U.S. Geological Survey
Water-Resources Investigations Report 02-4228

Prepared as part of the
NATIONAL WATER-QUALITY ASSESSMENT PROGRAM

Download this report as a PDF file (17.1 MB)

Download a free copy of Acrobat Reader

Quality of Shallow Ground Water in Areas of Recent Residential and Commercial Development, Wichita, Kansas, 2000

By Larry M. Pope, Breton W. Bruce, Patrick P. Rasmussen, and Chad R. Milligan

FOREWORD

Foreword
Abstract
Introduction

Purpose and Scope
Acknowledgments

Description of Study Area

Climate and Precipitation
Land and Water Use
Hydrogeologic Setting

Methods of Investigation

Site Selection and Well Installation
Sample Collection and Analysis
Quality-Control Data
Ancillary Data

Sediment Characteristics
Aquifer Properties
Age Dating
Quality of Shallow Ground Water

Physical Properties
Dissolved Solids and Major Ions
Nutrients and Dissolved Organic Carbon
Trace Elements
Pesticides
Volatile Organic Compounds

Relation to Land Use
Summary and Conclusions
References
Supplemental Information

FIGURES

1-5. Maps showing:

Figure 1. Extent of High Plains aquifer, study area, and monitoring wells sampled as part of an assessment of shallow ground-water quality in areas of recent residential and commercial development, Wichita, Kansas
Figure 2. Land use and areas of recent residential and commercial development in study area, 1999
Figure 3. Surficial geology in study area
Figure 4. Ground-water levels and approximate direction of ground-water flow in unconsolidated deposits in study area, December 1985-January 1986
Figure 5. Occurrence of anaerobic conditions in shallow ground water from monitoring wells in areas of recent residential and commercial development, 2000

Figure 6. Graph showing method reporting limits, sample concentrations and median values, and drinking-water standards for dissolved solids and major ions detected in shallow ground water from 30 monitoring wells in areas of recent residential and commercial development, 2000
Figure 7. Map showing distribution of dissolved-solids concentrations in shallow ground water from monitoring wells in areas of recent residential and commercial development, 2000
Figure 8. Graph showing major ion composition of shallow ground water from monitoring wells in areas of recent residential and commercial development, 20
Figure 9. Graph showing method reporting limits, sample concentrations and median values, and drinking-water standards for nutrients and dissolved organic carbon detected in shallow ground water from 30 monitoring wells in areas of recent residential and commercial development, 2000
Figure 10. Map showing distribution of nitrate as nitrogen concentrations in shallow ground water from monitoring wells in areas of recent residential and commercial development, 2000
Figure 11. Graph showing method reporting limits, sample concentrations and median values, and drinking-water standards for selected trace elements in shallow ground water from 30 monitoring wells in areas of recent residential and commercial development, 2000
Figure 12. Map showing number of pesticide compounds detected in shallow ground water from monitoring wells in areas of recent residential and commercial development, 2000
Figure 13. Map showing occurrence of the herbicide atrazine and (or) its degradation product deethylatrazine in shallow ground water from monitoring wells in areas of recent residential and commercial development, 2000
Figure 14. Map showing number of volatile organic compounds detected in shallow ground water from monitoring wells in areas of recent residential and commercial development, 2000
Figure 15. Graph showing comparison of percentage of detection of atrazine or deethylatrazine in shallow ground water recharged through 19 areas of agricultural and 10 areas of residential and (or) commercial land use in Wichita study area

TABLES

Table 1. Selected well-construction details for monitoring wells installed in areas of recent residential and commercial development, Wichita, Kansas
Table 2. Laboratory analysis methods for analyzed water-quality constituents
Table 3. Estimated aquifer properties at monitoring-well sites in areas of recent residential and commercial development, Wichita, Kansas
Table 4. Apparent dates of recharge calculated from analyses of chlorofluorocarbons for and concentrations of dissolved oxygen in water samples from 30 monitoring wells installed in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 5. Statistical summary and U.S. Environmental Protection Agency drinking-water standards for physical properties in water from 30 monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 6. Statistical summary and U.S. Environmental Protection Agency drinking-water standards for dissolved solids and major ions in water from 30 monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 7. Statistical summary and U.S. Environmental Protection Agency drinking-water standards for selected nutrients in water from 30 monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 8. Statistical summary and U.S. Environmental Protection Agency drinking-water standards for selected trace elements in water from 30 monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 9. Statistical summary and U.S. Environmental Protection Agency drinking-water standards for pesticide compounds detected in water from 30 monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 10. Statistical summary and U.S. Environmental Protection Agency drinking-water standards for volatile organic compounds detected in water from 30 monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 11. Apparent or estimated date of recharge, residence time, flow velocity, and distance to potential recharge areas for shallow ground water from 30 monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 12. Apparent or estimated recharge dates and land use of potential recharge areas at date of recharge for shallow ground-water samples from 30 monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 13. Water-quality constituents analyzed in shallow ground-water samples from monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 14. Results of physical and chemical analyses of shallow ground-water samples from monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000
Table 15. Results of pH, organic-carbon, and particle-size analyses of sediment samples collected during installation of shallow ground-water monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000

CONVERSION FACTORS, ABBREVIATIONS, AND DATUM

The use of brand, trade, or firm names in this report is for identification purposes only and does not constitute endorsement by the U.S. Geological Survey.

Quality of Shallow Ground Water in Areas of Recent Residential and Commercial Development, Wichita, Kansas, 2000

By Larry M. Pope, Breton W. Bruce, Patrick P. Rasmussen, and Chad R. Milligan

Abstract

INTRODUCTION

The High Plains Regional Ground-Water Study began in June 1998 and represents a modification of the traditional NAWQA study design in that the ground-water resource is the primary focus of investigation (Dennehy, 2000). The High Plains aquifer is a nationally important water resource that underlies about 174,000 miî in parts of eight Western States (fig. 1). About 27 percent of all irrigated land in the United States is in the High Plains, and about 30 percent of all the ground water used for irrigation in the United States is pumped from this aquifer. In addition, the aquifer system provides drinking water to 82 percent of the people who live within the aquifer boundary (Dennehy, 2000).

NAWQA ground-water studies include a component designed to assess the occurrence of water-quality constituents under areas of specific land use. The general objective of land-use studies is to assess the natural factors and human-related activities that affect the quality of recently recharged (generally less than 10 years old) shallow ground water that underlies key types of land use within each NAWQA study area. In 2000, the USGS began a ground-water-quality study of the unconsolidated deposits underlying areas of recent (1960-96) residential and commercial development in the Wichita, Kansas, area.

Studies of the potential effects of urban land use on shallow ground-water quality have two specific objectives: (1) Assess the water quality in recharge to shallow ground water underlying areas of recent residential and commercial development in large metropolitan areas, and (2) determine if ground-water-quality characteristics are related to the overlying urban land use (Squillace and Price, 1996).

Shallow depths to water, lack of a geologic barrier to slow downward migration of contaminants, and large hydraulic conductivity of unconsolidated deposits in and near the city of Wichita contribute to the potential for degradation of water quality in the High Plains aquifer by overlying land uses (Petri and others, 1964). Residential and commercial land uses and nearby agricultural land are potential sources of contaminants to the shallow ground water.

Purpose and Scope

The purpose of this report is to provide an assessment of water quality in recharge to shallow ground water underlying areas of recent residential and commercial development, Wichita, Kansas, and to relate, to the extent possible, ground-water quality to overlying land use (fig. 2). Water samples from 30 monitoring wells installed for this study (fig. 1) were analyzed for about 170 water-quality constituents, including chlorofluorocarbons (CFCs), physical properties, dissolved solids and major ions, nutrients and dissolved organic carbon, trace elements, pesticide compounds, and volatile organic compounds (VOCs). Many of these constituents are regulated in public drinking-water supplies by the U.S. Environmental Protection Agency (USEPA). The constituents will be discussed relative to USEPA drinking-water standards as a frame of reference and in relation to natural hydrologic factors and human-related activities.

Acknowledgments

DESCRIPTION OF STUDY AREA

The study area is located in Sedgwick County, south-central Kansas (fig. 1), and includes recently developed (1960 to 1996) residential and commercial areas of the city of Wichita and adjacent urban areas (fig. 2). The study area is located mostly within the easternmost extension of the High Plains aquifer and the alluvial and terrace deposits of the Arkansas River Valley in the Arkansas River Lowlands physiographic province (Schoewe, 1949). Land surface is flat within the Arkansas River Valley but becomes gently rolling in the uplands to the west and east of the study area. The Arkansas River and associated tributary streams generally flow south or southeast in the study area.

Climate and Precipitation

Land and Water Use

Land use in the study area is mostly associated with urban development. Residential and commercial land use occurs in areas near the 30 monitoring wells (fig. 2); however, peripherial to these urban areas are agricultural lands. Most of the area currently (2001) included in the city of Wichita has been agricultural land at some time. The growth and expansion of Wichita historically has been into agricultural areas. However, despite this urban encroachment, agriculture is still a dominant feature in Sedgwick County. In 1998, of the 640,000 acres in Sedgwick County, 86 percent (553,000 acres) was associated with the 1,570 farms in the county (Kansas Department of Agriculture and U.S. Department of Agriculture, 1999).

The current (2001) volume of ground-water withdrawn and used in the study area is not readily available; however, ground-water use information on a county-wide basis provides an approximation of water use within the study area. Of the 58,000 acre-ft of ground water withdrawn in Sedgwick County in 1997, 46 percent was used for irrigation, 24 percent for municipal use, 23 percent for industrial use, 6 percent for domestic self-supplied use, and 1 percent for a combination of recreation and stock use (Joan Kenny, USGS, written commun., 1999).

Within the study area (fig. 1), the vast majority of residents are supplied drinking water from the Wichita public-water supply. This public supply consists of about 60- to 70-percent surface water (reservoir water) with the remaining percentage coming from ground-water wells located in the Wichita well field in the High Plains aquifer located between the cities of Valley Center and Burrton (fig. 1) (Jerry Blain, city of Wichita Water and Sewer Department, oral commun., 2000). However, there are some residents who get drinking water from domestic water-supply wells completed in the alluvial and (or) terrace deposits of the Arkansas River Valley (fig. 3). Water from the alluvial deposits in the study area also is used as a supplementary public-water supply for Wichita. This supplementary ground-water supply is withdrawn from a small well field located near the confluence of the Arkansas and Little Arkansas Rivers (fig. 1) (Vernon Strasser, city of Wichita Water and Sewer Department, oral commun., 2002).

Wastewater from areas of residential and commercial development in the study area generally is discharged through municipal sanitary-sewer systems. However, some areas of onsite residential wastewater disposal (septic systems) do occur. Septic systems are used in the residential areas near monitoring wells 19 and 30 (fig. 1) (Brian Fisher, city of Wichita Department of Environmental Health, oral commun., 2002), well 23 (Dave Harper, city of Haysville, oral commun., 2002), and well 24 (Rob Younkin and La Donna Lawrenz, city of Wichita Water and Sewer Department, oral commun., 2002), and wells 17 and 18 west of Valley Center (Linda Heberling, city of Valley Center, oral commun., 2002) are in areas of past septic-system usage. The part of the study area currently (2001) affected by onsite disposal of residential wastewater is spatially limited, and similarly, the potential for adverse effects on ground-water quality from wastewater probably are limited.

Hydrogeologic Setting

The hydrogeologic setting of the study area has been described in detail by Williams and Lohman (1949), Petri and others (1964), and Lane and Miller (1965). A brief summary of those descriptions is presented here. Generally, unconsolidated deposits, primarily of Quaternary age, occur over bedrock in the study area (fig. 3). These deposits, both in and outside of the High Plains aquifer, consist of alluvial and terrace deposits of fine-to-coarse sand and gravel with interspersed thin layers of silt and (or) clay. Specifically, the deposits adjacent to the Arkansas River consist of fine-to-coarse sand and fine-to-very coarse gravel containing only minor amounts of silt and clay (Nal, fig. 3). The sand and gravel grade upward into clayey silt. The sand and gravel are mostly quartz fragments, but the gravel contains large amounts of pink feldspar and other minerals typical of the Rocky Mountains from which this material was eroded (Lane and Miller, 1965).

Sand and gravel in the unconsolidated deposits adjacent to the Arkansas River form the principal ground-water reservoir in and near Wichita and are the source of most of the ground water used. Saturated thickness of the sand and gravel ranges from less than 40 ft near Haysville (fig. 1) to about 180 ft near Maize (Hansen, 1987).

Recharge to the unconsolidated deposits is mainly by local precipitation with an annual average recharge of about 3 in. (Hansen, 1987). Annual fluctuations in the water table of 2 to 4 ft are common; however, fluctuations of as much as 10 ft are not unusual. The water table is 30 ft or less below land surface in most of the study area and generally slopes southeastward (parallel to the Arkansas River) at 5 to 7 ft/mi (Petri and others, 1964). Properly constructed wells in the Quaternary-age unconsolidated deposits of the Arkansas River Valley may yield as much as 1,500 gal/min or more (Lane and Miller, 1965).

METHODS OF INVESTIGATION

Site Selection and Well Installation

Selection of sites for installation of monitoring wells was a multiple-step process. First, target areas of recently developed residential and commercial land use were delineated. These target areas generally were defined as the intersection of the alluvial aquifer and areal extent of recently developed residential and commercial land uses in the study area. Second, the target areas were refined by excluding (1) areas within 1,000 m of heavy industries such as manufacturing or construction facilities, refineries of raw materials, transportation centers, or airports, and (2) areas where at least 75 percent of a 500-m buffer around potential well sites would not be in the target area. These target areas are outlined and identified in figure 2 as areas of recent residential and commercial development. Third, potential well sites were selected using an equal-area, randomized, grid-based computer program (Scott, 1990). This program was used to select 30 primary well-installation sites each with one or two secondary locations within the refined target areas. The secondary locations were identified in the event that a well-installation site could not be located within 250 m of the primary location. The computer program also provided at least a 750-m separation distance between all well locations. The locations of the 30 monitoring wells installed for the land-use study described in this report are shown in figure 1.

The installation of monitoring wells followed procedures outlined by Lapham and others (1995). All 30 monitoring wells were installed between March 24 and April 28, 2000, using a 4.25-in. inside-diameter hollow-stem auger. Well casings were 2-in. diameter, flush-threaded, schedule-40 polyvinyl chloride (PVC) pipe. Well screens also were constructed of schedule-40 PVC and were 5 ft in length with mill-slot perforation openings (0.01 in. wide). The filter pack (packing material around the well screen) consisted of silica sand above which about a 2-ft thick layer of bentonite pellets was installed. An annular seal of bentonite grout was emplaced above the bentonite pellets to within about 3 ft of land surface where a surface seal of concrete was installed. Selected well-construction details are listed in table 1. All drilling equipment was steam cleaned between monitoring-well installation sites.

All monitoring wells installed for this land-use study were completed at depths (below land surface) of 47 ft or less (table 1). Completion depths were determined by depth to water. One of the purposes of well construction was to ensure suitability for future (10 or more years) sampling for water-quality trend analysis. Therefore, the placement of the well screen relative to possible future water-level declines was an important consideration. Generally, the goal at well installation was to place the top of the well screen about 5 ft below the current (2000) water level. The top of well screens for all 30 monitoring wells were installed at a median depth below current (2000) water levels of about 6 ft. Depths to water ranged from about 4 to 35 ft, with a median depth of about 16 ft.

After installation, water levels in the monitoring wells were allowed to "recover" for about 2 weeks before development. Well development is a procedure to enhance flow of water to the well, to remove sediment that are artifacts of well installation, and to yield water representative of the aquifer being sampled. Well development mitigates artifacts associated with drilling such as changes in aquifer permeability, sediment distribution, and ground-water chemistry (Lapham and others, 1995). For the land-use study described in this report, well development consisted of pumping with a portable, low-volume submersible pump at three depth intervals in the water column. These intervals, near the top, middle, and bottom of the water column, were pumped until turbidity readings in the discharged water from each interval were less than 10 NTU (nephelometric turbidity units). Turbidity is a measure of the clarity of the water. The water levels in the monitoring wells again were allowed to "recover" for 2 weeks before water-quality samples were collected.

Sample Collection and Analysis

Sampled wells were first purged of standing water. During the initial pumping period, measurements of specific conductance, pH, water temperature, and dissolved oxygen were monitored every 5 minutes in a closed-cell, flow-through chamber until stable readings were obtained. Turbidity also was measured every 5 minutes using a portable turbidity meter. Once stable readings of these physical properties were obtained, water flow inside the laboratory was redirected to the clean sampling chamber where sample water was collected immediately for analysis. Constituents analyzed in water samples collected from each well are listed in table 13 in the "Supplemental Information" section at the back of this report.

Samples for the analysis of chlorofluorocarbons (CFCs) were collected in triplicate using a procedure described by Busenberg and Plummer (1992) that prevents exposure of the samples to air (a potential source of CFC contamination). The sampling procedure uses a closed water-transport path between the sampling pump through refrigeration-grade copper tubing to a valve system that allows filling and flushing of 62-mL borosilicate ampoules with water, creating a headspace with CFC-free ultrapure nitrogen, and permanently sealing the ampoule neck by heat fusion. Samples were analyzed at the USGS Chlorofluorocarbon Laboratory in Reston, Virginia, by gas chromatography to a detection limit of about 0.3 pg/kg of water. Results of CFC analyses are subsequently used in the calculation of potential dates at which the water sampled was recharged to the aquifer.

Samples to be analyzed for major ions, nutrients, and trace elements were filtered through a 0.45-mm pore-size disposable-capsule filter and collected in precleaned plastic bottles that were rinsed onsite with filtered ground water. Samples to be analyzed for concentrations of major cations and trace elements were preserved to less than pH 2.0 standard units using ultrapure nitric acid. A filtered, unpreserved sample was collected for major anion analysis. Additionally, a filtered sample was collected for onsite titration of carbonate alkalinity. An unfiltered sample for laboratory measurements of specific conductance and pH was filled directly in the sampling chamber and was not preserved. Nutrient samples were filtered (0.45-mm pore size) into onsite-rinsed brown plastic bottles and were chilled on ice, with no preservation, and were delivered to the laboratory within 24 hours for analysis.

To avoid contact between water sampled for dissolved organic carbon (DOC) and any methanol-rinsed sampling equipment (a possible source of DOC contamination), DOC samples were collected directly from the pump discharge tube close to the wellhead. The sample was processed through a stainless-steel, pressure-filtration funnel equipped with a 0.45-mm pore-size silver filter. Trace concentrations of silver imparted by the filter served as a biocidal preservative for the DOC sample. Water samples were forced through the silver filter using purified nitrogen gas at a pressure not exceeding 15 lb/inî. As with other organic samples, the water was collected in a cleaned and baked amber glass bottle and immediately chilled on ice and were delivered to the laboratory within 24 hours for analysis.

Water samples collected for pesticide analyses were passed through a methanol-rinsed, stainless-steel filter chamber containing a cleaned and baked 0.7-mm pore-size glass-fiber filter. All samples analyzed for organic constituents were collected in amber-colored glass bottles and immediately chilled on ice and were delivered to the laboratory within 24 hours for analysis. Unfiltered water for analysis of VOCs was collected in precleaned 40-mL septum vials with no headspace and preserved to less than pH 2.0 standard units using one to five drops of specially prepared 1:1 hydrochloric acid.

Analysis of all water-quality samples was performed at the USGS National Water-Quality Laboratory (NWQL) in Lakewood, Colorado, according to methods listed in table 2. Results of these analyses are listed in table 4 in the section on "Age Dating" and in table 14 in the "Supplemental Information" section at the back of this report.

Quality-Control Data

Aluminum and zinc were each detected in two of three field-blank samples at mean concentrations of 10 and 4 Îg/L, respectively. Barium and strontium were each detected in one field-blank sample at concentrations of 1 and 2 Îg/L, respectively. None of the other trace elements, dissolved solids and major ions, nutrients, pesticide compounds, or VOCs listed in table 13 (in the "Supplemental Information" section at the back of this report) were detected at concentrations greater than the analytical method reporting limit. The analytical results of field-blank samples indicate that the decontamination procedures used during this study were adequate and that onsite and laboratory contamination of environmental samples was minimal.

Three sets of replicate environmental samples were collected sequentially during this study for all analyses except pesticide compounds and VOCs. Replication of pesticide compounds and VOCs was accomplished during the field matrix-spiking process wherein replicate environmental samples were injected (spiked) with known concentrations of selected pesticide compounds or VOCs.

Analytical variation between reported concentrations of replicate environmental samples was calculated for the major ions, nutrients, dissolved organic carbon, and trace elements listed in table 13. The variation, as a percentage, between constituent concentrations of replicate environmental samples was calculated with the equation:

where   A is concentration from one replicate analysis, and
           B is concentration from the other replicate analysis.
Analytical variation was not calculated if one or both measured concentrations for a replicate pair were less than analytical method reporting limits.

Mean percentage variation between replicate analyses was less than 5 percent for all constituents except fluoride (7 percent), ammonia (6 percent), ammonia plus organic nitrogen (10 percent), orthophosphate (10 percent), dissolved organic carbon (15 percent), aluminum (13 percent), strontium (11 percent), and zinc (14 percent). Generally, however, the results of replicate environmental sample analyses indicate an acceptable degree of laboratory precision and reproducibility.

Analytical recovery bias and variability were assessed for the pesticides listed in table 13 using analyses of three field matrix-spike samples collected during this study. Replicate environmental samples were spiked with known amounts of pesticides and analyzed according to methods listed in table 2.. Percentage recoveries were calculated with the equation:

where   FMS is the analytically determined concentration in the field matrix-spike sample,
              CS is the calculated spike concentration, and
              PS is the analytically determined concentration in the primary environmental sample.

Mean percentage recoveries for pesticides listed in table 13 generally ranged from 26 percent (disulfoton) to 179 percent (carbaryl) with a mean percentage recovery for all 47 pesticide compounds of 102 percent. Most mean percentage recoveries (64 percent of the pesticides listed in table 13) were between 90 and 110 percent and included pesticide compounds that might be used in an urban environment and potentially may occur in shallow ground water. These potentially urban-related pesticides and percentage recoveries included atrazine (109 percent), dieldrin (99 percent), metolachlor (109 percent), pendimethalin (93 percent), prometon (105 percent), and simazine (99 percent). Two additional urban-related pesticide compounds, deethylatrazine (a degradation product of atrazine) and tebuthiuron, had mean percentage recoveries of 75 and 124 percent, respectively. These data indicate that sample processing and analytical methods provided a reasonable approximation of expected pesticide concentrations for most of the pesticide compounds listed in table 13.

Percentage recoveries for VOCs generally were less than those for pesticides. Mean percentage recoveries for the VOCs (table 13) ranged from 47 percent (3-chloropropene, 2,2-dichloropropane, and methyl iodide) to 106 percent (2-butanone). The average mean percentage recovery for all 85 VOCs was 66 percent, meaning that data in this report may underestimate true concentrations and frequencies of detection of VOCs by an average of 34 percent.

Ancillary Data

SEDIMENT CHARACTERISTICS

Sediment pH values (table 15 in the "Supplemental Information" section at the back of this report) were near neutral (pH value of 7.0 standard units) to alkaline (pH value greater than 7.0 standard units) at all 30 monitoring-well sites (fig. 1). pH values ranged from 6.8 to 8.4 standard units with a median value of 7.6 standard units. Therefore, pH was not small enough to promote the larger solubilities of most minerals and nutrients and degradation of some organic compounds in water recharged at the locations of the 30 monitoring wells.

Organic carbon can serve as adsorption sites for chemicals such as nutrients (Lefroy and Craswell, 1997) and pesticides (Van Es, 1990). The amount of organic carbon in sediment determines its potential for chemical adsorption. Sediment with large organic-carbon content has a small potential for chemical leaching and, therefore, mitigates the movement of some water-quality constituents to the water table. Organic-carbon content of 1 percent (dry weight) is considered small (Pepper and others, 1996). Organic-carbon content of sediment or soils derived primarily from weathering of bedrock usually ranges from 0 to 4 percent but may be as large as 20 to 30 percent (Bauder, 1999).

Organic-carbon content in the sediment from the 30 monitoring-well sites (fig. 1) was small (table 15). Organic-carbon content ranged from less than 0.02 percent at many well locations to 0.5 percent at well 4 (table 15, fig. 1). These small percentages probably indicate that chemical adsorption of water-quality constituents in water recharged through sediment at the monitoring-well sites would be minimal and that the movement of constituents to the shallow water table would not be impeded.

Sediment particle size can affect water quality in an underlying aquifer by increasing or decreasing the ability of water and associated water-quality constituents to percolate through the material rather than run off over the land surface. Coarse-textured material (large percentage of sand and (or) gravel) generally has large potential for leaching contaminants to ground water and small potential for surface-runoff loss to streams and lakes (Van Es, 1990).

Sediment particle sizes (table 15) at most of the 30 monitoring-well sites (fig. 1) were sand size or larger (larger than 0.062 mm in diameter). Overall, the percentage of silt- and (or) clay-size particles (less than 0.062 mm in diameter) for all 30 monitoring-well sites was small (median of 3.4 percent). Most of the silt- and (or) clay-size particles were associated with the unsaturated zone (upper sampling interval) at each monitoring-well site. Samples collected in this zone had a median percentage of silt and (or) clay of 23.4 percent. In contrast, the median percentage of silt and (or) clay in the saturated zone (below the water level) was 1.6 percent. Part of the difference in silt and (or) clay median percentages between the saturated and unsaturated zones may be attributed to sediment sampling procedures. Water draining from the split-spoon sampler from samples collected in the saturated zone might have carried off some fine sediment particles. However, this phenomenon could not be solely responsible for the large difference in median percentages between saturated and unsaturated zone samples. Data in table 15 indicate that sediment in the study area generally is coarse-textured and probably has a large potential for leaching contaminants to ground water.

AQUIFER PROPERTIES

Most of the water-bearing unconsolidated deposits in the Wichita study area consist of sand- and gravel-size particles (greater than 0.062 mm in diameter; table 15 at the back of this report). Estimates of hydraulic conductivity for these deposits have been made in previous ground-water investigations in and near the study area (Spinazola and others, 1985; Myers and others, 1996). Average hydraulic conductivity values of 350 and 750 ft/d were estimated for the terrace deposits (Nt, fig. 3) west of the Arkansas River alluvium and alluvial and terrace deposits (Nal, fig. 3) along the Arkansas River, respectively.

Hydraulic gradient is defined as the change in hydraulic head per unit of distance measured in the direction of steepest change (Heath, 1983). The hydraulic gradient in an unconfined aquifer is the slope of the water table. The slope of the water table determines direction of ground-water flow and is a component in calculating ground-water flow velocity.

The hydraulic gradient for most of the area where monitoring wells were installed for the study described in this report is about 0.001 with the possible exception of the area where monitoring wells 1 and 11 are located, where the ground-water flow direction is not definitely known (fig. 4). Generally, the direction of ground-water flow in the area of the monitoring wells is from the northwest to the southeast.

Most water-bearing material (consolidated or unconsolidated rocks) consists of mineral particles surrounded by open spaces. The porosity of a water-bearing material may be expressed as the ratio (dimensionless) of the volume of openings in a rock to the total volume of the rock (Heath, 1983). Ground water moves into, through, or is stored in these openings. Porosity, therefore, is an important determination in describing other ground-water characteristics such as flow velocity (Nielson, 1991). Porosity of unconsolidated deposits depends on the range in grain size (sorting) and on the shape of the rock particles but not on their size. Fine-grained materials tend to be better sorted and thus tend to have the largest porosity (Heath, 1983). For example, porosity of clay typically ranges from 0.40 to 0.70, whereas porosity of sand ranges from 0.15 to 0.48 and for gravel, from 0.25 to 0.40 (Nielson, 1991). On the basis of the information for sediment samples collected from the saturated zone (deeper of the paired samples from each well location) presented in table 15, an average porosity of 0.30 was estimated for the alluvial aquifer in the study area.

The rate of movement of ground water (flow velocity) is an important characteristic in studies of ground-water quality because, when used with estimated age of the ground water, location of potential recharge areas can be estimated. The land-use characteristics of these recharge areas may help in the evaluation of water-quality characteristics of the water samples collected from the 30 monitoring wells (fig. 1).

Ground-water flow velocity is a function of hydraulic conductivity, hydraulic gradient, and porosity and can be calculated with the equation (Heath, 1983, p. 25):

where   v        is ground-water flow velocity, in feet per day;
           K        is the hydraulic conductivity, in feet per day;
           n        is the porosity of the aquifer material, dimensionless; and
          dh / dl is the hydraulic gradient, dimensionless.

Ground-water flow velocities at each of the 30 monitoring wells (fig. 1) were calculated using either a hydraulic conductivity of 350 or 750 ft/d (refer to previous discussion of hydraulic conductivity), a hydraulic gradient of 0.001, and a porosity of 0.30. Ground-water flow velocities at the 30 monitoring wells were estimated at either 1.2 or 2.5 ft/d (table 3).

AGE DATING

Results of CFC analyses of water samples from the 30 monitoring wells (fig. 1) and calculated apparent recharge dates (table 4) indicate that the recharge ages of the collected samples ranged from about 1979 (monitoring well 2) to about 1990 (monitoring wells 5, 15, and 30) with an average apparent recharge age of about 1986 (calculated on the basis of dates from 18 monitoring wells). Apparent recharge dates listed in table 4 are averages of the highlighted (shaded) reported CFC dates from the three replicated samples analyzed for each well. Not all of the possible reported dates were used to calculate average dates because some CFC concentrations were considered contaminated, others were considered degraded because of anaerobic conditions (concentration of dissolved oxygen less than 0.5 mg/L), and others appeared to be unusually old such as at monitoring well 1. Recharge dates reported as "modern" (less than 10 years old) in table 4 may indicate recent recharge but were not quantifiable.

The average reported recharge date (1964) of the 12 samples (table 4) with dissolved-oxygen concentrations less than or equal to 0.5 mg/L was 22 years older than the average apparent recharge date (1986) for the other 18 samples, which indicates that CFC degradation probably did occur in the anaerobic parts of the aquifer. Only "non-less-than" dates (table 4) were used to calculate the average reported recharge date for the 12 anaerobic samples.

QUALITY OF SHALLOW GROUND WATER

Physical Properties

Physical properties were measured in water from each of the 30 monitoring wells (fig. 1) sampled during this study (table 14 at the back of this report). Measurements made at the time of sample collection included specific conductance, pH, water temperature, turbidity, dissolved oxygen, and alkalinity. A statistical summary and USEPA standards for physical properties in water from 30 monitoring wells sampled during this study are presented in table 5.

Water from most wells had specific conductance values less than 1,000 ÎS/cm (microsiemens per centimeter at 25÷C). Water from eight wells (6, 13, 20, 21, 24, 25, 28, 29) had specific conductance values greater than 1,000 ÎS/cm. Specific conductance values for these eight wells ranged from 1,020 to 2,590 ÎS/cm. The eight wells generally are located near and along the west side of the Arkansas River (fig. 1). Specific conductance describes the ability of water to conduct an electrical current and provides an indication of ion concentrations or dissolved solids. As specific conductance of the water increases, so does the ion concentration (Hem, 1985).

The pH of water from all except three sampled wells was within the USEPA SMCL range of 6.5 to 8.5 standard units in drinking water. The exceptions were water from wells 8 and 15 (pH of 6.4 standard units) and well 16 (pH of 6.3 standard units). pH values less than 7.0 are indicative of acidic water, and values greater than 7.0 are indicative of alkaline water.

The turbidity (cloudiness) of water from most wells was low, which generally indicates not only an acceptable aesthetic appeal and a lack of obvious contamination but also that the wells were developed properly and that samples were representative of the aquifer water. However, water from well 10 had a turbidity of 8.8 nephelometric turbidity units (NTU) (table 14) and was the only well where the water exceeded the USEPA drinking-water standard of 5.0 NTU. The second largest turbidity was 3.4 NTU in water from well 17. The relatively large turbidity in water from well 10 probably was an artifact of recent installation and infrequent pumping. Turbidity has no health effects but can interfere with disinfection and provide a medium for microbial growth.

Water from the wells had a median dissolved-oxygen concentration of 2.6 mg/L (table 5). Water from three wells (21, 22, 24) had no detectable dissolved oxygen (less than 0.1 mg/L). Water from nine other wells had dissolved-oxygen concentrations of 0.5 mg/L or less (table 14). Most of the wells (7 of 12) with water under anaerobic conditions (0.5 mg/L or less of dissolved oxygen) were clustered in a relatively small area north of Haysville between the Wichita-Valley Center Floodway and the Arkansas River (fig. 5).

Oxygen enters ground water through recharge of oxygen-enriched water that percolates down through the unsaturated zone where it may react with oxidizable material encountered along the flow path of the water. The principal reacting materials are organics and reduced inorganic minerals such as pyrite and siderite (Hem, 1985). The small dissolved-oxygen concentrations in water from many of the wells sampled during this study may represent natural and localized, oxygen-demanding (reducing) conditions in the unsaturated zone or may be an indication of contamination from the distribution or disposal of organic material associated with human activities.

Human-related organic material can originate from urban sources such as leachate from domestic septic systems and sanitary landfills, the use of land for sewage disposal, refuse dumps for disposal of organic compounds, the burial of containers with organic compounds, leakage from liquid-waste storage ponds, accidental spills along roads and highways, and the use of pesticides. Additional sources of organic material may include artifacts of past agricultural activities such as crop production, distribution of animal manure, and confined animal-feeding operations. Other types and sources of oxidizable material include synthetic fertilizers used on residential lawns, parks, golf courses, or around commercial areas.

Although the percentage of organic carbon in the unconsolidated, sediment collected at the 30 well sites was small (table 15 at the back of this report), larger percentages of organic carbon may be associated with surficial soils (depths less than about 5 ft) in the study area. The surficial soils were not sampled for this study, and the organic-carbon content of these soils are not readily available from other sources. However, it is expected that organic-carbon content of the soil profile would be larger in the surficial soils than in the deeper sediment sampled during this study because of proximity to organic materials such as plant residues, waste products, pesticides, and others sources of organic contamination. Freeze and Cherry (1979, p. 245) summarized the concept of oxygen depletion in recharge water, "It is reasonable to expect that the consumption of oxygen in the soil zone will vary depending on numerous factors, such as the soil structure, porosity and permeability, nature and depth distribution of organic matter, frequency of infiltration events, depth to water table, and temperature." They further stipulated that in areas of appreciable dissolved oxygen in shallow ground water in sandy deposits that it "...is probably a result of low contents of organic matter in the soil and rapid rates of infiltration through the soil."

Alkalinity as CaCO₃ (calcium carbonate) in water is a measure of the acid-buffering capacity of a filtered water sample. Alkalinity as CaCO₃ ranged from 68 to 570 mg/L with a median concentration of 265 mg/L in the ground-water samples collected (table 5). Water with excessive alkalinity as CaCO₃ may have a distinctive unpleasant taste, and irrigation water with excessive alkalinity may increase the pH of the soil solution, leach organic material, decrease permeability of the soil, and impair plant growth (Hem, 1985).

Dissolved Solids and Major Ions

The major constituents of dissolved solids in shallow ground-water samples collected from areas of recent residential and commercial development in Wichita were the cations (positively charged ions) calcium, magnesium, potassium, silica, and sodium and the anions (negatively charged ions) bicarbonate, bromide, chloride, fluoride, and sulfate. A statistical summary and USEPA drinking-water standards (where applicable) for dissolved solids and major ions in water from 30 monitoring wells sampled during this study are presented in table 6. Method reporting limits, sample concentrations and median values, and associated drinking-water standards for dissolved solids and major ions are presented in figure 6.

Concentrations of dissolved solids in water from the 30 monitoring wells sampled ranged from 229 (well 16) to 1,630 mg/L (well 6) (tables 6 and 14, fig. 6). The distribution of dissolved-solids concentrations in water is shown in figure 7. The median dissolved-solids concentration was 492 mg/L, only slightly less than the 500-mg/L SMCL established by the USEPA (U.S. Environmental Protection Agency, 2000a). Water from 14 wells (47 percent) exceeded the SMCL for dissolved solids (fig. 7).

The distribution of dissolved solids in shallow ground water in areas of recent residential and commercial development was similar to that determined for the Wichita area by Lane and Miller (1965) and Bevans (1989). Water from wells closest to the Arkansas River generally had the largest dissolved-solids concentrations (fig. 7). Water with dissolved-solids concentrations less than 500 mg/L generally came from wells farthest west of the Arkansas River in the Cowskin Creek area.

The largest dissolved-solids concentrations determined during this study probably are related to highly mineralized water moving out of the Arkansas River into the shallow, unconsolidated deposits along the river. This movement may occur naturally when the river stage is higher than the ground-water table or may be locally enhanced by pumpage of water for municipal, industrial, and irrigation purposes (Lane and Miller, 1965). The source of highly mineralized water in the Arkansas River probably is from saline ground water discharged from Permian rock along the Arkansas River about 60 mi northwest of Wichita and, to a lesser extent, from brine disposal related to salt and oil production near Burrton in western Harvey County (Bevans, 1989).

Calcium bicarbonate water is the dominant ground-water type in areas of recent residential and commercial development (fig. 8). However, this water type may be modified locally by the mixing of sodium chloride type water along the western side of the Arkansas River. Sodium chloride type water generally is present in alluvium and terrace deposits that occur along the western side of the Arkansas River north of Kansas Highway 96 (fig. 1). The source of this sodium chloride type water is the Arkansas River (Bevans, 1989).

Associated with the variability in or mixing of water types, the concentrations of chloride (390 mg/L) and sulfate (310 mg/L) in water from well 6 both exceeded the 250-mg/L SMCLs for these constituents. Both of these concentrations were larger than concentrations in water from the other wells (fig. 6, table 14). The probable source of the relatively large concentration of chloride may be the infiltration of sodium chloride water from the Arkansas River into adjacent alluvium and terrace deposits. Although there are natural sources of chloride in the study area to produce the concentrations in shallow ground water determined during this study, the discharge of human, animal, or industrial wastes and irrigation return flows may add chloride to ground water. The relatively large concentration of sulfate may be from the oxidation of pyrite and other sulfides (common in sedimentary rocks) in well-oxygenated water or the dissolution of evaporite deposits such as gypsum (calcium sulfate).

Few, if any, human-health effects are associated with chloride or sulfate in the concentration range determined during this study; however, concentrations larger than 250 mg/L may produce an objectionable taste in drinking water. Chloride may impart a salty taste to drinking water and may accelerate the corrosion of metals used in water-supply systems. Sulfate in drinking water may impart a bitter taste and act as a laxative on unacclimated users (Hem, 1985).

Nutrients and Dissolved Organic Carbon

Concentrations of most nutrients in water from the 30 wells sampled during this study were small (table 14 at the back of this report). Concentrations of ammonia as nitrogen ranged from less than 0.02 mg/L (not detected) in water from 20 wells (fig. 9; table 7) to 0.18 mg/L (well 22, table 14). Of the 10 wells with detectable concentrations of ammonia, water from seven (wells 4, 22, 24-28) were under anaerobic conditions (fig. 5). With the exception of monitoring well 4, wells 22 and 24-28 were in the cluster of wells with anaerobic water north of Haysville. Generally, ammonia is more common in ground water under reducing (anaerobic) conditions because the lack or small concentrations of dissolved oxygen stimulates the denitrification process. Denitrification refers to the biological reduction (loss of oxygen) of nitrate (NO₃^-) to ammonia (NH₃) or the ammonium ion (NH₄⁺). In the absence of dissolved oxygen, heterotrophic bacteria in soil and water may use nitrate as an oxygen source during respiration (Canter, 1997).

Little organic nitrogen (carbon compounds containing nitrogen such as proteins, peptides, amino acids, and nucleic acids) was detected in water from wells in the study area. Most of the detections of organic nitrogen [(ammonia + organic nitrogen) -ammonia, table 14] were from wells with water under anaerobic conditions. Of the 11 detections of organic nitrogen (wells 6, 8, 13, 22, 24-30), eight were in anaerobic water. Generally, the metabolism of organic compounds is an oxidation process, and the lack of oxygen will inhibit the rate of this process.

Concentrations of nitrite plus nitrate nitrogen (hereinafter referred to as nitrate or nitrate, as nitrogen) in water from the 30 wells sampled during this study varied throughout the study area (fig. 10). Concentrations of nitrate ranged from less than 0.05 (wells 22, 24, 27) to 10 mg/L (wells 12, 30) with a median concentration of 2.2 mg/L. Although water from none of the wells sampled during this study had concentrations greater than the 10-mg/L drinking-water MCL, water from 15 wells (50 percent) showed enrichment when compared to a 2.0-mg/L national background nitrate concentration (Mueller and Helsel, 1996). This enrichment probably is the result of human-related activities.

The 2.2-mg/L median nitrate concentration determined for water from the 30 monitoring wells sampled during this study was slightly larger than the 1.6-mg/L median concentration determined from a national study of shallow ground water beneath urban land (Nolan and Stoner, 2000). About one-half (53 percent) of shallow ground-water studies in urban and agricultural areas in the United States reported median nitrate concentrations in well water larger than the national background concentration of 2.0 mg/L (U.S. Geological Survey, 1999). Of the urban studies, 40 percent had median nitrate values larger than the national background concentration.

Most phosphorus in water from wells in the study area occurs as orthophosphate (fig. 9). Orthophosphate constituted a median percentage of 94 percent of the total phosphorus concentration in water from the 30 wells sampled in the study area. Orthophosphate (PO₄^-3) together with condensed forms of phosphate that include metaphosphate (P₂O₆^-2), pyrophosphate (P₂O₇^-4), and tripolyphosphate (P₃O₁₀^-5) constitute the major inorganic forms of phosphorus in natural water (Patnaik, 1997). However, the condensed phosphates are unstable in water and in time revert to orthophosphate (Hem, 1985). Concentrations of orthophosphate ranged from less than 0.01 (well 28) to 0.32 mg/L (well 1).

Most uncontaminated ground water has only small concentrations of dissolved organic carbon (DOC). In the absence of organic contamination such as leachate from landfills or spills or leaks from areas of oil and gas production, the source of DOC in ground water likely would be the end products of the biochemical cycle in the soil profile above the aquifer. These end products, generally, are humic and fulvic acids associated with the biological degradation and decay of soil-organic material or organic amendments (manure) applied to the soil. If present in large enough quantities, however, organic solutes composing DOC may form complexes that affect trace-element solubility, participate in redox reactions, and affect both physical and chemical properties of solid-liquid or liquid-gas interfaces (Hem, 1985). No drinking-water-quality standard has been established for DOC.

Concentrations of DOC were small in water from most of the 30 wells sampled during this study (fig. 9, table 7). Generally, concentrations less than 10 mg/L are representative of natural aquifer ranges (Nielsen, 1991, p. 545). However, two relatively large concentrations of 49 and 100 mg/L were measured in water from well 1 and well 16, respectively. The potential contamination evident in water from wells 1 and 16 is probably a localized issue and may originate as leachate from septic systems, waste disposal, or spills.

Trace Elements

Trace elements normally occur in natural water in small concentrations even though some trace elements are naturally abundant. For instance, although aluminum is the second-most abundant element (behind silica) in igneous and sedimentary rocks of the Earth's crust, it rarely occurs in solution in natural water in concentrations greater than a few hundred micrograms per liter (Hem, 1985). The occurrence in ground water of most of the trace elements listed in table 13 (at the back of this report) typically results from natural physical and chemical processes. Physical processes include those associated with the mineral composition and origin, transport, and deposition of the aquifer material. Chemical processes include the action of water, oxygen, carbon dioxide, and other acidic components that cause the chemical breakdown or dissolution of minerals containing trace elements (Bricker and Jones, 1995, p. 1-20). Generally, the occurrence and chemical speciation of trace elements in ground water are determined by (1) hydrogen ion availability (defined by pH); (2) the presence and concentrations of inorganic ligands such as carbonate, sulfate, sulfide, and chloride; (3) the presence and concentrations of organic complexing agents (primarily humic and fulvic acids); (4) free electron availability (reducing or oxidizing conditions); and (5) the ionic strength and cation distribution of the water (Allard, 1995, p. 151-176).

Human activities may affect trace-element concentrations in shallow ground water. For instance, organic and inorganic compounds of copper have been used extensively in agricultural pesticide sprays, and their wide dispersal in cropland areas has the potential to enrich copper concentrations in shallow ground water. However, the source and distribution of trace elements probably are more dominant in urban areas where trace elements such as aluminum, chromium, copper, iron, lead, nickel, and zinc are used as structural or decorative components of buildings, exterior structures and trim work, in automobiles, and as protective coverings or coatings against corrosion and oxidation. Large quantities of some trace elements have been released with effluent discharge from industrial activities, the burning of fossil fuels, and dispersed in automobile exhaust as a result of their addition to gasoline in the refining process. This potentially wide environmental distribution of trace elements creates the possibility for dissolution of some of these elements in water that, ultimately, may recharge a shallow aquifer.

Most trace elements in water from the 30 wells sampled were detected only in small concentrations that were less than established MCLs table 8, fig. 11). Antimony, beryllium, cadmium, lead, and silver were not detected. Chromium (two detections) and cobalt (seven detections) were detected infrequently. Aluminum, copper, molybdenum, selenium, and zinc were detected more frequently, but none of the detected concentrations were larger than established or proposed drinking-water standards. Barium and nickel were detected in water from all 30 wells, but none of the detected concentrations exceeded drinking-water standards.

Of the 30 monitoring wells sampled, concentrations of iron in water from six wells (20 percent) were larger than the 300-Îg/L USEPA SMCL established for drinking water (fig. 11). Concentrations of iron in water from these six wells ranged from 630 (well 28) to 3,700 Îg/L (well 24) (table 14 at the back of this report), probably associated with reducing conditions. Under reducing conditions, iron generally occurs in ground water in the ferrous iron (Fe⁺²) oxidation state, largely as the result of the reduction (loss of oxygen) of ferric oxyhydroxides (Bricker and Jones, 1995). Iron also is present in organic wastes and in plant debris in soil, and microbial activities may affect the occurrence of iron in water. Iron is an essential element in the metabolism of plants and animals. However, if present in excessive amounts, iron forms red oxyhydroxide precipitates that strain laundry and plumbing fixtures and, therefore, is an objectionable impurity in domestic and industrial water supplies (Hem, 1985). The occurrence of iron in water from wells sampled during this study probably is from natural sources.

Water from 12 monitoring wells (40 percent) sampled had concentrations of manganese larger than the USEPA SMCL of 50 Îg/L (fig. 11, table 14). Concentrations of manganese in water from these 12 wells ranged from 61 (well 9) to 1,000 Îg/L (well 4). As was the case with iron, the occurrence of excess concentrations of manganese in shallow ground-water samples was associated mostly with reducing environments. Water from only three wells (1, 9, 11) with concentrations of manganese larger than the SMCL had substantial dissolved-oxygen concentrations (larger than 3.5 mg/L). For the other nine wells, dissolved-oxygen concentrations were equal to or less than 0.3 mg/L. Under reducing conditions, manganese may be released (dissolved) from organic and manganese complexes or from manganese oxides. Manganese is an undesirable impurity in water supplies because of a tendency to deposit black manganese dioxide stains (Hem, 1985). The occurrence of manganese in ground water in the study area is probably from natural sources. A previous investigation of ground water in Sedgwick County concluded that most concentrations of manganese in ground water that exceeded the SMCL are from alluvium and terrace deposits, which may indicate that the manganese is derived from organic matter in the soil and then is leached by percolation of precipitation (or irrigation) through the unconsolidated sand and gravel into the ground water (Bevans, 1989).

In addition to the concentrations of iron and manganese that exceeded SMCLs, the concentrations of arsenic in water from one monitoring well (well 4) and uranium in water from four monitoring wells (fig. 11, table 14), exceeded their proposed MCLs of 10 Îg/L (U.S. Environmental Protection Agency, 2001a) and 30 Îg/L (U.S. Environmental Protection Agency, 2000b), respectively. The 13-Îg/L arsenic concentration in water from well 4 probably was the result of reducing conditions (dissolved-oxygen concentration 0.3 mg/L) in the ground water at this site. Potential health effects from excessive arsenic concentrations in drinking water include cardiovascular disease, diabetes, anemia, and an increased risk of cancer (U.S. Environmental Protection Agency, 2001a).

Concentrations of uranium that exceeded the proposed MCL ranged from 31 (well 1) to 76 Îg/L (well 6). Reducing conditions were evident in all but well 1 of the four wells with uranium concentrations that exceeded the proposed MCL. No known human-related source of uranium exists in the study area. Therefore, the occurrence of uranium in shallow ground water probably results from natural sources.

Exposure to radionuclides such as uranium in drinking water results in an increased risk of cancer (U.S. Environmental Protection Agency, 2000b). This increased risk is the result of "ionizing radiation" from radioactive particles (alpha particles) emitted by the decay of uranium-238. Alpha particles may ionize atoms in cells of living tissue that could damage chromosomes and lead to unnatural cellular reproduction. In addition to this potential "radiotoxicity," damage to the kidneys may occur through exposure to uranium itself ("chemical toxicity").

Some sources of trace elements in shallow ground water from areas of recent residential and commercial development could originate from human-related activities. However, the generally small concentrations that were measured for most trace elements probably reflect natural sources.

Pesticides

A total of 47 pesticide compounds from several classes of herbicides and insecticides that included triazine, organochlorine, organonitrogen, organophosphorus, and carbamate compounds and three pesticide degradation products (table 13 at the back of this report) were analyzed for this study. Of the 30 wells sampled, water from 22 (73 percent) had detectable concentrations of one or more of 8 of the 47 compounds (fig. 12, table 14), A statistical summary and USEPA drinking-water standards (where applicable) for these eight detected pesticide compounds are presented in table 9. The herbicide atrazine or its degradation product deethylatrazine were detected most frequently (water from 21 wells, 70 percent); other detected pesticides included the insecticide dieldrin (in water from one well) and the acetanilide herbicide metolachlor (in water from three wells), the organonitrogen herbicides pendimethalin and tebuthiuron (in water from one well each), and the triazine herbicides prometon (in water from one well) and simazine (in water from nine wells, 30 percent) (table 9).

Atrazine was detected in water at concentrations ranging from 0.005 (wells 6 and 16) to 0.14 Îg/L (wells 7 and 18), and deethylatrazine was detected in water at estimated concentrations ranging from 0.002 (well 6) to 0.08 Îg/L (well 16). The occurrence of atrazine and deethylatrazine in shallow ground water generally was restricted to the western and northern part of the study area. Atrazine or deethylatrazine was not detected in water from the cluster of monitoring wells (wells 22, 24-29) in the southern part of Wichita (north of Haysville), in water from well 17 west of Valley Center in the northern part of the study area, or in water from well 1 in the western part of the study area (fig. 13).

All detected concentrations of atrazine and simazine were less than their MCLs of 3.0 and 4.0 Îg/L, respectively (U.S. Environmental Protection Agency, 2000a); MCLs have not been established for the other detected pesticide compounds. However, nonenforceable lifetime Health Advisory Levels (HALs) have been established for metolachlor (100 Îg/L), prometon (100 Îg/L), and tebuthiuron (500 Îg/L) (U.S. Environmental Protection Agency, 2000a). All detected concentrations for these pesticide compounds were less than respective HALs. Although concentrations of detected pesticide compounds were small (relative to established MCLs and HALs), the synergistic effects on human health of these concentrations and long-term exposure to multiple pesticide compounds are unknown.

Volatile Organic Compounds

Water samples from 30 monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, were analyzed for 85 VOCs (table 13 at the back of this report). Of the 30 wells sampled, 13 samples (43 percent) had detectable concentrations of one or more VOCs (fig. 14, table 14 at the back of this report). A total of eight different VOCs were detected in water from these wells (table 10). Median concentrations for all eight VOCs were less than respective analytical method reporting limits. Chloroform, however, was the most frequently detected VOC (table 10). Chloroform was detected in water from seven wells (1, 5, 8, 11-14; 23 percent). Detected concentrations of chloroform ranged from an estimated 0.07 (wells 5 and 11) to 1.7 Îg/L (well 13) and were less than the 80-Îg/L MCL for drinking water. Other VOCs detected included 1,1,1-trichloroethane (in two samples), bromodichloromethane (in two samples), tetrachloroethylene (in four samples), toluene (in one sample), trichloroethylene (in two samples), cis-1,2-dichloroethylene (in two samples), and tert-Butyl methyl ether (also known as methyl tert-Butyl ether, MTBE) (in two samples).

With the exception of the 9.0-Îg/L concentration of tetrachloroethylene in water from well 10 (table 14), all concentrations of VOCs detected in the shallow ground-water samples collected during this study were less than established MCLs. The 9.0-Îg/L concentration of tetrachloroethylene was 80 percent larger than its 5.0-Îg/L MCL.

The VOCs detected in ground-water samples collected during this study have many and varied uses. Bromodi-chloromethane and chloroform are trihalomethane (THM) compounds that probably are formed when drinking-water sources are chlorinated during the disinfection process. The chlorination process halogenates naturally occurring organic substances (mainly fulvic and humic acids; Rook, 1977) in the source water to produce THMs. The formation of THMs is the result of the action of chlorine (OCl^-) on the fulvic and humic acids (Thurman, 1985). The bromide-substituted VOCs such as bromodichloromethane are thought to result from parallel bromination reactions initiated by the action of chlorine on background concentrations of bromide ions, which is present in most natural water (Boyce and Hornig, 1983). The occurrence of THMs in shallow ground water may be the result of using chlorinated drinking water as an irrigation source for lawns and gardens and (or) the reaction of residual chlorine in the irrigation water with naturally occurring soil fulvic and humic acids.

Tetrachloroethylene and trichloroethylene are used in dry-cleaning operations and as solvents for fats, greases, and waxes. The VOCs 1,1,1-trichloroethane and cis-1,2-dichloroethylene are used in the degreasing of metals and as constituents in solvents and glues and household products for stain removal. Toluene is a component of gasoline and a solvent for paints and coatings, lacquers, and adhesives (Verschueren, 1996). The occurrence of these compounds in shallow ground water may result from extensive use in an urban environment and methods of disposal or improper handling.

MTBE has been used as an octane enhancer in gasoline in the United States since the late 1970s. MTBE is the most widely used fuel oxygenate. In 1995, 8.0 billion kg (almost 9 million tons) were produced in the United States (American Chemical Society, 1996). The occurrence of MTBE in shallow ground water probably results from gasoline spills or leaking storage tanks.

RELATION TO LAND USE

Ground-water flow direction in the study area is from the northwest to the southeast (fig. 4) and, generally, is parallel to the Arkansas River. Flow velocity was estimated previously (table 3) at 2.5 ft/d in the alluvial and terrace deposits (Nal, fig. 3) along the Arkansas River and 1.2 ft/d in the terrace deposits (Nt, fig. 3) west of the Arkansas River alluvium. The age at which sampled water was recharged was calculated through analyses of CFC concentrations (table 4).

Potential recharge areas for the water sampled from 30 monitoring wells during this study were determined by first calculating the distance to those potential areas. These distances (in miles) (table 11) were calculated by multiplying the ground-water residence time (in years) by flow velocity (in feet per day; table 3) by 365.25 (days per year) by a unit conversion factor of 0.000189. Ground-water residence time was calculated by subtracting apparent recharge dates from 2000 (year of sampling). The average apparent recharge date (1986) of reported CFC-determined dates (table 4) was used as an estimated recharge date for water from monitoring wells with contaminated or otherwise unreliable CFC results. Distances to potential recharge areas for the 30 monitoring wells sampled in this study (table 11) ranged from 0.8 (wells 5 and 15) to 2.8 mi (well 19) upgradient of the well locations (fig. 4).

Land use in potential recharge areas was determined by locating areas upgradient (northwest) of the monitoring-well locations (fig. 4) using the distances indicated in table 11 and then determining the land use of these areas at the apparent or estimated time of recharge. The size of these potential recharge areas was arbitrarily selected to be an area with a radius of 1,000 ft around a point measured the calculated distances upgradient from monitoring-well locations. Dominant land use in these areas at the time of recharge (table 12) was determined from a combination of sources that included (1) USGS 7.5-minute topographic maps photographically updated in 1982, (2) digital land-use data (city of Wichita, Kansas, 2000), (3) dates of sewer-main installation (LaDonna Lawrenz and Kerry Gibson, city of Wichita Water and Sewer Department, oral commun., 2002), (4) casting dates on street curbs and side walks, (5) information from area residents, and (6) author knowledge and onsite inspection of the areas.

Agricultural land use was the dominant land use of potential recharge areas (table 12) for 16 (53 percent) of the 30 monitoring wells in areas of recent residential and commercial development (fig. 1). Residential and (or) commercial land use was the dominant land use of potential recharge areas for 10 (33 percent) of the wells, and potential recharge areas for three (10 percent) of the wells were believed to be in a transitional state of development from agricultural to residential and (or) commercial land use. The potential recharge area for one well (well 20) was believed to be in an area of extensive sand-dredging (sand-pit) operations. The results presented in table 12 provide possible explanations for previously discussed ground-water-quality characteristics, such as pesticides and VOC concentrations, for some of the 30 monitoring wells in areas of recent residential and commercial development.

One or more pesticide compounds were detected in water from 22 (73 percent) of the 30 monitoring wells in areas of recent residential and commercial development (fig. 12). Atrazine or deethyl-atrazine were detected in water from 21 (70 percent) of the 30 monitoring wells (fig. 13). These pesticide detection rates for shallow ground water in the Wichita area are substantially larger than the 49-percent detection rate of one or more compounds reported nationally in shallow ground water in urban areas (U.S. Geological Survey, 1999). Therefore, the possibility of the occurrence of some pesticide compounds, particularly atrazine and deethylatrazine, being an artifact of past agricultural use was investigated.

Atrazine has been the most extensively used herbicide in corn and grain-sorghum production in Kansas for the past 40 years. It is applied annually to about 79 percent of the corn and 82 percent of the grain sorghum (Devlin and others, 2000). In urban areas, atrazine has been used for nonselective weed control on industrial or noncropped land (U.S. Environmental Protection Agency, 1989) and for selective weed control on warm-season turfgrasses such as bermuda and zoysia (Fagerness and others, 2001). Historically, however, approximately 95 percent of the atrazine applied in the United States is used in corn and grain-sorghum production, predominately in the Mississippi River Basin where about 82 percent of the Nation's corn acreage is planted (CIBA-GEIGY Corporation, 1992).

Author discussions with professional lawn-care specialists in the Wichita area indicated that, although atrazine has been used historically in turfgrass application, it has not been used extensively or intensively applied in the Wichita area. The majority of atrazine use has been on bermuda and zoysia grass in the southern part of the study area (the area near monitoring wells 22-29; fig. 1). Furthermore, in recent years, cool-season grasses such as fescue and blue grass have become popular in the developing areas of west Wichita (the area near monitoring wells 1-16, 20, 21). These grasses are intolerant of atrazine. Additionally, Sedgwick County has not used atrazine in controlling weeds along roadways for about 20 years (Joseph O. Brunk, Sedgwick County Noxious Weed Department, oral commun., 2002). Historical use of atrazine by or in the city of Wichita was not available, but much of the atrazine used may have been associated with golf-course maintenance, all of which are downgradient of well locations.

Atrazine or deethylatrazine was detected mainly in shallow ground water recharged through agricultural areas or in areas of transition from agricultural to residential and (or) commercial land use (water from monitoring wells 1-11, 14-19, 21, 30; table 12). Water recharged through agricultural areas contained a detectable concentration of atrazine or deethylatrazine in 89 percent of the samples (fig. 13, table 12). For the purpose of this discussion, water recharged through areas in transition from an agricultural to a residential and (or) commercial land use (monitoring wells 9-11) was considered to be agriculturally affected.

In contrast to the 89-percent detection rate of atrazine or deethylatrazine in shallow ground water recharged through agricultural areas, these chemicals were detected in only 30 percent (fig. 15) of shallow ground water (water from monitoring wells 12, 13, 22-29) recharged through residential and (or) commercial areas (fig. 13, table 12) even though these wells had some of the shallowest depths to water in the study area (table 1). Because shallow ground water from 70 percent of the wells with water recharged through residential and (or) commercial areas was under anaerobic (reducing) conditions (dissolved-oxygen concentrations 0.5 mg/L or less) (fig. 5), the smaller detection rate of atrazine or deethylatrazine may be the result of these reducing conditions and not the result of differences in land use.

Literature describing anaerobic degradation of atrazine is not extensive; however, available literature (Gu and others, 1992; DeLaune and others, 1997; Papiernik and Spalding, 1997; Larsen and others, 2001) indicates that degradation of atrazine or deethylatrazine is no more frequent or rapid in soils or ground water under anaerobic conditions than under aerobic conditions. Therefore, the absence of atrazine or deethylatrazine in the majority of water from wells where the source of recharge was through residential and (or) commercial areas probably reflects limited use of atrazine in these areas. Furthermore, the major source of these chemicals in shallow ground water of the study area probably was the result of agricultural use. The possibility that the occurrence of agricultural herbicides in shallow ground water in urban areas may be an artifact of past or nearby agricultural use also was discussed by Barbash and others (1999, p. 23).

VOCs were detected in water recharged through both agricultural areas and residential and (or) commercial areas. At least one VOC was detected in 42 percent of water samples from monitoring wells recharged through agricultural or transitional areas (tables 12 and 14) with chloroform the most frequently detected (26 percent). Other VOCs detected in water recharged through agricultural areas included 1,1,1-trichloroethane, bromodichloromethane, tetrachloroethylene, toluene, and MTBE (table 14). Similarly, at least one VOC was detected in 50 percent of water samples recharged through residential and (or) commercial areas (table 12); however, chloroform was not the single-most frequently occurring VOC. Tetrachloroethylene was detected in 30 percent, and chloroform, trichloroethylene, and cis-1,2-dichloroethylene each were detected in 20 percent of the water samples recharged through residential and (or) commercial areas (tables 12 and 14). Other VOCs detected in these water samples included 1,1,1-trichloroethane, bromo-dichloromethane, and MTBE (table 14).

The frequency of occurrence of at least one VOC in water recharged through agricultural areas and residential and (or) commercial areas was similar, 42 and 50 percent, respectively. When detected, twice as many VOCs (on average) were detected in water recharged through residential and (or) commercial areas than in water recharged through agricultural areas.

The occurrence of chloroform in water from monitoring wells in residential and commercial areas where chlorinated drinking water was used to irrigate lawns was discussed in the section on "Volatile Organic Compounds". Chloroform could have been formed by the action of chlorine (OCl^-) on naturally occurring fulvic and humic acids in the soil. However, when chloroform occurrences in water from monitoring wells were evaluated relative to potential recharge areas (table 12) chloroform detections occurred mostly (71 percent) in water recharged through agricultural or transitional land uses (wells 1, 5, 8, 11, 14; fig. 1). Chlorinated municipal drinking-water use in agricultural areas for irrigation is unlikely.

The occurrence of chloroform in water recharged through agricultural areas may be the result of one or a combination of the following: (1) the atmospheric distribution of chloroform from nearby urban areas, (2) the use of some pesticides that contain chloroform as an artifact of their manufacturing process, or (3) chloroform may have entered the shallow ground-water system and mixed with recharge from agricultural areas as it flowed under areas of residential or commercial land use where chlorinated municipal water was used for lawn and garden irrigation. As of 1993, Wichita was one of the five major chloroform manufacturing centers in the United States with an estimated annual production capacity of 145 million lb (Agency for Toxic Substances and Disease Registry, 1997). This production capacity may increase atmospheric concentrations of chloroform relative to nonproduction areas. Chloroform is used in the production of some pesticides (Verschueren, 1996). The possibilities of mixing of water from different land-use recharge areas would complicate the evaluation of land-use effects on the quality of shallow ground water in areas of recent residential and commercial development.

SUMMARY AND CONCLUSIONS

REFERENCES CITED

SUPPLEMENTAL INFORMATION

• Table 13. Water-quality constituents analyzed in shallow ground-water samples from monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000

   

• Table 14. Results of physical and chemical analyses of shallow ground-water samples from monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000

   

• Table 15. Results of pH, organic-carbon, and particle-size analyses of sediment samples collected during installation of shallow ground-water monitoring wells in areas of recent residential and commercial development, Wichita, Kansas, 2000

   

For additional information contact:

Larry M. Pope
U.S. Geological Survey
4821 Quail Crest Place
Lawrence, KS 66049-3839
Telephone: (785) 832-3548
Fax: (785) 832-3500
Email: mpope@usgs.gov