Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Hyclrogeochemical Studies at a Landfill in Delaware 10 INTRODUCTION MARY JO BAEDECKER U.S. Geological Survey MICHAEL A. APGAR Department of Natural Resources and Environmental Control, Delaware AB STRACT Plumes of leachate have migrated downgradient from the Army Creek landfill in Delaware, a site used for disposal of industrial and municipal refuse during the 1960s. A series of contaminant-recovery wells was installed in the early 1970s between the landfill and downgradient water-supply wells to intercept and remove contaminated water. Beneath the landfill and immediately downgradient of the landfill large amounts of iron and manganese are dissolved, organic matter is oxidized and reduced, oxygen is consumed, ammonia is adsorbed, and nitrate is reduced. Farther downgradient some of the reactions are altered: iron and manganese precipitate, less organic matter is oxidized and reduced, and more ammonia is removed by ion exchange. Farther downgradient the water chemistry is controlled largely by mixing. Measurements of chemical con- stituents in water near the landfill show that most constituents have decreased in concentration over an 8- yr period (1973-1981). As leachate moves downgradient, the attenuation of contaminants is controlled by the following processes: (1) operation of the recovery-well system that removes contaminated water and reverses the local flow, (2) dilution of leachate with native groundwater, (3) mixing of anaerobic leachate with oxygenated water that facilitates the decomposition of organic matter and precipitation of metals, and (4) interaction of contaminants with aquifer materials. Although the recovery-well system has intercepted and removed most of the contaminated water and the quality of water from most downgradient monitor wells has improved, the leachate remains anaerobic, large amounts of gases are generated, and refractory organic compounds are present. Continued pumping of contaminant-recovery wells has lowered the head in the aquifer and wasted a large amount of uncontaminated groundwater. The movement of contaminated groundwater from waste-dis- posal sites has become a major problem in many parts of the world. Disposal of liquid and solid wastes in landfills creates an environment where organic and inorganic compounds in the waste or those generated in the landfill are subject to movement in the hydrologic regime. The problems are more pronounced in humid and temperate climates because infiltration of rainfall is high to moderate, and isolation of leachate from the under- 127 lying aquifers for long periods of time is difficult if not impos- sible. In these areas landfills become anaerobic and organic materials decay largely through fermentation, thus creating many intermediate organic compounds that affect the ground- water chemistry differently than those originally buried. The degree to which leachate should be contained, removed and treated, or allowed to move at a controlled rate through natural materials is a basic question that must be considered in making management decisions. In addition to understanding the hydrogeology of an area and the physical processes, such
128 as dilution, dispersion, and filtration, an understanding of the biological and geochemical processes is necessary to describe and predict the movement of contaminants. The principal chemical reactions and processes in groundwater at disposal sites include biological decay, precipitation and dissolution of minerals and other inorganic compounds, sorption of chemical constituents, leaching of sediments, ion exchange, generation and diffusion of gases, and movement of dissolved species. The purposes of this chapter are to review groundwater con- tamination at a landfill, examine and interpret hydrologic and chemical data available from 1973 to 1981, describe the hy- drogeologic processes and chemical reactions that have affected the groundwater chemistry, and discuss the results in terms of long-range planning and management of contaminated aqui- fers. The landfill studied is the Army Creek landfill, previously known as the Llangollen landfill, in New Castle County, Del- aware. Complete chemical analyses for water sampled from the landfill, from immediately downgradient, and from supply-well fields within a mile of the landfill have been published earlier (Baedecker and Back, 1979a, 1979b). Considerable data have been collected by New Castle County and the state of Delaware since 1972 (unpublished files). The chemical constituents discussed here in detail are C1- (because it is considered a conservative parameter), pH, dis- (because it is considered a conservative parameter), pH, dis- carbon, and volatile organic compounds. Except for C1- these species are sensitive to oxidation-reduction reactions. The mix- ing of highly oxygenated groundwater and anaerobic leachate has a significant impact on the water chemistry. Variations in concentrations of chemical species downgradient of the landfill are controlled by changes in leachate composition, changes in the flow path ofthe leachate, physical attenuation, and chemical reactions. Statement of Problem The Army Creek landfill covers about 24 hectares and contains more than 1,500,000 m3 of refuse. The refuse was deposited from 1960 to 1968 in an abandoned quarry from which 6 to 9 m of sand and gravel had been removed. The excavation con- FIGURE 10.1 Geohydrologic section of so landfill area (from Apgar, 1975). in A: ,_ -30 lL -60 -90 MARY JO BAEDECKER and MICHAEL A. APGAR tinned until either the water table or a red-clay zone was en- countered. Many types of waste were buried, including solid and liquid industrial wastes and municipal refuse. During the 1960s major well fields were developed in a confined aquifer 900 to 1200 m south and east of the landfill. Pumpage from the well fields lowered water levels and increased the downward movement of water from the landfill to the aquifer. In 1970 the landfill was covered with sandy material and the property was purchased by the county for a park. Two years later a water-quality problem was detected in a nearby domestic well. It was concluded that the absence of or the removal of the red-clay layer in places permitted leachate to migrate into the underlying aquifer (Apgar, 1975~. The leachate, highly con- taminated in both organic and inorganic compounds, was not adequately diluted or purified by filtration before it entered the aquifer. By 1972, it was confirmed that leachate had spread extensively throughout the confined aquifer and was moving south toward major supply wells. Description of Site The Army Creek landfill lies in the Atlantic Coastal Plain and includes both unsaturated and saturated zones. The surficial sands of Pleistocene age, the Columbia Group (Iordan, 1976), are not thick enough in the immediate area to be developed for water supply. Underlying the Columbia sand is the Cre- taceous Potomac Formation, which overlies Precambrian rocks. The Potomac Formation consists of silt and clay interbedded with quartz sand and some gravel (Pickett, 19701. The upper sand of the Potomac Formation thickens to the southeast and forms one of the most productive confined aquifers (Potomac aquifer) in the state (Figure 10.11. The upper confining layer (Potomac clay) is thin and may be absent in places close to and beneath the landfill. The major well fields developed in the Potomac sand south and east of the landfill are capable of pro- ducing 25 x 103 m3/day of water. Pumping from the well fields has lowered water levels and increased the rate of water move- ment through the aquifer and increased the downward move- ment of water from the landfill to the Potomac sand. To limit the spread of contaminants, a coordinated effort was LANDFI L L CONTAM I NANT RECOVERY WELL SUPPLY WELL n . ~~ lowers ~ ~ . rewater table Surface sand Artesian head ~ ' - - I_ Potomac clays _ Potomac sand aquifer APPROXIMATELY 1000 METERS ~
Hydrogeochemical Studies at a Landfill in Delaware r Wilmington if\ Study \~`Area DELAWARE L; ID 75° 37' 75° 36' LANDFILL · °AMOCO · ·25 0 ~ 1 ~ 1 1 ~ AWC-7~ ARTESIAN 0 200 400 600 METERS G-3 EXPLANATION · LANDFILL · RECOVERY · MONITOR WELLS WELLS WELLS O SUPPLY WELLS FIGURE 10.2 Location of wells sampled. Wells numbered are re- ferred to in text. undertaken to reduce pumping from supply wells and to initiate a program of recovery-well pumping between the landfill and the supply wells (Figure 10.2~. Pumping of the contaminant- recovery wells caused a local cone of depression, which reduced contaminant movement toward the supply wells. The recovery- well program was begun in 1973 and expanded in 1974 to the 10 wells that operate today, although not all recovery wells have been operating continuously. The recovery wells dis- charge into Army Creek, which flows into the Delaware River. The direction of flow in the Potomac aquifer has been altered significantly over the past several decades. Elevation of the potentiometric surface in the 1950s, before significant devel- opment of the aquifer, was about 6 m above sea level, and the flow direction was toward Delaware Bay to the southeast. In the 1960s, after well fields were developed for public water supply and industrial use, the direction of flow was to the south and east fFigure 10.3A. The supply wells yield water from the confined sand in the Potomac aquifer from depths of 45 to 60 m. Operation of the recovery-well system reversed the di- rection of flow locally away from the supply wells (Figure 10.3(B)~. Most of the recovery wells are screened over 12- to 25-m in- tervals and are completed 25 to 43 m below land surface. The flow pattern indicates that water upgradient from the landfill moves through and beneath the landfill, where it encounters leachate, and is then discharged, in part, by the recovery wells downgradient from the landfill. Part of the discharge from the recovery wells is also native groundwater from south and east of the landfill. Continued pumping for water supply has permitted contam- inants to move south ofthe recovery-well system tFigure 10.3(C)~. Hydrologic and chemical data indicate that leachate does not migrate at a continuous rate from the landfill but rather moves in pulses. The amount of leachate generated at any one site depends on the amount of recharge to the landfill. It is esti- 129 mated that 70 percent of the landfill leachate originates as infiltrating percolation on the site, and the remaining 30 per- cent results from lateral inflow where the refuse is below the water table. These estimates are based on an evaluation of aquifer transmissivity, measurements of the hydraulic gradient in the aquifer adjacent to the landfill, and estimates of rainfall entering the landfill. During periods of low precipitation, the water table declines and less leachate is generated. Another factor that affects the rate of leachate movement in the aquifer is variation in pumpage from wells. Greater pumpage lowers water levels and creates steeper vertical hydraulic gradients from the landfill to the underlying aquifer. The steeper gra- dients result in more leachate movement into the aquifer and more rapid movement toward the wells. In 1980 the Amoco chemical plant closed, and its well field 900 m southeast of the landfill was used only for maintenance after 1981. Since then, the direction of groundwater flow is southward ~ F igure 10. 3 ~ D ~ ~ . Three major well fields the county's contaminant recovery system and those operated by the Amoco Chemical Corporation and the Artesian Water Company account for all pumping in the area. Total pumpage in the immediate vicinity of the landfill has ranged from 20 x 103 m3/day in 1974 to 12 x 103 m3/day in 1981 (Figure 10.41. Amoco's pumpage has gradually declined from 5.3 x 103 m3/day in 1971 to 0.8 x 103 m3/day in 1981, whereas Artesian's rate reached a maximum of 12.5 x 103 m3/day in 1972 and a minimum of 5.3 x 103 m3/day in 1976. The contaminant-recovery-well system was in operation by late 1973 and operated at maximum capacity in 1974 at a withdrawal rate of 9.5 x 103 m3/day. Since then, pumpage of the recovery wells has declined to an average of 3.8 x 103 m3/ A ~ O ~ O Landfill ~ Amoco Recove A esian eat] well field ~ well field 400 M FIGURE 10.3 Schematic of the direction of groundwater flow: A, before the recovery wells were in operation; B. after the recovery wells were in operation; C, after continued pumping for several years; and D, after Amoco well field closed. Dash-lined arrows show movement of contaminants in C and D.
130 21.0 ~ 19.5 18.0 16.5 15.0 13.5 - E 12.0 Eli lo ~ 10.5 ME 9-0 7.5 6.0 4.5 3.0 1.5 ol l l MARY JO BAEDECKER and MICHAEL A. APGAR 1 1 1 1 1 1 1 1 1 - 1 1 /\/ \ TOTAL _ ~x\ / X/ \ \X~ X _~°/-~ `, WATER G~. · ~ I ~ '-- ' / \,& - ~ 1 1 1 1 1 1 1 1 1971 1972 1973 'x Q \~` AMOCO CHEW. ~ -~ - `. _ _x no 1974 1975 1976 YEAR FIGURE 10.4 Pumpage (cubic meters per day) versus time in the immediate vicinity of the landfill. LANDFILl ,~\, ~ . _ · _ ~_./ FIGURE 10.5 Lines of equal C1- concentration (50 mg/L) for the years 1973 to 1977, and 1981. 1977 1978 1979 1980 1981 day in 1981 because the wells are operating less efficiently and some have been closed permanently. The location of the wells and their pumpage have influenced the extent of plume migration and concentrations of chemical constituents within the plume, as shown by the 50 mg/L iso- chlor maps for the landfill area (Figure 10.5~. The leachate plume migrated significantly farther downgradient in 1974 than in 1973. The effect of the recovery wells in controlling the leachate plume was observed by 1975 and in 1976 when the extent of plume migration was significantly diminished. How- ever, the plume again extended downgradient in 1977 after pumpage of the recovery-well system fell below that of the public water-supply wells (Figure 10.41. Although for the past 5 yr the position of the 50 m0L isochlor has changed little, the location of the plume and the relative pumpage of the Artesian Water Company and the contaminant-recovery well are in delicate balance. The distribution of C1- in the major leachate plume south of the landfill showed the most striking change before 1977 (Figure 10.61. From 1973 to 1976 the concentration of C1- at recovery-well site 29 decreased from 380 to 50 mg/L. Farther downgradient at well sites 33, 25, and 52, the C1- concentra- tions also declined during this time but to a lesser extent. The decline in C1- concentrations was caused largely by installation of the recovery-well system, which operated most efficiently from 1974 to 1977. Concentrations of C1- in the water standing
Hydrogeochemical Studies at a Landfill in Delaware 400 300 o, E 200 cat 100 - \\ F9\ \\/ N\ - X~, Amp_ _ 25_o, O ~ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ~ 1 1 MONTH 1 - 6 7 - 12 1 - 6 7 - 12 1 - 6 7 - 12 1 - 6 7 - 12 1 - 6 7 - 12 1 - 6 7 - 12 1 - 6 7 - 12 1 - 6 7 - 12 1 - 6 7 - 12 YEAR 1973 ~ 1974 1 1975 1 1976 1 1977 1 1978 1 1979 1 1980 1 1981 FIGURE 10.6 Average concentration of C1- in groundwater from wells 29, 33, 25, and 52 for 1973 to 1981. in the refuse have declined significantly as a result of previous leaching, and this may, in part, account for the decline down- gradient. In 1977, C1- concentrations increased at all sites and since then have remained relatively constant between 25 and 80 mg/L. Pumpage from Artesian Water Company well fields has exceeded pumpage from the contaminant-recovery-well system since 1977. Alternatives to operation of a recovery-well system to control the generation and movement of leachate were considered by county and state officials for returning the aquifer to full use. Leachate production can be halted only by preventing water from entering the landfill or by removing the refuse. Methods that were evaluated to curtail generation of leachate include covering the landfill with a synthetic surface and isolating the leachate by installation of a barrier or draining ditches. Un- fortunately, these solutions are costly and unlikely to eliminate the problem of leachate migration to the aquifer. Removal of the refuse and incinerating it or disposing of it at another site, in addition to being costly, are politically difficult and envi- ronmentally disruptive. The cost of isolating the refuse from contact with water was estimated to be in the same range as the cost of moving the landfill. These estimates ranged from $12 million to $16 million in 1974 (Nielsen, 1974~. To date, installation and operation of the recovery-well system, moni- toring groundwater quality, and feasibility studies have cost about $4 million. Surface grading and runoff control could be 131 used to limit infiltration, although compaction of the refuse and subsidence of cover material are perpetual problems. Some of the recovery wells are currently being relocated on the landfill, which may remove contaminants more efficiently and, in the process, waste less uncontaminated groundwater. The latter remedial measures were estimated in 1978 at $4 million to $6 million in capital and $1 million in annual maintenance costs (Anonymous, 19781. GEOCHEMICAL ASPECTS Sampling Techniques Water samples were collected on the landfill with a metal bailer from 1973 to 1976 and by a peristaltic pump from 1976 to 1981. Monitor wells were sampled with a submersible pump. Re- covery and supply wells were sampled at spigots while the wells were pumping. Vandalism and the closing of some re- covery wells made it difficult to sample the same wells through the period of study. Wells were pumped or bailed before sam- pling to remove water standing in the well casing. Temperature, pH, alkalinity, and dissolved O2 were mea- sured in the field. A sample was filtered in the field through a 0.45-,um filter for analyses of major anions and cations. Sam- ples for analyses of iron and manganese were filtered through
132 a 0.1-~m filter to remove colloidal oxidized species. Analyses of CH4 and organic compounds were made on unfiltered sam- ples. It proved to be difficult to obtain reproducible dissolved organic carbon data because of the high biological activity in leachate-contaminated water. For this study, organic samples were either chilled and analyzed immediately or preserved by the addition of HgCl2 or Ag and chilled until analyzed. Dissolved Oxygen, pH, and Alkalinity The concentration of dissolved O2 ranges from 9.0 mg/L (near saturation) in uncontaminated zones to no detectable O2 in contaminated water in and near the landfill. The native ground- water is highly oxygenated; however, water moving through the landfill has all of its O2 removed by respiratory processes. The available oxidizable material in the landfill far exceeds the amount of O2 in the recharge water. Data from 1977 and 1981 show two major plumes of leachate moving south of the landfill (Figure 10.7~. Between the two reducing plumes is a more highly oxygenated area, which has expanded over the 4-yr period. Less contaminated water moves in the aquifer beneath the narrow central part of the landfill, probably because the refuse source is smaller. Also, the aquifer may be thinner or less permeable in this area. Water-level data indicate that the oxygenated area between the reducing plumes may be re- charged locally; however, the data are equivocal. The extent of the anaerobic zone southeast of the landfill diminished in 1981 as compared with 1977 as a result of the closing of the Amoco plant well field in 1980. Large amounts of O2 are removed from the native ground- water by the oxidation of organic matter and dissolution of metals. In fact, more O2 than is available in native groundwater is needed to account for the amount of iron precipitated and total inorganic carbon (CO32-, HCO3-, H2CO3) in solution. After all molecular O2 is utilized, organic matter is both oxi- dized and reduced by anaerobic fermentative processes to pro- duce CO2 and reduced organic species, such as CH4. The pH of uncontaminated water in the area is between 5.0 and 5.5, which is typical for a poorly buffered water in clay and sand layers that lack calcareous material and are extensively weathered. In contrast, water from the anaerobic zone has a pH between 6.5 and 6.9. Within this zone the pH at any one site has changed little from 1977 to 1981; the average change in pH is 0.09 for five locations for which data are available. The major controls on pH in anaerobic leachate are (1) deg radation of organic matter producing large amounts of CO2, which forms carbonic acid and lowers pH; (2) production of NH4+ and CH4 and reduction of Fe and Mn oxides, all of which consume hydrogen and raise the pH; and (3) exchange of hy- drogen on clay, which may lower or raise the pH. Downgra- dient from the anaerobic zone the pH values of contaminated water range from 5.2 to 6.8. Within this transition zone the average change in pH at each site is 0.26 for the same 4-yr period. The pH varies more in the transition zone because it reflects the subtleties of mixing of the contaminated water with native groundwater. Measurements of alkalinity in contaminated water consist MARY JO BAEDECKER and MICHAEL A. APGAR 1 977 _34/ ·8.0 1981 LANDFI LL ·6.9 r~r\ \ 9.4 \ \ ~ 6.34 ) ·1 7 t~ 6.2- ~ / ·1.7 _' / 3.° - / °° ~s// /3.4 .0.0/ ~ ~ /,~- / \ ~6.7 / \ ~0 "'` ,3.0 ·6.2 400 M ·7.0 FIGURE 10.7 Distribution of dissolved O2 in groundwater. Lines of equal concentration drawn for 0.0, 3.0, and 6.0 mg/L. largely of measurements of HCO3-. Although, at some sites on the landfill, organic acids contribute to the total alkalinity, downgradient this contribution is minimal (Baedecker and Back, 1979a). The main control of alkalinity concentration is CO2 generated from respiratory processes that consume O2 rather than dissolution of calcareous material. Thus, O2 is consumed in the landfill, and large amounts of HCO3- are generated. As O2 concentrations increase away from the landfill, HCO3- con- centrations decrease. The changes in concentrations of O2 and HCO3- due to mixing and chemical reactions are discussed in more detail later. Uncontaminated oxygenated water (6 mg/L of O2) has HCO3-
Hydrogeochemical Studies at a Landfill in Delaware concentrations of less than 10 mg/L. Most of the total inorganic carbon (CO32-, HCO3-, and H2CO3) is in the form of carbonic acid (including aqueous CO2), as reflected in the low pH of the water. Transitional or partially oxygenated water (0.3 to 6.0 mg/L of 02) has HCO3- concentrations between 12 and 75 ma/ L, whereas anoxic water has HCO3- concentrations of >100 mg/L. With a few exceptions, alkalinity concentrations changed little from 1977 to 1981. The largest change was a decrease in alkalinity at site S-1 on the landfill, a site that has consistently produced the most concentrated leachate. Previous results showed that a large part of the alkalinity at site S-1 was organic acid anions rather than from HCO3-. More recently, the con- tribution from organic acids has greatly decreased, and this may, in part, account for the decrease in alkalinity from 4450 m0L to 1570 mg/L over the 4-yr period. Major Constituents The relative abundances of the major inorganic constituents in water, Ca2+, Mg2+, Na+, K+, HCO3-, C1-, NO3-, and SO42-, are significantly different in the anaerobic zone and the native groundwater. The average concentrations as percentage of mil- liequivalents/liter (meq/L) and standard deviation of major in- organic constituents are shown in Figure 10.8 for three water types, based on dissolved O2 concentrations. Native ground- water unaffected by leachate is shown in group A as a Ca-Mg type with C1- as the major anion. Sulfate concentrations are quite low in both the leachate and natural water. Rainfall is the major source of C1- in the native groundwater; rainfall and oxidation of iron sulfide are the main sources of SO42-. Con- centrations of all constituents are quite low in the natural water, which has dissolved solids (DS) of 50 to 90 mg/L. This mixed- cation type of water, in which no cation contributes more than 50 percent to total cations, is characteristic of much water in recharge areas of the Atlantic Coastal Plain. Group C, the anaerobic zone, includes those samples from the landfill and the recovery wells immediately downgradient. This water is a sodium bicarbonate type, and in 1981 the DS ranged from 250 to 2000 mg/L. During 1977 to 1981 the DS at S-1 showed the most significant change from 6400 mg/L in 1977 to 2000 mg/L in 1981. This is consistent with other data that show that the concentrations of all constituents in the leachate at S-1 have decreased. Other sites within this zone have DS values that were lower from 1977 to 1981 by about 16 percent. Concentrations of constituents other than those shown in Figure 10.8, including NH4+, Fez+, Mn2+, CH4, and organic compounds, are high in this water type. In fact, NH4 + and Fez+ account for 14 to 50 percent of the total cations in the anaerobic zone. The high values of HCO3-, CH4, and NH4+ result from degradation of organic matter by fermen- tation. A major source of cations (Mg2+, Ca2+, Na+, K+) may be their exchange from clays by NH4+ generated in the leachate. The high concentrations of Fez+ and Mn2+ most likely result from their mobilization when water, which has a reducing po- tential due to the presence of organic material, comes in contact 133 ~V~§ Percent of Constituents FIGURE 10.8 Inorganic chemical composition of groundwater for (A) the native groundwater, high in 02; (B) transition zone, low in 02; and (C) the anoxic zone. Average concentrations and standard deviation in percentage of total milliequivalents per liter for cations and anions (modified from Baedecker and Back, 1979a). either with metals in the refuse or with oxidized Mn-bearing minerals and ferric oxide cements in the sand and clay. Water in the transition zone (B) is intermediate between water types A and C and is a mixed cation-anion type. The water chemistry in this zone results largely from mixing of leachate with native groundwater, with only minor changes as a result of chemical reactions. The values for DS in this zone ranged from 50 to 160 mg/L in 1981, which is about 30 percent lower than those measured in 1977.
134 Organic Compounds Organic carbon concentrations in the landfill ranged from 77 to 260 mglL in 1981. A marked decline was observed at landfill site S-1 from 3700 mg/L in 1977 to 260 mg/L in 1981. Organic carbon concentrations decrease rapidly downgradient from the landfill to low values of <0.1 mglL in uncontaminated water (Table 10.11. Organic carbon concentrations vary more with time and sampling procedures than concentrations of inorganic constituents. Much of the organic matter is oxidized as the leachate migrates, and some is adsorbed on mineral surfaces. Sorption is expected to be greater if sediments in contact with the water have a high organic content and if the compounds sorbed have low solubilities and polar functional groups (Griffin and Chian, 1979; Means et al., 1980~. Although the clay is weathered and the organic material is low, some lignitic silt was found in Potomac Formation cuttings when the wells were installed. Many of the organic compounds in the water are polar, such as organic acids and chlorinated compounds, and may sorb more readily on sediments. Because organic acids are highly reactive, they disappear rapidly downgradient. By con- trast, the chlorinated hydrocarbons are more refractory and may persist for a long time. Volatile organic compounds (VOC) were identified in the water samples by purging the water with inert gas, trapping the compounds on a Tenax column, and eluting them into a gas chromatography-mass spectrometry system (Pereira and Hughes, 1980~. The compounds with confirmed identification, that is, analyzed with standards run under the same conditions, are listed in Table 10.1. The column 3 (total VOC) is the sum of all compounds identified, including those for which identi- f~cation is tentative. Aromatic hydrocarbons are the most prev- alent of the compounds definitely identified. The same com- pounds are not present throughout the landfill, as shown by comparison of analyses from sites A-4 and S-1. The total con- centration of VOC decreases rapidly away from the landfill and MARY JO BAEDECKER and MICHAEL A. APGAR recovery wells but does not drop much below 10 ~g/L. In an earlier study of VOC in the landfill area by DeWalle and Chian (1980), S-1 reportedly had a much higher concentration of VOC (982,000 ~g/L) compared with recent values (840 1lg/L) re- ported here. Although it is difficult to compare analyses from different laboratories, concentrations of VOC in water collected in 1981 are significantly lower than those in water collected in 1977 for the study by DeWalle and Chian. Another factor that may have affected the concentration of VOC is lowering of the water table at site S-1, which was 5 m below land surface in June 1981 as compared with previous water depths of 3.6 m. As a result, a much smaller volume of refuse was saturated in 1981. The lower values for specific organic compounds in the landfill leachate and the water downgradient over a 4-yr period are consistent with lower total organic carbon values and suggest that organic compounds are being removed from the aquifer. High concentrations of CH4 in the landfill result from mi- crobial degradation under anoxic conditions. CH4 concentra- tions have decreased from 22 mg/L in 1977 to 14 mg/L in 1981 in water collected at site A-4 in the landfill. However, at site S-1, which is located in the newest part of the landfill, CH4 concentrations have increased from 10 mg/L in 1977 to >20 mg/L in 1981. This difference may indicate that the newer part of the landfill near S-1 may have reached the state of high CH4 production at a later time than the zone near A-4. Nicholson et al. (in press) have shown that the gas initially produced in landfills is predominantly CO2, whereas at later stages the pre- dominant gas is CH4. Expressed in partial pressures, CH4 con- centration of 20 mg/L is about 0.74 atm at 16°C, which means it constitutes 74 percent of the total gases in solution and is probably near maximum concentration. Concentrations of CH4 decrease rapidly downgradient and are below the limit of de- tection in the supply wells. However, CH4 has continued to move downgradient and is present in water containing as much as 1.7 mg/L of O2 TABLE 10.1 Dissolved Organic Carbon and Volatile Organic Compounds (VOC) in Groundwater at Army Creek Landfill VOC (~g/L) Wells DOC Total VOC (mg/L) (,ug/L) Number of Methylene Compounds Chloride Benzene Toluene Ethyl Chloro- Trichloro- Ethanea Benzene benzene ethene Dioic Acid Landfill S-1 260 843.5 26 76.8 3.0 35.0 224.7 A-4 77 843.0 34 3.7 140.0 0.7 92.8 16.1 207.7 MonitorlRecovery 29 33 25 52 Supply G-3 AWC-7 13 3.4 2.7 1.0 0. 0. 242.4 29.8 26.7 15.4 18.4 8.9 21 18 13 11 0 1.8 6.9 2.2 2.3 0.1 0.4 19.5 .3a .6a 2.4 0.4 0.2 7.2 0.2 3.1 aIdentification tentative.
Hydrogeochemical Studies at a Landfill in Delaware move downgradient and is present in water containing as much as 1.7 mg/L of O2 Chemical Mass Balance Although most of the leachate is removed downgradient by the recovery wells, some contaminated water travels beyond the recovery-well system, and the behavior and fate of this water with its constituents are considered in more detail. The con- taminants are attenuated by mixing, by chemical reactions, and by physical processes as the leachate moves downgradient. The controlling reactions and processes that alter the chemical com- position of leachate as it moves along a groundwater-flow path were simulated by a chemical mass-balance model. The first calculation in the model is to evaluate the role of mixing of the contaminated water with native groundwater. The second cal- culation is to simulate chemical reactions to generate a chemical composition of water identical to that obtained by chemical analyses. Chemical modeling identifies a set of plausible chem- ical reactions but does not provide a unique set of chemical reactions. The extent of attenuation by mixing was calculated using C1- as a conservative parameter. The flow path of contaminants, based on the distribution of chemical species, is from the land- fill to well 29 and between the following wells: 29 to 33, 33 to 25, and 25 to 52 (Figure 10. 2~. Site A-4 was used as the leachate end member because its water chemistry has been relatively consistent over a 5-yr period and because in recent years the wells have been destroyed on the eastern part of the landfill and recent data are not available. Chemistry of the native groundwater, the other end member, is based on the analyses of water from uncontaminated supply wells. Ratios of the mix- ing components were determined by assuming that C1- con- centrations along the path are controlled only by mixing of contaminated water with native groundwater. For other con- stituents, deviations of observed concentrations from those cal- culated by using the mixing ratios represent an enrichment or depletion of species due to chemical reactions or processes. It must be assumed that the concentrations of constituents have been constant at any one point along the flow path during the time required for water from the landfill to reach the outermost well. The flow time was estimated to be 2 yr. This discussion is limited to abundances of species that are related to the presence of organic matter and that are sensitive to the availability of O2 and therefore subject to oxidation- reduction processes. To construct a model to explain the sources and sinks of constituents, limitations were placed on the pos- sible reactions, which are given in Table 10.2. For example, sources of Fe and Mn were limited to natural oxide coatings, Fe2O3 or MnO2, even though another possible source is the material deposited within the landfill. In the model, Mn2+ was precipitated as MnO2 or MnCO3, and Fez+ was precipitated either as Fe(OH)3 or FeCO3, depending on saturation values at each site. NO3- was reduced to N2, although it can be reduced to either N2O or NH4+ under certain conditions. Be- cause nitrification does not occur in the anaerobic water, NH4+ was removed by cation exchange on clay. Organic matter of the composition CH2O was oxidized to CO2; and organic matter 135 TABLE 10.2 Chemical Reactions Used in Model (1) (2) (3) ~ ~ ~ ~ (4) 4Fe2+ + O2 + 1OH2O ~ 4Fe(OH)3 + 8H+ (5) 2Mn2+ + O2 + 2H2O > 2MnO2 + 4H+ (6) 4NO3 + 5CH2O ~ 5CO2 + 3H2O + 40H- + 2N2 (7) CH2O + O2 ~ CO2 + H2O (8) C8NH~70~ + 2H2O ~ NH3 + 2C2H6O + 4CO2 + 3H2 (9) CO2 + 4H2 ~ CH4 + 2H2O (10) Cation-clay + NH4 ~ NH4 -clay + cation Fe2O3 + 2H+ + CH2O ~ 2Fe2+ + 2H2O + CO2 2MnO2 + 4H+ + CH2O ~ 2Mn2+ + 3H2O + CO2 Fez+ + 2HCO~ ~ FeCO~ + H~CO~ containing nitrogen, represented in a general formula by C8NH,708, was oxidized to CO2, reduced to NH4+, and par- tially reduced to C2H6O. Several pathways are known for the formation of CH4, such as via an acetate intermediate; however, for this model CH4 was formed by the hydrogenation of CO2. Although the stoichiometry assigned to organic matter is not unequivocal, it is consistent with reactions known to occur in anaerobic environments. CH2O is an oxidized form of organic material as in carbohydrates, C8NH~7O8 is an oxidized form of organic material with an C:N ratio of 8:1, and C2H6O is a partially reduced form of organic material as in alcohols. After the mixing ratios were determined, the amounts of reactants and products in the reactions were calculated using, in part, the computer program BALANCE (Parkhurst et al., 19821. Electrons were balanced, but the concentrations of H+ and OH- were not calculated because they are components of water. Thus, the sequence of mass-balance calculations for each step along the flow path is as follows: Chemical composition of water at Point A Chemical composition Gain or loss of water after mixing of species due water from Point A + to chemical with native groundwater reactions (Table 10.2) v Chemical composition of water at Point B The chemical compositions of water at Points A and B were determined analytically, and the chemical composition of water after mixing was calculated from mixing ratios. Finally, the amounts of reactants and products for the chemical reactions were calculated to balance the abundances of species. In the first step (Table 10.3, Part A), the concentrations of dissolved constitutents in the leachate are altered by mixing 21.9 percent leachate from well A-4 with 78.1 percent native groundwater. The concentrations of all constituents decrease with the exception of O2 and NO3-, which increase because these constituents are higher in the native groundwater than in the landfill water. The results of the chemical mass balance show that from the landfill (A-4) to the recovery well (29) iron and manganese oxides dissolve, organic matter is oxidized and reduced, O2 is consumed, NH4+ is exchanged, and NO3- is
136 TABLE 10.3 Chemical Mass Balance along Flowpath at Army Creek Landfill (mmoles/L) l A. Water at After Water at Species Well A-4 Mixing a Chemical Reactions Well 29 5;CO2 26.59 6.95 CH4 0.875 0.192 NO3 0.029 0.054 NH4 8.150 > 1.788 + Fez+ 0.269 0.062 Mn2+ 0.002 0.0005 O2 0.0 0.261 C1- 9.027 2.20 a21.9~o Well A-4 Water + 78.1% NGW. 0.274 Fe2O3 dissolves 0.020 MnO2 dissolves 0.261 O2 consumed 0.655 C~NH~70~ consumed 0.612 CH2O consumed 0.002 NH4+ exchanges 1.310 C2H6O generated 0.027 N2 generated 1.709 He generated 10.12 0.256 0.027 2.445 0.609 0.020 0.0 2.20 - C. Water at After Water at Species Well 33 Mixing Chemical Reactions Well 25 SCOT 4.45 3.44 0.0006 MnO2 dissolves 3.28 SCOT 3.28 2.49 0.119 O2 consumed 2.23 CH4 0.034 0.023 0.067 O2 consumed 0.007 CH4 0.007 0.004 0.127 CH2O consumed 0.038 NO3 0.003 0.023 0.089 CH2O consumed 0.003 NO3 0.003 0.028 0.008 NH4 exchanges 0.021 NH4 0.136 > 0.091 + 0.077 NH4 exchanges > 0.014 NH4 0.014 > 0.008 + 0.0006 Fe(OH)3 ppt > 0.00 Fez+ 0.001 0.0016 0.0012 Fe(OH)3 ppt 0.0004 Fez+ 0.0004 0.0015 0.0009 MnO2 ppt 0.0009 Mn2+ 0.001 0.0011 0.011 N2 generated 0.0017 Mn2+ 0.0017 0.0011 0.004 N2 generated 0.0002 O2 0.013 0.120 0.016 CH4 outgasses 0.053 O2 0.053 0.172 0.35 CO2 outgasses 0.053 C1- 1.38 1.02 0.25 CO2 outgasses 1.02 C1- 1.02 0.65 0.65 C66.7% Well 33 Water + 33.3% NGW. ~57.7% Well 25 Water + 42.3% NGW. NOTE: Native groundwater (NOW) composition: SCOT = 1.43; CH4 = 0.0; NO3 = 0.0613; NH4 = 0.0; Fez+ = 0.031; Mn2+ = 0.0002; O2 = 0.3344; C1- = 0.28. MARY JO BAEDECKER and MICHAEL A. APGAR B. Water at After Water at Species Well 29 Mixing Chemical Reactions Well 33 SCOT 10.29 6.41 0.130 O2 consumed 4.45 CH4 0.256 0.147 0.179 CH2O consumed 0.034 NO3 0.027 0.042 1.266 NH4 exchanges 0.003 NH4+ 2.445 ~ 1.402 + 0.350 FeCO3 ppt ~ 0.136 Fe2 + 0.609 0.351 0.010 MnO2 ppt 0.0009 Mn2+ 0.020 0.012 0.020 N2 generated 0.0015 O2 0.0 0.143 1.78 CO2 outgasses 0.013 C1 - 2.20 1.38 0.113 CH4 outgasses 1.38 b57.4% Well 29 Water + 42.6% NGW. D. Water at After Species Well 25 Mixing . Water at Chemical Reactions Well 52 reduced. As a result of these processes, water of the chemical composition found at well 29 is simulated. In addition, a par- tially reduced form of organic matter (C2H6O), N2, and H2 is generated. The dissolution of Fe2O3 and MnO2 by reactions (1) and (2) (Table 10.2) and reduction of NO3- by reaction (6) produce 0.35 mmole/L of I:CO2 of which 0.06 mmole/L was used to form CH4. The limit on the amount of organic matter oxidized by re- action (7) is the amount of O2 that needs to be removed from solution, whereas the amount of organic matter used by re- action (8) is controlled by the amount of CO2 needed at site 29 in addition to that produced by the reactions discussed above. These reactions f(7) and (8~] produce another 2.62 mmoles/ L of CO2. Of the total amount of H2 produced, a small portion, 0.26 mmole/L, is used to form CH4, which leaves 1.71 mmoles/ L of H2. Because H2 was not found in the gases it was probably removed from solution by organic reactions such as CnH2n + removed from solution by organic reactions such as CnH2n + was outgassed. In the above reaction, where n = 2, C2H4 (ethylene) would be hydrogenated to C2H6 (ethane), both of these compounds were identified in the leachate. However, it is equally probable that CH4 is formed and subsequently out- gassed. On the landfill, significant quantities of CH4 can escape and O2 can be replenished before the leachate reaches the confined aquifer. This also means that more organic matter can be oxidized than is indicated in the model as long as the prod- ucts outgas. The nitrogen in the organic matter is reduced, forming NH4+. Most of the NH4 + remains in solution, and the excess of the mass balance is exchanged on clay. The reactions in the next sequential step downgradient be- tween well 29 and well 33 are significantly different from those discussed above. In this step (Table 10.3, Part B), the concen- trations of dissolved constituents are altered by mixing 57.4 percent water from well 29 with 42.6 percent native ground- water. In contrast to the previous step, ICO2, CH4, Fez+, and Mn2+ are in excess in the mixed water and must be removed. Fez+ precipitates as FeCO3, Mn2+ as MnO2, and CO2 and CH4 outgas. As in the previous step, NH4+ and O2 are removed by ion exchange and oxidative processes, respectively. It is nec- essary to oxidize only 0.18 mmole/L of organic matter to sim- ulate the water chemistry; however, these reactions are difficult to quantify, and, as discussed above, it is possible that more organic matter is oxidized and reduced and the end products (CO2 and CH4) outgas. Because the confining layer is thin close to the landfill and the partial pressures of CO2 and CH4 are high, outgassing is possible in this part of the aquifer. Iron precipitates as either siderite or amorphous iron oxide based on thermodynamic saturation data calculated by the chemical equilibria computer program WATEQF (Plummer et al., 19761. The saturation index flog (ion activity product/equilibrium con- stant)] is 0.97 for siderite and 1.43 for amorphous iron oxide. However, siderite is more stable than amorphous iron oxide under these conditions, with a pH of 6.6, calculated Eh of approximately 0.2 V, ICO2 concentration of 10-2 mole, and Fe concentration of 1O-4 mole. In the next step (Table 10.3, Part C), the concentrations of dissolved constituents are altered by mixing 66.7 percent of water from well 33 with 33.3 percent native groundwater. Like- wise, for the following step (Table 10.3, Part D), 57.7 percent of water from well 25 is mixed with 42.3 percent native ground-
Hydrogeochemical Studies at a Landfill in Delaware water. In these steps, from well 33 to well 25, and well 25 to 52, the water chemistry is simulated largely by mixing. In contrast to the sites upgradient, little NH4+ is exchanged, less O2 is removed, and little solution or precipitation of minerals occurs. The amounts of Fez+ and Mn2+ transferred (0.001 mmole/L) are low and close to the limits of precision of the determinations. Small amounts of organic matter are oxidized, and 0.2 to 0.4 mmole/L of CO2 outgasses. It is unlikely that CO2 would outgas through a confining layer of this thickness (more than 30 m), even though the PcO2s of water (0.05 atm at site 25 and 0.04 atm at 52) are greater than PCO2 of the atmo- sphere. Also, the model shows that CH4 was generated be- tween sites 25 and 52, which is not possible in the presence of oxygen. A possible explanation is that CO2 and CH4 concentrations predicted by mixing are not valid at these sites and that gases are present in solution from an earlier reducing front that ex- tended downgradient. Alternatively, a microreducing environ- ment may exist that permits gases to be generated in the aqui- fer. Less CO2 is generated by considering other equations for the oxidation of organic matter. Reaction (7) (Table 10.2) uses CH2O as a form of organic material, whereas using a more reduced form of organic material such as C2H6O or CH4 yields less CO2 for the same amount of O2 consumed. This is shown by the following equations, in decreasing oxidation state of organic compounds: CH2O + O2 > CO2 + H2O C2H6O + 3O2 > Scot + 3H2O CH4 + 2O2 > CO2 + 2H2O iO2/CO2=l~ fO2/CO2= 1.51 tO2/CO2 = 21 (10. 1) (10.2) (10.3) Thus, oxidation of CH4 rather than CH2O yields only half as much CO2. All of these equations are valid for this site. How- ever, because a large amount of the ICO2 in excess is the difference between mixed water and final water, it is almost certain that the mixing ratios for steps C and D, calculated on the basis of C1- concentrations are in error. The mass-balance calculations suggest that between the land- fill and the first recovery well large amounts of organic matter undergo fermentation reactions and Fe- and Mn-containing oxides are mobilized. Fe and Mn are higher at the recovery site than the landfill because (1) the landfill materials have been leached of these metals since the material was emplaced, or (2) the leachate comes in contact with more oxide coatinp;s as it moves downgradient. Most of the NH4 exchange occurs downgradient of the landfill between wells 29 and 33, which is the zone where the clay-confining layer becomes thicker. The leachate that moves beyond the recovery well is subject to oxidation as it mixes with native groundwater. Iron and manganese precipitate; concentrations of nitrate increase be- cause it is not subject to reduction. Farther downgradient, with the exception of O2 removal and ICO2 generation, only small quantities of constituents are involved in chemical reactions. CONCLUSIONS Chemical reactions and processes in contaminated water do not remain constant with time. Early diagenesis of landfill 137 materials produces high concentrations of ~C02, NO3, and SO42-. During later stages, degradation of organic matter pro- ceeds until available sources of O2 are consumed and the landfill becomes anaerobic. The main products formed are ~C02, NH4+, H2S, CH4, Fez+, Mn2+, and partially reduced organic matter. Sulfide may generate earlier than CH4 or may be suppressed if methanogenic bacteria dominate. As the plume migrates through the aquifer, these processes change. Although CH4 and NH4 are not formed in the presence of 02, they may remain in small concentrations after the water is partially ox- ygenated, whereas Fez+ and Mn2+ will precipitate. Also, many organics, especially the chlorinated compounds, are insensitive to the presence of O2 Although zones of water with different chemical composi- tions develop by the oxidation and reduction of organic ma- terial, the boundaries and extent of these zones are controlled by competing rates of reaction and hydrologic transport. The fronts of these zones are transient because of the wide range in groundwater velocities and the dispersive character of the aquifer materials. If the flow is extremely slow in the aquifer relative to reaction rates, the plume will be attenuated more rapidly and the boundaries between zones may be noniden- tifiable. However, if the flow is fast relative to reaction rates, the zones will migrate farther downgradient and their bound- aries will be more distinct. Plumes of leachate have migrated downgradient from the Army Creek landfill. The recovery-well system installed be- tween the landfill and major downgradient water-supply wells has successfully intercepted and removed most of the contam- inated water. At the Army Creek landfill, the attenuation of contaminants depends on the following: (1) operation of the recovery-well system that removes contaminated water and reverses the local flow, (2) dilution of leachate with native groundwater, (3) mixing of anaerobic leachate with oxygenated water that facilitates the decomposition of organic matter and precipitation of metals, and (4) interaction of contaminants with aquifer materials. Measurements of the major chemical constituents in water immediately downgradient from the landfill show that most constituents have decreased in concentration since 1973, when the recovery system was put in operation. However, the leach- ate remains anaerobic, and large amounts of gases are being generated. With the exception of site S-1, concentrations of major constituents in landfill water have changed little with time. Site S-1 is in the newest part of the landfill and before 1981 the extremely high concentrations of constituents in water may be due to the mobilization of reactive materials and high rates of microbial activity. At site A-4, within the landfill, the pH, alkalinity, C1-, NH4+, and CH4 measurements were sur- prisingly constant during the past 4 yr. Although the major inorganic constituents have changed little, the organic carbon content and the number of organic compounds identified in the leachate have greatly decreased. The refractory nature of many organic compounds and their tendency to remain coated on aquifer materials may cause a contamination problem for long periods of time. Organic contamination may be a threat even after the aquifer is returned to prelandfill conditions. Concentrations of constituents in the leachate are deter- mined by the nature of the waste, by the amount of recharge
138 water that passes through the landfill, and by factors that con- trol groundwater velocity. Hydrologic and chemical data during an 8-yr period indicate that the major changes beyond the anoxic-oxic boundary are controlled by physical processes rather than chemical reactions. Thus, the extent of leachate movement within the aquifer is determined by factors that influence groundwater velocity, which include the rates of pumping re- covery and supply wells. The recovery-well system has effectively retarded the move- ment of contaminants from Army Creek landfill to the Potomac aquifer. However, to achieve this retardation, pumping of water for supply has been curtailed and a significant amount of un- contaminated groundwater is wasted by pumping the recovery wells. Continued heavy pumping has lowered the head in the aquifer 15 m since the mid-1960s. As a result, saltwater intru- sion is a threat to the eastern portion of the aquifer (Sundstrom, 1974~. At this landfill, it is necessary to continue operation of the contaminant-recovery wells. If recovery wells are located close to the landfill, more remote wells can be abandoned and less uncontaminated water will be wasted. Using the aquifer alone to dilute and attenuate contaminants is an alternative, but it involves risks when the downgradient water must be potable. However, an evaluation of the aquifer material as an attenuating medium could be made in other areas where low concentrations of contaminants would not be objectionable in water downgradient. AC KN OWLE D G M E NTS We are most grateful to Sharon Lindsay, Joe Chemerys, and Mike Brooks, of the U.S. Geological Survey, and to Charles I'Anthony and Sandra Robinson, of the Delaware Technical Services Section of the Department of Natural Resources and Environmental Control (DNREC), for providing chemical anal- yses for this study; to Sharon Lindsay and Mike Custer (DNREC) for field assistance; and to Lisa Hamilton (DNREC) for assis- tance with compilation of data. We are especially indebted and grateful to Ron StouBer, who conducted field sampling and provided great insight to this study from 1975 to 1980. We are indebted to William Back, Warren Wood, and Leonard Kon- ikow for clarifying discussions and review of the manuscript. This work was supported in part by U.S. Geological Survey Interagency Energy-Environment agreement #EPA-81-D- X0523. MARY JO BAEDECKER and MICHAEL A. APGAR REFERENCES Anonymous (1978). Army Creek landfill roundtable (November 17-18, 1977) Summary proceedings, New Castle County Areawide Waste Treatment Management Program, New Castle County, Delaware. Apgar, M. A. (1975). We can't afford to let this happen again, Delaware Conservationist 19, 19-22. Baedecker, M. J., and W. Back (1979a). Hydrogeological processes and chemical reactions at a landfill, Ground Water 17, 429-437. Baedecker, M. J., and W. Back (1979b). Modern marine sediments as a natural analog to the chemically stressed environment of a landfill, J. Hydrol. 43, 393-414. DeWalle, F. B., and E. S. K. Chian (1980). Detection of trace organics in well water near a solid waste landfill, J. Am. Water Works Assoc. 72, 206-211. Griffin, R. A., and E. S. K. Chian (1979). Attenuation of water-soluble polychlorinated biphenyls by earth materials, Ill. State Geol. Surv. Div., Environ. Geol. Notes 86. Jordan, R. R. (1976). The Columbia group (Pleistocene) of Delaware, A. M. Thompson, ea., in Guidebook 3rd Annual Field Trip: Petro- leum Exploration Society of New~ York, U. of Delaware, Newark, pp. 103-109. Means, J. C., S. G. Wood, J. J. Hassett, and W. L. Banwart (1980). Sorption of polynuclear aromatic hydrocarbons by sediments and soils. Environ. Sci. Technol. 14 1524-1528. Nicholson, R. V., J. A. Cherry, and E. J. Reardon (in press). Hydro- geological studies of a sandy aquifer at an abandoned landfill: Hy- drochemical patterns and processes in the contaminant plume, J. Hydrol. Niessen, W. R. (1974). Leachate control strategy for Llangollen landfill, New Castle County, Delaware Preliminary feasibility study, Roy F. Weston, Inc. Parkhurst, D. L., L. N. Plummer, and D. C. Thorstenson (1982). BALANCE A computer program for calculating mass transfer for geochemical reactions in groundwater, U.S. Geol. Surv. Water Re- sour. Invest. 82-XX. Pereira, W. E., and B. A. Hughes (1980). Determination of selected volatile organic priority pollutants in water by computerized gas chromatography Quadrupole mass spectrometry, J. Am. Water Works Assoc. 72, 220-230. Pickett, T. E. (1970). Geology of Chesapeake and Delaware Canal Area, Geol. Map Series 1, Delaware Geol. Surv. Plummer, L. N., B. F. Jones, and A. H. Truesdell (1976). WATEQF A Fortran IV version of WATEQ, a computer program for calculating chemical equilibrium of natural waters, U.S. Geol. Surv. Water Resour. Invest. 76-13. Sundstrom, R. W. (1974). Water resources in the vicinity of a solid waste landfill in the Midvale-Llangollen Estates Area, New Castle County, Delaware, Water Resources Center, U. of Delaware, New- ark.