percentage ofleachate water in these Memphis aquifer samples. Of the four tracers used, barium and strontium seemed to behave conservatively. Saturation indices suggest that strontium did not precipitate as a mineral phase along the flow path. Barium concentrations approached saturation, such that barite could have precipitated in mixtures com.posed of 25 to 100 percent contaminated alluvial aquifer water. Use of barium as a tracer indicated that the leachate component ranged between 5 and 7 percent; use of stron.tium indicated that the leachate component ranged between 397 WLLFOIA4312-001 -0023337 10 and 37 percent m contaminated Memphis aquifer samples. Appendix PHREEQE output of saturation indices (log [Ion Activity Product/Ksp PC]) for barite, celestite, and strontianite in initial solutions (from Tables I and 2) and model-generated mixtures. Positive values indicate precipi.tation, negative values indicate dissolution. Percent mix represents the percent of contaminated alluvial aquifer water mixed with background Memphis aquifer water. 100% contaminated 100% Memphis Mineral alluvial aquifer aquifer Barite 0.399 -0.360 Celestite -3.015 -3.775 Strontianite -4.355 -5.140 40% 30% 20% 10% 9% Mineral mix mix mix mix mix Barite 0.136 0.063 -0.031 -0.159 -0.175 Celestite -3.287 -3.361 -3.454 -3.581 -3.596 Strontianite -4.649 -4.726 -4.821 -4.947 -4.963 8% 7% 6% 5% 1% Mineral mix mix mix mix mix Barite -0.191 -0.208 -0.227 -0.246 -0.335 Celestite -3.612 -3.629 -3.647 -3.665 -3.750 Strontianite -4.979 -4.995 -5.031 -5.067 -5.116 References Baedecker, M. J. and W. Back. 1979. Hydrogeological processes and chemical reactions at a landfill. Ground Water. v. 17, no. 5, pp. 429-437. Bolton, K. A. and I .. J. Evans. 1991. Elemental composition and speciation of some landfill leachates with particular reference to cadmium. Water, Air, and Soil Pollution. v. 60, pp. 43-53. Borden, R. C. and T. M. Yanoschak. 1990. Ground and surface water quality impacts of North Carolina sanitary landfills. Water Resources Bulletin. v. 26, no. 2, pp. 269-277. Bradley, M. W. 1991. Ground-water hydrology and the effects of vertical leakage and leachate migration on ground-water quality near the Shelby County landfill, Memphis, Tennessee. U.S. Geological Survey Water-Resources Investigations Report 90-4075. 42 pp. Dewalle, F. B. and E.S.K. Chian. 1981. Detection of trace organics in well water near a solid waste landfill. Journal of the American Water Works Association. v. 73, pp. 206-2ll. Domenico, P. A. and F. W. Schwartz. 1990. Physical and Chemical Hydrogeology. John Wiley and Sons, New York, NY. p. 78. MacFarlane, D. S., J. A. Cherry, R. W. Gillham, and E. A. Sudicky. 1983. Migration of contaminants in a ground.water at a landfill: A case study. I. Groundwater flow and plume delineation. Journal of Hydrogeology. v. 63, pp. 1-29. Parkhurst, D. L., D. C. Thorstenson, and L. N. Plummer. 1980. PHREEQE~Acomputer program for geochemical calcu.lations. U.S. Geological Survey Water-Resources Investi.gations Report 80-96. 193 pp. Parks, W. S. 1990. Hydrogeology and preliminary assessment of the potential for contamination of the Memphis aquifer in the Memphis area, Tennessee. U.S. Geological Survey Water-Resources Investigations Report 90-4092. 39 pp. Parks, W. S. and J. E. Mirecki. 1992. Hydrogeology, ground.water quality, and potential for water-supply contamina.tion near the Shelby County landfill in Memphis, Tennessee. U.S. Geological Survey Water-Resources Investigations Report 91-4173. 79 pp. Reinhard, M., N. L. Goodman, and J. F. Barker. 1984. Occur.rence and distribution of organic chemicals in two landfill leachate plumes. Environmental Science and Technology. v. 18, pp. 953-961. Russell, G. M. and A. L. Higer. 1988. Assessment of ground.water contamination near Lantana landfill, southeast Florida. Ground Water. v. 26, no. 2, pp. 156-164. Stumm, W. andJ. J. Morgan. 1981. Aquatic Chemistry (2nd ed.). John Wiley and Sons, New York, NY. pp. 232-233. U.S. Environmental Protection Agency. 199la. Hazardous waste management system: Amendments to interim status standards for downgradient ground-water monitoring well locations at hazardous waste sites. Federal Register. v. 56, no. 246, October 9, 1991, pp. 66365-66369. U.S. Environmental Protection Agency. 1991 b. Guidelines for the land disposal of solid wastes. Code of Federal Regulations. 40 CFR part 240.204 Water quality. p. 266. U.S. Environmental Protection Agency. 1991 c. Criteria for classi.fication of solid waste disposal facilities and practices. Code of Federal Regulations. 40 CFR part 257.3-4 Ground water. p. 349.