U.S. Geological Survey
Open-File Report 97-235

Baseline Data-Collection and Quality-Control Protocols and Procedures for the Equus Beds Ground-Water Recharge Demonstration Project Near Wichita, Kansas, 1995-96

By A.C. Ziegler and L.J. Combs

Prepared in cooperation with the
CITY OF WICHITA

CONTENTS

Abstract
Introduction

Description of Equus Beds Ground-Water Recharge Demonstration Project
Description of Study Area
Purpose and Scope of This Report

Data-Collection Objectives and Intended Use of Baseline Data

Precipitation, Streamflow, and Water Levels
Water Quality
Aquifer Testing

Data-Collection and Quality-Control Protocols and Procedures

Precipitation, Streamflow, and Water Levels
Water Quality

Sample Containers, Preservation, and Holding Times
Sample Custody
Sample Packaging and Shipping to the National Water-Quality Laboratory

Aquifer Testing

Test Well
Discharge of Pumped Water
Monitoring Wells and Piezometers
Instrementation and Measurements
Record Keeping
Field Supervision
Water Quality Testing

Laboratory Procedures
Data Review, Validation, and Reporting
References Cited
Appendix A

FIGURES

• Figure 1. Map showing location of Equus Beds Ground-Water Recharge Demonstration Project near Wichita, Kansas, and areas where chloride concentrations in ground water exceed 250 milligrams per liter.
• Figure 2. Diagram showing Equus Beds Ground-Water Recharge Demonstration Project participants and responsibilities.
• Figure 3. Map showing land use and location of data-collection sites in study area.
• Figure 4. Schematic diagrams showing location of data-collection sites in detailed areas A-C as shown in figure 3.
• Example of surface-water field sheet
  1. Surface-water field sheet, page 1
  2. Surface-water field sheet, page 2
  3. Surface-water field sheet, page 3
  4. Surface-water field sheet, page 4
• Example of ground-water field sheet
  1. Ground-water field sheet, page 1
  2. Ground-water field sheet, page 2
  3. Ground-water field sheet, page 3
  4. Ground-water field sheet, page 4
• Example of U.S. Geological Survey National Water-Quality Laboratory analytical services request form
  1. Analytical services request form, page 1
  2. Analytical services request form, page 2

TABLES

1. Data-collection sites used during baseline data collection for the Equus Beds Ground-Water Recharge Demonstration Project, 1995-96
   1. Data-collection sites
   2. Data-collection sites--continued
2. Key water-quality constituents analyzed for all samples
3. Key water-quality constituents analyzed for comparison of total and dissolved concentrations
4. Key-plus water-quality constituent analysis for dissolved inorganic constituent concentrations and bacteria
   1. Dissolved inorganic constituent concentrations and bacteria
   2. Dissolved inorganic constituent concentrations and bacteria--continued
5. Key-plus water-quality constituent analysis for total inorganic constituent concentrations and bacteria analyzed for comparison of total and dissolved concentrations
   1. Total inorganic constituent concentrations and bacteria
   2. Total inorganic constituent concentrations and bacteria--continued
6. Key-plus water-quality constituents and limited U.S. Environmental Protection Agency Maximum Contaminant Level analysis for dissolved concentrations of selected pesticides
   1. Limited analysis for dissolved concentrations of selected pesticides
   2. Limited analysis for dissolved concentrations of selected pesticides--continued
7. Key-plus water-quality constituents and limited U.S. Environmental Protection Agency Maximum Contaminant Level analysis for total organonitrogen pesticides
8. Limited U.S. Environmental Protection Agency Maximum Contaminant Level analysis for dissolved concentrations of pesticides
   1. Limited analysis for dissolved concentrations of pesticides
   2. Limited analysis for dissolved concentrations of pesticides--continued
9. Limited U.S. Environmental Protection Agency Maximum Contaminant Level analysis for total recoverable concentrations of organochlorine and carbamate pesticides
10. Limited U.S. Environmental Protection Agency Maximum Contaminant Level analysis for total recoverable volatile organic compounds
    1. Limited analysis for total recoverable volatile organic compounds
    2. Limited analysis for total recoverable volatile organic compounds--continued
    3. Limited analysis for total recoverable volatile organic compounds--continued
11. Full U.S. Environmental Protection Agency Maximum Contaminant Level analysis for dissolved radionuclides
12. Full U.S. Environmental Protection Agency Maximum Contaminant Level analysis for total recoverable concentrations of organochlorine and organophosphate pesticides
13. Full U.S. Environmental Protection Agency Maximum Contaminant Level analysis for total recoverable concentrations of acid and base/neutral organic compounds
    1. Full analysis for total recoverable concentrations of acid and base/neutral organic compounds
    2. Full analysis for total recoverable concentrations of acid and base/neutral organic compounds--continued
14. Schedule for baseline and event sampling from February through September 1995
15. Schedule for baseline, event, and aquifer-test sampling from October 1995 through September 1996
    1. Schedule for baseline, event, and aquifer-test sampling from October 1995 through September 1996
    2. Schedule for baseline, event, and aquifer-test sampling from October 1995 through September 1996--continued
16. Sample bottles, treatment, and preservatives for key, key-plus, limited, and full U.S. Environmental Protection Agency analyses of water-quality constituents
    1. Sample bottles, treatment, and preservatives analyses of water-quality constituents
    2. Sample bottles, treatment, and preservatives analyses of water-quality constituents--continued
    3. Sample bottles, treatment, and preservatives analyses of water-quality constituents--continued
17. Water-quality sampling and analysis for 30-day aquifer test

Abstract

Introduction

The water supply for the city of Wichita in south-central Kansas currently (1997) comes from two primary sources--the Wichita well field completed in the Equus beds aquifer (a part of the regional High Plains aquifer system) and Cheney Reservoir (fig. 1). Because of the area's expected population growth and regionalization of the water system, the available water supply needs to be increased to meet future water demands. Recharge of the Equus beds aquifer is one alternative being considered to meet these increased demands.

The High Plains States Groundwater Recharge Demonstration Program is a cooperative effort between the U.S. Department of Interior, Bureau of Reclamation, U.S. Geological Survey, and U.S. Environmental Protection Agency. The purpose of the program is to study the potential for artificial ground-water recharge in 17 Western States and to demonstrate artificial-recharge technologies under a variety of hydrogeologic conditions. The results of demonstration projects will assist in determining the physical, chemical, and economic feasibility of artificial recharge.

Description of Equus Beds Ground-Water Recharge Demonstration Project

The primary purpose of the Equus Beds Ground-Water Recharge Demonstration Project is to evaluate two ground-water recharge and recovery techniques-surface-spreading basins and direct-injection recharge wells. Evaluation of these techniques will include description of recharge effects on ground-water quality, determination of operation and maintenance requirements for the two recharge methods, and identification of problems associated with long-term infiltration of recharge water. This project also will evaluate the potential of deterring the migration of saltwater (as identified by chloride concentrations exceeding 250 milligrams per liter) northwest of the existing Wichita well field (fig. 1) as identified by Burns and McDonnell (1995) and Myers and others (1996). The results of the demonstration project will be used to determine the feasibility and design and operating criteria for a full-scale recharge project. A detailed description of the Equus Beds Ground-Water Recharge Demonstration Project is provided by Burns and McDonnell (1995).

Project participants and responsibilities are outlined in (figure 2). The city of Wichita and Bureau of Reclamation cosponsor the overall project. Burns and McDonnell (consulting engineers, Kansas City, Missouri) manage the project for the city of Wichita and are responsible for the aquifer testing. The U.S. Geological Survey (USGS) is responsible for ensuring the collection and interpretation of water-quality and quantity data. Analysis of water-quality samples will be done by the city of Wichita laboratory, the USGS laboratory in Lawrence, Kansas, and the USGS National Water-Quality Laboratory in Arvada, Colorado.

Description of Study Area

The source of recharge water for the demonstration project is the Little Arkansas River between Halstead and Sedgwick, Kansas (fig. 1). Water will be recharged during the demonstration project whenever water levels in the Little Arkansas River are above base flow. Base flow has been determined to be when the discharge exceeds 42 cubic feet per second at USGS streamflow-gaging station 07143672 (fig. 3) near Halstead.

Purpose and Scope of This Report

Data-Collection Objectives and Intended Use of Baseline Data

Precipitation, Streamflow, and Water Levels

Precipitation and streamflow measurements and water-level data from both surface and ground water will be used to determine the vertical and horizontal movement and quantity of water in the river, at the planned recharge sites (near Wichita Well No. 4 and Wichita Well No. 36, (fig. 3), and in the aquifer. Precipitation streamflow were measured continuously from February 1995 through September 1996 at four USGS streamflow-gaging stations (07143665, 07143672, 07144100, and 07144200, (fig. 3 or fig. 4). Surface-water levels were measured continuously at sampling sites 07143680, 07143950, and 07144100 (fig. 4). Ground-water levels were measured continuously at three monitoring-well string sites located approximately perpendicular to the Little Arkansas River at detailed areas A, B, and C (fig. 4). There are at least five monitoring wells in each string site (fig. 4). Manual calibrations of automated and electronic data loggers were made on a monthly basis.

Water Quality

Aquifer Testing

The objective of aquifer testing was to determine the water quality of the induced surface water at the time pumping began at the aquifer-test well and throughout a 30-day and extended 75-day pumping period. Water-quality data were used to evaluate the effects of induced surface water in the ground-water system and to define the potential decrease or increase in constituent concentrations. Water-level data were used to determine the transmissivity, specific yield, hydraulic-conductivity, and riverbed-conductance values at detailed data-collection area A (TH-04-95) near Halstead, Kansas (fig. 4). A large-diameter (24-inch inside diameter) test well and 22 piezometers were installed to collect these data. These data will be used to design future recharge facilities, to determine the effect of the river on test-well water quality, and to evaluate the available quantity of induced surface water for recharge.

Data-Collection and Quality-Control and Procedures

Precipitation, Streamflow, and Water Levels

Precipitation gages were located at all of the streamflow-gaging stations in the study area except Little Arkansas River at SW 84th Street near Sedgwick, Kansas (station 07143950). Tipping-bucket rain gages were used to measure precipitation in 0.01-inch increments. Precipitation data were recorded every 15 minutes and totalled for each day. These data were transmitted by a data-collection platform (DCP) to a satellite and then to the USGS computer in Lawrence, Kansas. Precipitation data were verified against nearby National Weather Service precipitation gages located in Halstead and Wichita, Kansas (fig. 1).

Stream water-surface elevation (stage) was determined at six streamflow-gaging stations along the Little Arkansas River (Appendix A, table 1a and 1b, and figs. 3 and 4) with nonsubmersible, pressure transducers and was measured to the nearest 0.01 foot. The atage was recorded relative to an arbitrary datum, which has been referenced to the elevation of the gage datum (Appendix A, table 1a and 1b). Stage data were electronically recorded and transmitted by DCP. The data then were transmitted by satellite to a downlink site and then to the computer at the USGS office in Lawrence, Kansas. These data were recorded every 15 minutes and transmitted at least every 4 hours. The data obtained with the pressure-transducer equipment were verified at least monthly with physical measurements of water-surface elevations from an established reference mark on a bridge or by a permanently fixed wire-weight gage (Buchanan and Somers, 1968). Elevations were surveyed to the reference marks and gages and were verified annually. Methods used to determine streamflow are described in Carter and Davidian (1968) and Buchanan and Somers (1969).

Four of the six streamflow-gaging stations (07143665, 07143672, 07144100, and 07144200) were operated as continuous streamflow or discharge stations, and stage-discharge ratings were developed and maintained for these sites. The remaining two gaging stations (07143680 and 07143950) continuously recorded the water elevation or stage of the stream. Streamflow or discharge measurements were made at each station at least monthly by determining the cross-sectional area of the stream and measuring the flow velocity for at least 10 vertical transects in the cross section. Methods for discharge measurements are described by Buchanan and Somers (1969). A stage-discharge relation was developed on the basis of discharge measurements and the stage of the stream at the time of measurement. Methods used to develop the stage-discharge relations and to compute continuous discharge records for streamflow are described by Kennedy (1983, 1984).

Monitoring wells used in this study were constructed of polyvinyl chloride pipe. Well depth for most wells is shallow (about 50 feet below land surface), except for the aquifer-test well (depth 136.5 feet) and well TH-04-PD5 (depth 117 feet). The test well is screened from 71 to 132 feet below land surface; other monitoring wells typically are screened in the lowermost 10 feet of the casing.

Water levels in the monitoring wells were recorded to the nearest 0.01 foot at 15-minute intervals and transmitted from the same DCP as the colocated streamflow-gaging stations at all of the sites listed in Appendix A, table 1a and 1b, except sites TH-10-95, TH-02-95, TH-06-95, and TH-12-95, which were measured only at the time of sampling. Water-level sensing equipment consisted of submersible transducers that transmitted the water level to the DCP. Water levels were recorded and then transmitted every 4 hours to the USGS office in Lawrence, Kansas. The data from the transducers were verified at least monthly by manual measurements, and the computer data corrected on the basis of the results of the manual measurements. Water levels were measured manually at all sites by measurement to the water surface with either a steel or electric tape from a known reference-point elevation on the top of the well casing. Methods used for measurement of ground-water levels are described by Stallman (1971).

Water Quality

Water-quality data were collected during February 1995 through September 1996 to document a baseline for the quality of water in the Little Arkansas River and the quality of ground water in the adjacent Equus beds aquifer. Selected samples were analyzed for the constituents listed in Appendix A, tables 2 through 13. The sampling frequency and constituents for analysis are presented in Appendix A, tables 14 and 15. At least one sample per month was collected from the two surface-water sites (07143672 and 07144100), and quarterly samples were collected from six wells and analyzed for a subset of the EPA MCL list (Appendix A, tables 2 and 4) and other indicator constituents. Water-quality samples were collected once each quarter during the first three quarters of the project and analyzed for all of the constituents listed in Appendix A, tables 2-13. Chloride and triazine herbicides were expected to be the primary constituents of concern.

Daily water samples from the Little Arkansas River and test well were collected and analyzed for triazine herbicide concentrations during the aquifer test at site 07143672. On the basis of the quality-assurance/quality-control (QA/QC) verified results from the aquifer test and from analytical results from the baseline sampling, all inorganic and organic constituents with values greater than the MCL in these surface- water samples will be monitored for the remainder of the demonstration project. Subsequent data-collection activities will determine the possible effects of the withdrawal of test-well water and water from the river and of recharging the water 2 to 3 miles away in a different part of the aquifer.

Baseline data collected were of a definable and documented quality with respect to the precision and accuracy of the measurement (Appendix A, tables 1-12), representativeness of the sampled media, comparability between sites and sampling periods, and completeness of data (Appendix A, tables 2-13). Use of approved EPA or USGS onsite and laboratory methods helped ensure the precision and accuracy of the data. Representativeness of the samples was defined by the number of samples collected at the surface- and ground-water locations. The representativeness of each sample collected was described by the methods used for collection of samples. Detection limits, method detection limits, or reporting limits were used to quantify concentrations in water samples at the levels of at least 20 percent of the currently published MCL for each constituent analyzed except for antimony, beryllium, thallium, 1,2 dibromo-3-chloropropane, heptachlor, heptachlor epoxide, lindane, toxaphene, benzo(a-k)anthenes, bis(2-ethlyhexyl) phthalate, and chrysene (Appendix A, tables 2-13). Antimony and beryllium had reporting levels sufficient to quantify at the level of the MCL. The reporting limits for the organic constituents listed above were greater than the MCL; however, these constituents were not expected to be detected. Reporting limits for constituents without MCLs are listed in (Appendix A, tables 2-13).

Onsite and laboratory quality-control samples included equipment and trip blanks, field spikes, reference water samples, replicate samples, and concurrent samples. Equipment blanks are collected by passing certified inorganic and organic blank water through the sample-collection and processing equipment. About 5 percent of samples collected were equipment blanks. Trip blanks are collected by filling the sample bottles with certified blank water and are analyzed at the end of the trip. About 1 percent of the samples collected were trip blanks. Field spikes were collected by adding a known quantity of analyte to the sample. Field spikes were about 1 percent of the sample total. Reference water samples are samples with known quantities of analytes in the sample. About 1 percent of the samples analyzed were reference water samples. Data from equipment and trip blanks, field spikes, and reference water samples were used to define the bias from the introduction of contamination at any stage in the sample-collection and analysis process.

Replicate samples are samples collected in an essentially identical manner as the environmental samples so that the replicate sample should have concentrations equivalent to the environmental sample. Replicate split samples are samples collected from the same compositing container as the environmental sample. Concurrent samples are a type of replicate sample where samples are collected at the same time and placed in separate compositing containers. Replicate and concurrent samples were used to estimate the variability of the sample-collection and analysis process. Replicate and concurrent samples were about 2 percent of the sample total.

Each laboratory analyzing the samples had their own blanking and spiking protocols, usually 10 percent of the sample total. Consistent onsite and laboratory methods ensured comparability for all samples unless the data were not within acceptability limits as defined in Appendix A, tables 2-13. The goal was to obtain at least 90-percent completeness of all data collected onsite and all data resulting from analysis of the samples.

Before environmental sample collection, sampling equipment for both surface and ground water were cleaned with a nonphosphate-containing detergent, rinsed with deionized water, acid-rinsed with dilute hydrochloric acid (if equipment contained no metal parts), and rinsed again with deionized water. Equipment blanks were collected before each environmental sampling to document the cleanliness of the equipment. At the beginning of environmental sample collection, instruments for the measurement of specific conductance, pH, water temperature, and dissolved oxygen were calibrated, and the calibration was checked after sampling at each site. A more detailed discussion of the specific procedures for equipment cleaning, collection of environmental samples, collection of blanks, and equipment calibration is provided by Guy and Norman (1970), Edwards and Glysson (1988), Ward and Harr (1990), Wells and others (1990), Horowitz and others (1994), Koterba and others (1995), and Puls and Barcelona (1996). Onsite notes were recorded on surface-water field sheets (fig. 5a, 5b, 5c, and 5d), during sample collection at each site and included the site location, date and time of collection, collectors' names, calibration information for the onsite instrumentation, methods and equipment used for collection, stream and weather conditions at the time of sampling, bottles collected, preservatives used, volumes purged from wells, and onsite measurements.

During the anticipated "spring flush" of herbicides from cropland in the study area, automated samplers collected two samples per day at surface-water sites (07143672 and 07144100) and were analyzed for triazine herbicides by enzyme-linked immunosorbent assay (ELISA) (Thurman and others, 1990). Selected samples collected with autosamplers were verified by gas chromatography/mass spectrometry (GC/MS) analysis for triazine herbicide concentrations greater than 3.0 µg/L (micrograms per liter). When the river stage became high enough to inundate the autosamplers, the samplers were removed, and daily grab samples were collected. Equipment blanks and trip blanks were 6\x11percent of the automated sample total. Replicate and concurrent samples were 2 percent of the automated sample total. About 2 percent of the automated sample total were replicates split from the sample compositing container. These samples were submitted as blind to the laboratory. Acceptance criteria are similar to those described in the section "Data Review, Validation, and Reporting."

Autosample bottles were collected from the sampler, labeled, chilled to 4 oC, and transported to the USGS office in Wichita, Kansas, for processing. Samples were filtered through 0.45-mm (micrometer) glass-fiber filters that had been baked at 450 °C to eliminate organic contamination. Filters were rinsed three times with 100 mL (milliliters) of deionized water, and then 50 mL of sample water were passed through the filters and discarded. Two 125-mL bottles were filled for analysis, depending on the availability of water from the autosampler. These samples then were chilled to 4 °C and shipped to the USGS laboratory in Lawrence, Kansas, for analysis. Results of samples collected by automated samplers were compared to results from samples collected using equal-width sample-collection techniques.

Surface-water samples were collected by two USGS personnel using depth- and width-integrating techniques (Ward and Harr, 1990; Wells and others, 1990) with samplers constructed of polytetrafluoroethylene (PTFE). Samples for analysis of inorganic and radiochemical constituents were composited in a polyethylene churn splitter, and aliquots were withdrawn and preserved for the analysis of total recoverable concentrations according to requirements for the analytical methods (Appendix A, table 16). Samples for analysis of filtered (dissolved) concentrations of inorganic and radiochemical constituents were filtered from the churn splitter through a 0.45-mm pore-size cellulose acetate filter and preserved according to the requirements for the analytical methods (Appendix A, table 16). Filters and equipment were rinsed with 1 liter (about 1 quart) of deionized water before filtering the water sample. Bacteriological samples were collected from the midpoint of the stream in a sterile container. Samples for analysis of volatile organic compounds (Appendix A, table 10) were collected from the midpoint of the stream and sealed under water. Samples for analysis of the remaining organic constituents were composited in a stainless-steel container, and aliquots were collected for analysis of total recoverable or filtered concentrations. Samples for analysis of filtered concentrations of organic constituents were filtered through a glass-fiber filter previously baked at 450 °C. All organic sample aliquots were collected in 1-liter glass amber bottles previously baked at 450 °C. Caps for organic sample bottles were lined with PTFE. Caps for inorganic samples were lined with noncontaminating plastic. All equipment used in the filtering process was constructed of PTFE. Samples were processed and preserved inside isolation chambers to prevent ambient air contamination.

Ground-water samples were collected by three USGS personnel using methods described by Wood (1976), Koterba and others (1995), and Puls and Barcelona (1996). Samples were collected by using a noncontaminating submersible pump constructed of PTFE and stainless steel. At least five well volumes were purged before samples were collected. During purging, specific conductance, pH, water temperature, turbidity, and dissolved oxygen were measured and recorded on ground-water field sheets (fig. 6a, 6b, 6c, and 6d). Samples were collected only if the values of these properties and constituents were within 10 percent for three consecutive readings made at least 5 minutes apart and if the turbidity was less than 10 NTU (nephlometric turbidity units). The flow rate during sampling was 0.5 to 1.0 liter per minute (0.13 to 0.26 gallon per minute), similar to the method described by Puls and Barcelona (1996). Purged water was conveyed far enough away from the wells so as not to affect the recharge area. Ground-water sampling generally occurred from upgradient to downgradient sites or from sites with the smallest concentrations of water-quality constituents to those with the largest concentration. Between sites, the sampling equipment was cleaned as described previously. Samples were collected in-line and processed and preserved inside isolation chambers to prevent ambient air contamination. The general order for filling of sample bottles was organic, inorganic, radiochemical, and bacteriological.

Sample Containers, Preservation, and Holding Times

All sample containers and preservatives were obtained from the USGS Water-Quality Supply Unit in Ocala, Florida. All containers and preservatives were quality assured by processing blank water for each bottle type and analyzing for possible contamination. All bottles and caps used for the collection of metal and radiological concentrations were acid rinsed to prevent contamination. Bottles for nutrients were made of nonlight-penetrating material to prevent sample degradation by ultraviolet (UV) light. Bottles for organic analyses were made of amber glass to prevent UV light degradation and baked at 450 °C to eliminate organic contamination. Bacteriological sample bottles were sterilized using an autoclave to prevent contamination. Sample bottles, holding times, preservatives for key, key-plus, limited, and full EPA water-quality constituent analyses are listed in Appendix A, table 16.

Preservation was performed, and holding times of samples were assessed according to the method requirements of the city of Wichita laboratory (city of Wichita, written commun., 1994, revised 1996), U.S. Environmental Protection Agency (1995, p. 643-645), the USGS laboratory in Lawrence, Kansas (Thurman and others, 1990), and the USGS National Water-Quality Laboratory in Arvada, Colorado. Methods used for analyses were those required by EPA or USGS as noted in Appendix A, tables 2 through 13. Metals and radiochemical samples were preserved with concentrated nitric acid. Samples for anions and other inorganic constituents were not preserved. Samples for nutrients were either chilled or preserved with concentrated sulfuric acid. Organic samples were chilled. If holding times were exceeded, the analyses were used only in a semiquantitative manner and were reported as having exceeded the holding time.

Sample Custody

Samples analyzed by the city of Wichita laboratory and the USGS laboratory in Lawrence, Kansas, were transported or shipped directly to the laboratory and logged in. A copy of the field sheet (fig. 5a, 5b, 5c, and 5d or fig. 6a, 6b, 6c, and 6d) accompanied the samples, and the receipt of the samples and date and time of receipt were acknowledged on the form. Samples analyzed at the USGS National Water-Quality Laboratory (NWQL) in Arvada, Colorado, were shipped in a sealed cooler by an overnight carrier. Inside the cooler, a form outlining the request for analytical services and requestor for the services was sealed in a plastic bag. After receipt at all of the laboratories, the internal laboratory information management system tracked the custody of the samples.

An example of the NWQL analytical services request form is shown in figure 7a and 7b. This form uniquely identifies the sample, the office requesting the analysis, the determinations requested, onsite measurements, and any special sample information. Each sample and its request form were assigned a unique NWQL identification number that encodes the date of receipt, the sample type, and a serial number. The request form was routed to the data-processing section at NWQL, which entered all sample information, analytical requests, and analytical results into a data base. Analytical results were sent to the submitting office after all analytical work was completed.

Sample Packaging and Shipping to the National Water-Quality Laboratory

Aquifer Testing

A 30-day aquifer test began on April 2 and concluded on May 2, 1996, at the test well at site TH-04-95 near Halstead, Kansas (fig. 4). A subsequent 75-day aquifer test began on May 13 and ran through July 24, 1996. These tests will be used to help determine aquifer and riverbed hydraulic characteristics and will be used to evaluate aquifer-river interaction and to investigate ground-water-quality changes resulting from an extended period of induced infiltration during which river water is moving into the aquifer.

The aquifer test involved construction of a test well and installation of 22 piezometers surrounding the test well to measure changes in ground-water levels. The test well was constructed of stainless-steel screen and gravel pack with 24-inch diameter screen and casing. Screen length, slot size, and setting depth were based on analysis of site-specific drill-test data. The well was designed for a 900- to 1,000-gallons-per-minute capacity, depending on site geology. The pump was an electric-driven, vertical turbine type. Pumped water was discharged to the river after measurement of flow.

The piezometers were installed surrounding the test well to measure the response of the water-table surface during pumping. Strings of piezometers were constructed parallel and perpendicular to the river.

During the aquifer test, water-level measurements, pump discharge, river stage, specific conductance, pH, water temperature, and dissolved oxygen were recorded at least hourly. Ground-water-level measurements were recorded using a combination of electronic data-logging units and manual methods. The hydrogeologic characteristics evaluated with this test were transmissivity, specific yield, hydraulic conductivity, and riverbed conductance.

Water-level measurements were made using the methods described by Stallman (1971). Water-level measurements were verified with an electric or steel tape that had been previously verified to be accurate to the nearest 0.01 foot. Pump discharge was measured by a calibrated, totalizing water meter with a calibrated orifice weir, and an automated data-logging unit was used to record the discharge at 15-minute intervals.

Aquifer and water-quality testing procedures for the 30-day test of the deep part of the Equus beds aquifer are described in the following paragraph.

The test well was located approximately 200 feet north-northwest from the original boring at well EB-145-A1 (fig. 4). The geological conditions at the selected test-well site vary and were used to design the screened intervals for the test well. The boring at well EB-145-A1 encountered a shallow sand zone, a 15-foot thick silty-clay layer, and a deeper sand and gravel aquifer zone. Testing included a 30-day pumping test of the deep aquifer zone and use of clustered shallow and deep piezometers to measure aquifer response in the shallow and deep zones. Aquifer response in both the shallow and deep aquifer zones was recorded using automated data loggers. The general sequence of activities was:

1. Test hole drilling at well location.
2. Develop final well design and piezometer spacing.
3. Construct well and piezometers.
4. Develop well by surging and pumping.
5. Perform 24-hour acceptance test to verify well development and proper construction.
6. Set test pump and instrumentation.
7. Conduct 30-day aquifer test.
8. Evaluate water-level data and available water-quality data to determine if the deep aquifer zone is responding as a confined system. Review information with Groundwater Management District No. 2.
9. Conduct a 75-day pumping test during spring and summer to evaluate atrazine fate through induced infiltration and to confirm the river/aquifer interconnection.
10. Secure test-well site and equipment until construction of other demonstration recharge facilities near Wichita Well No. 4 and No. 36 (fig. 3) is complete.

Test Well

The test well was constructed as a full-scale production well. The conceptual well design was determined from the boring logs from well EB-145-A1 (fig. 3). Final design was determined from a pilot hole drilled as part of a well-construction contract. Final construction was with a 24-inch casing and 60 feet of 60-slot, stainless-steel screen from 71 to 131.5 feet below land surface, with a zone from 110 to 114 feet blanked out because of fine-grained sand. Screen setting and slot size were determined from grain-size determinations from split-spoon samples and boring logs from the pilot hole.

Discharge of Pumped Water

Monitoring Wells and Piezometers

Six monitoring wells were installed as part of a line of monitoring wells perpendicular to the river and were instrumented by the USGS at site TH-04-95 shown in detailed data-collection area A (fig. 4). Water-level data from these monitoring wells were collected at 15-minute intervals with data loggers. The monitoring wells and piezometers were completed with 10 feet of 2-inch diameter PVC screen. Most monitoring wells and piezometers were constructed in deep-shallow clusters except for the six shallow monitoring wells located near the test well. The boring for the deep monitoring well (PD5) was completed first and electric logged. Screen settings for both the deep and shallow monitoring wells and piezometers were determined on the basis of geologic and electric logs from the deep monitoring well (PD5) and the test well.

Monitoring-well and piezometer distances originally were established as multiples of the aquifer saturated thickness. Some distance adjustments were required after the pilot hole was drilled and actual well design was completed.

Instrumentation and Measurements

Selected piezometers were instrumented with transducers connected to a data-logging unit to record water-level measurements at 15-minute intervals. The water level in the pumping test well was measured manually using an electric sounding tape. Water levels in the instrumented piezometers were checked weekly with the electric sounding tape to verify automated readings. Automated water-level measurements were collected frequently at the beginning of the aquifer test (1 minute or less). After approximately 4 to 6 hours, the automated measurements were collected at 1-hour intervals for the remainder of the test. Six to eight hours of manual water-level measurements were made at selected monitoring wells and piezometers during the beginning of the aquifer test and at initial recovery after the aquifer test to provide backup for the automated system. Water levels for the monitoring wells at detailed data-collection area A (TH-04-95, fig. 4) instrumented by USGS were retrieved every week.

The drilling contractor installed and maintained an orifice weir for measuring flow from the test well. A transducer was connected to the electronic data logger to measure the water level behind the orifice in addition to a manual manometer.

Record Keeping

All manual depth-to-water measurements were maintained on ground-water field sheets located at a gage house at area A (fig. 4). An example field sheet is shown in fig. 6a, 6b, 6c, and 6d. Other information was maintained in a test diary at the site and was available for all authorized personnel to record site visits and findings. A master record was maintained at the Groundwater Management District No. 2 office in Halstead, Kansas, and updated daily. The master record contained photocopies of the field sheets, originals of the field diary, and electronic files of all retrievals from electronic data loggers.

Field Supervision

Water-Quality Testing

The schedule for water-quality sampling and analysis for the 30-day aquifer test is shown in Appendix A, table 17. Daily 2-liter grab samples of river water and test-well water were collected with automated samplers. Specific conductance, pH, water temperature, and dissolved oxygen were measured manually twice per week when samples were manually collected. These constituents were recorded hourly from the test well with the multiprobe data-logging unit.

Laboratory Procedures

Organic and radiochemical analyses were performed by the USGS National Water-Quality Laboratory (NWQL) in Arvada, Colorado. Methods for analysis are described in Fishman and others (1994). Specific analytes and methods used are referenced in Appendix A, tables 2-13. Organic constituents include volatile organic compounds, pesticides, and acid and base/neutral organic compounds. The NWQL participates in many quality-assurance evaluations, including EPA drinking-water certification testing. The general quality-assurance plan for the laboratory is described by Pritt and Raese (1995). Friedman and Erdmann (1982) describe the internal quality assurance and quality control used at the NWQL.

All laboratory methods included the use of internal blind duplicates, spikes, and standard reference samples as part of each laboratory's quality-assurance plan and standard operating procedures. The method detection limits and minimum reporting limits of all of the laboratories were sufficient to quantify concentrations of constituents for MCLs or HALs and are listed in Appendix A, tables 2-13. Acceptance criteria and percentage completeness goals also are listed in these tables.

Data Review, Validation, and Reporting

Surface- and ground-water-quality data were validated by reviewing the data-collection operations and field notes and by determining that they were appropriate for the hydrologic conditions. Validation included the review of onsite and laboratory quality-control sample results from equipment blanks, trip blanks, standard reference water samples, and field spikes to isolate any contaminants that may have been introduced onsite or in laboratory methods. Results of internal spike recoveries were within the guidelines for the methods, usually 80- to 120-percent recovery (Appendix A, tables 2-13). Validation was determined by review of holding times, comparing the quality-control sample results to the criteria for precision and accuracy for onsite and laboratory results (Appendix A, tables 2-13). For most constituents, an onsite or laboratory replicate analytical result within 10 percent was acceptable. Precision was defined from replicate measurements by the following equation:

Acuracy of onsite and laboratory methods were defined by results of analyses of field and laboratory spikes and results of standard reference water samples. Accuracy for spikes was defined by the following equation:

Accuracy for reference water samples was defined by the following equation:

Acceptable precision and accuracy for each analyte are presented in Appendix A, tables 2-13. In tables 2-13, the relative standard deviation is given for selected constituents on the basis of the laboratory analytical technique.

Relative standard deviation, in percent, was used in tables 2-13 because these values were based on results from three or more replicate results. The relative standard\x11deviation (RSD) was defined by the following equation:

The standard deviation (S) is defined by the following equation:

Representativeness of samples to meet the objective of determining the baseline water quality was accomplished through the selection of sampling sites, sampling methods, timing of sampling, and validation and completeness of the data. Comparability of the data was evaluated on the basis of the representativeness of the samples and comparison of individual sampling results to the results from all samples collected.

Aquifer-test data were reviewed, verified, and analyzed by a Burns and McDonnell hydrogeologist. Analysis of the data was made considering any fluctuation of river stage, pumping rates, and weather conditions. Narrative descriptions of the aquifer test, data collection, data analysis, and results were contained in a summary memorandum. The information also was included in the first annual report (Burns and McDonnell, written commun., 1996) following completion of the analysis.

Water-quality data were reviewed by USGS upon receipt from the laboratories. If any constituent concentrations in the recharge water exceeded EPA MCLs, the project sponsor was notified immediately. If concentrations exceed the project "action levels," the city of Wichita may decide to execute a plan for mitigation.

The city of Wichita, in consultation with the Kansas Department of Health and Environment, EPA, and Bureau of Reclamation, developed a mitigation plan (Burns and McDonnell, written commun., 1996) on the basis of evaluation of the baseline data. Both planned recharge sites include a large-capacity city of Wichita well capable of pumping recharged water from the well field to the Wichita water-treatment plant if contamination were to occur. Additionally, the minor quantities of water and concentrations of constituents in the water that will be recharged in this demonstration project are expected to be diluted to acceptable levels by the water present in the aquifer.

References Cited

Appendix A

• Table 1a. Data-collection sites used during baseline data collection for the Equus Beds Ground-Water Recharge Demonstration Project, 1995-96

• Table 1b. Data-collection sites used during baseline data collection for the Equus Beds Ground-Water Recharge Demonstration Project, 1995-96--Continued

• Table 2. Key water-quality constituents analyzed for all samples

• Table 3. Key water-quality constituents analyzed for comparison of total and dissolved concentrations

• Table 4a. Key-plus water-quality constituent analysis for dissolved inorganic constituent concentrations and bacteria

• Table 4b. Key-plus water-quality constituent analysis for dissolved inorganic constituent concentrations and bacteria--Continued

• Table 5a. Key-plus water-quality constituent analysis for total inorganic constituent concentrations and bacteria analyzed for comparison of total and dissolved concentrations

• Table 5b. Key-plus water-quality constituent analysis for total inorganic constituent concentrations and bacteria analyzed for comparison of total and dissolved concentrations--Continued

• Table 6a. Key-plus water-quality constituents and limited EPA MCL analysis for dissolved concentrations of selected pesticides

• Table 6b. Key-plus water-quality constituents and limited EPA MCL analysis for dissolved concentrations of selected pesticides--Continued

• Table 7. Key-plus water-quality constituents and limited EPA MCL analysis for total organonitrogen pesticides

• Table 8a. Limited EPA MCL analysis for dissolved concentrations of pesticides

• Table 8b. Limited EPA MCL analysis for dissolved concentrations of pesticides--Continued

• Table 9. Limited EPA MCL analysis for total recoverable concentrations of organochlorine and carbamate pesticides

• Table 10a. Limited EPA MCL analysis for total recoverable volatile organic compounds

• Table 10b. Limited EPA MCL analysis for total recoverable volatile organic compounds--Continued

• Table 10c. Limited EPA MCL analysis for total recoverable volatile organic compounds--Continued

• Table 11. Full EPA MCL analysis for dissolved radionuclides

• Table 12. Full EPA MCL analysis for total recoverable concentrations of organochlorine and organophosphate pesticides

• Table 13a. Full EPA MCL analysis for total recoverable concentrations of acid and base/neutral organic compounds

• Table 13b. Full EPA MCL analysis for total recoverable concentrations of acid and base/neutral organic compounds--Continued

• Table 14. Schedule for baseline and event sampling from February through September 1995

• Table 15a. Schedule for baseline, event, and aquifer-test sampling from October 1995 through September 1996

• Table 15b. Schedule for baseline, event, and aquifer-test sampling from October 1995 through September 1996--Continued

• Table 16a. Sample bottles, treatment, and preservatives for key, key-plus, limited, and full U.S. EPA analyses of water-quality constituents

• Table 16b. Sample bottles, treatment, and preservatives for key, key-plus, limited, and full U.S. EPA analyses of water-quality constituents--Continued

• Table 16c. Sample bottles, treatment, and preservatives for key, key-plus, limited, and full EPA analyses of water-quality constituents--Continued

• Table 17. Water-quality sampling and analysis for 30-day aquifer test

For additional information write to:       Copies of this report can be purchased from:

District Chief                             U.S. Geological Survey
U.S. Geological Survey                     Information Services
4821 Quail Crest Place                     Box 25286
Lawrence, Kansas  66049-3839               Federal Center