EPA-600/R-93/107
April 1993
CASE STUDIES IN WELLHEAD PROTECTION AREA
DELINEATION AND MONITORING
by
Beth A. Moore
(Co-Author and Editor)
Lockheed Environmental Systems & Technologies Company
Las Vegas, Nevada 89119
City of Stevens Point Water Department, Wisconsin
Littleton Water Department, Massachusetts
South Dakota Geologic Survey, Vermillion, South Dakota
BCI Geonetics, Inc., Laconia, New Hampshire
City of Springfield Utilities, Missouri
April 1993
Contract Number 68-CO-0049
Project Officer
Steven P. Gardner
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas, Nevada 89193-3478
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY .
LAS VEGAS, NEVADA 89193-3478
Printed on Recycled Paper
-------
NOTICE
This report is the result of research supported by the U.S. Environmental Protection Agency's
Environmental Monitoring Systems Laboratory in Las Vegas, Nevada, as part of its efforts to provide
technical assistance to state, tribal, and local governments on the implementation of the Wellhead
Protection Program. The specific methods and approaches contained in this document have
undergone peer review but do not constitute official Agency endorsement or policy recommendations.
The Office of Research and Development provides this information to help solve complex technical
problems related to refined delineation and ground-water monitoring of wellhead protection areas in
various hydrogeologic settings. Further assistance is available from the Environmental Monitoring
Systems Laboratory in Las Vegas, the Office of Ground Water and Drinking Water in Washington,
D.C, and the ground-water offices in the ten U.S. Environmental Protection Agency regions.
n
-------
TABLE OF CONTENTS
Executive Summary
Chapter 1
Monitoring Methodology for Wellhead Protection
Chapter 2
Wellhead Protection Program and Monitoring System Design,
Stevens Point, Wisconsin
Chapter 3
Monitoring System Design for Wellhead Protection, Littleton, Massachusetts
Chapter 4
Wellhead Protection and Monitoring Options for Sioux Falls Airport
Wellfield, South Dakota
Chapter 5
Wellhead Protection Area Delineation and Monitoring Strategies for a Fractured Bedrock
Aquifer, Dover, New Hampshire
111
-------
-------
EXECUTIVE SUMMARY
BACKGROUND
The increasing contaminant threat to public water supply wells has created a new political and
technical awareness of ground-water protection programs. Recognizing the need for conjunctive
management of contaminant sources and public water supplies to prevent, or minimize, ground-water
quality degradation, Congress amended the Safe Drinking Water Act in 1986 to include Section 1428.
This section mandated the development of the Wellhead Protection Program (WHPP), which
established a legal framework to protect public water supply wells, wellfields, and springs from
contamination.
An important technical element of WHPP implementation is wellhead protection area
delineation. A wellhead protection area (WHPA) is defined as the surface and subsurface area
surrounding a well, wellfield, or spring, through which contaminants may pass and reach the ground
water contributing to the supply source. Criteria and methods for WHPA delineation are given in
several U.S. Environmental Protection Agency (EPA) guidance documents. Ground-water
monitoring is one of many management tools available for protecting public drinking water supplies.
Ground-water parameters are monitored (1) to assess source-control measures, (2) to monitor
compliance with drinking water standards at sites other than the wellhead, and (3) to provide advance
warning of contaminants in ground water. Ground-water monitoring may enhance source
characterization, WHPA delineation, and new water supply evaluation.
This technical assistance document, "Case Studies in Wellhead Protection Area Delineation
and Monitoring," was prepared by Lockheed Environmental Systems & Technologies Company under
contract to the U.S. Environmental Protection Agency (EPA), Environmental Monitoring Systems
Laboratory at Las Vegas (EMSL-LV). The document provides technical information to
municipalities, states, and the EPA Regions in their implementation of WHPPs. The primary goals
of this document are to present a monitoring methodology for WHPAs and to exemplify this
methodology in five unique case study settings.
MONITORING METHODOLOGY WELLHEAD PROTECTION
In 1989, EMSL-LV engaged in cooperative research with five carefully selected municipalities
to develop proposed, long-term monitoring programs for their existing WHPAs. The product of the
cooperative research contains two types of information:
(1) A recommended methodology for planning and implementing a wellhead protection
monitoring program which emphasizes source assessment, correct WHPA delineation,
and hydrogeologic characterization (Chapter 1)
-------
(2) Five case study narratives (Chapter 1, Appendix A-l, and Chapters 2 through 5) used
to exemplify the monitoring methodology for different hydrogeologic and contaminant
source settings (Table 1)
TABLE 1. CHARACTERISTICS OF THE CASE STUDY RESEARCH SITES
Municipality
Hydrogeologic Setting
Characterization Tasks .
Stevens Point,
Wisconsin
Littleton,
Massachusetts
Sioux Falls,
South Dakota
Dover,
New Hampshire
Springfield,
Missouri
Unconfined aquifer
Unconfined aquifer, recharge
from Spectacle Pond
Unconfined aquifer; recharge
from the Big Sioux River
Fractured-bedrock aquifer,
discrete flow system
Mature karst (porous
limestone) aquifer, conduit
flow system
Flow system modeled with FLOWPATH
(two-dimensional, steady state); Point and
nonpoint sources assessed
Flow system modeled with FLOWPATH
(two-dimensional, steady state);
MODFLOW (three-dimensional, steady
state); and FLOWCAD (two-dimensional,
transient); Industrial and commercial point
sources assessed
Flow system modeled with FLOWPATH
(two-dimensional, steady state); Big Sioux
River assessed as a line source; Point and
nonpoint sources present
Flow system characterized with lineament
analysis, structural-control mapping,
aquifer testing, dye tracing, and borehole
geophysics; A few commercial, point
sources, and natural sources
Flow system determined by watershed
boundaries, dye tracing, and flow analysis;
Point and nonpoint sources assessed
Chapter 1 contains a methodology which is intended to serve as a guide for WHPP
implementors in establishing technically-defensible, reliable, and effective ground-water monitoring
programs for wellhead protection. This methodology emphasizes saturated zone monitoring. The
first four case study narratives are presented in the document (Chapters 2 through 5) in order of
increasing hydrogeologic complexity (aquifer heterogeneity). The exception to this organization is the
Springfield, Missouri, case study, which is presented in abbreviated form in Appendix 1-A of
Chapter 1.
Basic hydrogeology concepts and equations are reviewed in Chapter 1, including: ground-
water systems and flow, conceptual hydrogeologic models and flow nets, and accurate delineation and
monitoring in different hydrogeologic settings. The spectrum of Unconfined to confined aquifer
conditions is discussed in relation to porous, granular aquifers; fractured-bedrock aquifers; and karst
aquifers.
VI
-------
Physical and chemical parameter monitoring apply to wellhead protection. Three types of
ground-water monitoring are useful in managing WHPAs: ambient trend, source assessment, and
early-warning detection monitoring. Ambient trend monitoring detects the temporal and spatial
trends in physical and chemical quality of the ground-water system. Source assessment monitoring
evaluates the existing or potential impacts on the physical or chemical ground-water system from a
proposed, active, or abandoned contaminant source. Early-warning detection monitoring is conducted
upgradient from the wellhead, based on known travel times, to trigger a contingency response to
prevent public exposure to contaminants. These types of monitoring are incorporated to measure or
detect contaminants in aquifers, and should not be mistaken as preventative or remedial measures.
Source assessment is a critical first step in designing an effective monitoring program. Target
monitoring parameters for early-warning detection and source assessment are selected from a
comprehensive list of known and suspected contaminants associated with land-use activities and
practices. Optimal monitoring sites may be determined, reflecting prioritization of sources. An
inventory of common sources of contamination within and in proximity to WHPAs is included.
The monitoring methodology is divided into three phases:
• Phase I: WHPP Elements and Scoping Tasks
• Phase II: Research Monitoring Program
• Phase III: Wellhead Protection Monitoring Program
Phase I WHPP elements and scoping tasks include: designating roles and a management framework,
preliminary WHPA delineation, and source assessment. To support the research monitoring task, an
initial information base of ancillary and monitoring data should be compiled and reviewed to
determine data limitations and gaps. The strategy is to maximize information content; to define
monitoring objectives; and to conduct field studies with the least, but still adequate number of
monitoring points. Existing monitoring sites identified in this phase can be incorporated in the long-
term monitoring network. Phase I generally requires a 3- to 6-month period for completion.
Phase II is aptly named the Research Monitoring Program, or the phase of acquiring
information pertaining to how the subsurface system operates and of formulating interpretations.
Research monitoring is conducted to improve, or verify, elements of the hydrogeologic conceptual
model. A technically-defensible conceptual flow model ensures a more protective and reliable
monitoring program. Research monitoring for wellhead protection includes baseline water quality
characterization, aquifer testing and characterization, refined or verified WHPA delineation, and
ground-water flowpath determination to relate sources to the water supply well or spring. The
product of research monitoring is a proposed long-term monitoring program that may be partly
implemented in Phase II. Phase II may require 1 to 1-1/2 years for completion, depending on the
complexity of the site hydrogeology and the quality of the initial information base.
The by-product of Phases I and II is a proposed wellhead protection monitoring program,
Phase III. Generally, the program is submitted as a plan to be implemented in stages, as labor and
financial resources become available. The plan should include an organization chart, a source
assessment map and list, and a map depicting the WHPA and protective zones, as well as a
VII
-------
description of the delineation criteria and method(s). General and specific objectives for ambient
trend, source assessment, and early-warning detection monitoring should be detailed. Each objective
should justify the selection of monitoring sites, parameters, and frequencies.
The locations of existing and recommended monitoring sites in the proposed network should
be shown on a map. A formal identification system with a minimum set of data elements should be
used to label each site. The integrity of the design and construction of each existing site should be
considered prior to inclusion in the monitoring network to ensure data quality. New sites that require
installation should be described in detail, concerning completion depth, open or screened interval,
schematic design, and construction materials, as well as the methods of installation, development, and
testing. Physical and chemical parameters to be monitored at select frequencies should be listed and
technically justified. Monitoring site information should be stored in an automated data base for
convenient and safe storage, update, and retrieval. Each monitoring program should formulate a
minimum set of quality assurance and control objectives to match the objectives of the wellhead
protection monitoring program.
A 15-step approach for the design of a wellhead protection monitoring program is depicted as
a flowchart in Figure 1. The monitoring program should be reviewed and improved in an iterative
process over the life span of the WHPP. The organization of the case studies research in Chapters 2
through 5 follows the logical outline of Phases I, II, and III.
STEVENS POINT, WISCONSIN, CASE STUDY (Chapter 2)
The city of Stevens Point is located in central Wisconsin and has a population of
approximately 23,000. The source of the city water supply is from the Airport and Iverson wellfields.
These wellfields pump an average of 5 million gallons of water per day from a shallow, unconfined
aquifer. The aquifer is composed of coarse, unconsolidated sediments deposited by meltwater during
the Wisconsin glaciation. The preliminary wellhead protection zones for the combined wellfields were
based on estimates of the zone of influence (ZOI), the 5-year time-of-travel (TOT) zone (analytically
determined), and the recharge area. In the review process, the validity of the 5-year TOT calculation
was questioned, and the WHPA was never promulgated.
An extensive, historical source assessment was conducted within the B Zone of the preliminary
WHPA using aerial photographic interpretation techniques combined with conventional methods such
as surveys of directories, local and state records, visual inspections, monitoring data, and so on. Point
and nonpoint sources were identified, ranked, and prioritized for management and regulation.
Existing contaminant sources were given highest priority. Potential sources were then prioritized
based on source type, quantity, hazard, and location.
A network of 55 monitoring sites (single monitoring wells, well nests, and a multi-level well)
were used to measure water levels, to sample ground water, and to conduct aquifer tests in the
unconfined aquifer. Of the total network, three single wells and four well nests represent new
monitoring points installed for this research. Aquifer parameter results from slug, constant-discharge,
and recovery tests indicate a range of hydraulic conductivity values for three distinct geologic settings:
820 to 1,700 feet per day (ft/d) for the buried valley, 220 to 240 ft/d for ourwash plains, and 2 to 3 ft/d
for bedrock highs.
Vlll
-------
MONITORING
PROGRAM
DESIGN
CONDUCT SOURCE
ASSESSMENT WITHIN
WHPA
\/
COMPILE & ORGANIZE
EXISTING INFORMATION BASE
IDENTIFY DEFICIENCIES
OF INFORMATION BASE
DETERMINE DATA &
PROCESSING NEEDS FOR
SYSTEM CHARACTERIZATION
ESTABLISH
MONITORING OBJECTIVES
DETERMINE
MONITORING PARAMETERS
\/
IDENTIFY EXISTING MONITORING
SITES BASED ON OBJECTIVES
DETERMINE SAMPLING
FREQUENCIES FOR
MONITORING PARAMETERS
ASSESS NEED FOR
NEW MONITORING SITES
ESTABLISH NEW
MONITORING SITES
_v
IMPLEMENT
MONITORING PROGRAM
REVIEW & INTERPRET
MONITORING RESULTS
INCORPORATE INTERPRETATIONS
IN CHARACTERIZATION ASSESSMENTS
UPDATE MONITORING OBJECTIVES,
NETWORK DESIGN, & PROGRAM
R ITERATE
flONITORING
PROCESS
Figure 1. Flow chart of the 15-step monitoring methodology for wellhead protection areas.
IX
-------
Hydrochemical data indicate that nitrogen concentrations, a key indicator of contamination,
have increased over time. Currently, nitrogen concentrations in the monitoring network range from
less than 0.2 to 26.0 milligrams per liter. Other indicators of ground-water degradation include iron
and manganese from organic-rich soils located along the Plover River, chloride in proximity to roads
where de-icing occurs, and previous volatile organic compound contamination at the airport and
several underground storage tanks.
A two-dimensional, ground-water flow model (FLOWPATH) was used to delineate the 5- and
10-year TOT zones for the Airport and Iverson wellfields. In comparison, the previous, analytically
derived B Zone is larger, however, the 5-year TOT zone from FLOWPATH extends farther to the
east due to the effects of pumping at the Iverson wellfield and the presence of bedrock highs.
A long-term ground-water monitoring network is proposed for the Airport and Iverson
wellfields consisting of 34 existing and proposed wells. Nine new well locations are proposed to fill
data gaps in the existing network, primarily along the boundaries of the 5- and 10-year TOT zones.
Wells in the long-term monitoring network should be sampled twice a year in April and September
for indicator parameters. Water levels should be recorded each time a well is sampled. Compliance
monitoring networks are recommended for point sources of highest priority.
The wellhead protection contingency plan consists of three components: (1) reaction to the
early-warning detection system based on preventive action and state drinking water limits, (2) spill
response, and (3) new water-supply development and implementation.
LITTLETON, MASSACHUSETTS, CASE STUDY (Chapter 3)
The town of Littleton is located approximately 35 miles northwest of Boston in northeastern
Massachusetts. The daily water demand for the town's population of 7,300 is from 800,000 to
1,500,000 gallons per day from four production wells. Techniques for refined delineation and long-
term monitoring of the WHPA surrounding Production Well Number 5 (PW-5) are discussed in this
report. Production Well Number 5 is completed at a depth of 167 feet within saturated, stratified
valley-fill deposits. The aquifer is unconfined and receives significant surface-water recharge
(20 to 25%) from nearby Spectacle Pond and Bennetts Brook.
Land-use activities within the WHPA cover a range of commercial, industrial, and to a lesser
degree, agricultural operations. Collectively, these land-use activities pose potential contamination
threats to the aquifer, including heavy metals, volatile organic compounds, pesticides, and nutrients.
Baseline monitoring results indicate that ground-water quality within the capture zone of PW-5 is
currently unaffected by source operations. Sodium is the only exception. Slightly elevated levels of
sodium in surface water and the shallow aquifer are attributed to roadway de-icing. Manganese and
iron concentrations are elevated throughout the recharge area of PW-5, primarily because of their
occurrence in wetland sediments and glacial deposits. The levels of these parameters have increased at
PW-5 for several years and may warrant treatment in the future.
The PW-5 WHPA consists of three protection zones delineated using a combination of
numerical ground-water flow models (FLOWPATH, FLOWCAD, and MODFLOW) and
hydrogeologic mapping. Zone I is the 400-foot sanitary protective radius mandated by the State of
-------
Massachusetts. Zone II is the most critical management area and was delineated conservatively as the
union of three numerical capture zone solutions. These numerical solutions incorporate two- and
three-dimensional flow, as well as steady-state and transient flow conditions. Local and regional
ground-water flow simulations are based on the results of short- and long-term aquifer testing.
Zone II generally represents the steady-state capture zone for PW-5 that corresponds
approximately to the 400-day travel-time contour. Flowpath simulations indicate that Zone II extends
to the bottom of the aquifer and is constrained by bedrock and glacial till. Within Zone II, three
existing wells and two new wells are proposed for inclusion in the monitoring network for early-
warning detection and source assessment purposes. These wells lie along either the 150-day or the
300-day travel-time contours. Screened intervals for the new monitoring wells were chosen based on
results from MODPATH computer flow simulations. Monitoring parameter groups for these wells
include general water quality, site-specific, and physical parameters. Recommended monitoring
frequencies for these parameter groups vary from quarterly to annually, depending on the travel-time
distance from the monitoring well to PW-5 and the monitoring well depth.
Zone III is defined as the upgradient area of the aquifer that contributes to Zone II and
extends to the watershed boundary. Zone III is monitored at two surface-water locations, one at the
inflow and one at the outflow of Spectacle Pond. In addition, Zone III is monitored biannually at
existing compliance networks around waste management and industrial sites. Monitoring parameters
for the compliance wells include general water quality, site-specific, and physical parameter groups.
The Littleton WHPP incorporates contingency planning. Catastrophic releases initiate a spill-
response plan that involves many departments and agencies. In the event of contamination of PW-5
or another production well, Littleton has sited a new production well. The proposed well site is
approved by the State, and protection Zones I, II, and III are delineated. The adjacent town of
Boxborough shares the recharge area to the proposed well. Boxborough has adopted complementary
strategies with Littleton to ensure its water quality protection.
SIOUX FALLS, SOUTH DAKOTA, CASE STUDY (Chapter 4)
The city of Sioux Falls is located in the southeast corner of South Dakota The Big Sioux
aquifer is the primary source of water for about 125,000 persons in the Sioux Falls metropolitan area.
One of the municipal wells in the Big Sioux aquifer, the airport wellfield, is underlain by surficial,
glacial outwash deposits. The Big Sioux River is located directly west of the airport wellfield and flows
south over and through the outwash, draining approximately 4,000 square miles of upstream land.
The city's wells pump most of their water directly from the aquifer and a small quantity from
the Big Sioux River. However, the river is hydraulically connected to the aquifer, and recharge from
the river in the airport wellfield area is significant. In 1988, approximately 79% of the recharge to the
airport wellfield aquifer was induced from the river due to wells pumping. Induced flow from the river
to the aquifer is demonstrated by decreased flow in the river during low recharge periods.
XI
-------
This research was conducted to evaluate (1) the hydraulic connection between the Big Sioux
River and the adjacent aquifer, and (2) the potential impact of the river on aquifer water quality. In
the broader perspective, additional goals included refined delineation of the wellfield protection area
and design of a long-term water quality monitoring program.
Drilling logs indicate that the thickness of the aquifer in the wellfield area ranges from 20 to
50 feet. Aquifer testing results yield an average hydraulic conductivity value of 800 ft/d and a
transmissivity value of approximately 21,000 square feet per day for the aquifer.
Many potential point sources of contamination exist in the study area. These include:
industrial and commercial areas, the South Dakota Air National Guard facility, a petroleum pipeline,
the Sioux Falls Regional Airport, and a decommissioned municipal landfill. The threat of
contamination from these sources is underscored by the recent history of contaminant releases in the
area.
To estimate ground-water travel times in the study area, aquifer testing, dye tracing, and
ground-water modeling were employed. During aquifer testing, two dye injections were made. The
first dye was injected in a well approximately 40 feet north of the pumping well. Detectable dye
concentrations first arrived at the pumped well after about 12 hours. The second dye was injected in a
well near the edge of the river, approximately 140 feet north of the pumping well. Detectable dye
concentrations from the second injection site first arrived at the pumped well in 7 to 9 days. Aquifer
testing and dye-tracing results indicate that a -contaminant could travel from the river to the wellfield
in less than 9 days.
A two-dimensional, steady-state model (FLOWPATH) was used to generate time-related
capture zones for the municipal wells and to simulate contaminant travel times. One-, two-, and five-
year capture zones were calculated for each of the municipal wells in the airport wellfield. Modeling
of simulated spill sites from several of the potential point-source contamination areas indicates that
contaminants entering the aquifer at areas to the north and south of the well field could reach the
municipal wells in 1 to 2 years.
The City of Sioux Falls and Minnehaha County have delineated wellhead protection areas by
using the hydrogeologic-mapping method. Wellhead protection ordinances are designed to impose
guidelines and restrictions on new land uses, or proposed changes in existing use, in order to protect
the aquifer water quality.
A wellhead protection monitoring program at the airport wellfield is proposed to document
ambient water quality conditions and to serve as an early-warning detection system. Line-source
monitoring is proposed to monitor the Big Sioux aquifer and the diversion canal for contaminants
that could potentially enter the aquifer. Point-source and nonpoint-source monitoring are proposed
to monitor water quality between the airport wellfield and potential sources. The categories of
parameters for monitoring are general water quality, VOCs, trace metals, pesticides, and nutrients.
Sampling frequencies for each of the categories were selected as a function of the type of source to be
monitored.
XII
-------
Contingency planning is warranted to establish emergency responses to contaminant releases
at the surface of the aquifer and in the river. Alternative water supply development must also be
continued as part of the contingency planning effort.
DOVER, NEW HAMPSHIRE, CASE STUDY (Chapter 5)
Dover is a city of 26,000 people located in the seacoast region of New Hampshire. To meet
the increasing water supply demands of the future, the city embarked on a water exploration effort in
a fractured-bedrock aquifer at the Blackwater Brook site. A test well was installed to a depth of 400
feet in the bedrock aquifer as part of the ground-water exploration program. A wellhead protection
area and ground-water monitoring strategy were established for the test well. This study describes how
the conceptual hydrogeologic model for the site was developed and refined.
The bedrock aquifer consists primarily of quartz monzonite and metasedimentary rocks that
interfinger along a fractured, faulted contact zone trending north 60 degrees east (N60°E). A
N5-10°W trending lineament and fracture zone intersects the N60°E zone at the site. The bedrock
aquifer is directly overlain by Pleistocene-age sands and gravels. These sediments are overlain by low-
permeability marine clay and lodgement till. It is estimated that 20% of the water produced from the
bedrock aquifer is derived from overburden sediments in the watershed area.
Four overburden and bedrock well pairs constitute the present monitoring network for the test
well. Two well pairs lie along the N60°E faulted contact zone, and two well pairs lie along the
perpendicular N30°W trend. The test well and four of five bedrock wells airlift in excess of
150 gallons per minute. Few contaminant threats exist near the site. Baseline sampling indicates that
minor, elevated levels of iron, manganese, and radon pose the only water quality problems at present.
Test drilling and borehole surveys (caliper, video camera, acoustic televiewer, thermal-pulse
flowmeter, and hydrophysical logging) indicate that fracturing and ground-water flow are highly
discrete. Flow occurs at isolated, definable depths rather than uniformly along the length of the
borehole. Hydrophysical logging indicates that the borehole water is distinctly layered with respect to
the fluid electrical conductivity parameter. Most borehole water is produced by moderately to steeply
dipping fractures and fracture zones that intersect the wells.
Aquifer testing and dye-trace results indicate that the N30°W and N60°E directions have
higher aquifer transmissivities relative to the surrounding bedrock matrix. Drawdow,n contours are
elongate about the N30°W well alignment, suggesting preferred flow in this direction. Dye-trace
results indicate more rapid travel of injected dye along the N30°W direction than the N60°E direction.
Dye traveled 152 feet in 130 minutes (the time of first arrival of the dye) from injection in a bedrock
monitoring well along the N30°W trend to the test well, which was pumped at 200 gallons per minute.
This represents a velocity of 1,680 ft/d. Dye injected in a bedrock monitoring well located 596 feet
from the test well arrived there in 148 hours, indicating a velocity of 96 ft/d along the N60°E direction.
Flowmeter and acoustic televiewer surveys indicate that a moderately west-dipping fracture
zone provides interconnection between the test well and bedrock well R2 along the N30°W trend.
Lacking discrete flow information beyond the test well and well R2, statistical fracture descriptions
become good approximations of flowpaths at increasing distances from the site. Therefore, prominent
Xlll
-------
fracture peaks along the N5-10°W and N60°E trends represent the most probable flow directions
within the bedrock fracture system at Blackwater Brook The N60°E trend is substantiated by the
existence of the faulted, fractured contact zone along this strike. Evidence to suggest preferred flow
along the N5-10°W direction is structural and hydrogeologic. Structural control is inferred by strong
expression of the lineament on several platforms of photography and in outcrop fracture trends.
Enhanced transmissiviry along the N30°W direction is attributed to the proximity and similar
orientation of the N5-10°W fracture zone.
A quadratic equation is derived from accepted hydrogeologic relationships (Darcy's Law arid
the Thiem equation). In this equation, ground-water travel time (determined using the time of first
arrival of dye at the test well) is directly proportional to the square root of distance from the test well.
Constants of proportionality for the quadratic relationship are calculated for the N30°W and the
N60°E directions based on dye-trace velocities. Distances for the 200-day and 1,000-day TOT
thresholds are then calculated for the two fracture zone directions: N5-10°W and N60°E.
Three wellhead protection zones are delineated within the recharge area for the test well using
a variety of criteria and methods. Zone I is the state-mandated 400-foot sanitary radius. Zone IIA
consists of two 1,000-foot-wide "arms" along the N5-10°W and N60°E directions, extending to the
200-day TOT distances. Zone IIB is the area within a smooth curve connecting the outer boundaries
of Zone IIA, producing an oval shape. Zone III is the upgradient area contributing to the 1,000-day
TOT distance modified by hydrogeologic features. Recommended regulation of the wellhead
protection zones varies from complete control and restriction of activities in Zone I to public
education in Zone III.
A major component of wellhead protection program management is long-term ground-water
monitoring. Under present conditions, monitoring of the test well and existing monitoring wells will
focus on a moderate effort to assess ambient water quality and physical parameters. After the
production well is developed, the monitoring frequency and list of monitoring parameters increases.
Proposed frequencies, parameters, and new sites for monitoring derive from technical and
management goals. Action levels are proposed to trigger contingency responses.
xiv
-------
EPA-600/R-93/
April 1992
CHAPTER 1
MONITORING METHODOLOGY FOR WELLHEAD PROTECTION
by
Beth A Moore
Lockheed Environmental Systems & Technologies Company
Las Vegas, Nevada 89119
April 1993
Contract Number 68-CO-0049
Project Officer
Steven P. Gardner
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas, Nevada 89193-3478
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89193-3478
-------
NOTICE
This report is the result of research supported by the U.S. Environmental Protection Agency's
Environmental Monitoring Systems Laboratory in Las Vegas, Nevada, as part of its efforts to provide
technical assistance to state, tribal, and local governments on the implementation of the Wellhead
Protection Program. The specific methods and approaches contained in this document have
undergone peer review but do not constitute official Agency endorsement or policy recommendations.
The Office of Research and Development provides this information to help solve complex technical
problems related to refined delineation and ground-water monitoring of wellhead protection areas in
various hydrogeologic settings. Further assistance is available from the Environmental Monitoring
Systems Laboratory in Las Vegas, the Office of Ground Water and Drinking Water in Washington,
D.C, and the ground-water offices in the ten U.S. Environmental Protection Agency regions.
-------
ABSTRACT
A methodology for planning and implementing a wellhead protection monitoring program is
formulated and demonstrated at unique case study sites. This methodology emphasizes saturated zone
monitoring and is intended to serve as a guide for WHPP implementors. Careful implementation of
this methodology will enable managers and scientists to establish technically defensible, reliable, and
effective ground-water monitoring programs for wellhead protection.
Basic hydrogeology concepts and equations are discussed as they pertain to ground-water
systems and flow, conceptual hydrogeologic models and flow nets, and accurate delineation and
monitoring in different hydrogeologic settings. The spectrum of unconfined to confined aquifers is
discussed in relation to porous, granular aquifers; fractured-bedrock aquifers; and karst aquifers.
Physical and chemical parameter monitoring apply to wellhead protection. Three types of
ground-water monitoring are useful in managing wellhead protection areas ~ ambient trend, source
assessment, and early-warning detection monitoring. Ambient trend monitoring detects the temporal
and spatial trends in physical and chemical quality of the ground-water system. Source assessment
monitoring evaluates the existing or potential impacts on the physical or chemical ground-water
system from a proposed, active, or abandoned contaminant source. Early-warning detection
monitoring is conducted upgradient from the wellhead, based on known travel times, to trigger a
contingency response to prevent public exposure to contaminants in aquifers; they should not be
mistaken as preventative or remedial measures.
A long-term monitoring program for wellhead protection is developed in three phases. Phase
I, Wellhead Protection Program Elements and Scoping Tasks, includes: determining technical roles
and a management framework, preliminary wellhead protection area delineation, and source
assessment. This phase may require 3 to 6 months for completion. Phase II is referred to as the
Research Monitoring Program, or the phase of acquiring information pertaining to how the
subsurface system operates and of formulating interpretations. Research monitoring includes baseline
water quality characterization, aquifer testing and characterization, refined or verified wellhead
protection area delineation, and ground-water flowpath determination to relate to the water supply,
well, or spring. Phase II may require 1 to 1-1/2 years for completion.
The by-product of completing the first two phases is a proposed Wellhead Protection
Monitoring Program, Phase III. Generally, the Program is submitted as a plan to be completed in
stages, as labor and financial resources become available. The program should detail the roles and
duties of the implementation team, a source assessment map and list, and a map depicting the
wellhead protection area and protection zones. Each objective should justify the selection of
monitoring sites, parameters, and frequencies.
-------
The locations of existing and recommended monitoring sites in the proposed network should
be shown on a map. A formal identification system with a minimum set of data elements should be
used to label each site. The integrity of the design and construction of each existing site should be
considered prior to inclusion in the monitoring network to ensure data representation. New sites that
require installation should be described in detail, concerning: completion depth, open or screened
interval, schematic design, and construction materials, as well as the method of installation,
development, and testing. Physical and chemical parameters to be monitored at select frequencies
should be listed and technically justified. Monitoring site information should be stored in an
automated data base for convenient and safe storage, update, and retrieval. Each monitoring
program should formulate a minimum set of quality assurance and quality control objectives to match
the objectives of the wellhead protection monitoring program.
A step-wise approach for the design of a wellhead protection monitoring program is given,
incorporating Phases I, II, and III. The monitoring program should be reviewed and improved over
the life span of the Wellhead Protection Program as an iterative process.
1-iv
-------
TABLE OF CONTENTS
Abstract 1-iii
Figures 1-vii
Tables 1-ix
Abbreviations 1-xi
Acknowledgements 1-xii
Wellhead Protection Area Monitoring 1-1
Introduction 1-1
Purpose and Scope of the Document .- 1-2
Organization of Chapter 1 1-6
Hydrogeologic Concepts 1-7
Ground-water Systems 1-7
Ground-water Flow 1-10
Accurate Delineation and Monitoring in Different Hydrogeologic Settings 1-12
Wellhead Protection Monitoring Considerations and Objectives 1-15
Source Assessment 1-17
Monitoring Methodology 1-18
Phase I: WHPP Elements and Scoping Tasks 1-18
Phase II: Research Monitoring Program 1-20
Phase III: Wellhead Protection Monitoring Program 1-32
Step-wise Approach for the Design of a Wellhead Protection Monitoring Program .... 1-33
References 1-35
Additional Bibliography 1-43
Appendix 1-A 1-47
1-v
-------
-------
LIST OF FIGURES
Number
1-1 Conceptual wellhead protection area and monitoring scenario 1-2
1-2 Location of the five case study research sites within the United States 1-5
1-3 Saturated zone within the subsurface 1_6
1-4 Aquifers, confining units, and the hydrologic cycle 1-8
1-5 Flownet in (A) plan view and (B) cross section, depicting horizontal and
vertical hydraulic gradients 1_9
1-6 Conceptual diagram depicting Darcy's Law 1-11
1-7 The pumping cone-of-depression and wellhead protection area terminology 1-13
1-8 Fractured-bedrock aquifer in Dover, New Hampshire, resulting in discrete flow
zones 1_15
1-9 Zone of contribution for a water supply spring in a carbonate, karst terrain 1-16
1-10 Flowchart of the 15-step monitoring methodology for wellhead protection areas ... 1-22
1-11 The fate of organic contaminants in the subsurface 1-24
1-12 Organic contaminant spreading within the unsaturated zone because of water-table
fluctuations _
1-vii
-------
-------
LIST OF TABLES
Number
1-1 Characteristics of the Case Study Research Sites
1-2 Common Sources of Ground-Water Contamination in Wellhead Protection Areas .
1-3 Methodology for Assessing Point Sources of Contamination
1-4 Wellhead Protection Monitoring Program Development
1-5 Common Chemical Constituents in Ground Water ;
1-6 Monitoring Parameters for Categories of Sources in Proximity to Wellhead
Protection Areas
Page
1-4
1-19
1-20
1-21
1-26
1-29
1-ix
-------
-------
ABBREVIATIONS AND SYMBOLS
ABBREVIATIONS
BOD biochemical oxygen demand
COD chemcial oxygen demand
DNAPL dense, nonaqueous phase liquid
DO dissolved oxygen
EMSL-LV Environmental Monitoring Systems Laboratory, Las Vegas, Nevada
EPA U.S. Environmental Protection Agency
LNAPL light, nonaqueous phase liquid
mg/1 milligram per liter
PCB polychlorinated biphenyl
ppm part per million
SDWA Safe Drinking Water Act
SHPA springhead protection area
TDS total dissolved solid
TOC total organic carbon
TSS total suspended solid
TU tritium unit
VOC volatile organic compound
WHPA wellhead protection area
WHPP Wellhead Protection Program
ZOC zone of contribution
ZOI zone of influence
SYMBOLS
A cross-sectional area
dh/dl head loss per unit distance
Eh redox potential
I hydraulic gradient
K hydraulic conductivity
n porosity
Q discharge per time
q specific discharge
v average ground-water velocity
1-xi
-------
ACKNOWLEDGEMENTS
This document was prepared for the U.S. Environmental Protection Agency (EPA),
Environmental Monitoring Systems Laboratory in Las Vegas, Nevada (EMSL-LV), under Contract
Number 68-CO-0049 to Lockheed Environmental Systems and Technologies Company (LESAT). We
gratefully acknowledge Steven Gardner (the EMSL-LV Project Officer) and Joseph D'Lugosz
(EMSL-LV) who approved this cooperative research effort in 1989, and provided strong moral and
financial support through its completion. Ron Hoffer (U.S. EPA Office of Water), and many other
dedicated individuals, nurtured and built the Wellhead Protection Program—the success of which
supported this research. Dr. Philip Berger (U.S. EPA Office of Water), Dr. Jeffrey Rosenfeld (South
Florida Water Management District), Douglas Heath (U.S. EPA Region I), Michael Wireman
(U.S. EPA Region VIII), and John Rotert (LESAT) offered excellent technical input and critical
review to the research scope and direction through its duration. The authors are indebted to the
following individuals who provided formal review comments: Dr. Charles Kreitler (University of
Arizona), Douglas Heath (U.S. EPA Region I), and Dr. Philip Berger (U.S. EPA Office of Water).
Theresa Brown (University of Wisconsin-Madison) and Alan Cathcart (Littleton Water Department)
provided thorough and critical reviews of Chapter 1.
Special thanks are extended to the many people at LESAT, the U.S. EPA, and elsewhere, who
helped to publish this manuscript. John Nicholson and Carolyn Cameron (LESAT) provided tireless
and competent technical writing and editing support. Steve Garcia and Elizabeth Morehouse
(LESAT), along with Robin Roth, contributed excellent graphics to the entire manuscript. Shalena
Fendzlau and Lillian Steele (LESAT) are graciously acknowledged for their patience and expertise in
word processing support. Annette King (LESAT) thoughtfully donated her time for general support
and report reproduction. The publication staff at the U.S. EPA, Center for Environmental Research
and Information, are responsible for final document publication and are acknowlegded for their
efforts.
1-xii
-------
WELLHEAD PROTECTION AREA MONITORING
INTRODUCTION
The increasing contaminant threat to public water supply wells has created a new political and
technical awareness of ground-water protection programs. Recognizing the need for conjunctive
management of contaminant sources and public water supplies to prevent, or minimize, ground-water
quality degradation, Congress amended the Safe Drinking Water Act (SDWA) in 1986 to include
Section 1428. This section mandated the development of the Wellhead Protection Program (WHPP),
which established a legal framework to protect public water supply wells, wellfields, and springs from
contamination.
Elements of the WHPP include, at a minimum (U.S. EPA, 1989b):
• Specifying the roles and duties of cooperating agencies
• Delineating the wellhead protection area (WHPA) for each well, wellfield, or spring
• Identifying sources of contamination within and in proximity to each WHPA
• Developing management approaches to protect the water supply within WHPAs from
contamination
• Developing contingency plans in response to well, wellfield, or spring contamination
• Siting new wells properly to maximize yield and to minimize contamination
• Ensuring public participation
An important technical element of WHPP implementation is WHPA delineation. A wellhead
protection area is defined as the surface and subsurface area surrounding a well, wellfield, or spring,
through which contaminants may pass and reach the ground water contributing to the supply source
(Figure 1-1). Criteria and methods for WHPA delineation are discussed in the following guidance
documents: U.S. EPA, 1987a; Wisconsin Geological and Natural History Survey, 1991; and Bureau of
Economic Geology, Universiy of Texas-Austin, 1991.
Ground-water monitoring is one of many management tools available for protecting public
drinking water supplies (U.S. EPA, 1989b). Ground-water monitoring of physical and chemical
parameters may enhance source characterization, WHPA delineation, and new water supply
evaluation. Ground-water parameters are monitored for the following reasons:
• To assess source-control measures
• To monitor compliance with drinking water standards at sites other than the wellhead
(municipal supply monitoring already required by the SDWA)
• To provide advance warning of contaminants in ground water
1-1
-------
MONITORING
WELL
LANDFILL
WHPA
WATER WELL-
WATER
TABLE
VGROUND-WATER
CONTAMINATION PLUME
Figure 1-1. Conceptual wellhead protection area and monitoring scenario.
This technical assistance document, "Case Studies in Wellhead Protection Area Delineation
and Monitoring," was developed to provide technical information to municipalities, states, and the U.S.
Environmental Protection Agency (EPA) Regions in their implementation of WHPPs. The primary
goals of this document are to present a monitoring methodology for WHPAs and to exemplify this
methodology in different case study settings. Developing a long-term monitoring program for a
WHPA involves much more than locating wells by intuition and "seat of the pants" sampling
programs. Reliable and effective monitoring requires significant forethought and investigation. This
document discusses monitoring objectives and recommends a sequence of steps associated with
planning and implementing a long-term program, referred to as the "research monitoring phase."
Correct delineation of the WHPA and a basic understanding of ground-water flowpaths to the well or
wellfield are prerequisites for proper monitoring network design and sampling. Therefore, the
methodology presented here emphasizes subsurface site characterization and verification of the
WHPA prior to long-term monitoring program design.
PURPOSE AND SCOPE OF THE DOCUMENT
A brief description of the evolution of this document is appropriate to define its purpose and
scope. The EPA's Office of Drinking Water and Ground Water (the client office) and the
Environmental Monitoring Systems Laboratory in Las Vegas, Nevada (EMSL-LV), began cooperative
research in 1988 to develop guidance in effective ground-water monitoring for wellhead protection.
1-2
-------
This research was intended to produce a complementary document to "Guidelines for Delineation of
Wellhead Protection Areas" (U.S. EPA, 1987a). A nation-wide survey of existing WHPPs in 1989
indicated a number of fledgling programs; most were struggling through the delineation process and
legal enforcement of the WHPA. A literature survey uncovered numerous references regarding
ground-water monitoring planning, methodologies, technologies, and interpretive approaches. As a
result of these surveys, it was determined that general guidance on monitoring WHPAs would
duplicate the existing literature and be of little use to the intended audience. By studying existing
WHPPs, it became apparent that each program reflected unique municipal and state legal, economic,
technical, and physical factors or constraints. In light of these findings, EMSL-LV engaged in
cooperative research with five carefully selected municipalities in 1989 to conduct refined delineation of
existing WHPAs and to develop long-term monitoring programs.
The product of the cooperative research is this document, which contains two types of
information:
(1) A recommended methodology for planning and implementing a wellhead protection
monitoring program, which emphasizes source assessment, correct WHPA delineation,
and hydrogeologic characterization (Chapter 1)
(2) Five case study narratives exemplifying the application of the monitoring methodology
for different hydrogeologic and contaminant source settings (Table 1-1; Figure 1-2):
Stevens Point, Wisconsin (Chapter 2)
Littleton, Massachusetts (Chapter 3)
Sioux Falls, South Dakota (Chapter 4)
Dover, New Hampshire (Chapter 5)
Springfield, Missouri (Chapter 1, Appendix A-l)
The methodology presented in Chapter 1 serves as a guide for WHPP implementors in establishing
technically-defensible, reliable, and effective ground-water monitoring programs for wellhead
protection. This methodology emphasizes saturated zone monitoring, or strategies used to monitor
the zone beneath the water table where interconnected openings are filled with water (Figure 1-3).
References and an additional bibliography list are provided at the end of Chapter 1 regarding
applicable technologies, as well as analytical and interpretive approaches used to establish ground-
water monitoring programs.
Of the 66 municipal programs evaluated for participation in this research, five were chosen
based on the strength of their existing WHPP, the hydrogeologic setting, the type of sources present,
and the qualifications of the participating scientists to conduct research. When this research began in
1989, the WHPP was relatively well established in the eastern states, and less so in the western states.
Consequently, few arid settings were evaluated as potential case study research sites, and none are
included in this document. Distinguishing characteristics of the five case studies and hydrogeologic
techniques used to characterize the flow systems are summarized in Table 1-1. The first four case
study narratives are presented in the document (Chapters 1 through 4) in order of increasing
hydrogeologic complexity. The exception to this organization is the Springfield, Missouri, case study,
1-3
-------
TABLE 1-1. CHARACTERISTICS OF THE CASE STUDY RESEARCH SITES
Municipality
Hydrogeologic Setting
Characterization Tasks
Stevens Point,
Wisconsin
Littleton,
Massachusetts
Sioux Falls,
South Dakota
Dover,
New Hampshire
Springfield,
Missouri
Unconfined aquifer
Unconfined aquifer, recharge
from Spectacle Pond
Unconfined aquifer; recharge
from the Big Sioux River
Fractured-bedrock aquifer,
discrete flow system
Mature karst (porous
limestone) aquifer, conduit
flow system
Flow system modeled with FLOWPATH
(two-dimensional, steady state); Point and
nonpoint sources assessed
Flow system modeled with FLOWPATH
(two-dimensional, steady state);
MODFLOW (three-dimensional, steady
state); and FLOWCAD (two-dimensional,
transient); Industrial and commercial point
sources assessed
Flow system modeled with FLOWPATH
(two-dimensional, steady state); Big Sioux
River assessed as a line source; Point and
nonpoint sources present
Flow system characterized with lineament
analysis, structural-control mapping,
aquifer testing, dye tracing, and borehole
geophysics; A few commercial, point
sources, and natural sources
Flow system determined by watershed
boundaries, dye tracing, and flow analysis;
Point and nonpoint sources assessed
which is presented in abbreviated form in Appendix 1-A of Chapter 1. As a general rule, the
hydrogeologic settings presented in this document become progressivley more heterogeneous, or
incorporate additional flow factors, such as the nearby river in the case of Sioux Falls, South Dakota.
This organizational approach will enable nontechnical readers to progressively increase their
vocabulary and knowledge of hydrogeology and wellhead protection as they read each section.
The Stevens Point, Wisconsin, case study (Chapter 2) is typified by an unconfined, sand and
gravel aquifer (modeled as a two-dimensional flow system) overlain by both urban point and nonpoint
sources. The watersheds in Littleton, Massachusetts (Chapter 3), consist of stratified-drift, valley-fill
deposits that are primarily under unconfined conditions and receive significant recharge from surface
ponds (modeled as two- and three-dimensional flow systems). Point sources are the focus of
assessment and regulation in Littleton. Sioux Falls, South Dakota (Chapter 4), derives its municipal
drinking water from a surface-water and ground-water flow system consisting of induced flow from the
Big Sioux River through the unconfined, sands and gravels of the Big Sioux aquifer. Three types of
sources are present within the Sioux Falls municipal WHPA, listed here in order of importance: the
Big Sioux River as a nearby line source to the wellfield, urban point sources, and rural nonpoint
sources. Dover, New Hampshire (Chapter 5), is developing a new water supply well in a fractured,
faulted zone within the metasedimentary bedrock; the bedrock is characterized by discrete flow zones.
1-4
-------
CO
I
I
1-1
o
o
o
1
60
1-5
-------
«*•
RIVERS AND LAKES ^>\
EC
til
1
Q
Z
3
O
DC
0
DC
111
Q
Z
=>
^
a
Ul
!* in
tcz
3 O
5°
(0
a
Ul
Kill
>fV
t
INTERMEDIATE ZONE '
1
_^ m
r
• ^M !•!•!
n«— Y ~~™ ^^ ^™" T™" ™H ~T* "•* ™9* ™^ *~"~ ^™" 7™~ *^* ^™ ")™ ""^ ^t "V*
1 > I ! | CAPILLARY FRINGE ) I I } I 1 I (
i L I J ' { M i i i i U » ' U i
(WATER TABLE)
GROUND WATER
WELL
Figure 1-3. Saturated zone within the subsurface. (Modified after Pettyjohn, 1987, p. 4.)
By design, sources are rare within and in proximity to the new WHPA for Dover. However, some
common point sources exist, such as gas stations and residential developments. Springfield, Missouri
(Chapter 1, Appendix 1-A), is situated in a mature karst setting. The city receives 25% of its drinking
water from a large capacity spring, Fulbright Spring, that issues from the base of a pure limestone
formation. The springhead protection area (SHPA) contains both point and nonpoint sources of
potential contamination.
ORGANIZATION OF CHAPTER 1
Chapter 1 begins with a discussion of basic hydrogeology concepts and equations pertaining to
flow systems, WHPA delineation, and ground-water monitoring for different aquifer systems
(unconfined, confined, fractured-bedrock, and karst aquifers). The monitoring methodology includes a
discussion of general and specific objectives as well as factors to be considered in program planning.
Source assessment is addressed as a critical first step in the design of an effective monitoring program.
Monitoring program design and development are divided into three phases and discussed in detail, as
follows:
• Phase I: WHPP Elements and Scoping Tasks
• Phase II: Research Monitoring Program
• Phase III: Wellhead Protection Monitoring Program
Chapter 1 concludes with a step-wise approach for the design of a wellhead prtoection monitoring
program.
1-6
-------
HYDROGEOLOGIC CONCEPTS
Basic hydrogeologic concepts related to ground-water flow, wellhead protection, and
monitoring network design are discussed in this section. In less technical terms, an aquifer can be
defined as "a natural zone below the surface that yields water in suffuciently large amounts to be
important economically" (Davis and DeWiest, 1966). Aquifers may exist within unconsolidated
sedimentary deposits, porous sandstone, karstic limestone, fractured bedrock, and other geologic
materials. The most common subdivision of aquifers is based on stratigraphic setting and hydraulic
head relationships: unconfined versus confined aquifers (U.S. EPA, 1987a). In reality, this is not a
subdivision, but rather, a continuum from unconfined to semi-confined to confined conditions.
Another common way to categorize aquifers is with respect to porosity and flow velocities, or
potential ground-water travel times. Ground-water flow in conduits and fractures differs radically
from flow in granular aquifers (Quinlan, 1989). Travel times in karst and fractured-bedrock aquifers
may be as rapid as hours to days—considerably briefer than in porous, granular aquifers (U.S. EPA,
1987a). Consequently, these more susceptable aquifers should be evalutated differently regarding
characterization, delineation, and monitoring methods for wellhead protection. Karst and fractured
bedrock aquifers can also exist under unconfined or confined conditions. To simplify discussions in
this section, aquifers will be referenced according to four basic categories: unconfined, confined,
fractured-bedrock, and karst aquifers. The hydrogeologic factors governing this categorization and
their effect on WHPA delineation approaches and monitoring strategies are briefly reviewed.
Ground-Water Systems
Subsurface water beneath the water table in soils and geologic formations that are fully
saturated is generally referred to as ground water (Freeze and Cherry, 1979). Technically speaking,
aquifers are saturated, permeable geologic units that transmit and yield water in usable quantities to a
well or a spring. Ground water is an integral part of the hydrologic cycle, accounting for the
movement of water between the ocean, atmosphere, and land (Figure 1-4; modified after Canter et al.,
1987).
A ground-water system is comprised of the underlying aquifers and less permeable units,
known as aquitards or confining beds. These units partly govern the dynamics of flow between areas
where water enters (recharges) and leaves (discharges) the ground-water system (Figure 1-4). Correct
identification of the recharge and discharge areas for a public water well, wellfield, or spring is critical
for wellhead protection because of increasing surficial contamination and land-use for waste disposal.
A ground-water system stores and transmits water as a function of its porosity (pores and
fracture openings) and hydraulic conductivity (water-transmitting capability of the geologic unit),
respectively (Pettyjohn, 1987). Ground-water movement, defined hi terms of flow velocity and
direction, is dictated by hydraulic gradients (slope of the water table or potentiometric surface;
Figure 1-4) and hydraulic conductivities.
In a ground-water system, the height to which water will rise in a well (usually given in units of
elevation) is the water table, hydraulic, or potentiometric head at a specific point in the aquifer
(Figure 1-4). By depicting hydraulic heads on a map (Figure 1-5A) or in cross section (Figure 1-5B)
1-7
-------
D
OC
1
ti
ci.
•3
t3
S3
C9
"8
!
i
6*
•a-
o
•S
I
^
a
S5
1
f
1-8
-------
INfUVQ 3AO8V 133d
_• ^ T— O)O)O>O>O>00
• •
o
o_
o
cvf
UJ
U
tn
O
N
tr
o
x
g
o
UJ
o
CE
O
0s
.23
2
1
1
1
s
•£3
2
J
a.
fi
00
1-9
-------
for a given depth in the aquifer, the hydraulic gradient can be determined. Ground water moves
generally downgradient, from recharge areas of high hydraulic head, or potential, to discharge areas of
lower hydraulic head.
In general, ground-water recharge areas are much larger in areal extent than discharge areas in
an unconfined system (Pettyjohn, 1987). Recharge is usually intermittent and is controlled by
precipitation events. Conversely, discharge occurs as a more continuous process at wells, springs, and
surface-water bodies, as well as from evaporation and transpiration (by plants) losses.
Ground-Water Flow
Ground-Water Velocity-
Henry Darcy is recognized for his contribution to present day ground-water knowledge and
flow theory. In 1856, Darcy's Law (Equation 1 and Figure 1-6) was formulated relating ground-water
discharge (Q, volume per time) directly to the hydraulic gradient (I or dh/dl, head loss per unit
distance) and cross-sectional area (A, distance squared) through which flow occurs by a constant,
hydraulic conductivity (K, velocity or distance per time):
-------
UNIT ELEMENT
OF AQUIFER-
STREAMLINES
REPRESENTING
LAMINAR FLOW
CONFINED AQUIFER
UNIT PRISM OF AQUIFER
Figure 1-6. Conceptual diagram depicting Darcy's Law. (Modified after Pettyjohn, 1987, p. 10.)
Equation 2 which defines the specific discharge, q. Specific discharge is the rate of flow (discharge per
time, Q) per unit area (A), and it bears the units of a velocity term, distance per time. However,
specific discharge is not used as a velocity term; it is more commonly referred as to as the Darcy flux
(Freeze and Cherry, 1979):
(Equation 2)
Combining terms and simplifying, Equation 3 is derived:
.
q = -KI
(Equation 3)
The average ground-water particle velocity through the pores, v, is found by incorporating the porosity
of the geologic unit (n, unitless ratio) in the denominator of Equation 3, yielding Equation 4:
V =
KI
n
(Equation 4)
1-11
-------
Ground-water flow velocity and travel-time calculations are used simply as average estimates of
contaminant or water particle movement through an aquifer. Realizing that factors such as dispersion
and retardation can complicate solute transport predictions, the actual movement of contaminants
within the aquifer may be several times slower or faster than the average advective rate. In complex
settings, such as mature karst, travel times for ground water and contaminants vary from hours to
weeks in response to differing recharge and flood-pulse events (Quinlan, 1989; Moore et al., 1990).
Conceptual Hydrogeologic Models and Flow Nets—
The success of wellhead protection and detection monitoring is based, in part, on a good
technical understanding of (1) the potential and existing sources of contamination, and (2) the
hydrogeology of the ground-water supply system. Therefore, an accurate hydrogeologic conceptual
model of the ground-water supply system should be developed to support wellhead protection
implementation and management. The cause and effect relationships of how the aquifer system works
(conceptual model) are expressed by:
• Aquifer water quality and contaminants
• Hydrogeologic units and boundaries
• Aquifer parameters and characteristics
• , Ground-water flow dynamics such as recharge and discharge areas, directions,
velocities, and flow-system scales (local, intermediate, and regional flow)
• Integration of the flow field created by pumping of the well, or wellfield, over a
specific time period and the natural, regional flow gradient
A useful tool to depict flow dynamics is a flow net (Figure 1-5). Specifically, a flow net shows
the direction of ground-water movement and can be used to estimate the quantity of ground-water in
transit through the aquifer. One set of contour lines, equipotential lines, is drawn connecting lines of
equal water-table, hydraulic, or potential head. The other set, flow lines, represents probable paths of
ground-water flow through the aquifer.
A flow net, or similar cross-sectional representation of ground-water flow in the water-supply
system, should be a prerequisite to designing and establishing a long-term monitoring network for
wellhead protection. Hydrochemical characterization of strategic portions of the aquifer used for well
production (for example, recharge and discharge areas) is often beneficial to corroborate physical
interpretation of the flow system. Without clear definition of how potential and existing contaminant
sources are physically and chemically interrelated to wells or springs, via flowpaths, ground-water
supplies cannot be protected by management, regulation, or early-warning detection monitoring.
Correct delineation of the WHPA and effective monitoring are by-products of an accurate
conceptual model and flow simulation, done with either analytical or numerical methods. Managers
and scientists must do the necessary groundwork to ensure protection through proper well siting,
representative data base development, and effective allocation of budgetary and labor resources.
Accurate Delineation and Monitoring in Different Hydrogeologic Settings
Ground water for public water supplies is derived from wells and springs. Withdrawal of
ground water from a well or wellfield perturbs the natural gradient of the ground-water system
creating a "cone-of-depression" in the water table around the well or wellfield (Figure 1-7). An
1-12
-------
(A) VERTICAL
PROFILE
GROUND-WATER
DIVIDE
LAND
SURFACE
PREPUMPING
WATER LEVEL
CONE OF
DEPRESSION
BEDROCK
SURFACED
(B) PLAN
VIEW
GROUND-WATER
DIVIDE
LEGEND
5 WATER TABLE
0-YEAR ZONE OF TRANSPORT
•* DIRECTION OF GROUND-WATER FLOW
ZOC-ZONE OF CONTRIBUTION
ZOI-ZONE OF INFLUENCE
ZOT-ZONE OF TRANSPORT
Figure 1-7. The pumping cone-of-depression and wellhead protection area terminology.
(U.S. EPA, 1987b.)
1-13
-------
unconfined aquifer is a subsurface geologic unit in which water only partly fills the aquifer and the
water table is in direct contact with the atmosphere. Where an aquifer is bounded by a relatively
impermeable unit, is it is said to be confined (Figure 1-4). The shape of the cone-of-depression in an
aquifer of low transmissivity (the product of hydraulic conductivity and saturated aquifer thickness) is
steep and localized about the well, as opposed to a flatter and more regional cone in an aquifer of
high transmissivity. Transmissivity influences the shape of the cone-of-depression more than the
storativity parameter (Freeze and Cherry, 1979).
The projection of the cone-of-depression at the land surface is called the zone of influence
(ZOI). The entire area recharging the ZOI, from some criteria-determined (for example, time,
distance, drawdown, or boundary; U.S. EPA, 1987a), upgradient point to the downgradient, inflection
point (with respect to the well), is called the zone of contribution (ZOC), or capture zone. The ZOI
fells within a portion of the ZOC for a sloping water table, and corresponds to the ZOC for a flat
water table. It is important to determine the ZOC correctly to ensure wellhead protection because any
contaminants introduced within this zone may reach the well or wellfield.
In some fractured-bedrock aquifers, ground-water flow is structurally controlled by directional
fractures and faults, resulting in limited, but very productive (discrete) flow zones within the well
borehole (Figure 1-8; Griswold et al., 1990). Consequently, simple analytical approaches (U.S. EPA,
1987a) for accurate ZOC determination may not apply. Specialized techniques for fracture and fault
identification and flow characterization include: geologic and structure-control mapping, fracture trace
analysis, surface geophysical surveying, borehole geophysical logging, and dye-trace testing. The use of
these techniques to successfully delineate WHPAs was demonstrated in Lancaster, Pennsylvania
(geologic and structure-control mapping and fracture trace analysis; Ogiela and Moore, 1991), and in
Dover, New Hampshire (surface geophysics, borehole geophysics, and dye tracing; Griswold et al.,
1990; Vernon et al., 1992, Chapter 5).
In carbonate terrains such as Springfield, Missouri (Chapter 1, Appendix A-l), where springs
issue from fractures, formation boundaries, or bedding planes (Figure 1-9), the ZOC may not be so
easily determined. In simple cases where contributing ground-water divides correspond to surface-
water divides, the ZOC determination may be based on surface-water boundary criteria (Moore et al.,
1990; U.S. EPA, 1987a). Conversely, in more complex flow regimes, contributing areas and divides
may need to be determined or verified using fracture trace analysis, potentiometric-surface mapping,
aquifer testing, and dye-tracing methods.
In areas typified by Darcian-flow conditions, such as in homogeneous, porous granular
aquifers, standard methods and computer model evaluations for monitoring site selection or well
placement may be accurate, yielding optimal sites and reliable data. However, in more heterogeneous,
complex ground-water flow terrains, such as fractured-bedrock and karst aquifers, monitoring sites for
pollutants should be sited or verified using more appropriate, and often sophisticated, techniques.
Finally, to test the dependability of monitoring sites in these regimes, dye tracing may be used to verify
interconnection to sources under a variety of flow conditions (Quinlan, 1989).
1-14
-------
TEST WELL
GLACIAL
OUTWASH,
OUTCROP
BERWICK
FORMATION
QUARTZ
MONZONITE
FRACTURE
ZONE
Figure 1-8. Fractured-bedrock aquifer in Dover, New Hampshire, resulting in discrete flow zones.
(Griswold et al., 1990.)
WELLHEAD PROTECTION MONITORING CONSIDERATIONS AND OBJECTIVES
Physical and chemical parameter monitoring help to document ground-water trends that can
improve hydrogeologic characterization and enhance protection. Todd et al. (1976) wrote: "...the
objective of a monitoring program should be to collect, manage, and analyze the data on ground water
quality and the sources and causes of ground water pollution, and other information - geologic,
hydrologic, and econmonic - necessary to enable the EPA and the other state(s) involved to fulfil their
statutory responsibilites as regards protection of ground water quality..." Historical monitoring records
(long-term baseline data) provide a reference base to minimize duplication of efforts, to assess
program effectiveness, and to optimize program changes. In addition, an extensive monitoring data
base enables the analyst to utilize more rigorous techniques to better locate monitoring sites.
Monitoring objectives should be identified and kept in perspective for the duration of the
monitoring program. Objectives should be neither too broad nor to limiting. Ideally, a complete
objective should address three concerns (Canter et al., 1987). First, it is important to identify the
specific reasons for monitoring or the usage of the data. For example, are monitoring data being
acquired to identify unknown contaminants from a suspected source or for baseline water quality
assessment? Secondly, determine the available types of monitoring access points, such as wells,
1-15
-------
(A) VERTICAL
PROFILE
WATER
SUPPLY
SPRING
A1
(B) PLAN VIEW
LEGEND
O SINKHOLE
• WATER SUPPLY SPRING
-X--*- SURFACE STREAM
L_ CONDUIT SYSTEM
2 WATER TABLE
I . .1 LIMESTONE
ZOC = ZONE OF CONTRIBUTION
Figure 1-9. Zone of contribution for a water supply spring in a carbonate, karst terrain.
1-16
-------
springs, surface-water bodies, and caves. Thirdly, define how the monitoring information will spatially
represent the system. For example, are the data being collected to determine the type(s) and
concentration(s) of pollutants in the subsurface? Alternatively, is the intent of the data to depict the
spatial distribution of a contaminant plume?
Both physical and chemical parameter monitoring apply to wellhead protection. The reasons
for ground-water chemical monitoring with regard to wellhead protection were brielfly listed on
page 1-1. Physical monitoring examines the behavior of the aquifer over time to enhance
understanding of the impact of water withdrawal, well and aquifer performance, and management.
Three types of ground-water monitoring apply to wellhead protection: ambient trend, source
assessment, and early-warning detection monitoring (Canter et al., 1987). Ambient trend monitoring
detects the temporal and spatial trends in physical and chemical quality of the ground-water system.
Source assessment monitoring examines the existing or potential impacts on the physical or chemical
ground-water system from a proposed, active, or abandoned contaminant source. Early-warning
detection monitoring is conducted upgradient from the wellhead, based on known travel times, to
trigger a contingency response to prevent public exposure to contaminants. These types of monitoring
are planned to measure or detect contaminants in aquifers and should not be mistaken as preventative
or remedial measures (U.S. EPA, 1989b).
Elements that are common to all effective monitoring systems include an accurate
conceptualization of the sites and its environs, and hypotheses regarding contaminant migration.
Elements of the conceptual model include the sources, pathways, and sinks (or receptors). With
respect to the WHPP, the sink is the public water supply well, wellfield, or spring. During the initial
stages of program implementation, sources are identified, characterized, and prioritized. Concurrently,
WHPAs are delineated based on site-specific ground-water flow and contaminant transport
assessment. By integrating each component of the system to form an accurate conceptual model,
more efficient monitoring programs can be developed.
SOURCE ASSESSMENT
Contaminant source assessment is a critical first step in designing an effective monitoring
program; therefore, it is included as a Phase I task. Target monitoring parameters for early warning
detection and source assessment are best selected from a comprehensive list of known and suspected
contaminants associated with land-use activities and practices. Secondly, critical or optimal
monitoring sites may be determined, reflecting prioritization of sources. In this way, monitoring
objectives can be directed toward immediate concerns. Optimal monitoring parameters are
determined to enhance protection and characterization, to improve efficiency, and to limit the cost of
the program.
1-17
-------
An inventory of common sources of contamination within and in proximity to WHPAs is
given in Table 1-2 (U.S. EPA, 1990a). General guidance for conducting inventories of, prioritizing,
and characterizing sources for wellhead protection is given in the following U.S. EPA technical
assistance documents:
• Managing Ground Water Contamination Sources in Wellhead Protection Areas: A
Priority Setting Approach (1991a)
• A Guide for Conducting Contaminant Source Inventories for Public Drinking Water
Supplies (1991b)
• A Review of Sources of Ground-Water Contamination from Light-Industry (1990b)
• Tools for Local Governments (1989b)
Table 1-3 presents a methodology for assessing contaminant point sources (U.S. EPA, 1989c) that is
applicable to most sites and ground-water systems.
The source assessment process can take a qualitative approach (accomplished in days) versus
more quantitative approaches (accomplished in months or years), entailing source characterization,
remedial investigations, and a formal risk ranking and screening system. The approach taken is
dictated by (1) the threat of contamination that sources pose to the water supply, and (2) the unique
characteristics of the municipality such as WHPP objectives, financial resources, implementation
timing, regulatory framework, and labor resources.
MONITORING METHODOLOGY
Long-term monitoring programs were researched and are described in Chapters 2 through 5,
for four case study WHPAs. The proposed monitoring programs are expected to be implemented
over a 3- to 5-year-period, as money and resources are available and as they are allocated in a
prioritized manner. It is important to emphasize that wellhead protection monitoring is a long-term
management option for WHPP implementation. Therefore, the long-term monitoring program
should be reviewed and improved over the life span of the WHPP. As a result of the case study
monitoring research, a three-phase approach for monitoring program development is recommended
for implementors (Table 1-4). A flowchart is given in Figure 1-10 that outlines the 15-step
monitoring methodology. Monitoring for wellhead protection is generally initiated following the
source assessment, delineation, and initial regulation phases of program development.
Phase I: WHPP Elements and Scoping Tasks
Ground-water monitoring is a relatively expensive, time consuming, and technically demanding
task. For these reasons, establishing a scientifically sound and functional WHPP is recommended to
provide adequate protection of the public water supply. At the same time, monitoring planning and
information gathering can proceed. Basic elements of WHPP implementation include: designating
roles and a management framework, preliminary WHPA delineation, and source assessment. To
support the research monitoring task, a preliminary information base of ancillary data (geologic and
topographic maps, lithologic logs, aquifer properties and characteristics, etc.) and monitoring data
1-18
-------
55
§
B
§
PL,
55
1
H
1
Q
55
D
O
8
B
00
55
O
O
O
s
1
(A
8"
I
4-J
S
i
a
Waste Man
facilities
.S
1
S
E
•§
ea
bB C
C 03
Chemical manufacture, warehousi
distribution activities
Electrical and electronic products
"c3
"S
1
I
.•§
3
1
I
1
§
s
Ul
Hazardous w;
-"
manufacturing
S,
1
.ts
&
a
2
S2
eS tn
11
Municipal ini
Municipal lai
CA
Ul
Electroplaters and metal fabricate
Foundries
Boat yards
Car washes
«
.a
8
§
•a
istewater an
g sites
Municipal wa
Open burnin,
&
S
Machine and metalworking shops
Manufacturing and distribution si
S
c
Cemetaries
Constructio
S3
-t— »
3
2
g
si
•2 a
•*^ ai
15
•">.
£
1
1 a,
ii
ii
SI
•» I1
w ^
•5 o
II
?
•c
.§
&
5 (chemical ai
Golf course!
bJD
1
CL,
3
(U
e
1
&«
13
>->
-------
TABLE 1-3. METHODOLOGY FOR ASSESSING POINT SOURCES OF
CONTAMINATION (Modified after U.S. EPA, 1989c, Table 1-1, page 1-2)"
Step 1 Identify all existing and potential contaminant sources within and in proximity to
the wellhead protection area.
Step 2 Prioritize contaminant sources for characterization based on the degree of the
threat or the need for controls.
Step 3 Delineate the three-dimensional distribution of existing contamination to
accurately prioritize sources for characterization or remediation.
Step 4 Document how wastes or products are managed within the source to more
accurately predict the spatial and temporal variability of contaminant releases.
Important factors to consider include:
• Total quantity of the waste
• Locations of waste treatment, storage, and disposal
• Transportation of wastes
• Age of source
• Spatial and temporal variability of waste management
Step 5 Document the characteristics of the wastes at the source, considering:
• Waste product interactions
• Chemical compatibility with the containment system material
• Fate and transport within the subsurface
Step 6 Document the condition of the source to evaluate its capacity to contaminate the
ground water. An important item to consider is the need for or adequacy of
containment features.
Step 7 Assess the extent of contamination at the source, considering: location,
horizontal and vertical distributions, plume configuration, and media effects of
air, soil, surface-water, and ground-water quality.
(existing water quality, contaminant, and physical parameters) should be compiled and reviewed to
determine general information limitations and data gaps. The strategy is to maximize information
content; to define monitoring objectives; and to conduct field studies with the least, but still adequate
number of monitoring points, which can later be integrated into the long-term monitoring program
(Pfannkuch, 1982). Phase I generally requires a 3- to 6-month time period for completion.
Phase IT: Research Monitoring Program
Canter et al. (1987) describe research monitoring as "...the development of information on how
the subsurface system operates." The research monitoring phase of the planning process entails data
acquisition and interpretation to improve, or verify, elements of the hydrogeologic conceptual model.
A technically-defensible model ensures a more protective and reliable long-term monitoring program.
1-20
-------
TABLE 1-4. WELLHEAD PROTECTION MONITORING PROGRAM
DEVELOPMENT (Three-Phase Approach)
Phase I: WHPP Elements and Scoping
Tasks
WHPP implementation, roles, and
management framework
Preliminary WHPA delineation:
- Criteria
- Methods
Source identification and prioritization
Informational case compilation, review, and
limitations
Monitoring objectives:
- General objectives: what, where, and how
of data usage
- Specific objectives:
Ambient trend monitoring
Source assessment monitoring
Early-warning detection monitoring
Phase II: Research Monitoring Program
Water quality characterization
- Nature and fate of contaminants in the
subsurface
- General chemical and physical comparison
- Isotope analyses and age dating
- Indicator parameters
Aquifer testing and characterization methods
Refined WHPA delineation
- Criteria
- Methods
Flowpath determination
Phase III: Wellhead Protection
Monitoring Program
Introduction
Monitoring objectives
Optimal monitoring sites
- Location
- Aquifer depth
Site design, construction, and installation
Monitoring parameters and frequencies
Quality assurance and control considerations
Complimentary contingency planning
- Spill-response plan
- Water development site
1-21
-------
MONITORING
PROGRAM
DESIGN
\/
CONDUCT SOURCE
ASSESSMENT WITHIN
WHPA
DETERMINE SAMPLING
FREQUENCIES FOR
MONITORING PARAMETERS
COMPILE & ORGANIZE
EXISTING INFORMATION BASE
ASSESS NEED FOR
NEW MONITORING SITES
IDENTIFY DEFICIENCIES
OF INFORMATION BASE
\/
ESTABLISH NEW
MONITORING SITES
DETERMINE DATA &
PROCESSING NEEDS FOR
SYSTEM CHARACTERIZATION
\/
IMPLEMENT
MONITORING PROGRAM
ESTABLISH
MONITORING OBJECTIVES
REVIEW & INTERPRET
MONITORING RESULTS
DETERMINE
MONITORING PARAMETERS
INCORPORATE INTERPRETATIONS
IN CHARACTERIZATION ASSESSMENTS
IDENTIFY EXISTING MONITORING
SITES BASED ON OBJECTIVES
UPDATE MONITORING OBJECTIVES,
NETWORK DESIGN, & PROGRAM
W ITERATE
IONITORING
PROCESS
Figure 1-10. Flowchart of the 15-step monitoring methodology for wellhead protection areas.
1-22
-------
To determine effective, long-term monitoring strategies, the nature of the contaminant, the
mechanisms of contaminant propagation, and the physical properties of the site must be known and
understood (Pfannkuch, 1982). Complete site characterization and monitoring require extensive
knowledge of the temporal and spatial variability of the physical properties of the site.
Research monitoring for wellhead protection focuses on baseline water quality
characterization, aquifer testing and characterization, refined or verified WHPA delineation, and
ground-water flowpath determination to relate sources to the water supply well or spring. The
product of the research monitoring is a proposed long-term monitoring program that may be partly
implemented in Phase II. Case study research demonstrates that this phase may require 1 to 1-1/2
years for completion, depending on the complexity of the site hydrogeology, as well as the quality and
the co'mprehensiveness of the informational base.
Water Quality Characterization--
Baseline water quality and contaminant parameters need to be established to document
changes in the natural system due to seasonal or land-use impacts, and ultimately, degradation.
Surficial or shallow aquifers are vulnerable, exhibiting rapid and dramatic changes hi quality. Deeper
or confined aquifers exhibit a more constant chemical quality that often reflects geochemical reactions
between the ground water and the aquifer matrix.
In areas where cyclic fluctuations in water quality occur, or documentation of aquifer responses
is required for characterization purposes, it may be necessary to install a network of carefully sited,
nested wells (multiple well configuration with screened intervals at different depths in the aquifer) to
properly monitor discrete sections of the aquifer (Pettyjohn, 1987). To establish a cyclic pattern in
water quality, data may need to be acquired at a daily, weekly, or monthly frequency. Field-deployable,
rapid-turnaround monitoring methods are now available to facilitate frequent monitoring, particularly
ion-mobility probes, immunoassay kits, and a variety of test kits (Eschner et al., 1991). In situ, real-
time monitoring devices, such as fiber optic sensors, are still being developed in the laboratory and are
not readily available for field use. In addition, acquisition and use of these devices is quite costly.
Nature and Fate of Contaminants in the Subsurface-The fate of contaminants that reach the
ground water is partially controlled by the chemical volatility and solubility of the contaminant in
water (Figure 1-11; U.S. EPA, 1989a). Contaminants can be characterized as either water soluble
(miscible), partially soluble, or insoluble (immiscible). Landfill leachates, road salts, acid-mine
drainage, and liquid fertilizers are good examples of water-soluble contaminants, or "mixers."
Hydrocarbons, solvents, and other organic compounds are good examples of liquids that do not
readily dissolve in water and can exist as a separate phase. Generally, separate-phase fluids are
subdivided into two classes: (1) those that are lighter than water, light nonaqueous phase liquids
(LNAPLs or "floaters") and (2) those with a density greater than water, dense nonaqueous phase
liquids (DNAPLs or "sinkers") (Figure 1-11).
The migration of water-soluble pollutants within the saturated zone is primarily dictated by
local hydraulic gradients (Figure 1-11). As the contamination is transported, it will form a plume and
eventually be diluted due to physical processes such as advection, dispersion, retardation, filtration,
time, and distance of travel (Pettyjohn, 1987). Through advection, non-reactive contaminants are
transported and diluted at the average velocity that a water particle moves through the aquifer. There
is also a tendency for non-reactive contaminants to spread out from the path they would be expected
1-23
-------
MOVEMENT OF CONTAMINATION IN GROUND WATER
FLOATING PRODUCT
"FLOATERS"
LAND SURFACE
STORAGE TANK
RESIDUAL PRODUCT
VAPOR
. ENVELOPE
GROUND-WATER
FLOW
DISSOLVED.
PRODUCT
"MIXERS"
LANDFILL
LAND SURFACE
GROUND-WATER
FLOW
CONFINING LAYERXSS
GROUND-WATER
FLOW
DISSOLVED
ORGANICS
IMMISCIBLE
ORGANICS
Figure 1-11. The fate of organic contaminants in the subsurface. (U.S. EPA, 1989a.)
1-24
-------
take, as dictated by advection (Freeze and Cherry, 1979). This phenomenon is known as
hydrodynamic dispersion. It causes the concentration of contaminants to decrease with increasing
distance from the source as a function of velocity and the configuration of aquifer pore space. The
advance of the contaminant plume can be retarded relative to the ground-water velocity and solute
concentrations can be lowered if chemicals are sorbed onto or undergo chemical reactions with the
surrounding matrix. Natural filtration by the aquifer matrix removes suspended particles from the
water, such as iron, manganese, and other precipitates. Finally, in the case of organic compounds,
microbial degradation by organisms (bacteria, algae, fungi, yeasts, and protozoans) within the aquifer
can occur if the system is not overloaded by pollutants and appropriate nutrients are available (U.S.
EPA, 1989a). A good example of this process is the degradation of tetrachloroethylene to more toxic
vinyl chloride (Manahan, 1984).
Contaminants lighter than water (LNAPLs), such as gasoline, heating oil, kerosene, jet fuel,
and aviation gas tend to form a lens on the water table and a three-phase system within the
unsaturated zone: water, product, air and vapors (U.S. EPA, 1989a). Below the floating lens,
dissolved constituents such as benzene, toluene, and xylene are carried away, forming the
contamination plume. Water-table fluctuations can spread the floating contaminant over a greater
thickness of the aquifer (Figure 1-12). In addition, if a source leaks, stops, then leaks again, the
resultant floating lens may be layered with water, dissolved constituents, and the contaminant.
Insoluble contaminants denser than water (DNAPLs), such as chlorinated hydrocarbons
[trichloroethane, carbon tetrachloride, chlorophenols, chlorobenzenes, tetrachloroethylene, and
polychlorinated biphenyls (PCBs)], remain as a separate phase, moving under gravity through the
unsaturated and saturated zones (U.S. EPA, 1989a). DNAPLs can partition into a vapor phase within
the unsaturated zone and settle by gravity within the capillary fringe. Water flowing by can dissolve
the residual DNAPLs or vapors, creating a dissolved constituent plume within the aquifer. If large
amounts of the dense contaminant are released, residuals can penetrate the entire aquifer forming
pools in depressions (U.S. EPA, 1989a). In extreme cases, these residuals can move down the slope of
an impermeable boundary, spreading the contamination in directions that are independent of the
ground-water flow directions and gradients (Figure 1-11). Fine-grained strata can act as effective
barriers to the contaminants. Fractures and karst openings, on the other hand, can act as conduits,
allowing further penetration into deeper aquifers.
The nature, fate, and transport of contaminants should be considered in the design of
monitoring wells, networks, and remediation programs (1) to effectively determine the well position,
screen depth, and interval; (2) to provide accurate assessments of the extent and nature of the
contamination; and (3) to avoid, or minimize, spreading of the contaminant into clean areas of the
aquifer.
General Chemical and Physical Composition—Ground water contains natural chemical
constituents in solution. The type and quantity of these constituents depends upon the geochemical
environment, migration, and source of ground water (Todd et al., 1976). The most common chemical
1-25
-------
PRODUCT SOURCE
INACTIVE
PRODUCT AT
RESIDUAL
TOP OF
CAPILLARY
FRINGE
SATURATION
PRODUCT AT
RESIDUAL-
SATURATION
GROUND-WATER
FLOW
GROUND-WATER
FLOW
Figure 1-12. Organic contaminant spreading within the unsaturated zone because of water-table
fluctuations. (U.S. EPA, 1989a.)
constituents in ground water, in the 0.01 to 1,000 parts per million (ppm) or milligrams per liter
(mg/1) range, are the major cations, major anions, and undissociated constituents listed in Table 1-5
(Davis and DeWiest, 1966).
TABLE 1-5. COMMON CHEMICAL CONSTITUENTS IN GROUND WATER
Cations
Anions
Undissociated
Calcium
Magnesium
Sodium
Potassium
Iron
Carbonate
Bicarbonate
Sulfate
Chloride
Nitrate
Fluoride
Silica
Strontium
Boron
By analyzing and graphing the inorganic chemical composition of ground water,
interpretations can be made concerning the classification, origin, and source of the water, ground-
water flow paths, and interconnection between aquifers. Often these data are used to refute,
corroborate, or refine the hydrogeologic conceptual model of the study area. For visual inspection of
1-26
-------
water-chemistry data, several commonly used graphical methods are suggested such as bar, circle,
radial, or vector graphs, Stiff diagrams, and Piper or trilinear diagrams (Davis and DeWeist, 1966;
Freeze and Cherry, 1979; Hem, 1989).
Other general chemical indicators (parameters that have chemical units of measure) of water
quality are alkalinity (total), biochemical oxygen demand (BOD), chemical oxygen demand (COD),
dissolved oxygen (DO), hardness, pH, total organic carbon (TOC), and total dissolved solids (TDSs).
Physical properties (parameters that have physical units of measure) that often reflect water quality
include: color, odor, redox potential (Eh), specific conductivity, temperature, and total suspended
solids (TSSs).
Data from a complete inorganic chemical analysis can be used to check the accuracy of the
laboratory results by calculating the cation-anion balance (Hem, 1989). The sum of the cations, in
milliequivalents per liter, should equal the sum of the anions within an error range of less than 5 to
10 percent. This check should supplement the laboratory quality assurance plan, which covers the
laboratory's standard operating procedures for equipment use and maintenance, as well as operator
variation. A comparison of determined versus calculated values of TDS is another procedure for
checking analytical accuracy; the values should agree within a few mg/1. An approximate accuracy
check is possible by determining if the TDS value is from 0.55 to 0.75 times the specific conductance.
Finally, comparison of results from the same or similar sources is a good way of identifying errors of
transcription or analysis.
Isotope Analysis and Age Dating—Isotopes can be used to infer much about the age, source,
and movement of ground water. The radioactive isotopes tritium, carbon-14, and chlorine-36, which
are used to measure water age, are of primary interest. The stable isotopes, oxygen-18 and deuterium,
are used to understand the source of water and processes that have affected the water since it entered
the aquifer system (Davis, 1988; Davis and Bentley, 1982; Drever, 1982).
The most common use of tritium analysis is to identify ground water which contains recent
recharge water. The heavy isotope of hydrogen, tritium, has a short half-life of 12.3 years and occurs
in water from both natural and man-made sources. Tritium is produced naturally in the earth's
atmosphere by the interaction of cosmic-ray-produced neutrons with nitrogen (Freeze and
Cherry, 1979). Tritium concentrations in precipitation ranged from 5 to 15 tritium, units (TUs) prior
to the 1950s. As a result of nuclear testing from 1952 to 1963, tritium levels in the atmosphere
increased by three orders of magnitude relative to background levels (Schlosser et al., 1988).
Consequently, the relative abundance of tritium in ground water is used to classify "pre-bomb" versus
"post-bomb" water, to distinguish age zones within the recharge area of the flow system, and to
identify zones of ground-water mixing. As a general rule of interpretation, a TU of less than 3
• suggests water ages greater than 30 years, and a TU of greater than 20 indicates ground water
originating since 1961 (Hendry, 1988; Davis and Bentley, 1982)
Carbon-14 is useful for studying ground water that has been isolated from the atmosphere for
several hundred to thousands of years. Before water is recharged to the aquifer, the carbon content in
the water is balanced with the carbon content in the soil atmosphere. Once the recharge water enters
the ground-water system and is no longer in contact with the atmosphere, the carbon-14 content
decays with time in accordance with its half-life of 5,730 years. This creates a method for age dating
1-27
-------
ground waters up to approximately 50,000 years. Unless analytical results are corrected by known
carbon-13 to carbon-12 ratios within sampled ground water and the host rock, ages determined by the
carbon-14 method are relative and can only be used for comparison (Strait and Moore, 1982).
Chlorine-36 has been used for some ground-water studies (Rosenfeld, 1991). The advantages
of using chlorine-36 are (1) its long half-life of 300,000 years and (2) its tendency to travel with the
water and not to be significantly affected by sorption or mineralization reactions. Complications
associated with chlorine-36 age dating of ground water include the following:
• Knowing the original concentrations of chlorine-36 in the recharge water
• Estimating subsurface production of chlorine-36 and chloride contributions from
mineral dissolution and ground-water mixing
• Correcting for anthropogenic sources due to nuclear testing
Clorine-36 has also been used as a tracer of recent recharge water, similar to tritium, due to
production by nuclear tests (Bentley et al, 1986; Davis, 1988).
The concentrations of oxygen-18 and deuterium in ground water is established during
atmospheric precipitation as a result of.evaporation, condensation, freezing, melting, chemical
reactions, or biological processes. It is also highly dependent on temperature. Once water moves
beyond the upper part of the soil zone, oxygen-18 and deuterium concentrations become a
characteristic property of the ground water, enabling the source and mixing patterns of the ground
water to be determined (Freeze and Cherry, 1979). Such interpretations are based on departures of
isotope concentrations from relationships based on global precipitation surveys (Damsgaard, 1964).
Indicator Parameters—Indicator parameters have been the subject of much research and debate
for hazardous waste site characterization. Research conducted by Plumb (1992, 1991, 1987, 1985) and
Rosenfeld (1990a) lends strong evidence to the selection of volatile organic compounds (VOCs) as
standard screening parameters in proximity to hazardous waste sites. Plumb and Pitchford (1985)
documented that VOCs are the single most abundant class of organic contaminants in disposal-site
ground water. A VOC scanning approach is recommended for moderate- to high-risk industrial and
commercial sources in proximity to WHPAs. The VOC scan results would be used to establish the
need for more comprehensive organic analyses for less common compounds such as semi-volatiles,
pesticides, and PCBs. There are a limited number of situations where VOC monitoring may not be
applicable:
• Incinerated municipal waste ash landfills, where an inorganic monitoring strategy is
recommended (Plumb, 1990)
• Wood preservation and creosoting operations, where a semi-volatile monitoring
strategy is recommended (Rosenfeld and Plumb, 1991)
• Single, well-defined waste disposal facilities, where customized monitoring parameters
may be determined due to the presence of consistent sets of detected compounds
(Rosenfeld, 1990a)
In an industry-specific survey of ground-water contaminants, Rosenfeld (1990b) found that at
aerospace, metal manufacturing, and military facilities the predominant contaminants were VOCs. At
chemical and oil-refining facilities, the predominant organic type varied from site to site. Rosenfeld
1-28
-------
concluded that it may not be possible to utilize these industry-specific monitoring parameters in
regional investigations due to the overlap of compound distributions at different industries.
In summary, Table 1-6 lists recommended monitoring parameters for the five categories of
sources typically found in proximity to WHPAs (Table 1-2).
TABLE 1-6. MONITORING PARAMETERS FOR CATEGORIES OF SOURCES IN
PROXIMITY TO WELLHEAD PROTECTION AREAS
Source Category
Monitoring Parameters
Agricultural
Light, Commercial and Industrial Facilities
Residential
Waste Management
Naturally Occurring
General water quality*
Nutrients and microbiological1"
Pesticides*
VOC scan (optional)5
General water quality
Heavy metals§
VOC scan
Gross alpha and beta (optional)8
General water quality
Nutrients and microbiological
Heavy metals (optional)
VOC scan (optional)
General water quality
VOC scan/priority pollutant scan (optional)8
Heavy metals
Nutrients and microbiological (optional)
Gross alpha and beta (optional)
General water quality
Uranium and radon (optional)
Heavy metals (optional)
General water quality parameters include alkalinity (total), biochemical oxygen demand (BOD),
chemical oxygen demand (COD), color, dissolved oxygen (DO), hardness, major cations and
anions, odor, pH, redox potential (Eh), specific conductivity, temperature, total organic carbon
(TOC), total dissolved solids (TDSs), and total suspended solids (TSSs).
Nutrients and microbiological parameters include ammonia, bacteria, nitrate, nitrite, total nitrogen,
and viruses.
Selection of pesticides for sampling and analysis is application- and site-specific. Refer to Munch
et al. (1990) for a brief discussion of pesticide parameters and methods of analysis.
Refer to the following for a complete list of parameters and methods of analysis:
- VOCs: U.S. EPA, 1991d; 1990c; 1990d
- Heavy Metals: U.S. EPA, 1991e; 1990c
- Gross alpha and beta: U.S. EPA, 1991e; 1990c
1-29
-------
General water quality parameters include both the chemical and physical indicators discussed
in the General Chemical and Physical Composition section. Nutrients and microbiological parameters
include ammonia, bacteria, nitrate, nitrite, total nitrogen, and viruses. U.S. EPA references 1991d and
1990c provide the reader with a detailed listing of additional nutrient and microbiological parameters
for monitoring and associated methods of analysis. Pesticide parameter information is referenced in
Munch et al, 1990. VOC scan and parameter information is referenced in U.S. EPA (1991d, 1990c,
1990d). The heavy metal parameters consist of the inorganics excluded from the general water quality
and nutrients groups such as aluminum, arsenic, asbestos, barium, cadmium, chromium, lead, mercury,
selenium, silver, and so on (U.S. EPA, 1991e; 1990c). The selection of parameters from the heavy
metals group for monitoring is, again, use- and site-specific. Parameters and methods of analysis for
the priority pollutant and gross alpha and beta groups are referenced in U.S. EPA (1991d and 1990d)
and U.S. EPA (1991e and 1990c), respectively.
Aquifer Testing and Characterization Methods—
Often, review of the existing informational base identifies gaps in hydrogeologic knowledge
and data that must be addressed in order to design the monitoring program. To fill these gaps, field
investigations are necessary. These include well and piezometer installation, ground-water sampling,
screening samples for water quality, aquifer testing, areal water-level monitoring, structural and
geologic mapping, fracture trace analysis, surface and borehole geophysical studies, and dye tracing.
Hydrologic investigations generally include well or piezometer installation so that aquifer testing and
ground-water sampling can be performed. Aquifer testing involves analyzing the change in water
levels (hydraulic or piezometric head data), with respect to time, by ground-water withdrawal from
wells. From these field investigations, conducted in amenable hydrogeologic regimes (primarily
Darcian-flow conditions), the following site-specific information can be obtained (Pettyjohn, 1987):
Position and thickness of aquifer or confining units
Aquifer and confining unit properties such as transmissivity, storativiry, and porosity
Position and nature of aquifer boundaries
Location and amounts of ground-water withdrawal
Location, type, and quantity of water-quality or contaminant constituents
The reader is referred to other publications, which discuss issues in detail beyond the scope of
this section. These publications address well design and installation (Aller et al., 1989; ASTM, In
publication; Campbell and Lehr, 1973; Driscoll, 1986; Nielsen, 1991, U.S. EPA, 1986; U.S. EPA, 1977;
U.S. EPA, 1975); ground-water sampling (Barcelona et al., 1989; Barcelona et al., 1985; Canter et al.,
1987; Garrett, 1988; Nelson and Ward, 1981; Scalf et al., 1981; U.S. EPA, 1987b; U.S. EPA, 1986): as
well as aquifer testing and analysis (Davis and DeWiest, 1966; Fetter, 1980; Ferris et al., 1962; Freeze
and Cherry, 1979; Lohman, 1979; Kruseman and de Ridder, 1990; Stallman, 1971; Walton, 1962).
Characterizing aquifer properties and predicting the movement of contaminants in the
subsurface is a more difficult undertaking in anisotropic, heterogeneous media, such as stratified drift,
fracture-dominated, or karst terrains. The single most important parameter to accurately determine
for characterization purposes is hydraulic conductivity, in both the lateral and vertical directions
(Moltz et al., 1990). To this end, Moltz et al. (1990) describe and recommend several techniques to
quantify variations in vertical hydraulic conductivity, including tracer, flowmeter, dilution, and multi-
1-30
-------
level slug tests.
Complementary techniques that describe either borehole, secondary-porosity features
(fractures, joints, and bedding planes) or borehole flow and direction include: the standard suite of
borehole geophysical logs (Benson et al, 1983; Keys and MacCary, 1971); wellbore fluid-replacement
logging (GZA GeoEnvironmental, Inc., 1991; Pedler et al., 1990; Pedler et al., 1989; Tsang et al., 1990,
Vernon et al., 1993); acoustic televiewer logging (Vernon et al., 1993; Paillet et al., 1987); and dye
tracing (Quinlan, 1989; Mull et al., 1988). To extrapolate and integrate borehole flow information at a
more local or regional scale, additional techniques are recommended, such as: fracture-trace analysis
(Lattman and Parizek, 1964; Ogiela and Moore, 1991; Parizek et al., 1990; Parizek, 1976), geologic
and structure-control mapping, and surface geophysics (micro-gravity, electrical, magnetic, and seismic
methods to locate subsurface fractures and faults). In karst regimes where conduit flow (flow through
large openings and caves) occurs, fracture-trace analysis, dye tracing, and select geophysical methods
are particularly helpful in defining the areal ground-water flow system.
Hydrogeologic characterization serves as the technical basis for refined WHPA delineation,
flowpath identification, and ultimately, improved monitoring network and program design. It is
important to remember that the characterization effort should not become an obstacle to completing
the monitoring program (Todd et al., 1976). During the research monitoring phase, the informational
base is incomplete, relative to the size of the resultant data base. Monitoring should be viewed as an
iterative process, where data are gathered to continually augment and improve the existing
informational base.
Refined Wellhead Protection Area Delineation— J
Following detailed hydrogeologic characterization, existing delineation criteria and methods for
the preliminary WHPA should be reviewed and improved (U.S. EPA, 1987a). Generally, increasing
knowledge of the hydrogeologic situation enables a more quantitative and sophisticated delineation
method to be employed, resulting in a more accurate WHPA, and often, a reduction in area. A
reduction in the WHPA usually reduces management and regulatory oversight, yielding labor, time,
and cost savings.
Flowpath Determination—
. With more comprehensive coverage and understanding of the ground-water flow system, a
flownet of the WHPA can be constructed using analytical (Freeze and Cherry, 1979; Fetter, 1980) or
numerical (Anderson and Woessner, 1992; Water Science and Technology Board et al., 1990; Van der
Heidje and Beljin, 1988) methods. A flownet is useful in selecting existing monitoring points or siting
new ones to intersect flowpaths from known or potential contaminant sources. In addition, a more
effective screen depth and interval can be determined by combining flownet interpretations with a
knowledge of the fate and transport of the contaminant type (water-soluble versus insoluble
contaminants) in the subsurface.
In non-Darcian flow conditions, such as fractured-bedrock or karst regimes, it may be
beneficial to simulate contaminant pathways and travel times by dye tracing. Monitoring for
pollutants in discrete-flow or conduit-flow systems is most reliably done at locations (such as wells,
springs, caves, and streams) shown by dye tracing to be connected to sources, possibly under a variety
of flow conditions (Quinlan, 1989). Conversely, ambient trend monitoring is best done in areas
determined to be geochemically similar to the target area and where dye traces confirm no hydrologic
1-31
-------
connections.
In conclusion, an accurate, optimally-sized WHPA in conjunction with flowpath assessment,
improves the long-term monitoring program in the following ways:
• Effective selection and location of monitoring sites
• Effective selection of monitoring well screen depths
• More reliable monitoring data base
• Cost savings associated with a minimum number of well installations and ground-
water samples
• Overall, more protective monitoring program
Phase 171: Wellhead Protection Monitoring Program
The by-product of Phases I and II is a proposed wellhead protection monitoring program.
Generally, the program is drafted as a plan to be implemented in stages, as labor and financial
resources become available. If the technical groundwork for the monitoring program is done
correctly, the plan can be drafted in two weeks to a month. A brief discussion of each program
element (Table 1-4) is given as follows.
Introduction--
The plan should include a preface summarizing a brief history of the local WHPP inception
and development. The purpose of the monitoring program should be integrated with respect to the
management structure, designating the roles and duties of personnel responsible for implementing the
monitoring program. A map depicting the location of prioritized sources should be included,
accompanied by a list detailing the type of the source and suspected or known contaminants. A map
depicting the WHPA (and protective zones, if applicable), as well as a description of the delineation
criteria and method(s) should be given as background information to clarify objectives and the
selection of monitoring sites.
Monitoring Objectives—
The plan should include general and specific objectives for ambient trend, source assessment,
and early-warning detection monitoring. Each objective should be justified in terms of monitoring
sites, parameters, frequencies, and quality assurance.
Monitoring Sites—
The plan should include a map depicting the location of existing and recommended
monitoring sites for inclusion in the network. A formal identification system can be adopted to label
monitoring sites, which may include: wells, piezometers, boreholes, springs, seeps, caves, excavations,
sinkholes, and streams that can be shown to be hydraulically connected to the ground water. A
minimum set of data elements should be used to identify the site, such as geographic location, political
regime, source identifiers, and individual site characteristics (U.S. EPA, 1991c). Guidance and criteria
for monitoring site selection based on specific objectives are given through example in Chapters 2
through 5 and in Appendix 1-A.
Site Design, Construction, and Installation—
The integrity of the design and construction of existing sites (wells, for example) should be
1-32
-------
considered for inclusion in the monitoring network, to ensure that representative data can be obtained,
such as water levels and water quality. New sites that require installation should be described in detail.
Descriptions should include information on completion depth, open or screened interval, schematic
design, and construction materials, as well as the method of installation, development, and testing. All
monitoring site information should be stored in an automated data base for convenient and safe
storage, update, and retrieval.
Monitoring Parameters and Frequencies-
Physical and chemical parameters to be monitored at select frequencies should be listed and
technically justified. Specifically, monthly monitoring of water levels may be justified by cyclic water-
table fluctuations. Similarly, quarterly monitoring of indicator parameters may be justified by
compliance regulations. More frequent monitoring (biweekly) may be justified for statistical reasons.
Quality Assurance and Control Considerations--
Quality assurance refers to a system for ensuring that all information, data, and resulting
decisions compiled under specific ground-water monitoring tasks are technically sound, statistically
valid, and properly documented (Canter et al., 1987). Quality control provides the mechanisms to
attain the quality assurance goals. Quality assurance and quality control apply to laboratory
measurement data and field operations, such as monitoring site construction, sampling and
decontamination procedures, sample custody, and equipment calibration. In general, quality assurance
goals are driven by project objectives such as client concerns, compliance monitoring, court
admissibility of data, statistical significance, and experimental design. Each monitoring program
should formulate a minimum set of quality assurance and control objectives to match the objectives of
the wellhead protection monitoring program. At a minimum, this should include laboratory standard
operating procedures and a field operations plan.
Complimentary Contingency Planning—
In hydrogeologic regimes characterized by rapid ground-water velocities (.25 to 0.5 miles per
hour) and travel times (hours to days in some fractured-bedrock and karst aquifers), early-warning
detection monitoring is impractical and unachievable. Contaminants pass through the system
essentially unaltered. In situations such as this, a spill-response plan (Quinlan, 1986) is recommended
to eliminate or reduce the amount of contamination entering highly-vulnerable aquifers directly from
the surface. Situational monitoring is detailed by Pfannkuch (1982) in response to accidental or
unanticipated spills. Finally, contingency planning requires that a new water-development area be sited
in the event that the existing water supply is rapidly diminished or contaminated.
STEP-WISE APPROACH FOR THE DESIGN OF A WELLHEAD PROTECTION
MONITORING PROGRAM
In conclusion, a strategy for the design of a monitoring program for wellhead and springhead
protection is presented in step-wise format (Figure 1-10). Used as a guide, this strategy may require
modification to accommodate the spectrum of flow systems in many aquifers: diffuse- to conduit-flow.
It synthesizes several excellent technical approaches for ground-water monitoring program design
(Canter et al., 1987; Pfannkuch, 1982; Quinlan, 1989; and Todd et al., 1986).
1-33
-------
(1) Identify, prioritize, and characterize all existing and potential contaminant sources
within and in proximity to the wellhead or springhead protection area.
(2) Compile and organize all existing baseline data and ancillary information.
(3) Identify deficiencies in the existing informational base.
(4) Identify additional data acquisition and processing needs to enable system
characterization.
(5) Establish monitoring objectives to improve the informational base and fill the data
gaps.
(6) Determine physical, chemical, and biological parameters to be monitored.
(7) Identify existing sites for background, pollutant, and water-quality monitoring to meet
specific objectives.
(8) Determine optimal sampling frequencies for monitoring parameters considering
spatial, temporal, and quality assurance and control considerations.
(9) Assess the need for new monitoring points.
(10) Establish new monitoring sites.
(11) Implement the monitoring program.
(12) Review and interpret monitoring results.
(13) Integrate monitoring results to revise source, water quality, hydrogeology, and
flowpath transport assessments.
(14) Modify monitoring objectives, network design, and the program approach, .
appropriately.
(15) Iterate the monitoring process.
1-34
-------
REFERENCES
Aller, L., T. W. Bennett, G. Hackett, R. J. Petty, J. H. Lehr, H. Sedoris, D. M. Nielsen, and J. E.
Denne. 1989. Handbook for the Design and Installation of Ground-Water Monitoring Wells. EPA-
600/4-89/034. U.S. Environmental Protection Agency, Environmental Monitoring Systems
Laboratory, Las Vegas, Nevada. 398 pp.
American Society for Testing and Materials. In publication. Proposed Recommended Practice for
Design and Installation of Ground-Water Monitoring Wells in Aquifers. American Society for
Testing and Materials, Philadelphia, Pennsylvania. 15 pp.
Anderson, M. P. and W. W. Woessner. 1992. Applied Groundwater Modeling: Simulation of Flow
and Advective Transport. Academy Press Inc., San Diego, California. 381 pp.
Barcelona, M. J., J. P. Gibb, J. A Helfrich, and E. E. Garske. 1985. Practical Guide for Ground-
Water Sampling. EPA/600/2-85/104. U.S. Environmental Protection Agency, R. S. Kerr
Environmental Research Laboratory, Ada, Oklahoma. 169 pp.
Barcelona, M. J., H. A. Wehrmann, M. R. Schock, M. E. Sievers, and J. R. Karny. 1989. Sampling
Frequency for Ground-Water Quality Monitoring. U.S. Environmental Protection Agency,
Environmental Monitoring Systems Laboratory, Las Vegas, Nevada. 191 pp.
Bentley, H., F. Phillips, and S. Davis. 1986. Clorine-36 in the Terrestrial Environment. In Handbook
of Environmental Isotope Geochemistry. Fritz, P. and J. Fontes (Editors). The Terrestial
Environment, Volume 2. pp. 427-480.
Benson, R. C, R. A. Glaccum, and M. R. Nwoel. 1983. Geophysical Techniques for Sensing Buried
Wastes and Waste Migration. U.S. Environmental Protection Agency, Environmental Monitoring
Systems Laboratory, Las Vegas, Nevada.
Bureau of Economic Geology, University of Texas-Austin. 1991. Wellhead Prtoection Strategies for
Confined-Aquifer Settings. EPA 570/9-91-008. U.S. Environmental Prtoection Agency, Office of
Ground Water and Drinking Water, Washington, D.C. 168 pp.
Campbell, M. D. and J. H. Lehr. 1973. Water Well Technology. McGraw Hill, New York, New
York. 681 pp.
Canter L. W., R. C. Knox, and D. M. Fairchild. .1987. Ground Water Quality Protection. Lewis
Publishers, Inc., Chelsea, Michigan. 562 pp.
Damsgaard, W. 1964. Stable Isotopes in Precipitation. Tellus, 16. pp. 436-468.
1-35
-------
Davis, S. 1988. Natural Radionuclides in Ground Water. Journal of Geological Education 36: 24-29.
Davis, S. and H. Bentley. 1982. Daring Ground Water - A Short Review. In: Nuclear and Chemical
Dating Techniques. Currie, L. (Editor). American Chemical Society Symposium, Volume 176.
pp. 187-222.
Davis, S. N. and R. J. M. DeWiest. 1966. Hydrogeology. Wiley, New York. 463 pp.
Domenico, P.A. 1972. Concepts and Models in Groundwater Hydrology. McGraw-Hill Book
Company, New York, New York. 405 pp.
Drever, J. L 1982. The Geochemistry of Natural Waters. Prentice-Hall, Inc., Englewood Cliffc, New
Jersey. 388 pp.
Driscoll, F. G. 1986. Ground Water and Wells. Johnson Division, Saint Paul, Minnesota. 1089 pp.
Eschner, E. 1991. Monitoring Technologies for Wellhead Protection. (Editor and Co-Author).
Internal report. U.S. Environmental Protection Agency, Environmental Monitoring Systems
Laboratory, Las Vegas, Nevada.
Ferris, J. G., D. K Knowles, R. H. Brown, and R. W. Stallman. 1962. Theory of Aquifer Tests,
Ground-Water Hydraulics. U.S. Geological Water-Supply Paper 1536-E. U.S. Government Printing
Office, Washington, D.C pp. 69-174.
Fetter, C W. Jr. 1980. Applied Hydrogeology. Charles E. Merrill Publishing Company, Inc.,
Columbus, Ohio. 714 pp.
Freeze, R. A. and J. A. Cherry. 1979. Groundwater. Prentice-Hall, Inc., Englewood Cliffe, New
Jersey. 605 pp.
GZA GeoEnvironmental, Inc. 1991. Hydrophysical Logging of Wellbores BW-1, BW-3, BW-6, and
BW-7, Blackwater Brook Site, Dover, New Hampshire. GZA GeoEnvironmental, Inc., Newton
Upper Falls, Massachusetts. 31 pp.
Garrett, P. 1988. How to Sample Ground Water and Soils. National Water Well Association,
Dublin, Ohio. 37pp.
Griswold, W. J., J. H. Vernon, and B. A. Moore. 1990. Developing a Wellhead Protection Area and
Monitoring Strategy For a Fractured Bedrock Aquifer, Dover, New Hampshire. In: Ground Water
Management Journal, 3: 131-145, Proceedings of the Eastern Regional Ground Water Issues
Conference, Springfield, Massachusetts. National Water Well Association, Dublin, Ohio.
Hem, J. D. 1989. Study and Interpretation of the Chemical Characteristics of Natural Water. U.S.
Geological Survey Water-Supply Paper 2253, Edition 3. U.S. Government Printing Office,
Washington, D.C 363pp.
Hendry, M. 1988. Do Isotopes Have a Place in Ground-Water Studies? Ground Water 26: 410-415.
1-36
-------
Keys, W. S. and L. M. MacCary. 1971. Applications of Borehole Geophysics to Water-Resources
Investigations. Techniques of Water-Resources Investigations of the United States Geological Survey,
Book 2, Chapter El. U.S. Government Printing Office, Washington, D.C 126 pp.
Kruseman, G. P. and N. A. de Ridder. 1990. Analysis and Evaluation of Pumping Test Data.
Edition 2. International Institute for Land Reclamation and Improvement, Wageningen, The
Netherlands.
Lattman, L. H. and R. R. Parizek. 1964. Relationship Between Fracture Traces and the Occurrence
of Ground-Water In Carbonate Rocks. Journal Hydrology 2: 73-91.
Lohman, S. W. 1979. Ground-Water Hydraulics. U.S. Geological Survey Professional Paper 708.
U.S. Government Printing Office, Washington, D.C 70 pp.
Manahan, S. E. 1984. Environmental Chemistry. Edition 4. Brooks/Cole Publishing Company,
Monterey, California. 612 pp.
Moltz, F. J., O. Guven, J. G. Melville, I. Javandel, A. E. Hess, and F. L. Paillet. 1990. A New
Approach and Methodologies for Characterizing the Hydrogeologic Properties of Aquifers.
EPA/600/2-90/002. U.S. Environmental Protection Agency, R. S. Kerr Environmental Research
Laboratory, Ada, Oklahoma.
205 pp.
Moore, B. A., J. T. Witherspoon, L. L. Bullard, T. J. Aley, and J. K Rosenfeld. 1990. Strategy for
Delineation and Detection Monitoring of the Fulbright Springhead Protection Area, Springfield,
Missouri. In: Ground Water Management 1: 447-461, Proceedings of the 1990 Cluster of
Conferences, Kansas City, Missouri. National Water Well Association, Dublin, Ohio.
Mull, D. S., T. D. Liebermann, J. L. Smoot, and L. H. Woosley, Jr. 1988. Applications of Dye-
Tracing Techniques for Determining Solute-Transport Characteristics of Ground Water in Karst
Terranes. EPA 904/6-88-001. U.S. Environmental Protection Agency Region 4, Atlanta, Georgia.
103 pp.
Munch, D. J., R. L. Graves, R. A. Maxey, and T. M. Engel. 1990. Methods Development and
Implementation for the National Pesticide Survey. Environmental Sciences and Technology 24(10):
1446-1451.
Nelson, J. D. and R. C Ward. 1981. Statistical Considerations and Sampling Techniques for Ground-
Water Quality Monitoring. Ground Water 19(6): 617-625.
Nielsen, D. M. 1991. Practical Handbook of Ground-Water Monitoring. (Editor). Lewis Publishers,
Inc., Chelsea, Michigan. 717 pp.
Ogiela, L. G. and B. A. Moore. 1991. Fracture Trace Analysis for Wellhead Protection Area
Delineation, East Lancaster County, Pennsylvania. Remote Sensing Report. U.S. Environmental
Protection Agency, Environmental Monitoring Systems Laboratory, Las Vegas, Nevada. 73 pp.
1-37
-------
Paillet, F. L., A. E. Hess, and E. Hardin. 1987. Characterization of Fracture Permeability with High-
Resolution Vertical Flow Measurements During Borehole Pumping. Ground Water 25(1): 28-40.
Parizek, R. R., P. M. Lavin, R. J. Greenfield, R. A. Weiss, C. A. Shuman, and M. L. Moran. 1990.
Nature and Hydrologic Significance of Fracture Traces, Lineaments, and Fracture Zones Related to
Ground-Water Monitoring. EPA CR813660-01-0. Internal report. U.S. Environmental Protection
Agency, Environmental Monitoring Systems Laboratory, Las Vegas, Nevada. 165 pp.
Parizek, R. R. 1976. Lineaments and Groundwater. In: Interdisciplinary Applications and
Interpretations of EREP Data Within the Susquehanna River Basin. ORSER-SSEL Technical
Report 2-76. NASA, Johnson Space Center, Houston, Texas. 90 pp.
Pedler, W. H., J. J. Barvenik, C F. Tsang, and F. V. Hale. 1990. Determination of Bedrock Hydraulic
Conductivity and Hydrochemistry Using a Wellbore Fluid Logging Method. In: Proceedings of the
Fourth National Outdoor Action Conference on Aquifer Restoration, Groundwater Monitoring and
Geophysical Methods. National Water Well Association, Dublin, Ohio. 10 pp.
Pedler, W. H., J. J: Barvenik, G. W. Gardner, and D. W. Urish. 1989. Characterization of
Hydraulically Connective Fractures by Geophysical Logging After Fluid Emplacement.
In: Proceedings of the Second Annual Hazardous Waste Materials Management Conference.
Rosemont, Illinois, pp. 121-129.
Pettyjohn, W. A. 1987. Protection of Public Water Supplies from Ground-Water Contamination.
Noyes Data Corporation, Park Ridge, New Jersey. 177 pp.
Pfannkuch, H. 1982. Problems of Monitoring Network Design to Detect Unanticipated
Contamination. Ground Water Monitoring Review 2(1): 67-76.
Plumb, R. H., Jr. 1992. The Importance of Volatile Organic Compounds as a Disposal Site
Monitoring Parameter. Chapter 7. In: Groundwater Contamination and Analysis at Hazardous
Waste Sites. Lesage, S. and R. E. Jackson (Editors). Marcel Dekker, Inc., New York, New York.
pp. 173-197.
Plumb, R. H., Jr. 1991. The Occurrence of Organic Appendix IX Constituents in Disposal Site
Ground Water. Ground Water Monitoring Review 11(2): 157-164.
Plumb, R. H., Jr. 1990. Development of a Ground-Water Monitoring Strategy for Incinerated
Municipal Waste Landfills. EPA/600/X-90/110. Internal report. U.S. Environmental Protection
Agency, Environmental Monitoring Systems Laboratory, Las Vegas, Nevada.
Plumb, R. H., Jr. 1987. A Practical Alternative to the RCRA Organic Indicator Parameters.
In: Proceedings of HAZMACON 87, Hazardous Materials Management Conference, Santa Clara,
California. Association of Bay Area Governments, Oakland, California, pp. 135-150.
1-38
-------
Plumb, R. H., Jr. 1985. Disposal Site Monitoring Data: Observations and Strategy Implications.
In: Proceedings of the Second Canadian/American Conference on Hydrogeology, Hazardous Wastes
in Ground Water: A Soluble Dilemma, Banff, Alberta, Canada. National Water Well Association,
Dublin, Ohio. pp. 69-77.
Plumb, R. H. Jr. and A. M. Pitchford. 1985. Volatile Organic Scans: Implications for Groundwater
Monitoring. In: Proceedings of the Conference on Petroleum Hydrocarbons and Organic Chemicals
in Ground Water - Prevention, Detection, and Restoration, Houston, Texas. National Water Well
Association, Dublin, Ohio. pp. 207-222.
Quinlan, J. F. 1989. Ground-Water Monitoring In Karst Terranes: Recommended Protocols and
Implicit Assumptions. Internal report. U.S. Environmental Protection Agency, Environmental
Monitoring Systems Laboratory, Las Vegas, Nevada. 79 pp.
Quinlan, J. F. 1986. Recommended Procedure for Evaluating the Effects of Spills of Hazardous
Materials on Ground Water Quality in Karst Terranes. In: Proceedings of the Environmental
Problems in Karst Terranes and Their Solutions Conference, Bowling Greene, Kentucky. National
Water Well Association, Dublin, Ohio. pp. 183-195.
Rosenfeld, J. K. 1991. Applications of Isotope Geochemistry for Wellhead Prtoection Area
Delineation. Unpublished report. Lockheed Engineering and Sciences Company, Las Vegas, Nevada.
10pp.
Rosenfeld, J. K 1990a. Ground-Water Contamination at Hazardous Waste Disposal Facilities.
In: Ground Water Management 1: 237-250, Proceedings of the 1990 Cluster of Conferences, Kansas
City, Missouri. National Water Well Association, Dublin, Ohio.
Rosenfeld, J. K 1990b. Industry-Specific Ground-Water Contamination. In: Proceedings of the
Conference on Minimizing Risk to the Hydrologic Environment, Las Vegas, Nevada. American
Institute of Hydrology, Minneapolis, Minnesota, pp. 93-111.
Rosenfeld, J. K and R. H. Plumb Jr. 1991. Ground-Water Contaminants At Wood Treatment .
Facilities. Ground Water Monitoring Review 11(1): 133-140.
Scalf, M. R., J. F. McNabb, W. J. Dunlap, R. L. Cosby, and J. S. Fryberger. 1981. Manual of Ground-
Water Quality Sampling Procedures. National Water Well Association, Dublin, Ohio. 93 pp.
Schlosser, P., M. Stute, H. Dorr, C. Sonntag, and K Munnich. 1988. Tritium/3He Dating of Shallow
Ground Water. Earth Planetary Letters 89: 353-362.
Stallman, R. W. 1971. Aquifer-Test Design, Observations, and Data Analysis. U.S. Geological
Survey Techniques of Water-Resources Investigations, Book 3, Chapter Bl. U.S. Government
Printing Office, Washington, D.C. 26 pp.
Strait, S. R. and B. A. Moore. 1982. Geohydrology of the Rattlesnake Ridge Interbed in the Gable
Mountain Pond Area. Rockwell Hanford Operations, Richland, Washington. 73 pp.
1-39
-------
Todd, D. K, R. M. Tinlin, K. D. Schmidt, and L. D. Everett. 1976. Monitoring Groundwater Quality:
Monitoring Methodology. EPA-600/4-76-026. U.S. Environmental Protection Agency,
Environmental Monitoring Systems Laboratory, Las Vegas, Nevada. 154 pp.
Tsang, C R, P. Hufschmied, and F. V. Hale. 1990. Determination of Fracture Inflow Parameters
With a Borehole Fluid Conductivity Logging Method. Water Resources Research 26(4): 561-578.
U.S. Environmental Protection Agency. 1991a. Managing Ground Water Contamination Sources in
Wellhead Protection Areas: A Priority Setting Approach. U.S. Environmental Protection Agency,
Office of Ground Water and Drinking Water, Washington, D.C 252 pp.
U.S. Environmental Protection Agency. 1991b. A Guide for Conducting Contamination Source
Inventories for Public Drinking Water Supply Protection Programs. 570/9-91-014. U.S.
Environmental Protection Agency, Office of Ground Water and Drinking Water, Washington, D.C.
53 pp.
U.S. Environmental Protection Agency. 1991c. Definitions for the Minimum Set of Data Elements
for Ground-Water Quality. Final Draft. U.S. Environmental Protection Agency, Office of Ground
Water and Drinking Water, Washington, D.G 57 pp.
U.S. Environmental Protection Agency. 1991d. USEPA Contract Laboratory Program, Statement of
Work for Organics Analysis, Multi-Media, Multi-Concentration. Document Number OLMO1.0 and
Revisions.
U.S. Environmental Protection Agency. 1991e. USEPA Contract Laboratory Program, Statement of
Work for Inorganic Analysis, Multi-Media, Multi-Concentration. Document Number ILMO2.1 and
Revisions. U.S. Environmental Protection Agency, Office of Emergency and Remedial Responses,
Washington, D.G
U.S. Environmental Protection Agency. 1990a. Guide to Ground-Water Supply Contingency
Planning for Local and State Governments. EPA-440/6-90-003. U.S. Environmental Protection
Agency, Office of Ground-Water Protection, Washington, D.G 83 pp.
U.S. Environmental Protection Agency. 1990b. A Review of Sources of Ground-Water
Contamination from Light Industry. EPA-440/6-90-005. U.S. Environmental Protection Agency,
Office of Ground-Water Protection, Washington, D.C. 83 pp.
U.S. Environmental Protection Agency.. 1990c. Fact Sheet: Drinking Water Regulations under the
Safe Drinking Water Act. U.S. Environmental Protection Agency, Office of Drinking Water,
Washington, D.G 83pp.
U.S. Environmental Protection Agency. 1990d. USEPA Contract Laboratory Program, Statement of
Work for Low Concentration Water for Volatile Organics Analysis. Document Number OLVO1.0
and Revisions. U.S. Environmental Protection Agency, Office of Emergency and Remedial
Responses, Washington, D.G
1-40
-------
U.S. Environmental Protection Agency. 1989a. Transport and Fate of Contaminants in the
Subsurface, Seminar Publication. EPA/625/4-89/019. Center for Environmental Research Information
and R. S. Kerr Environmental Research Laboratory, Ada, Oklahoma. 50 pp.
U.S. Environmental Protection Agency. 1989b. Wellhead Protection Programs: Tools for Local
Governments. EPA-440/6-89-002. U.S. Environmental Protection Agency, Office of Ground-Water
Protection, Washington, D.C 50 pp.
U.S. Environmental Protection Agency. 1989c. Practical Guide for Assessing and Remediating
Contaminated Ground Water. Unpublished draft. U.S. Environmental Protection Agency, Office of
Solid Waste, Washington, D.G 88 pp.
U.S. Environmental Protection Agency. 1987a. Guidelines for Delineation of Wellhead Protection
Areas. EPA-440/6-87-010. U.S. Environmental Protection Agency, Office of Ground-Water
Protection, Washington, D.C.
U.S. Environmental Protection Agency. 1987b. Bibliography of Ground-Water Sampling Methods.
EPA/600/X-87/325. U.S. Environmental Protection Agency, Environmental Monitoring Systems
Laboratory, Las Vegas, Nevada. 43 pp.
U.S. Environmental Protection Agency. 1986. RCRA Ground-Water Monitoring Technical
Enforcement Guidance Document. OSWER-9950.1. U.S. Environmental Protection Agency, Office
of Waste Programs Enforcement, Washington, D.C. 208 pp.
U.S. Environmental Protection Agency. 1977. Procedures Manual for Ground Water Monitoring at
Solid Waste Disposal Facilities. EPA/530/SW-611. U.S. Environmental Protection Agency, Office of
Solid Waste, Washington, D.C. 269 pp.
U.S. Environmental Protection Agency. 1975. Manual of Water Well Construction Practices. EPA-
570/9-75-001. U.S. Environmental Protection Agency, Office of Water Supply, Washington, D.C.
156 pp.
Van der Heijde, P. and M. S. Beljin. 1988. Model Assessment For Delineating Wellhead Protection
Areas. EPA-440/6-88-002. U.S. Environmental Protection Agency, Office of Ground-Water
Protection, Washington, D.C. 33 pp.
Vernon, J. H., F. L. Paillet, W. H. Pedler, and W. J. Griswold. 1993. Application of Borehole
Geophysics in Defining a Wellhead Protection Area for a Fractured Crystalline Bedrock Aquifer. The
Log Analyst. 48 pp.
Walton, W. C. 1962. Selected Analytical Methods for Well and Aquifer Evaluation. Illinois State
Water Survey, Bulletin 49. Urbana, Illinois. 81 pp.
Water Science and Technology Board; Committee on Ground Water Modeling Assessment;
Commission on Physical Sciences, Mathematics, and Resources; and National Research Council.
1990. Ground Water Models - Scientific and Regulatory Applications. National Academy Press,
Washington, D.C. 303 pp.
1-41
-------
Wisconsin Geological and Natural History Survey. 1991. Delineation of Wellhead Protection Areas
in Fractured Rocks. EPA 570/9-91-009. U.S. Environmental Protection Area, Office of Ground
Water and Drinking Water, Washington, D.C. 144 pages.
1-42
-------
ADDITIONAL BIBLIOGRAPHY
Campbell, M. D. and J. H. Lehr. 1973. Water Well Technology. McGraw Hill, New York, New
York. 681pp.
Caswell, B. 1989. Realities of Wellhead Protection. Water Well Journal 43(4): 36-39.
Conservation Foundation. 1987. Groundwater Protection. The Conservation Foundation,
Washington, D.C 240 pp.
Committee on Ground Water Quality Protection; Water Science and Technology Board; Commission
on Physical Sciences, Mathematics, and Resources; and the National Research Council. 1986.
Ground Water Quality Protection: State and Local Strategies. National Academy Press, Washington,
D.C. 309pp.
DeMartinis, J. M. 1989. Small-Scale Retrospective Ground Water Monitoring Studies for Agricultural
Chemicals: Study Design and Site Selection. Ground Water Monitoring Review 9(4): 167-176.
D'ltri, F. and L. G. Wolfson. 1988. Rural Groundwater Contamination. Lewis Publishers, Inc.,
Chelsea, Michigan. 416 pp.
Everett, L. G. and K D. Schmidt. 1978. Establishment of Water Quality Monitoring Programs.
In: Proceedings of the American Water Resources Association, San Francisco, California. Technical •
Publications Series, Minneapolis, Minnesota.
Fairchild, D. M. 1988. Ground Water Quality and Agricultural Practices. Lewis Publishers, Chelsea,
Michigan. 400 pp.
Fawcett, R. S. 1989. Pesticides in Ground Water - Solving the Right Problem. Ground Water
Monitoring Review 9(4): 5-8.
Gibbons, R. D. 1990. A General Statistical Procedure for Ground-Water Detection Monitoring at
Waste Disposal Facilities. Ground Water 28(2): 235-243.
Hampton, N. F. 1967. Monitoring Groundwater Quality: Data Management. EPA-600/4-76-019.
U.S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory, Las Vegas,
Nevada. 62 pp.
Henning, R. J., S. Dhamotharan, and S. Alfred. 1983. Monitoring Data for Other Than Regulatory
Purposes. Ground Water Monitoring Review 3(1): 37-41.
Holden, P. W. 1986. Pesticides and Groundwater Quality: Issues and Problems in Four States.
National Academy Press, Washington, D.C. 124 pp.
1-43
-------
Jaffe, M. and F. DiNovo. 1987. Local Groundwater Protection. American Planning Association,
Chicago, Illinois. 262 pp.
Jones, R. L. and A. L. Bostian. 1989. Developing Management Practices for Preventing Residues of
Agricultural Chemicals in Drinking Water Wells. Ground Water Monitoring Review 9(4): 75-78.
Keith, S. J., L. G. Wilson, H. R. Fitch, and D. M. Esposito. 1983. Sources of Spatial-Temporal
Variability in Ground-Water Quality Data and Methods of Control. Ground Water Monitoring
Review 3(2): 21-32.
Kohen, J. W. and G. H. Stanko, Jr. 1988. Groundwater Monitoring. Environmental Science &
Technology 22(11): 1262-1264.
LeGrand, H. E. 1987. Radon and Radium -Emanations From Fractured Crystalline Rocks - A
Conceptual Hydrogeological Model. Ground Water 25(1): 59-69.
Mademint, D. R. 1993. Handbook of Hydrology. (Editor). McGRaw-Hill. Inc. New York, New
York.
Moore, B. A., A. H. Cathcart, and S. C Danos. 1990. Wellhead Protection Area Delineation and
Monitoring Strategy, Littleton, Massachusetts. In: Proceedings of the Conference on Minimizing
Risk to the Hydrologic Environment, Las Vegas, Nevada. American Institute of Hydrology,
Minneapolis, Minnesota, pp. 182-197.
Morrison, R. D. 1983. Ground Water Monitoring Technology: Procedures, Equipment, and
Applications. Timco Manufacturing, Inc., Prairie Du Sac, Wisconsin. Ill pp.
Nacht, S. J. 1983. Ground-Water Monitoring System Considerations. Ground Water Monitoring
Review 3(2): 33-39.
Nelson, J. D. and R. C Ward. 1981. Statistical Considerations and Sampling Techniques for Ground-
Water Quality Monitoring. Ground Water 19(6): 617-625.
Novotny, V. and G. Chesters. 1981. Handbook of Nonpoint Pollution: Sources and Management.
Van Nostrand Reinhold Company, New York, New York. 555 pp.
Page, G. W. 1987. Planning for Groundwater Protection. Academic Press, Inc., Orlando, Florida.
387pp.
Perazzo, J. A., R. C Dorrier, and J. P. Mack. 1984. Long-Term Confidence in Ground Water
Monitoring. Ground Water Monitoring Review 4(4): 119-123.
Plumb, R. H., Jr. In review. Assessment of Volatile Organic Scans As An Alternative RCRA
Indicator Parameter.. Ground Water Monitoring Review. 25 pp.
Plumb, R. H., Jr. 1987. A Comparison of Ground Water Monitoring Data from CERCLA and
RCRA Sites. Ground Water Monitoring Review 7(4): 94-100.
1-44
-------
Quinlan, J. R, R. O. Ewers, and M. S. Field. 1988. How to Use Ground-Water Tracing to "Prove"
that Leakage of Harmful Materials from a Site in a Karst Terrane Will Not Occur. In: Proceedings
of the Second Conference on Environmental Problems in Karst Terranes and Their Solutions,
Nashville, Tennessee. National Water Well Association, Dublin, Ohio. pp. 289-302.
Rajagopal, R. 1989. Large Data Bases and Regional Ground-Water Quality Assessment - An Iowa
Case Study. Ground Water 25(4): 415-426.
Rajagopal, R. 1988. Influence of Outlying Observations on Selected Estimates of Parameters of
Distributions. Ground Water 26(3): 325-332.
Rajagopal, R. 1986a. The Effect of Sampling Frequency on Ground Water Quality Characterization.
Ground Water Monitoring Review 6(3): 65-73.
Rajagopal, R. 1986b. Conceptual Design for a Groundwater Quality Monitoring Strategy. The
Environmental Professional 8: 244-264.
Scheibe, T. D. and D. P. Lettenmaier. 1989. Risk-Based Selection of Monitoring Wells for Assessing
Agricultural Chemical Contamination of Ground Water. Ground Water Monitoring Review
9(4): 98-108.
Spruill, T. B. and L. Candela. 1990. Two Approaches to Design of Monitoring Networks. Ground
Water 28(3): 430-442.
Tinlin, R. M. 1981. A Methodology for Monitoring Ground-Water Quality Degradation. Ground
Water Monitoring Review 1(2): 27-32.
Tinlin, R. M. 1987. Monitoring Groundwater Quality: Illustrative Examples. EPA-600/4-76-036.
U.S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory, Las Vegas,
Nevada. 81 pp.
U.S. Environmental Protection Agency. 1987a. An Annotated Bibliography on Wellhead Protection
Programs. EPA-440/6-87-014. U.S. Environmental Protection Agency, Office of Ground-Water
Protection, Washington, D.C. 75 pp.
U.S. Environmental Protection Agency. 1987b. Surface Geophysical Techniques for Aquifer and
Wellhead Protection Area Delineation. EPA-440/12-87/016. U.S. Environmental Protection Agency,
Office of Ground-Water Protection, Washington, D.C. 49 pp.
U.S. Environmental Protection Agency. 1987c. Statistical Analysis of Ground-Water Monitoring
Data at RCRA Facilities, Interim Final Guidance. U.S. Environmental Protection Agency, Office of
Solid Waste, Washington, D.C.
U.S. Environmental Protection Agency. 1987d. Statistical Methods for Evaluating Ground-Water
Monitoring Data from Hazardous Waste Facilities. Federal Register, Proposed Rules, 52(163), Part
VII, 40 CFR Part 264. U.S. Environmental Protection Agency, Office of Solid Waste, Washington,
D.C.
1-45
-------
U.S. Environmental Protection Agency. 1987e. Transport and Fate of Contaminants in the
Subsurface. EPA/625-4-89/019. U.S. Environmental Protection Agency, Technology Transfer, Center
for Environmental Research Information, Cincinnati, Ohio. 149 pp.
U.S. Environmental Protection Agency. 1977. Procedures Manual for Ground Water Monitoring at
Solid Waste Disposal Facilities. EPA/530/SW-611. U.S. Environmental Protection Agency, Office of
Solid Waste, Washington, D.C 269pp.
Uhlrnan, K and L. R. Smith. 1990. LUST Busting: Inventory and Ranking of Leaking
Underground Storage Tank Incidents. In: Ground Water Management 1: 579-593, Proceedings of
the 1990 Cluster of Conferences, Kansas City, Missouri. National Water Well Association, Dublin,
Ohio.
Van der Heijde, P., Y. Bachmat, J. Bredehoeft, B. Andrews, D. Holtz, and S. Sebastian. 1985.
Groundwater Management: The Use of Numerical Models. American Geophysical Union,
Washington, D.C 180pp.
Ward, W. D., L. E. Gates, and K B. McCormack. 1990. Tools for Wellhead Protection: Control and
Identification of Light Industrial Sources. In: Ground Water Management 1: 579-593, Proceedings of
the 1990 Cluster of Conferences, Kansas City, Missouri. National Water Well Association, Dublin,
Ohio.
Willis, R. and W. Yeh. 1987. Groundwater Systems Planning & Management. Prentice-Hall, Inc.,
Englewood Cliffs, New Jersey. 416 pp.
Wolf, F. G. and S. M. Testa. 1985. Kriging: A Geostatistical Technique for Evaluating Groundwater
Monitoring Networks. Groundwater Contamination Control, pp. 168-172.
1-46
-------
APPENDIX 1-A
STRATEGY FOR DELINEATION AND DETECTION MONITORING
OF THE FULBRIGHT SPRINGHEAD PROTECTION AREA,
SPRINGFIELD, MISSOURI
1-47
-------
-------
STRATEGY FOR DELINEATION AND DETECTION MONITORING OF THE
FULBRIGHT SPRINGHEAD PROTECTION AREA, SPRINGFIELD, MISSOURI
Beth A. Moore, Lockheed Engineering & Sciences Company, Las Vegas, Nevada
John T. Witherspoon, City Utilities of Springfield, Springfield, Missouri
Loring L. Bullard, Watershed Committee of the Ozarks, Springfield, Missouri
Thomas J. Aley, Ozark Underground Laboratory, Protem, Missouri
Jeffrey K. Rosenfeld, Lockheed Engineering & Sciences Company, Las Vegas,
Nevada
ABSTRACT
Reliable monitoring for contaminants within the Fulbright springhead protection
area, which is situated in a mature karst terrane, is based on identification and
characterization of prioritized source areas, complete delineation of the recharge
area, and flow-route mapping of pollutants during variable recharge events. A
combination of hydrologic, geochemical, and geologic techniques are proposed in
this study to better delineate the protection area, to predict flow transit times, and
to predict pulse characteristics within the recharge system to develop an effective
detection monitoring system. Successful techniques include lineament analysis;
spring, sinkhole, and cave surveying; geologic and potentiometric-surface
mapping; dye tracing; flow-discharge quantification; and surface- and ground-
water quality assessment. Potential contaminant source burdens to the upper and
lower aquifers include agricultural chemicals and pesticides; septic system-
leachate; uncontrolled discharge to abandoned wells; storm-water runoff to
sinkholes and losing streams; spills along transportation routes; and leakage from
underground storage tanks, landfills, and industrial facilities. The strategy
presented in this paper may be generally applicable, with .necessary modifications,
to other karst terranes for the successful implementation of Wellhead Protection
Programs.
INTRODUCTION
The 1986 Amendments to the Safe Drinking Water Act established the Wellhead
Protection Program (WHPP) to protect ground water that contributes to public
water supply wells, wellfields, and springs. We focus here on one important
technical and management element of WHPP implementation: ground-water
monitoring. Ground-water monitoring of chemical and physical parameters may
enhance source characterization, wellhead protection area delineation, and new
water supply evaluation. A wellhead protection area (WHPA) is generally defined
as the surface and subsurface area surrounding a well or wellfield, supplying a
public water system, through which contaminants are reasonably likely to move
toward and reach such a well or wellfield. Moreover, ground water is monitored
to assess source-control measures, to monitor compliance with drinking water
standards, and to provide advance warning of contaminants in ground water.
1-49
-------
MAJOR KARST AREAS WITHIN THE UNITED STATES
Figure 1. Regional location of Springfield, Missouri, with respect to major karst
areas in the United States (After Mull et al., 1988).
This paper presents a strategy for the design of a monitoring program for the
Fulbright springhead protection area (SHPA) in Springfield, Missouri. A
community with a population of approximately 150,000, Springfield is located
within the karst region of the Ozark Highlands of the central United States
(Figure 1). The City has utilized the Spring as a significant source of its drinking
water for over 80 years. Fulbright Spring contributes from 20 to 25 percent to the
City's annual water supply. McDaniel Lake, Fellows Lake, and the James River,
as shown in Figure 2, each contribute approximately 25 percent to the annual
total. The technical approach described in this paper is being implemented by
the City Utilities to protect the Fulbright Spring recharge area for present and
future water supply needs.
In recent years, increasing emphasis has been placed on the complex and
unusual character of karst hydrology. Karst terranes are formed by the
dissolution of carbonate rock such as limestone, dolomite, and gypsum.
Characteristic topographic features include karst windows, sinkholes, losing
streams, caves, and springs. Some of these features are illustrated in Figure 3...
Karst ground waters are highly vulnerable to contamination owing to thin soils in
1-50
-------
co
CD
CO
T3
CD
CO
I
£
I
CO
co
co"
CD
(0
o
T3
CO
0
O)
Q.
CO
CD
O)
cL
CO
II
c ..a
-Q D>
O
_3
oi
1
o>
•••
u.
1-51
-------
SCHEMATIC OF GROUND-WATER FLOW IN MATURE KARST
-LOSING STREAM
DIFFUSE-FLOW AREA
SINKHOLES
CAVES
CONDUIT-FLOW AREA
Figure 3. Schematic of some characteristic features and flow types in karst
regimes (After Mull et al., 1988).
upland areas; rapid ground-water flow rates; and numerous, discrete recharge
points. This vulnerability has underscored the necessity for preventative
contamination measures, such as wellhead and springhead protection.
WELLHEAD PROTECTION MONITORING CONSIDERATIONS
Common to all effective monitoring systems is accurate conceptualization of the
site and its environs. Elements of the conceptual model include the source,
pathway, and receptor. During the initial stages of WHPP implementation, sources
are identified, characterized, and prioritized. Concurrently, WHPAs are delineated
based on site-specific ground-water flow and contaminant-transport assessment.
The receptor corresponds to a public water supply well, wellfield, or spring. By
integrating each component of the environmental system to form an accurate
conceptual model, more efficient monitoring programs can be developed.
Monitoring helps to document ground-water system trends. Properly interpreted,
trend analysis can improve hydrogeologic characterization and enhance
protection. Three types of ground-water monitoring apply to WHPP development:
1-52
-------
ambient-trend, source-assessment, and early-warning detection monitoring (Canter
et al., 1987). Ambient-trend monitoring addresses temporal and spatial trends in
the overall physical and chemical quality of a ground-water basin or subarea.
Source monitoring assesses the existing or potential impacts on the physical or
chemical ground-water system from a proposed, active, or abandoned
contaminant source. Early-warning detection monitoring is conducted at a
calculated distance from the wellhead, based on known travel times, to trigger a
contingency response to prevent public exposure to pollutants. These types of
monitoring are planned to measure or detect contaminants in aquifers, not as
preventative or remedial measures (U.S. EPA, 1989).
SPRINGFIELD CASE STUDY
Hvdroaeoloaic Setting
Springfield is situated on the Springfield Plateau, an erosional surface of the
Mississippian Limestone in southwest Missouri (Harvey, 1980; Pendergrass, 1983;
Thompson, 1986). The upland cap of the Plateau is composed of a very pure
limestone, the Burlington-Keokuk Formation, which is highly solutioned. Locally,
the upper aquifer exists within the Burlington-Keokuk, Elsey, and Pierson
limestone formations, as depicted in Figure 4. It ranges in thickness from 0 to
300 feet in Greene County. The water table is approximately 100 feet deep in
recharge areas and intersects the land surface at stream discharge areas (Imes,
1989). Above the water table, numerous dissolution channels exist where flow
ranges from diffuse to conduit. Major springs, such as Fulbright Spring, issue
from the base of the Burlington-Keokuk within the upper aquifer. Fulbright Spring
discharges from 2 to 8 million gallons per day (gpd), whereas, smaller yields of
30,000 gpd are typical for upper aquifer wells.
Underlying the upper aquifer is the Northview Formation (5 to 80 feet thick in
Greene County), a low-permeability shale and siltstone that separates the upper
and lower aquifers (Figure 4). In areas beneath Springfield/the Northview
Formation thins to 10 feet (Imes, 1989) and is potentially more permeable along
fracture zones. The lower Ozark aquifer consists of over 1,000 feet of dolomite
and sandstone formations capable of yielding up to 30 million gpd. Industrial and
municipal wells completed within the lower aquifer are at depths of 1,200 to 1,700
feet.
Predevelopment regional ground-water flow gradients within the upper and lower
aquifers were toward the west and southwest. Present ground-water flow
directions are significantly influenced by areas of local pumpage. A downward
hydraulic gradient exists between the upper and lower aquifers directly beneath
the City owing to pumpage. Potentiometric levels are as much as 300 feet lower
in the lower aquifer. The upper aquifer significantly recharges the lower aquifer
(Harvey, 1980; Imes, 1989). Consequently, ground-water monitoring of the
Fulbright springhead recharge area may help to protect the water quality of the
underlying Ozark aquifer.
1-53
-------
GEOLOGIC CROSS-SECTION OF THE FULBRIGHT SPRING AREA
Southwest
Northeast
PIERSON FM | **^~ | | BURLINGTON-KEOKUK LIMESTONE
UPPER
AQUIFER
1000-
800-
g
1
400-
ZOO-
SEA LEVEL 0-
_1_
^^•^
^^mm
•••• ••""• . i -i V '7 /
/
^^^-^ x" S / , COMPTON FM-y / COTTER DOLOMITE/
^-7^-
" /
/
/ ' s
'j *
fj '
*-*•
7 — '
/
/
P&,.
/
/
/
/
(-/>
/
/
/ / /
S / .
/ /
/ /
/ /
y "?' ;;
/ /
/ /
/ /
/ /
/ /
'/ '/ '/
/ / /
*-•* * * — ^r?
/ / y
•^T'T
/ / /
/ / /
/ / /
/ / /
'/ /
/ / / ••- /
/ / JEFFERSON CITY /
/ / DOLOMITE / /
/ / / / /
/
/
X
/ / / LOWER GASCONADE DOLOMITE
/ / / / / / /
/ / / / / / /
////////
/ / / EMINENCE DOLOMITE /
X///////
' // // ////
X / /DERBY-DOEHUN DOLOMITE /
/
.-r/~
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
4
/
/
/
s
/
/
J.
/
/
/
/
-4-
/
LOWER
AQUIFER
(OZARK)
'APPROXIMATE WATER TABLE
2000
"SCALE" (FEET)
4000
Figure 4. Typified hydrogeologic cross-section of the Fulbright springhead
protection area (SCS Engineers, 1988).
Fulbright Springhead Protection Area
The inferred Fulbright SHPA is depicted in Figures 2 and 5. The SHPA boundary
corresponds to the surface drainage divides of contributing watersheds and
includes sinkhole plains in inferred recharge areas. The boundary is approximate
owing to the complex surface-subsurface drainage system resulting from karst
development and structural controls such as faults, joints, and lineaments.
Ground-water flow to springs in the area often crosses topographic basin divides.
Numerous dye-tracing studies and additional mapping will be required to
accurately map the SHPA and to verify contributing local, intermediate, and
regional flow systems.
The Fulbright SHPA consists of seven separate watersheds, as shown in Figure 5.
Research in this study focuses on detailed hydrogeologic characterization of the
1-54
-------
o .
O
'3
8
o c
to £
OT O
•
co
5:
.s8
(0 D
g) o
CO »
.Qo
13 Q.
0} •*-*
^ (0
® c
C3) 8
"S ^
Q-*=
CO C
Si
:Q CO'
D 0}
o
10
o>
IT
1-55
-------
Fulbright Spring and Valley Mills connection and the western portion of the SHPA.
Dye traces have verified interconnection between the Fulbright Spring and Valley
Mills Reservoir, indicating travel times as short as three hours. Both the Valley
Mills and South Dry Sac Swallow Hole Watersheds lose most of their surface flows
to the Swallow Hole. Apparently, most of this water discharges at Fulbright and
Bitter Springs. Hydraulic connection between Fulbright Spring and these two
eastern watersheds is partly attributable to fault-controlled losing stream
segments.
The Northwest Springfield Karst, as well as the Spring Branch and Pea Ridge
Creek Watersheds, constitute the western portion of the protection area, as shown
in Figure 5. The Northwest Springfield Karst is characterized by intense sinkhole
development and near absence of surface streams. The Spring Branch and Pea
Ridge Creek Watersheds define the approximate southern boundary of the SHPA.
Fulbriaht SHPA Monitoring Objectives
This study focuses on refined delineation and monitoring system design in two
regions of the SHPA: (1) the Fulbright Spring and Valley Mills connection and (2)
the central and western watersheds. Rapid development will follow installation of
the Little Sac Trunk sewer line (Figure 5). Hydrogeologic characterization is
required here to support land use management planning necessitated by
urbanization. An early-warning detection monitoring system would probably not
be appropriate owing to rapid travel times (three hours to one day) and apparent
velocities (up to 0.5 mile per hour) (Pendergrass, 1983). However, source-
assessment monitoring, in combination with development screening or source
prohibition, may prove to be a practical management strategy for ground-water
protection. Ambient-trend monitoring is necessary in this area to delineate the
distributary drainage network and quantify the flow system at variable discharges.
Through better evaluation of flow times and pulse characteristics, contaminant-
specific monitoring may be achieved.
In the western watersheds, the need for and the feasibility of early-warning and
source-assessment monitoring will be evaluated. Ambient-trend monitoring will be
implemented. Early-warning detection monitoring may be relevant if ground-water
velocities range from days to weeks. Coordinated with a spill-response plan
(Quinlan, 1986), detection monitoring may be an effective management alternative
for determining compliance and triggering emergency cleanup. Assessment
monitoring of prioritized sources must be evaluated with regard to the adequacy
of the site data and the degree of the contamination threat.
Source Assessment
Source assessment is being conducted as a first step in the design of an effective
protection monitoring system for the Fulbright SHPA. All potential point and
nonpoint sources (NPSs) were inventoried and prioritized. Characterization of
prioritized sources will parallel design of the monitoring system.
1-56
-------
As is the case in most karst aquifers, bacteria and nitrates are the most common,
potential NFS pollutants in the study area. Agricultural runoff and septic system
leachate contribute significantly to their presence in ground water. The dominant
land use practice within the Fulbright SHPA is agriculture. Nutrient loading,
primarily from animal wastes, to McDaniel Watershed has accelerated lake
eutrophication and threatens ground water. Dye-trace studies conducted by Aley
and Thompson (1981) have shown that many springs in the county receive
effluent from septic systems.
As urbanization encroaches on the Fulbright Spring recharge area, deterioration of
water quality is anticipated (Pendergrass, 1983). Three water quality problems are
associated with land development in the Springfield area: (1) sewage effluent; (2)
urban runoff; and (3) leaks, spills, and discharges from commercial and industrial
facilities (Aley and Thompson, 1981). Expansion of the sewer system into rural
areas may help prevent some NFS contamination.
Sinkholes, losing streams, and abandoned wells provide direct access to the
ground-water system. One-half mile west of Fulbright Spring, a number of private
water supply wells were contaminated by gasoline constituents from an unknown
source. Illegal disposal into one of numerous sinkholes and an abandoned well at
a gasoline station site are suspected sources. Surface pathways facilitate
uncontrolled storm-water runoff to the upper and, possibly, lower aquifers. Storm-
water runoff typically contains higher concentrations of metals, agricultural
nutrients and pesticides, bacteria, and petrochemicals (Jenkins, 1988).
Potential point sources within and in proximity to the SHPA include landfills,
manufacturing plants, and hydrocarbon-related facilities, as shown in Figure 5.
Two sanitary landfills, where industrial wastes containing cyanides, heavy metals,
and acids were disposed of, are located in the northern portion of the SHPA. An
industrial landfill, where foundry sand was disposed of, is located within the Pea
Ridge Creek Watershed. Industrial and electronics equipment are the
predominant products of the manufacturing facilities located within the Northwest
Springfield Karst. Hydrocarbon-related facilities include a creosote-based wood
treatment plant, railroad yard, airport, asphalt .plant, petroleum products
distribution center, and numerous underground storage tanks (USTs).
Springhead Protection Boundary and Flow-Path Delineation
Numerous hydrogeologic studies have been conducted to characterize the
shallow aquifer system around Springfield. The most significant conclusions from
these studies are that "...(1) fluid flow is rapid, unrestricted, and capable of
substantial lateral transport, and (2) contaminants pass through the system
essentially unchanged" (Pendergrass, 1983). For karst terranes such as this,
delineation can only be done by dye tracing, guided by inference from accurate
potentiometric mapping (Mull et al., 1988).
1-57
-------
Figure 5 depicts a compilation of spectrofluorometric and visual dye-trace results
in proximity to the Fulbright SHPA. Interconnection to Fulbright Spring is evident
from the Northwest Springfield Karst. Pathways may exist from the Spring Branch
and Valley Mills Watersheds to the Spring. Recharge from the Pea Ridge Creek
Watershed is poorly understood, although losing stream sections are prevalent in
the upper watershed. Visually confirmed traces, originating from watersheds to
the south of the SHPA, suggest that recharge also occurs from intermediate and,
possibly, regional flow systems. Dye traces from the Northwest Springfield Karstp
indicate flow components to the north and west beyond the inferred SHPA
boundary.
Inspection of Figure 5 leads to several data deficiencies that will be addressed in
this study. Local flow systems in the western and central watersheds, as well as
the Fulbright Spring and Valley Mills connection, must be further delineated and
quantified. Through dye tracing and volumetric flow analyses at select locations
(springs, sinkholes, caves, and wells) and at variable flow conditions,
characteristics of the karst system can be determined, such as (Mull et al., 1988;
Quinlan, 1989):
• Areas of diffuse versus conduit flow
• Flow paths, apparent velocities, and residence times
• Point-to-point connections to aid in source assessment
• Qualitative indications of contaminant fate and transport along various flow
paths
• Local recharge boundaries for protection area delineation
To better characterize the Fulbright Spring and Valley Mills connection, two
physical monitoring programs will be initiated: (1) flow discharges at Fulbright
Spring, Sander Spring, and Valley Mills Reservoir and (2) water levels in a lower
aquifer well and shallow wells along the Little Sac Trunk sewer line (Figure 5).
Physical monitoring will focus on flood-pulse evaluation to establish low and
intermediate flow conditions. Complementary qualitative and quantitative dye
traces will be designed for recovery at the monitoring sites. A similar technical
approach will be implemented for the western and central watersheds of the
SHPA. However, dye-trace studies will focus on determining potential point-
source connections to Fulbright Spring, as well as local flow-path assessment.
Dye-tracing results, in conjunction with potentiometric mapping, will be used to
modify and improve the present conceptual model of flow within the upper aquifer.
While the two techniques are complementary, accurate dye-trace results tend to
be less ambiguous than interpretations based on potentiometric mapping
(Quinlan, 1989). Other supporting information to refine development of the
conceptual flow model include:
• Geologic unit and structural-control mapping
• Lineament analysis, both photo interpretation and ground verification
• Spring, cave, and sinkhole surveying
• Surface- and ground-water quality characterization
1-58
-------
An iterative process of data integration and corroboration will be implemented to
generate more accurate interpretations and effective monitoring systems.
Optimal location of water quality monitoring sites will be dependent on
identification of diffuse- versus conduit-flow areas and distributary-flow systems. In
areas of diffuse flow, standard methods and computer model evaluations for site
selection or well placement may be applicable. However, monitoring for pollutants
in conduit-flow systems will be done at locations shown by dye tracing to be
connected to sources under base- and flood-flow conditions (Quinlan, 1989).
Conversely, background monitoring will be done in areas determined to be
geochemically similar to the target area and where dye-traces confirm no
hydrologic connections. Knowledge of distributaries is important due to the
influence that the system can have on the areal and temporal dispersion of
contaminants at various recharge events (Quinlan, 1989).
Ground-Water Quality Assessment
Bacteriological data indicate that water quality has improved somewhat in many
parts of the recharge area corresponding to installation of a centralized sewer
system. Ozark aquifer ground water has never exhibited microbiological
degradation. Analyses of the spring water for VOCs, napthalene, and cyanide
indicate nondetectable to very minor concentrations. Other key indicators of
upper aquifer water quality are nitrate and turbidity levels.
A major objective of this study is to document and analyze present ground-water
quality trends to develop effective monitoring protection strategies. The storm-
related variability in water quality at monitoring sites will be measured much more
frequently than at standard intervals (for example, quarterly or semi-annually) to
ensure accurate characterization (Quinlan, 1989). The maximum value of some
parameters coincides with peak discharges; others culminate several days after
the maximum storm event (Quinlan 1989). Consequently, sampling at 1- to 6-hour
intervals may be required prior to and during peak discharge, while 4- to 24-hour
intervals may be appropriate later.
An automatic sampler will be installed at Fulbright Spring to monitor the storm-
related variability of most inorganic and select organic compounds. Automatic
sampling will only yield qualitative estimates of VOC contamination. Continuous
determination of dissolved oxygen, pH, specific conductance, and temperature will
be attempted with dedicated devices. Other sites within the SHPA will be similarly
monitored, but less frequently. Hydrochemical data will be analyzed to categorize
water types and to establish select spatial and temporal water quality trends.
Indicator parameters will be determined for early-warning detection and source-
assessment monitoring.
1-59
-------
Monitoring Parameters
The variety of potential contamination sources within the Fulbright Spring recharge
area suggest the need for a broad range of monitoring parameters. Chlorinated
solvents, heavy metals, volatile fuel hydrocarbons, and semi-volatile polynuclear
aromatic hydrocarbons were detected in the ground water collected from springs
and wells in the vicinity of some of the sources. This range of contaminants can
be associated with the different types of potential sources found in the area. The
chlorinated solvents and heavy metals are the predominant wastes from
machining, plating, and painting activities at the manufacturing facilities.
Polynuclear aromatic hydrocarbons are found in wood-treating and asphalt
operations, while the lighter fuel hydrocarbons are associated with the use of
gasoline and oils.
Chlorinated solvents, hydrocarbons, and heavy metals have been found in the
ground-water down gradient of the municipal landfills. These compounds are
thought to be related to the industrial waste disposed of at the landfills. In
addition, general chemistry parameters, such as specific conductivity, chloride,
sulfate, and nitrate can commonly be used to detect leakage from sanitary
landfills. The possibility of nonpoint contamination from septic systems and
agricultural sources suggests the need for nutrient and pesticide analyses.
In summary, contaminant detection monitoring may require complete analyses for
organic priority pollutants (volatile, semi-volatile, and pesticide compounds), heavy
metals, and general chemistry parameters. Comprehensive monitoring is
warranted to cover the range of potential contaminant sources within the SHPA.
Periodic analyses must be done at Fulbright Spring and other critical monitoring
locations during the initial phases of this wellhead protection effort to determine
background levels for these compounds.
CONCLUSION
This paper combines a strategy for the design of a monitoring system for
wellhead and springhead protection in karst terranes with a case study application
in Springfield, Missouri. Determination of target monitoring parameters requires
comprehensive identification, prioritization, and characterization of potential point
and nonpoint sources. Correct delineation in karst terranes can only be done by
dye tracing, guided by inference from accurate potentiometric mapping (Mull et
al., 1988). Complementary, and necessary, investigations include geologic and
structural-control mapping; lineament analysis; spring, cave, and sinkhole
mapping; and water quality characterization.
Ambient-trend, source-assessment, and early-warning detection monitoring
objectives should be established to clearly support technical and management
programs (Canter et al., 1987). Optimal and effective monitoring network design
is based on accurate characterization of the karst flow system. This is
accomplished through dye tracing and volumetric flow analysis at select locations
and at variable flow rates (Quinlan, 1989). The storm variability in water quality at
1-60
-------
monitoring sites should be measured as a basis for determining the areal and
temporal dispersion of contaminants during recharge events.
ACKNOWLEDGEMENT
This work was conducted under Task Directive 90G02 of Contract 68-03-3245
between the U.S. Environmental Protection Agency, Environmental Monitoring
Systems Laboratory, Las Vegas, Nevada, and Lockheed Engineering & Sciences
Company, Las Vegas, Nevada. Mr. Steven P. Gardner, the EPA Technical
Monitor, and Mr. James F. Quinlan (ATEC Associates, Nashville, Tennessee) are
gratefully acknowledged for their support and technical guidance in this research
effort.
NOTICE
Although the research described in this document has been funded wholly or in
part by the United States Environmental Protection Agency, it has not been
subjected to Agency review and, therefore, does not necessarily reflect the views
of the Agency and no official endorsement should be inferred.
REFERENCES
Aley, T. and Thompson, K.C., 1981, Hydrogeologic Mapping of Unincorporated
Greene County, Missouri, to Identify Areas Where Sinkhole Flooding and
Serious Groundwater Contamination Could Result from Land Development,
Ozark Underground Laboratory, Protem, Missouri, 11 p. + 5 plates.
Canter, L.W., Knox, R.C., and Fairchild, D.M., 1987, Chapter 8: Ground water
monitoring planning in Ground Water Quality Protection, Lewis Publishers, Inc.,
p. 325-361.
Harvey, E.J., 1980, Ground Water in the Springfield-Salem Plateaus of Southern
Missouri and Northern Arkansas, U.S. Geological Survey Water-Resources
Investigations 80-101, Rolla, Missouri, 66 p.
Imes, J.L, 1989, Analysis of the Effect of Pumping on Ground Water Flow in the
Springfield Plateau and Ozark Aquifers Near Springfield, Missouri, U.S.
Geological Survey Water-Resources Investigations Report 89-4079, Rolla,
Missouri, 63 p.
Jenkins, D.T., 1988, Development of storm water management criteria for sensitive
karst areas in north-central Florida, U.S.A. in Proceedings of the Second
Conference on Environmental Problems in Karst Terranes and Their Solutions
Conference, National Water Well Association, Dublin, Ohio, p. 333-343.
.1-61
-------
Mull, D.S., Liebermann T.D., Smoot J.L, and Woosley, L.H., Jr., 1988, Application
of Dye-tracing Techniques for Determining Solute-Transport Characteristics of
Ground Water in Karst Terranes, U.S. Environmental Protection Agency,
Ground-Water Protection Branch, Atlanta, Georgia, 103 p.
Pendergrass, G.J., 1983, Chapter I: The environment in Greene County in Report
of the Watershed Task Force for Springfield-Greene County, Mo., November
1983, City Utilities, Springfield, Missouri, p. 1-30.
Quinlan, J.F., 1989, Ground-Water Monitoring in Karst Terranes: Recommended
Protocols & Implicit Assumptions, U.S. Environmental Protection Agency,
Environmental Monitoring Systems Laboratory, Las Vegas, Nevada, 79 p.
Quinlan, J.F., 1986, Recommended procedure for evaluating the effects of spills of
hazardous materials on ground water quality in karst terranes in Proceedings
of the Environmental Problems in Karst Terranes and Their Solutions
Conference, National Water Well Association, Dublin, Ohio, p. 183-195.
SCS Engineers, 1988, Remedial Investigation and Environmental Assessment:
Fulbright and Sac River Landfill Sites, SCS Engineers, Reston, Virginia.
Thompson, K.C., 1986, Geology of Greene County, Missouri, Watershed
Management Coordinating Committee, Springfield, Missouri, 87 p.
U.S. Environmental Protection Agency, Office of Ground-Water Protection, 1989,
Wellhead Protection Programs: Tools for Local Governments, Washington,
D.C., 50 p.
BIOGRAPHICAL SKETCHES
Beth A. Moore is a Senior Hydrogeologist with Lockheed Engineering & Sciences
Company (Suite 157, 1050 E. Flamingo Road, Las Vegas, Nevada, 89119), which
is the technical support contractor for the U.S. EPA's Environmental Monitoring
Systems Laboratory in Las Vegas. Ms. Moore conducts research in the
delineation of wellhead protection areas and the design of monitoring systems for
Wellhead Protection Programs. She has a B.S. in Geology from the Pennsylvania
State University, an M.S. in Hydrology from the University of Idaho, and is
presently completing an Associate degree in Computer Science.
John T. Wrtherspoon is the Director of the Central Laboratory, Systems Operations
Department, for the City Utilities (301 E. Central, Springfield, Missouri, 65801). Dr.
Wrtherspoon has sixteen years of direct experience in environmental monitoring,
as well as water supply assessment, management, and regulation. He holds B.S.
and M.A. degrees in Biology from the Southwest Missouri State University and a
Ph.D. in Botany from the University of Montana.
1-62
-------
Loring L. Bullard is the Director of the Watershed Committee of the Ozarks (Room
23, 819 Boonville, Springfield, Missouri, 65802), an organization established to
manage and protect Ozark area public water resources. Mr. Bullard is actively
involved in surface- and ground-water monitoring, quality assessment, allocation,
and protection. He has a B.S. in Biology from the Central Missouri State
University; has attended the Water & Wastewater Technical School in Neosho,
Missouri; and is presently a Master's Candidate at the Southwest Missouri State
University.
Thomas J. Aley is a Hydrogeologist and the Director of the Ozark Underground
Laboratory (Protem, Missouri, 65733). Mr. Aley has twenty-four years of
experience in ground- and surface-water hydrology, pollution control
investigations, and land management issues with particular emphasis on soluble
rock landscapes. He holds B.S. and M.S. degrees in Forestry from the University
of California, Berkeley, and received additional training in hydrogeology at the
University of Arizona, Tucson, and the Southern Illinois University, Carbondale.
Jeffrey K. Rosenfeld is a Geochemist with Lockheed Engineering & Sciences
Company (Suite 157, 1050 E. Flamingo Road, Las Vegas, Nevada 89119). He is
involved primarily in the assessment of ground-water and soil contamination at
hazardous waste sites. Dr. Rosenfeld has a B.S. in Geology from M.I.T., and M.S.
and Ph.D. degrees in Geochemistry from Yale University.
1-63
-------
-------
EPA-600/R-93/
APRIL 1993
CHAPTER 2
WELLHEAD PROTECTION PROGRAM AND MONITORING SYSTEM DESIGN
STEVENS POINT, WISCONSIN
by
Theresa Brown
Project Coordinator ~~
Wellhead Protection Program
Stevens Point, Wisconsin 54481
Beth A. Moore
Lockheed Environmental Systems & Technologies Company
Las Vegas, Nevada 89119
Greg Disher
Administrator
City of Stevens Point Water Department
Stevens Point, Wisconsin 54481
John Gardner
Director of Community Development
Stevens Point, Wisconsin 54481
April 1993
Contract Number CR-816203-01-0
Project Officer
Steven P. Gardner
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas, Nevada 89193-3478
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89193-3478
-------
NOTICE
This report is the result of research supported by the U.S. Environmental Protection Agency's
Environmental Monitoring Systems Laboratory in Las Vegas, Nevada, as part of its efforts to provide
technical assistance to state, tribal, and local governments on the implementation of the Wellhead
Protection Program. The specific methods and approaches contained in this document have
undergone peer review but do not constitute official Agency endorsement or policy recommendations.
The Office of Research and Development provides this information to help solve complex technical
problems related to refined delineation and ground-water monitoring of wellhead protection areas in
various hydrogeologic settings. Further assistance is available from the Environmental Monitoring
Systems Laboratory in Las Vegas, Nevada, from the Office of Ground Water and Drinking Water in
Washington, D.C, and from the ground-water offices in the ten U.S. Environmental Protection
Agency regions.
2-ii
-------
ABSTRACT
The Stevens Point, Wisconsin, Wellhead Protection Program is an ongoing effort by the city's
Water Department to ensure a high-quality, economical, and lasting municipal water supply. The city
obtains all of its water from a shallow, unconfined, sand and gravel aquifer that is highly susceptible to
contamination. The major threats to ground-water quality are from urban, suburban, and agricultural
land uses that create both point and nonpoint sources of contamination. Typical ground-water
contaminants include nitrates from septic systems, fertilizers, and animal wastes; volatile organic
compounds from leaking underground storage tanks and spills; and naturally occurring iron,
manganese, and radioactivity. This study encompasses refined delineation of the Wellhead Protection
Area, hydrogeologic mapping, contaminant source assessment, and monitoring system design.
Completion of these tasks will serve as the basis for continued research in wellhead protection in
order to increase the understanding of the hydrogeologic system and to protect the ground-water
resource.
2-iii
-------
-------
CONTENTS
Abstract iii
Figures vii
Tables ix
Abbreviations, Symbols, and Conversion Factors xi
Acknowledgments xiiii
Background for the Stevens Point Case Study 2-1
Introduction 2-1
Location 2-2
Wellhead Protection Program Overview 2-2
Hydrogeologic Setting 2-5
Preliminary Wellhead Protection Area 2-9
Source Assessment 2-9
Source Identification , 2-9
Source Characterization and Prioritization 2-16
Research Monitoring Program 2-18
Data Base Limitations 2-18
Monitoring Objectives 2-18
Research Monitoring Tasks 2-19
Data Acquisition 2-19
Monitoring Wells 2-19
Physical Parameters 2-23
Chemical Parameters 2-32
Data Interpretation 2-33
Aquifer Parameters 2-33
Hydrochemical Assessment 2-36
Ground-Water Flow Modeling 2-41
Wellhead Protection Monitoring Program 2-51
Ambient Trend Monitoring 2-51
Monitoring Sites 2-51
Monitoring Well Construction 2-53
Monitoring Parameters and Frequencies 2-53
Monitoring Data Base 2-53
Compliance Monitoring 2-53
Monitoring Sites 2-53
Monitoring Well Construction 2-54
Monitoring Parameters and Frequencies 2-56
Monitoring Data Base 2-56
2-v
-------
CONTENTS, Continued
Contingency Plan 2-56
Conclusions • 2-59
Recommendations 2-61
References — I 2-63
Appendix 2-A. FLOWPATH Ground-Water Model Results for the
Stevens Point, Wisconsin, Aquifer
2-67
2-vi
-------
FIGURES
Numbe
2-1
2-2
2-3
2-4
2-5
2-6
2-7
2-8
2-9
2-10
r
Location of the city of Stevens Point within Portage County,
City wells at the Airport and Iverson wellfields Stevens Point M^isconsin
Surface geology map Stevens Point \Visconsin.
Hydrogeologic cross section northwest to southeast across the Airport wellfield,
Stevens Point ^^isconsin
Recharge area for the Airport and Iverson wellfields, Stevens Point, Wisconsin ....
Wellhead protection zones (A, B, and C) for the Airport and Iverson
wellfields Stevens Point \Visconsin •
Typical point sources south of the Airport wellfield Stevens Point, Wisconsin .....
Typical nonpoint sources north of the Airport wellfield, Stevens Point, Wisconsin . .
Research monitoring wells Stevens Point Wisconsin
Research monitoring well construction diagrams for (A) single, drilled well;
CB) well nest: and (C\ hand-driven well
Page
2-3
2-4
2-6
2-7
2-8
2-10
2-14
2-15
2-20
2-21
2-11
2-12
2-13
2-14
Logarithmic plot of drawdown versus time in observation well MW36 during
drawdown testing at city well 6 on June 5, 1990,
Stevens Point, Wisconsin
Semi-log plot of drawdown versus the log of time at observation well MW36
during drawdown testing at city well 6 on June 5, 1990,
Stevens Point, Wisconsin
Semi-log plot of the log of the relative change in head versus time for slug testing
at well MW36 on July 18, 1990, Stevens Point, Wisconsin
Semi-log plot of the log of the relative change in head versus time for slug testing
at well J160 on July 18, 1990, Stevens Point, Wisconsin
2-34
2-35
2-37
2-38
2-vii
-------
FIGURES, Continued
Number Page
2-15 Semi-log plot of the log of the relative change in head versus time for
slug testing at well K58 on July 17, 1990, Stevens Point, Wisconsin 2-39
2-16 Semi-log plot of the log of the relative change in head versus time
for slug testing at well MW13 on July 18,1990, Stevens Point, Wisconsin 2-40
2-17 Nitrogen (nitrite and nitrate) concentrations at city wells 6 through 9
from 1984 to 1990, Stevens Point, Wisconsin 2-42
2-18 Spatial distribution of nitrogen (N), iron (Fe), manganese (Mn), and chloride (Cl)
concentrations exceeding drinking water standards or Preventive Action Limits
(PALs) throughout the research monitoring network, Stevens Point, Wisconsin . . 2-43
2-19 Variations in nitrogen concentration with respect to well depth throughout
the research monitoring well network, Stevens Point, Wisconsin 2-44
2-20 Bedrock topography map, Stevens Point, Wisconsin 2-45
2-21 (A) Hydraulic head distribution map and (B) 5-year time-of-travel zone for
the Airport and Iverson wellfields, Stevens Point, Wisconsin 2-49
2-22 Comparison of B Zone and 5-year time-of-travel zone areas determined by
analytical and modeling (FLOWPATH) methods, respectively, for the
Airport and Iverson wellfields, Stevens Point, Wisconsin 2-50
2-23 Proposed wellhead protection monitoring network for the Airport and Iverson
wellfields, Stevens Point, Wisconsin 2-52
2-viii
-------
TABLES
Number
Page
2-1 Potential Contamination and Degredation Sources within the B Zone of the Preliminary
WHPA, Stevens Point, Wisconsin 2-12
2-2 Existing and Potential Contamination and Degredation Sources within the WHPA,
Stevens Point, Wisconsin 2-13
2-3 Prioritized, Potential Contamination and Degredation Sources within the Preliminary
WHPA, Stevens Point, Wisconsin 2-16
2-4 Research Monitoring Well Data from 1990, Stevens Point, Wisconsin 2-24
2-5 Aquifer Test Results for Geologic Terrains within the Sand and Gravel Aquifer ... 2-36
2-6 Aquifer Parameter Values and Their Application to the Ground-Water
Flow Model, FLOWPATH 2-48
2-7
2-8
Preventive Action Limits and State of Wisconsin Drinking Water
Standards
2-54
Compliance Monitoring Guidelines 2-57
2-ix
-------
-------
ABBREVIATIONS, SYMBOLS, AND CONVERSION FACTORS
ABBREVIATIONS
amsl above mean sea level
Alk alkalinity
ds cubic foot per second
COD chemical oxygen demand
Cond specific conductivity
CW city well
d day
DEBT N,N-diethyl-3-methylbenzamide
DNAPL dense, nonaqueous-phase liquid
DO dissolved oxygen
DOW depth of well
EMSL-LV Environmental Monitoring Systems Laboratory, Las Vegas, Nevada
EPA U.S. Environmental Protection Agency
ft foot
ft/d foot per day
ft2/d square foot per day
ft/s foot per second
ft^/s square foot per second
gal gallon
gpd gallon per day
Hard hardness
in inch
LESAT Lockheed Environmental Systems & Technologies Company
1 liter
min minute
Hg microgram
Hg/1 microgram per liter
|iS/cm microSiemen per centimeter
mg milligram
Mgd million gallon per day
mg/1 milligram per liter
MP Elev measuring point elevation
MW monitoring well
N/A not applicable
ND not detected
NSA nonsewered area
PAL preventive action limit
2-xi
-------
PCB polychlorinated biphenyls
Pest pesticide
ppm part per million
PVC polyvinyl chloride
SCH schedule
sec second
SVOA semi-volatile organic analytes
SWL Elev static water level elevation
TOT time of travel
TPH total petroleum hydrocarbon
UST underground storage tank
USGS U. S. Geological Survey
VOC volatile organic compound
WGNHS Wisconsin Geological and Natural History Survey
WHPA wellhead protection area
WHPP wellhead protection program
ZOI zone of influence
SYMBOLS
Ca
a
Fe
Mn
Na
N
NO2
N03
P
Zn
b
d
h
K
Q
R
RC
r
S
s
T
t
to
calcium
chloride
iron
manganese
sodium
nitrogen
nitrite
nitrate
phosphorus
zinc
aquifer thickness
thickness of river or lake bed
hydraulic head
hydraulic conductivity
vertical hydraulic conductivity
screen length
leakage factor
well discharge
well radius
correction factor
distance between wells
specific yield for an unconfined aquifer
drawdown
transmissivity
time
specified time during aquifer testing
, 2-xii
-------
CONVERSION FACTORS
Multiply
acre-foot
cubic foot per second
foot
foot per second
gallon
gallon
gallon
gallon per day
gallon per day per foot
gallon per day per square foot
inch
inch per year
mile
million gallons per day
square foot per minute
square foot per second
square mile
By
1230
0.0283
0.3048
0.3048
3.785
0.134
0.00379
0.000003528
0.000207
0.0408
0.0254
25.4
1.609
2.629
0.0929
0.0929
2.59
To Obtain
cubic meter
cubic meter per second
meter
meter per second
liter
cubic foot
cubic meter
cubic foot per second
square meter per day
meter per day
meter
millimeter per year
kilometer
cubic meter per minute
square meter per minute
square meter per second
square kilometer
2-xiii
-------
ACKNOWLEDGMENTS
This research was funded through the U.S. Environmental Protection Agency (EPA),
Environmental Monitoring Systems Laboratory at Las Vegas, Nevada (EMSL-LV), under Contract
Number CR-816203-01-0 to the Stevens Point Water Department, Stevens Point, Wisconsin. We
gratefully acknowledge the support of Steven Gardner (EMSL-LV) who served as the EPA Project
Officer. The research was started because of the combined efforts of Tom Osborne (Braun
Engineering) and Dr. Byron Shaw (Central Wisconsin Groundwater Center). Dr. Jean Bahr and her
field methods class (University of Wisconsin-Madison), Jo Ellen Seizer (Portage County Planning
Department), as well as Bill DeVita and other staff of the Environmental Task Force Laboratory,
University of Wisconsin-Stevens Point, provided valuable technical support throughout the project.
For their careful review of the manuscript and insightful comments, we thank the following technical
reviewers: Dr. Charles Kreitler (University of Arizona), Douglas Heath (U.S. EPA Region I), Tom
Osborne (Braun Engineering), and Dr. George Kraft (Central Wisconsin Groundwater Center). For
their hard work and contributions to the field effort, we are grateful to Dale Johnson and Larry
Slusarki (Stevens Point Water Department), and Marianne Ouren.
John Nicholson and Carolyn Cameron of Lockheed Environmental Systems & Technologies
Company (LESAT) provided technical writing and editing support for preparation of the manuscript.
Steve Garcia (LESAT) contributed excellent graphics to the report. Shalena Fendzlau (LESAT) is
graciously acknowledged for her patience and expertise in word processing support.
2-xiv
-------
BACKGROUND FOR THE STEVENS POINT CASE STUDY
INTRODUCTION
The Stevens Point, Wisconsin, Wellhead Protection Program (WHPP) is a continuation of the
city's efforts to maintain their ground-water supply and to protect their water quality. The purpose of
the local WHPP is to design and implement a protection plan for the city's production wells. The
four major components of the Stevens Point WHPP are a ground-water monitoring program, land-use
restrictions, an education program, and a contingency plan.
This investigation focuses on the design of a monitoring program for a shallow, unconfined,
highly permeable aquifer with point and nonpoint sources of ground-water contamination from urban
and agricultural activities. The purpose of the ground-water monitoring system is to detect
contamination before it reaches the city water supply. The monitoring system design is based on
refined delineation of the Wellhead Protection Area (WHPA) and the results of the contaminant
source assessment. The land-use restrictions and education program components of the WHPP may
limit or modify practices that endanger the water supply, thus reducing the probability of
contamination. A contingency plan is necessary to ensure that the city's water-supply needs can be
met if the protection plan fails.
The objectives of this study are to design and document the elements of each of the four
major components of the Stevens Point WHPP. Activities conducted in support of the project's
objectives include:
Compiling existing data
Identifying limitations of the existing data base
Contaminant source assessment
Prioritization of sources
Hydrogeologic mapping
Installation of additional monitoring wells
Aquifer testing
Ground-water sampling for chemical analysis
Flow and transport modeling
Refined delineation of the WHPA
Evaluation of management options
The need for a comprehensive WHPP in Stevens Point is warranted by the aquifer's sole-source
classification, its vulnerability to contamination, and a history of ground-water contamination proximal
to the city's well fields.
2-1
-------
LOCATION
Stevens Point is located in central Wisconsin near the center of Portage County (Figure 2-1).
Stevens Point is the largest city in the county, with a population of approximately 23,000. The city is
bordered by the town of Hull to the north and east, by the town of Plover and the village of Whiting
to the south, and by the towns of Linwood and Carson to the west (Figure 2-1).
In Portage County, more than 90% of all water used is obtained from ground-water supplies
(Portage County Planning Department, 1988). Agricultural irrigation is the largest ground-water
consumer, requiring approximately 10 trillion gallons of water per year or 67% of the total estimated
ground-water withdrawals (based on the water budget for 1979). Ground water is used for all
domestic and municipal drinking water supplies in the county.
Several ground-water contamination problems are common in the region, including high levels
of nitrates, pesticides, and volatile organic compounds (VOCs). The village of Whiting has purchased
water from the city of Stevens Point since 1979 when nitrate levels in the village wells exceeded 10
parts per million (ppm). Aldicarb, a pesticide used to control the potato beetle, was detected in 53
wells in Portage County in 1986 (Portage County, 1986). Restrictions on the application of this
pesticide were enacted by the Wisconsin state legislature in March 1986. Recently the State Board of
Agriculture Trade, and Consumer Protection proposed restrictions on the application of another
pesticide, atrazine, which has also been detected in drinking water from private wells in Portage
County Local contamination problems include (1) several private wells in the town of Hull
contaminated with gasoline, and (2) ground-water contaminated with aviation fuel at the Stevens Point
airport.
WELLHEAD PROTECTION PROGRAM OVERVIEW
Although the term "wellhead protection" was not yet conceived, the concept of wellhead
protection has been applied by the city of Stevens Point since the first municipal supply well was
installed in 1923. The original Iverson wellfield consisted of three wells installed between 1923 and
1938. The wells were installed near the Plover River to ensure an adequate supply of drinking water.
Although the primary concern was water quantity, water quality and its preservation were also
considered a priority in constructing these wells. The property surrounding the wells was designated
as park land to reduce the possibility of water-supply contamination, and the Iverson wells were
protected by permanent structures. Wellhead protection practices remained essentially the same for
the next 58 years. In the 1960s, two wells at the Iverson wellfield and four wells at the new Airport
wellfield were added to the city's water-supply system (Figure 2-2). These six wells comprise the city's
existing well system.
When the site for the Airport wellfield was chosen in 1965, the city purchased property
surrounding the wellfield property to provide a closely supervised buffer zone to minimize ground-
water contamination (Hickok and Associates, 1965). Ground-water quality of the production wells
was monitored regularly. Monitoring wells were not used until 1981. Hickok and Associates was
contracted to study the potential for contamination of the city's Airport wellfield in the event of a leak
in one of the airport's large underground fuel storage tanks. The Hickok and Associates (1981) study
2-2
-------
PORTAGE COUNTY
0 4000 8000
I I I I I
SCALE IN FEET
Figure 2-1. Location of the city of Stevens Point within Portage County,
State of Wisconsin.
2-3
-------
'-•J
AIRPORT
WELLFIELD
U.S. 10
IVERSON
WELLFIELD
8
0 4000 8000
I ' ' ' '
SCALE IN FEET
Figure 2-2. City wells at the Airport and Iverson wellfields, Stevens Point, Wisconsin.
2-4
-------
resulted in hydrogeologic mapping of the area from the airport to the Airport wellfield and the
installation of five observation wells downgradient from the airport's fuel storage tanks. Although the
fuel leak at the airport in 1988 was not detected by sampling the observation wells, the estimates of
contaminant transport velocities from the Hickok and Associates study prompted swift remedial
action by the city.
Increased public awareness of wellhead protection and the city's and county's efforts to
promote it were hastened by: (1) the closing of the town of Whiting's wellfield in 1979, (2) the
discovery of aldicarb in Portage County ground water in 1980 (Chesters et al., 1982), (3) the
publication of a statewide ground-water susceptibility map in 1987, and (4) the fuel leak at the Stevens
Point airport in 1988. The Stevens Point Water Department began funding research through the
University of Wisconsin - Stevens Point regarding recharge and water quality at both the Iverson and
Airport wellfields (for example, Renaud, 1987). The Portage County Planning and Zoning
Department developed a county ground-water management plan in 1988 that includes a county-wide
source assessment and proposes generic land-use restrictions for the zone of influence (ZOI), the
5-year time-of-travel (TOT) zone, and the recharge area of municipal wells in the county. The Stevens
Point WHPP is the next step in the evolution of the city's water management plan. This project
emphasizes the importance of a technically defensible WHPA, management tools available to protect
the ground-water supply, and the need for cooperation among government agencies.
The overall objectives of the WHPP project are as follows: (1) to conduct a detailed source
assessment for the 5-year TOT zone, (2) to rank the sources identified in the source assessment, (3) to
refine the delineation of the WHPA, (4) to design a ground-water monitoring well network, (5) to
document a contingency plan, and (6) to recommend a strategy for future research to improve the
WHPP.
Hydrogeologic boundaries seldom coincide with political boundaries. Consequently,
cooperation among participating government agencies is essential. The groups cooperating with the
Stevens Point Water Department in this study include the Stevens Point Community Development
Department, the Stevens Point Public Works Department, the Portage County Planning and Zoning
Department, the University of Wisconsin - Stevens Point Environmental Task Force, and the Central
Wisconsin Ground-Water Center.
HYDROGEOLOGIC SETTING
Geologically, Portage County consists of unconsolidated, glacial outwash sediments that were
deposited over crystalline rock, primarily granite, of Precambrian age. Scattered remnants of
Cambrian sandstone exist on a few bedrock highs. A geologic map of the county (Figure 2-3) shows
the location of bedrock and sandstone outcrops (Clayton 1986; Holt, 1965). In areas where the glacial
sediments are thin or not present, ground water is obtained from fractured crystalline bedrock. The
glacial sediments form the most productive aquifer in this region.
The city's production wells pump ground water from the sand and gravel aquifer. This aquifer
is a shallow, unconfined system composed of coarse, unconsolidated sediments deposited by meltwater
streams during the Wisconsin glaciation. In the study area, the glacial deposits are thickest (200 feet)
2-5
-------
POST
GLACIAL
ALLUVIUM
AND MARSH
DEPOSITS
• •„•«.>:- -o • - • •'
• ;«. • ? *-• *
;^>:••'i.OV.r
^'•''•'\°\'---°:.'-'^'-'.:
'vV° •'-•'•?;-"-°;;°:'-0;'
CAMBRIAN .
^pS PREGLACIAL
&SWSS3 RANnRTOMP
SANDSTONE
PRECAMBRIAN
Figure 2-3. Surface geology map, Stevens Point, Wisconsin. (Modified after Holt, 1965, Plate 1.)
2-6
-------
in the preglacial entrenched Wisconsin river valley, thinning to the west where the Precambrian age
crystalline bedrock crops out. A hydrogeologic cross section of the Airport wellfield, shown in
Figure 2-4, illustrates the location of the buried valley and the variability in the saturated thickness.
Regional ground-water flow is toward the Plover River (Figure 2-5). The city wellfields are
within a single flow system that is delineated by the drainage basin for the Plover River. The portion
of the flow system upgradient of the city's wells is 3 to 7 miles wide, extending from 1.5 miles
southwest of the Airport wellfield to more than 30 miles northeast of the wellfield. Locally, the
ground-water flow system is influenced by aquifer heterogeneities, changes in topography, and ground-
water sources and sinks (for example, recharge events, ponds, irrigation, and production wells).
The city produces an average of 5 million gallons of water per day (Mgd) from the aquifer.
The period of maximum pumping occurred in June 1988. In that month, the city wells pumped more
than 300 million gallons of ground water from the sand and gravel aquifer. The city obtains water
from six wells (numbers 4 through 9) in two separate wellfields (Figure 2-2). City wells 4 and 5 (CW4
and CW5) are located at the Iverson wellfield. City wells 6, 7, 8, and 9 comprise the Airport wellfield,
located approximately 1 mile north-northeast of the Iverson wellfield. City well 4 is used only during
peak demand periods or emergencies, because of high iron and manganese concentrations.
NORTHWEST
1,150
SOUTHEAST
SAND & GRAVEL
AQUIFER
PREGLACIAL
RIVER VALLEY
1 MILE
SCALE
V WATER TABLE
Figure 2-4. Hydrogeologic cross section northwest to southeast across the
Airport wellfield, Stevens Point, Wisconsin.
2-7
-------
LEGEND
-1250-
WATER-TABLE
CONTOUR
(CONTOUR INTERVAL 50 FEET;
DASHED WHERE INFERRED)
GROUND WATER ..
DIVIDE «...
'GROUND WATER ^
FLOW DIRECTION
WELLFIELDS
AREA
MODELED
*.___.
•i
I
CITY LIMITS I ' T|
Kft^iUMM^J
I
-N-
024
SCALE IN MILES
^STEVENS* POINT
Figure 2-5. Recharge area for the Airport and Iverson wellfields, Stevens Point, Wisconsin.
(Modified after Lippelt and Hennings, 1981.)
2-8
-------
The majority of the recharge area for the city's wellfields has a DRASTIC index rating of 180
to 200, indicating a very high relative pollution potential (Aller et al., 1987; Portage County Planning
Department, 1988). A statewide analysis by the Wisconsin Geological and Natural History Survey
(WGNHS) indicates that the soils in the recharge area have low to moderate attenuation ability, and
recommends that land-use activities in the county be carefully managed (Good and Madison, 1987).
Land use in the recharge area is generally intensive, comprised of a combination of sewered and
nonsewered residential, commercial, agricultural, and undeveloped property. The majority of the
recharge area is beyond the city's jurisdiction. Much of the land immediately to the north and east of
the Airport wellfield is in the town of Hull, a rapidly growing, high-density, nonsewered residential
area.
PRELIMINARY WELLHEAD PROTECTION AREA
Wellhead protection zones proposed by the city (Figure 2-6) are based on estimates of the
ZOI, the 5-year TOT zone, and the recharge area for both wellfields. The ZOI ("A" Zone) was
delineated assuming a fixed radius of 1,500 feet. The 5-year TOT zone ("B" Zone) was based on a
fixed radius of 2 miles, calculated using a ground-water velocity of 6 feet per day (ft/d), a hydraulic
conductivity of 1,200 ft/d (calculated from aquifer-test results of the city wellfield), and a porosity of
approximately 25%. The ground-water velocity was calculated based on a maximum hydraulic
gradient of 25 feet per mile. The western boundary of the B Zone was modified to conform to the
hydrologic boundary. The recharge area ("C" Zone) is the remainder of the flow system within the
county, upgradient of the 5-year TOT zone. The C Zone boundaries were mapped hydrogeologically
and correspond to the boundaries of the Plover River drainage basin (Lippelt and Hennings, 1981).
The draft Portage County Groundwater Management Plan (Portage County Planning
Department, 1988) proposed that the city adopt protection zones for the Stevens Point wellfields. In
the review process, the validity of the B Zone, based on analytical calculations of the 5-year TOT, was
questioned regarding the representativeness of hydraulic conductivity and porosity values for the
aquifer and the use of a maximum hydraulic gradient. Consequently, the Management Plan was
published without determination of the preliminary WHPA and protection zones. One objective of
the current study is to verify, or refine, the boundaries of the ZOI and 5-year TOT zone through
further hydrogeologic characterization and ground-water modeling studies. This goal is addressed by
water-level mapping, aquifer testing at the Airport wellfield and at other wells in the recharge area,
and by the use of a two-dimensional numerical ground-water flow model.
SOURCE ASSESSMENT
Source Identification
A general source assessment of the recharge area for the city's wellfields was conducted during
this study, and a detailed evaluation of sources in the 5-year TOT zone was made. Existing and
potential sources of ground-water contamination were identified, classified, and mapped. Contaminant
sources were identified by researching existing and historical land use. The research methods used
2-9
-------
AIRPORT
WELLF1ELD
IVERSON
WELLFIELD
WELLHEAD PROTECTION ZONES
A ZONE: WELLFIELD ZONE OF INFLUENCE
B ZONE: 5-YEAR TIME-OF-TRAVEL ZONE
C ZONE: RECHARGE AREA
Figure 2-6. Wellhead protection zones (A, B, and C) for the Airport and Iverson wellfields,
Stevens Point, Wisconsin. (Modified after Portage County Planning Department, 1988, Appendix C-8.)
2-10
-------
included listing known ground-water contamination sources, incorporating existing data bases on
potential source locations, making a visual survey of the study area for potential sources, sampling
monitoring wells, and conducting an historical source assessment.
Existing ground-water contamination problems in Portage County include aviation fuel
leakage at the airport in 1988; gasoline detected in three private wells in the town of Hull in 1990;
nitrates in the village of Whiting's municipal wells; and aldicarb and atrazine pesticides detected in
rural, private wells. Concentrations of nitrogen, manganese, and iron in ground-water samples from
some of the city's monitoring wells have exceeded primary and secondary drinking water standards.
Elevated levels of iron and manganese are attributed to natural sources, whereas nitrogen is
anthropogenic. Sources of high nitrogen levels in the recharge area include leachates from septic
systems, lawn fertilizers, crop fertilizers, and animal wastes. Gasoline contamination of ground water
may occur due to leaks from underground storage tanks (USTs), spills, improper disposal of
petroleum wastes, and ruptures in aboveground storage or transport vessels. Pesticide sources include
applications to crops or lawns and accidental releases from storage facilities during transport or
mixing operations.
Land-use practices were identified and mapped as potential sources of ground-water
contamination (Table 2-1). Initially, potential sources were located based on the source assessment
conducted as part of the county's ground-water management plan. During this study, a more detailed
assessment within the B Zone was conducted. Businesses that store, transport, generate, or dispose of
potential contaminants were located by using city directories, local emergency government files, and by
visual survey. In addition, the location of USTs within the recharge area was determined from an
inventory of underground petroleum-product storage tanks, maintained by the Wisconsin Department
of Industry, Labor, and Human Relations. The location of hazardous material spills was identified
using the Wisconsin Department of Natural Resources spills list for Portage County.
An historical source assessment was conducted in conjunction with the U.S. Environmental
Protection Agency's (EPA) Environmental Monitoring Systems Laboratory in Las Vegas (EMSL-
LV). Lockheed Environmental Systems & Technologies Company (LESAT), under contract to
EMSL-LV, identified and documented the historical evolution of potential contaminant sources in the
B Zone by analyzing aerial photographs dating back to 1938 (Finkbeiner et al., 1991). The aerial
photographic source assessment was supplemented with data obtained from historical maps and city
directories dating back to 1922.
Following the local source data base investigation and photographic analysis, an inventory list
of existing and potential contaminant sources was compiled (Table 2-2). These sources require
prioritization to focus protection and management efforts within the WHPA. Table 2-2 also lists
detected and suspected contaminants associated with the point and nonpoint categories. For the
purposes of this report, septic systems are categorized as both point and nonpoint sources. Individual
systems are considered point sources. However, areas with a high density of nonsewered residences
are classified as nonpoint sources.
2-11
-------
TABLE 2-1. POTENTIAL CONTAMINATION AND DEGREDATION SOURCES WITHIN
THE B ZONE OF THE PRELIMINARY WHPA, STEVENS POINT, WISCONSIN*
Agricultural
Commercial
Industrial
Abandoned, and poorly
constructed wells
Feedlots
Fertilizer and pesticide
application
Fertilizer and pesticide storage
Fuel storage
Pesticide mixing
Sludge and septic spreading
sites
Airports
Cemeteries
Dry cleaners
Exterminators
Firing ranges
Fuel and chemical storage
Furniture refmishers
Garages (repair shops)
Gas stations
Golf courses
Junkyards
Laboratories
Landscapers
Paint stores and painters
Plumbers and welders
Printers
Asphalt plants
Concrete plants
Electroplaters
Formulators
Fuel and chemical storage
Machine shops
Pipelines
Transformers (electrical)
Residential
Waste Management
Naturally Occurring
Fertilizer and pesticide
application
Fuel and chemical storage
Septic systems
Incinerators
Injection wells
Landfills
Recycling facilities
Snow dump sites
Storm sewer and dry wells
Waste water treatment
plants
Organic-rich wetland soils
(iron and manganese)
Granite bedrock (radon and
radioactivity)
*Modified after U.S. EPA, 1990, p. 33.
Examples of potential and existing sources close to the Airport wellfield are shown in Figures
2-7 and 2-8. Typical point sources include USTs, spills, a landfill, waste and snow disposal areas, light-
industrial and commercial facilities (such as gas stations, auto repair shops, and dry cleaners), dry wells
(drainage wells in nonsewered areas), and perforated drainage pipes. Nonpoint sources include crop
and pasture lands, septic systems, nonsewered residential areas, golf courses, and institutional lawn
areas.
Farther from the wellfield, the character of the sources shift from urban and suburban to rural
and agricultural. This shift is marked by the abundance of point sources in the urban and suburban
area to the predominance of nonpoint sources in the rural, agricultural areas. On a regional scale, the
widespread application of pesticides and fertilizers in agricultural areas is considered a threat to the
city's wells. Large quantities of hazardous substances stored or handled within the recharge area may
2-12
-------
£
1
g
o
CO
1
i-i
o
CO
13
•J3
I
£
O3 --S
•2 I
s 1
cfl C
8
"S3
U3
•> tn
II
f9
1
«
"8
1
I
II
8 8
a. o.
•O T3 T3
a a 2
CO cd 03
S g S g
- '
S g
"
-g
a1
11
e a
» *
0
'
l ll'lfl
&
•a «
D rt
11
°3
I
CO
T
1
-S
fh
I
•o
«
o
c
03
t
•8
!
8
CH
W
O
'a
&
o
-B
a
I
2-13
-------
Figure 2-7. Typical point sources south of the Airport welffield, Stevens Point, Wisconsin.
(Institutional lawn area and golf course are considered as nonpoint sources for this study.)
2-14
-------
Figure 2-8. Typical nonpoint sources north of the Airport wellfield, Stevens Point, Wisconsin.
(Feedlot area is considered a point source for this study.)
2-15
-------
pose a significant threat to the city's wells. Large point sources farther from the wellfield include
junkyards, municipal landfills, feedlots, a salt pile, and high-density, nonsewered residential areas (for
example, trailer parks).
Source Characterization and Prioritization
Sources were prioritized on the basis of their category (point versus rionpoint), the degree of
their threat (existing versus potential), their proximity to the city wellfields, and the quantity and type
of hazardous or toxic substances handled. A qualitative ranking of sources and contaminants resulting
from this evaluation is given in Table 2-3. Rather than assign a numeric ranking scale for each
characteristic, existing and potential sources were grouped by category from greatest to least concern.
Existing contaminant sources were given highest priority. Next, potential sources were ranked
based on a combination of factors: source type, quantity, hazard, and location. The highest priority
contaminant sources are USTs, agricultural areas, high-density, nonsewered residential areas,
landscaped areas, and feedlots. Existing VOC, hydrocarbon, nutrient, and pesticide contamination in
the ground water is believed to be attributed to these sources. Large spills of hazardous substances
within the 5-year TOT zone and all spills within the ZOI of the city wells are also of highest priority.
TABLE 2-3. PRIORITIZED, POTENTIAL CONTAMINATION AND DEGREDATION
SOURCES WITHIN THE PRELIMINARY WHPA, STEVENS POINT, WISCONSIN
Level of
Concern
Greatest
Great
Moderate
Least
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Contaminant Source
Large (>500 gal) USTs at the airport, gas stations, and garages; small USTs
(B Zone) from residential and commercial areas; spills (A Zjone)— existing
VOC and hydrocarbon contamination
Agriculture, residential, and commercial areas— existing nonpoint nutrient and
pesticide contamination
Landfills (B Zone)
Urban runoff (B Zone)
Naturally occurring sources; granite bedrock and organic-rich wetland
Spills of hazardous substances (B Zone)
Residential runoff and hazardous substance disposal (B Zone NSAs)
Landfills
Storage tanks (C Zone)
Large spills (C Zone)
Junkyards (C Zone)
Road de-icing runoff and storage areas (B Zone)
Small spills (C Zone)
Low-density NSAs (C Zone)
soils
Abbreviations: USTs = underground storage tanks; VOCs = volatile organic compounds;
NSAs = nonsewered areas.
2-16
-------
Iron and manganese ground-water degradation exists because of the presence of organic-rich soils
high in wetland areas along the Plover River. Reducing conditions in these soils increases metal
solubility, therefore increasing the concentrations of iron and manganese in ground water. These
constituents are of less concern because the drinking water standards for iron and manganese are
secondary, based on aesthetic considerations rather than primary health factors. Other sources of
great concern within the B Zone include landfills, perforated storm pipes and dry wells that drain
urban runoff, granite as a natural source, and spills. Large point sources in the C Zone [for example,
fuel storage tanks larger than 1,000 gallons (gal)] and de-icing salts in the B Zone are classified as
moderate threats. Point sources in the C Zone are farther from the wellfield, allowing for greater
dilution and reaction time if contamination occurs. Another factor in assigning moderate priority to
these sources is that the health hazards from sodium and chloride are less than those associated with
VOCs and pesticides. Of least concern are small, potential sources located outside the 5-year TOT
. zone, such as low-density, nonsewered residential areas (less than one home per 5 acres) and small
spills.
2-17
-------
RESEARCH MONITORING PROGRAM
The ground-water monitoring program has two components: research monitoring and a
proposed long-term monitoring program. The purpose of the first stage, research monitoring, is to
gather reliable physical and chemical information in order to develop an accurate conceptual model of
the hydrogeologic system. Specifically, research monitoring may lead to a refined WHPA delineation,
flowpath assessment, or better hydrochemical characterization of the aquifer and contaminant sources.
The second stage is monitoring of ground water in the WHPA for the long-term preservation of the
wellfield. The design of a reliable wellhead protection monitoring system is based on: (1) thorough
source assessment, (2) accurate delineation of the WHPA, and (3) hydrochemical characterization and
flowpath evaluation.
DATA BASE LIMITATIONS
In the first phase of research monitoring, the existing data base was reviewed to ascertain the
hydrochemical and hydrogeologic characteristics of the sand and gravel aquifer. Hydrogeologic
mapping and investigations of aquifer properties in the Stevens Point area include work by Holt
(1965), Hickok and Associates (1965 and 1981), Weeks and Stangland (1971), Lippelt and Hennings
(1981), Renaud (1987), and Donohue and Associates (1988). In addition, a number of hydrochemical
studies focused on the sand and gravel aquifer. Renaud (1987) studied the occurrence of manganese in
the vicinity of the Airport wellfield. Several studies on pesticide occurrence and transport through the
sand and gravel aquifer supplement the existing data base (for example, Harkin et al., 1984; Jones,
1987; and Portage County Planning Department, 1988). Other research on hydrochemistry included a
local study on nitrogen contamination from feedlots (Bowen, 1987) and a county map of soil
attenuation potential (Good and Madison, 1987).
All of these studies were useful in the development of a detailed conceptual model of the sand
and gravel aquifer. However, none focused on the combined physical and chemical hydrologic system
for the city wellfields. These studies present an incomplete description of the hydrogeologic system.
They either address in detail a small portion of the aquifer or discuss the entire regional hydrogeologic
system. Consequently, the existing data base requires supplementation before it can support the
objectives of the wellhead protection monitoring program.
MONITORING OBJECTIVES
The second step of the research monitoring program was to refine or verify the conceptual
hydrogeologic model. The goals associated with this step were to supplement and fill gaps within the
existing hydrochemical and hydrogeologic data bases. Deficiencies in the data bases include
information on: (1) the vertical and lateral distribution of hydraulic conductivity in the sand and
2-18
-------
gravel aquifer; (2) the local ground-water flow patterns; (3) the extent of the ZOI for the city
wellfields; and (4) the lateral, vertical, and temporal distributions of contaminants in the vicinity of the
Airport wellfield.
RESEARCH MONITORING TASKS
Tasks performed in support of the monitoring objectives included:
Installing new monitoring wells
Sampling existing and new monitoring wells for suspected contaminants
Measuring water-table elevations
Conducting aquifer tests to determine hydraulic conductivities
All hydrochemical sampling during the research monitoring phase was conducted to document spatial
and temporal trends in ground-water quality. Physical measurements were made to refine the WHPA,
assess predominant flowpaths, and improve the conceptual hydrogeologic model.
DATA ACQUISITION
Monitoring Wells
Existing wells—
During the research monitoring stage of this project, a network of 44 single monitoring wells,
10 well nests, and one multi-level well were used to. measure water levels, to sample ground water for
chemical analysis, and to conduct aquifer testing (Figure 2-9). Thirty-five of the single monitoring
wells, six of the well nests, and the multi-level well existed prior to this study.
The existing monitoring wells were installed for a variety of projects and reasons. Fourteen of
the existing monitoring wells (MW1 through MW11, and MW25 through MW27) are shallow wells
installed to measure water-table elevations and to monitor shallow hydrochemistry in the recharge
area north and west of the Airport wellfield. Similarly, 18 shallow monitoring wells (MW12 through
MW22, and MW29 through MW35) were installed to verify water levels and shallow hydrochemistry
on the east side of the Plover River. All of these wells were drilled using a hollow-stem auger. The
well construction for the single monitoring wells consists of 1.25-inch, flush-threaded, polyvinyl
chloride (PVC) casing, a 3-foot screen, natural, collapsed sand filter, native sand backfill, and a 2-foot
bentonite seal at the surface (Figure 2-10A).
One of the monitoring wells, MW24, is part of the monitoring system established at the
airport after the aviation fuel leak in 1988. Two existing monitoring wells, MW23 and MW28, have
galvanized-steel casings. One is an old test well at the Airport wellfield (MW23), and the other is an
old production well for the airport (MW28).
Well nests A through F and well G were installed to monitor vertical gradients and water-level
fluctuations, as well as manganese and iron concentrations between the Airport wellfield and the
Plover River. Well nest G was originally installed as two wells at different depths in one borehole
2-19
-------
MONITORING
H (SINGLE)
(NEST)
(MULTI-LEVEL)
Figure 2-9. Research monitoring wells, Stevens Point, Wisconsin.
2-20
-------
(A) SINGLE DRILLED WELL (MW 38)
PROTECTIVE CASING
LOCKING CAP
•GROUND SURFACE
-BENTONITE SEAL (2 FT)
-1.25-IN SCH 80 PVC CASING
-SAND BACKFILL
-(25 FT)
-COLLAPSED SAND
-4-IN BOREHOLE
3-FT WELL SCREEN
"(0.01-IN SLOT)
-WELL DEPTH (42 FT)
WELL NEST (K)
n
11
T
7
'/
/
•.•»:
t
*;!
:'•:
T9
'
•it*
O'JO;
'%
iv.°',
:•:;
*^-
56
t
*
*$
ft?
^
;-p
tf
=•>
;••'•
»y
'•'?•'•!
•ji^
__^_
PROTECTIVE
CASING WITH
LOCKING CAP
2-IN SCH 40 PVC
CASING
GROUND
BENTONITE
SEAL (5 FT)
£ = \ (34 FT)
NX
(37 FT)
6-IN BOREHOLE
-(58 FT)
COLLAPSED
SAND
5-FT WELL SCREEN
(0.01-IN SLOT)
•WELL DEPTH (79 FT)
PROTECTIVE CASING
"WITH LOCKING CAP
-GROUND SURFACE
.BENTONITE SEAL
(2 FT)
1.25-IN SCH 40 PVC
"CASING
3-FT WELL SCREEN
1-IN SLOT)
WELL POINT DEPTH
(10 FT)
Figure 2-10. Research monitoring well construction diagrams for (A) single, drilled well;
(B) well nest; and (C) hand-driven well.
2-21
-------
(Renaud, 1987). Later, the deeper well was decommissioned, leaving a single well at shallow depth.
Well nests A through F consist of two to three single wells at different depths (Figure 2-10B). These
wells are constructed of 1.25-inch, flush-threaded, PVC casing, a 3-foot screen, natural, collapsed sand
filter, native sand backfill, and a 5-foot bentonite seal at the surface. The multi-level well, IM1, was
positioned to monitor hydrochemical variations with depth, upgradient of CW5 (Figure 2-2). The
multi-level well is essentially a bundle piezometer, consisting of a series of 1.25-inch PVC main
structures (supporting core) with a series of 0.25-inch polytubes attached. The monitoring well was
designed to detect changes in hydrochemistry at relatively small vertical intervals.
New Wells-
New monitoring wells were installed for a variety of reasons. Three of the wells, MW36
through MW38 (Figure 2-9), were installed around CW6 (Figure 2-2) as observation wells during
aquifer testing. Five monitoring wells, MW39 through MW43 (Figure 2-9), were added on the east
side of the Plover River for water-table elevation measurements. Four well nests, J through M
(Figure 2-9), were installed to monitor hydrochemistry at depth at identified gaps in the existing
monitoring well network.
All of the water-level monitoring wells were constructed as shown in Figure 2-10A, using
1.25-inch PVC pipe, 3-foot screens with 0.01-inch slots, and a 2-foot bentonite seal at the top of the
well. All the new monitoring wells, with the exception of MW42, have PVC casing. Galvanized-steel
pipe was used for MW42 so that it could be driven to a greater depth to monitor vertical hydraulic
gradients. Monitoring wells 39 through 43 were installed as hand-driven well points (Figure 2-IOC)
because the well sites were physically inaccessible to a drilling rig. The three wells installed adjacent to
CW6 (MW36 through MW38) were drilled with a small, hollow-stem auger rig.
The well nests were installed with a large drilling rig by pushing 6-inch steel casing to bedrock,
cleaning the casing with a bailer, installing a 2-inch PVC pipe, then pulling up the steel casing to the
next depth (Figure 2-10B). At this depth, a 2-inch pipe was installed. Then, the casing was pulled up
to just below the water table and the last 2-inch pipe was installed before removal of the casing from
the borehole, addition of the native sand backfill, and addition of a 5-foot bentonite seal. Two-inch
PVC pipe and 5-foot to 10-foot screen lengths with 0.01-inch slots were used for well-nest
construction. The nests were designed with 2-inch pipe so that a submersible pump could be used to
adequately develop and purge the wells.
Well nests were installed where data gaps existed, both vertically and laterally in the system.
Well nest J was positioned to determine the hydrochemistry with depth near the center of the buried
river valley. This nest is the deepest in the monitoring network (160 feet). Well nest K was installed
to obtain more data between the airport and the Airport wellfield. Well nests L and M were
positioned to determine the hydrochemistry with depth between the Airport wellfield and two
nonsewered residential and commercial areas. In general, each well nest consists of three separate
wells screened at variable depths: one immediately above bedrock, one just below the water table, and
one halfway between the other two. Well nest J consists of six wells in two separate boreholes: one
borehole with three wells extends to bedrock; the other, with three wells, extends to half the bedrock
depth. The wells in the J nest boreholes were screened at various depths: one just below the water
table, one immediately above bedrock, and the intermediate wells at depths based on changes in
lithology.
2-22
-------
Construction diagrams for the three types of monitoring wells and nests are shown in
Figure 2-10A, B, and C. Saturated sand and gravel collapses around the well casing, creating a natural
filter pack for each well near the bottom of the borehole. Native sand is then used as backfill for the
open borehole up to the bentonite seal. Each monitoring well and well nest has a protective steel
casing with a locking cover. After the wells were installed, they were each developed by surging and
pumping until the issuing water was clear. The elevations of the mark point for each monitoring well
were surveyed to the nearest ±0.01 foot.
Physical Parameters
Two physical parameters, water level and specific conductivity, were determined each time a
monitoring well was sampled. Water levels were measured twice for mapping purposes in January and
March 1990 at all the monitoring wells within a 24-hour period. During this 24-hour period, no
precipitation occurred to recharge the aquifer system. During January, several of the wells were
inaccessible because of freezing or burial. Water-level measurements were obtained using a "popper"
attached to a fiberglass measuring tape. The "popper" consists of a brass weight with a concave
bottom surface that makes a sharp popping sound when it encounters the free-water surface in a well.
Water-level data are compiled with other well information and hydrochemical results in Table 2-4.
Hydraulic conductivity values were determined by means of slug tests (Bouwer and Rice, 1976;
Horslev, 1951) at select monitoring wells and a 24-hour drawdown (constant discharge) and recovery
test at CW6 (Ferris et al., 1962). Slug tests were conducted using a PVC slug, a data logger, and a
pressure transducer. The slug tests were performed by "dropping" the slug into the water ("slug in")
and recording the response of the aquifer as it returned to the original static water level. After the
water level stabilized, the slug was removed ("slug out"), and the response of the aquifer was recorded.
Both 1.25-inch and 2-inch wells were used for slug testing.
Multiple slug tests were conducted at MW36, J78, J118, J160, MW13, K58, and K79. (The
number following well nest designations J and K indicates the depth of the bottom of the well screen
within the nest.) These wells were selected to determine variations in hydraulic conductivity of the
sand and gravel aquifer in the vicinity of the Airport wellfield. Monitoring well 36 was selected as an
observation well for the drawdown and recovery test at CW6 so that comparisons could be made with
the slug-test results. Test comparisons were done to determine if the buried river valley has a
consistently higher hydraulic conductivity than other areas of the aquifer. Slug tests at well nest J were
conducted to verify the high hydraulic conductivity value obtained during the drawdown and recovery
test. Monitoring well 13 is located on a bedrock high. During sampling, it was noted that the
monitoring wells completed above bedrock highs could be purged dry with a bailer and would recover
slowly. Monitoring well 13 was selected for slug testing to derive the lower range of hydraulic
conductivity for the aquifer. Finally, well nest K was selected for testing to verify hydraulic
conductivities of the sand and gravel aquifer in areas other than the buried river valley and bedrock
highs.
2-23
-------
Ji
8
g
c<3
8
8
S
1 1
1
I
8
8
2-24
-------
1
a
V ,
I
8-
S
tet Q
O
i
Q
s'
s
S
«
•*
ol
8
eo w oo op
ssss
s
3
ts
1-'
S
0
1
o\ r^
r^
ts
3
1
s
i
8
0
§
S
O i-t i-l CO
^ C*l ^"
s
a
2-25
Q
Z
S
a
25
* oo
s
S
•I
-------
I
B
2
ffi
8
H
II?
•H-
s s
s
- S =i CJ
I
s
S
2-26
-------
•c
1
I
2
ter Q
s
I
§
I
8
S
S
"" —- x-s
Sis
fc 5:
2 « •
1
3
§5
3
i
0
sd
1
s
s
cs
I
3
s
^t
Q
1
»-4 r- m
vl 1^ H
O\ *-) rj-
ssl
a
§
(S
§
i
r- p\
«*! F*
o\ o\
oq oo
T-I _ . Ti- o
T-I S^ M CO
«
ss
1
S
85
§5
2-27
I
•a
-------
f
I
2
§
I
8
S
I 1
in oq
§
S
=3 o'
(Si
58
a s
a s
000
d d
S S
S
2-28
-------
!
f
g
I
£
i
I
§
I
S
PL
PC
H
£
•5"
S S
0 CJ
1
00
o
1
o ej
CO 0
•
Q
•*
S
0
§
SOD
O
s
0
1
M
§
o
1 i
CM
t-i
CM
0)
o
CS
CJ
0
S
1 1
s
1 1
S'
ei
SB
S
S S
5
.
8 3 S
in \d m
S S S
-r
OJ 00 5
^ C 2
sss
^ r» oo
i^ OO t""
H tj H
I** i^> r*
O 0 0
in m o\
oj vo t-;
CS **J <*1
t-- r^ r-
ss s
s s
s s
a
is4
-
1
s
63
Q
*
NO -^ 1-1
sis
sS«
F
3
S
§
a
g
2-29
-------
8
8
I
lit
8
a
I:
SRS3
s
o\ oo t»
^-
.8
vd
o o\
s
°o 3
23 ^
cu 6 -S-
253---
g
8
s
i
2-30
-------
I
•«
w
2
H
i
I
1
a
«
3'
3
^t
\—' O
a
I
* t •=•
saS
^<=-
1
s
vi o
»— i to
a
S
JS 1- 0
P^ f* C*l
£f>S
000
S
Is
2
S
2-31
-a-a
mi
?-i|f
1| -a-a
ja S ig -g
•a -s
-------
Drawdown and recovery testing at CW6 was performed in conjunction with the University of
Wisconsin - Madison. The drawdown test was performed using methods outlined by the U.S.
Geological Survey (USGS, 1977). City well 6 was chosen for drawdown and recovery testing because
it has the pumping capacity to produce a significant drawdown; it is located at the Airport wellfield,
and it has a convenient observation well nearby, MW23. Three additional observation wells, MW36
through MW38, were installed around CW6 for this aquifer test. At the end of the drawdown portion
of the test, water-level recovery rates were measured.
Chemical Parameters
Ground-water sampling for laboratory analysis was conducted using the methods established
by the Wisconsin Department of Natural Resources (1987). The samples were analyzed by state-
certified laboratories. The majority of samples were analyzed at the Environmental Task Force
Laboratory at the University of Wisconsin - Stevens Point; a few of the organic analyses were
conducted by Enviroscan hi Wausau, Wisconsin.
The first set of ground-water samples was collected in January and February 1990 (Table 2-4).
Ground-water samples from all the existing wells were analyzed for specific conductivity (Cond), pH,
sodium (Na), chloride (Cl), nitrate and nitrite as nitrogen (N), manganese (Mn), iron (Fe), zinc (Zn),
chemical oxygen demand (COD), alkalinity (Alk), and hardness (Hard). Some of these parameters
were selected to screen for potential contamination problems, including Na, Cl, N, Fe, Mn and Zn.
Others were selected to maintain continuity with the existing data base (COD and Hard) and to serve
as indicators of general water quality (Cond, Alk, and pH).
For the second set of samples collected in May and June 1990, select wells were chosen for
VOC and pesticide analyses (Pest) (Table 2-4). Dissolved oxygen (DO) was also added as a field-
monitored parameter for select wells during the second sampling round. Wells were selected for
sampling if they were positioned downgradient from or near a potential source. For example, MW1,
MW2, MW6, MW17, and MW18 are located in areas where there are numerous, small gasoline USTs;
MW12, MW15, MW16, MW25, and MW27 are downgradient from large, commercial USTs at
gasoline stations or the airport; and MW16, MW20, and MW28 are located in areas with dry wells.
Some wells located at the outside edge of the monitoring network, closest to the agricultural areas,
were sampled and analyzed for nitrogen- and phosphorous-based pesticides. A nitrogen- and
phosphorous-based pesticide analysis was chosen because many of the pesticides used in this area are
based in these chemicals, including atrazine, alachlor, metolachlor, and metribuzen. Also, during the
second set of tests, samples from selected wells were analyzed for the January parameter list, with the
exception of zinc and sodium. Wells selected for the second round of sampling included those with
high concentrations of a particular substance and those in optimal locations. Fewer wells were
sampled and fewer parameters were analyzed because of project budget limitations.
For the third set of analyses, during October 1990, the new well nests were sampled for the
inorganic parameters listed in the previous sample set (Table 2-4). The J nest wells were also sampled
for the nitrogen- and phosphorous-based pesticide analysis. The shallow wells in the K, L, and M
nests were sampled for VOCs.
2-32
-------
DATA INTERPRETATION
Aquifer Parameters
The drawdown test at CW6 on June 5, 1992, and observation data from MW36 confirmed the
results of the aquifer test by Hickock and Associates (1965). Analysis of a logarithmic plot of
drawdown versus time shown in Figure 2-11 (Neuman, 1974) indicates an aquifer transmissivity (T) of
1.41 fl^/s (Bahr, 1990). The semi-log plot analysis of drawdown versus tune for MW36 observation
data shown in Figure 2-12 (Cooper and Jacob, 1946) yields a T value of 1.10 tf/s (Bahr, 1990). Given
that the saturated aquifer thickness (b) at CW6 is 90 feet and that the hydraulic conductivity (K) is
equal to T/b, then the respective K values for the logarithmic and semi-log analyses are 1,300 and
1,100 ft/d. An average K value of 1,200 ft/d was chosen to represent the buried valley aquifer at CW6
(Table 2-5).
Slug-test data were interpreted using Horslev's (1951) empirical method. Slug tests performed
at MW36 yielded a K estimate of 1,100 to 1,700 ft/d (Figure 2-13; Table 2-5). The range in this
estimate results from uncertainty in the results of the slug test. The high degree of uncertainty is due
to the very rapid response of the aquifer to the displacement of fluid by the slug. Three sets of slug-in
and slug-out tests were conducted at MW36 (Figure 2-13). Of these three sets, one slug-in test was
timed correctly. The water levels in MW36 recovered completely in less than 0.04 minute, or
approximately 2.4 seconds. The highest estimate of hydraulic conductivity (1,700 ft/d) is obtained
from the slope of the early-time data (straight line) at a t0 of 0.0016 minute: that is, within the
first 0.01 minute of the test. The lower hydraulic conductivity (1,100 ft/d) estimate is obtained at a t0
of 0.0025 minute using a power regression fit (curved line) to the time data points up to and including
0.07 minute. After 0.01 minute, the water-level data begin to fluctuate. Despite the uncertainty in the
results, the slug test at MW36 yields a range of hydraulic conductivity values from 1,100 to 1,700 ft/d
that bracket the hydraulic conductivity obtained for the drawdown test of 1,200 ft/d.
Similar problems occurred with the slug testing at well nest J. Four sets of slug tests were
conducted at well J160. Two sets of tests were conducted at well J118 and two sets at well J78. Of
these eight total tests, only one, a slug-in test at J160, was timed properly. In this test, water levels
dropped to within 20% of the maximum displacement in less than 0.005 minute (Figure 2-14). When
the slope of the earliest-time data points is used, an estimate of hydraulic conductivity of 950 ft/d is
derived. From the later time data, a value of 820 ft/d is derived. This value indicates that the buried
channel has a characteristically higher hydraulic conductivity than the adjacent aquifer material, as
indicated by previous drawdown results. (The reader will note that well nest J is positioned near the
center of the buried valley.)
The slug-test results from well nest K (Figure 2-15) and from MW13 (Figure 2-16) are
considered more reliable because the aquifer did not recover as rapidly and the data are more accurate.
Slug tests at well K58 yielded a range of hydraulic conductivities from 220 to 260 ft/d for late- and
early-time data, respectively. This hydraulic conductivity range (Table 2-5) is typical for the range
reported for outwash deposits in other studies, from 100 to 500 ft/d (Faustini, 1985; Rothchild, 1982;
2-33
-------
on
O
CO
CO
>
>
-J
_l
LU
£
Z
O
i
DC
LU
CO
CQ
O
Cp
^"
L_
CO
LU
Z
o
Q
^^^
1
DC
Q
:
. "
• -
• *; ^ . -
• •*- >*—
•
• 00 o
• ^ 52.
• i^ .^^ ~
• •*- •***•
*t ° CO — CO "O
• co rr- CM ^ ^ —
• coco'iL 4±*->^--
OS^I T- » § :
O O 'fr ^ ^ O W .
I 1-: i-: i-1 1— ZZ-o-r-.
^ * II II II II II II H
o : •- • *
Q- :
x :
o :
i •
* * *
4- C •
«'E '•
0 O CM t- .
T^ T-: co to" |
II II II II ' 1
co O OT ~ \ —
CM pr ! ~
4W CO •
h-
N. *•
CM .
O •
«- • I
II •
^ ^
*
~
,,i i 1 I , 1 l.ii i 1 I I 1 l,,i I 1 1 i 1
°-
O
O
o
^
T—
O
T—
O
OLD CM T— IO CM "«~ IO CM -I~O
*- CD CD CD 0 9 °
000
'c^ *"
1 ^
c 5
*^ II
111 "jj"
^ CO
i- §2
>
2
0
h-
CD
1
o
cc
LL
LU
O
1
Q
LU
CC
LU
X
^
co
CO
CD
II
O
LU
CD
DISCHAR
LU
"co"
CM
1-
SMISSIVIT
z
cc
to
CM
d
II
CO
Q
LU
O
LL
O
LU
0_
CO
"m
+-*
M—
L_
1^
K
0
Q
O
O
g
i
DC
Q
X
O
O)
II
CO
CO
LU
0
X
cc
LU
LL
i
>w
*o5
^
•5
•fcrf
w
00
1
a
d
o
I
00
1
•o
&
«-H
1
0
1
1
*2
3
1
£
|
1
•o
•s
•4-t
"S.
o
1
6
Si
^
•
rH
C^l
o
1
• »l
y^pv
43
^
0
§
II
N
15
1
%
1
f*
1
0
r-1
I
v""'
.S
i
8
•*-•
.a
1
r-H
§
3
^
t4
3
ff
•s
unconfine
t-<
o
§
ts
a
S3
1
3
I
€
1
s
*
1
}
|
O
1
s
(U Uj s) NMOaMVHQ
tL,
2-34
-------
CO
CO
LU
I
tr
LU
CO
CQ
O
CO
LU
DRAWDO
W
fe£
•* co
-—. CO CO
o s^r
CO CO "~
o o o
CO CO t-
cvi.cvi T^
II II II
o
en
n
(O -Q
O
O
O
o
o
Ct->
M—
E £
c -^
— •*
^ II
UJ
CM
q
o
CO
•5
CO
CO
If
II
O
II
CO
II
CO
CO
CO
^
0
S
DISTANCE FRO
LU
CD
tr
WELL DISCHA
^^
t-
TRANSMISSIVI
o
SPECIFIC YIEL
0
z>
Q
Z
O
HYDRAULIC C
LU
cr
01
a 3
W ^ MH
iZ C O
S^
wT
« '3
(W U| s) NMOQMVaa
2-35
-------
TABLE 2-5. AQUIFER TEST RESULTS FOR GEOLOGIC TERRAINS WITHIN THE
SAND AND GRAVEL AQUIFER
Well
CW6/MW36
MW36
J160
K58
MW13
Terrain
Buried valley t
Buried valley
Buried valley
Outwash
Bedrock high
Aquifer Test
Drawdown —
logarithmic
Drawdown —
semi-log
Average
Slug — late time
Slug ~ early time
Slug — late time
Slug — early time
Slug — late time
Slug — early time
Slug — late time
Slug — early time
K(ft/d)
1,300
1,100
1,200
1,100
1,700
820 .
950
220
260
2.1
3.5
Holt, 1965; Jones, 1987). Three slug tests at MW13 resulted in a hydraulic conductivity range of 2.1 to
3.5 ft/d for the bedrock high area (Figure 2-16). During purging of the monitoring wells, significant
variability in hydraulic conductivity was noted in association with slower water-level recovery rates in
some wells. Slug tests quantified these observations.
The hydraulic conductivity values calculated from the results of the drawdown and slug tests
are summarized in Table 2-5. The lateral variation in hydraulic conductivity values indicates that the
system is not as homogeneous as previously assumed due to the presence of distinct geologic terrains
within the sand and gravel aquifer, including the buried valley, bedrock highs, and outwash plains.
The aquifer-test results were incorporated in the model of ground-water flow used to refine the
WHPA and to optimally site monitoring wells.
Hydrochemical Assessment
The results of the hydrochemical analyses for the monitoring wells are compiled in Table 2-4.
These data provide the city with the foundation for long-term ambient trend monitoring. Because
these are the initial hydrochemical data in the data base, the analysis of seasonal trends is premature
and the analysis of long-term trends is unfeasible. However, the hydrochemistry of the city wells has
been monitored over a long period of time, and some trends are evident.
2-36
-------
CO
CO
CO
LLJ
CD
CM
10
p =5-
<=> ^
' II J^
«r co ^
II II II
§
Q
I Z
ii
DC
LU
X
.a e
(OH-H / M-H)
QV3H Nl 39NVHO 3AI1V13U
2-37
-------
o o
in CM
O) 00
in en
CM CM
o o
o o
d o
O
CO
CO
LLJ
O
CO
m
CO
q
d
CO
q
o
10
CM
q
CD
CM
q
o
to
q
d
q
c>
in
o
o
It
V) '
c
1
LJJ
m
S c»
P g
o c>
II O
CM
CM T-
• •
O O
in
q
d
CM •«-
q q
ci d
2c_
§2g
II I! ^
DC _J ^
II II II
h-
O
I-S
£gi
_J LU DC
-J CC Q
38£
oc
LU
I
(°H-H / M-H)
(U) QV3H Nl 3ONVHO 3AI1V13U
2-38
-------
00
LO
CO
LLJ
H-
O
.s
1
3
ta
CO
CO •!-•
O "- TJ
c> in ^r
II II ^
DC _l ^
II II II
o
Q
«|i
2 m
Q _J
oc gj <
_| UJ CC
_l DC Q
HI O >-
^ CO X
.
II
b=i O
LU
DC
111
(U
O3
>o
T-l
C^
u
(°H-H /M-H)
QV3H Nl 3ONVHO 3AI1V13U
2-39
-------
CM
IO ^
P *- T?
O CO ^
II II &
CC _l *
II II II
O
co|8
i LU O
Q _i i]
tt ul <
—I LU CC
rrJ CC Q
yj o >
^ CO I
UJ
cc
uu
i
C)0
.(°H-H / M-H)
QV3H Nl 39NVHO 3AI1V13U
2-40
-------
Nitrogen (nitrate and nitrite as nitrogen) concentration trends in city wells 6 through 9 from
1984 to 1990 are depicted in Figure 2-17. Although the pattern and concentration for each well differ
and seasonal trends are difficult to discern without further analysis, there is a general increase in the
concentration of nitrogen over time. In fact, since 1984, the peak concentrations in nitrogen have
increased in all the city wells. Currently, nitrogen concentrations in the monitoring wells range from
less than 0.2 mg/1, in several wells, to 26 mg/1 in MW2 (Table 2-4). Monitoring well 2 is a shallow
Well located downgradient from a nonsewered trailer park (Figure 2-9). The location of monitoring
wells in which the ground water exceeds the State Preventive Action Limit (PAL), or drinking water
standard, for nitrogen are shown in Figure 2-18. All of these wells are shallow wells, either directly
downgradient from high-density, nonsewered residential areas or agricultural activities.
Nitrogen concentrations also vary with depth, as shown in Figure 2-19. The monitoring wells
with the highest nitrogen concentrations are shallow wells. In contrast, the deeper monitoring points
in the multi-level well, with the exception of IM1-155, have nitrogen concentrations between 5 and
10 mg/1. Assuming that the hydrochemistry of the deeper ground water represents a more regional
flowpath, the nitrogen in this portion of the aquifer may be derived from distant agricultural areas.
The lower concentration of nitrogen in IM1-155 may be a result of dilution or alteration.
Monitoring wells with high concentrations of iron, manganese, and chloride are shown in
Figure 2-18. The majority of the monitoring wells exceeding drinking water standards for iron and
manganese are in wetland areas with organic-rich soils located along the Plover River. The reducing
conditions associated with these environments increases metal solubility and may cause higher
concentrations of iron and manganese in ground water. All of the monitoring wells with elevated
chloride levels are shallow wells located near roads where de-icing salts are used during the winter.
Only one ground-water sample, from MW23, showed detectable levels of organic compounds
when analyzed for VOCs. Monitoring well 23 is a hand-driven, galvanized well point—one of the test
wells installed in the 1960s to investigate the Airport wellfield site. The source of the VOCs in this
well is believed to be a coating that was applied to the inside of the pipe by the manufacturer. With
the exception of DEBT (the active ingredient in mosquito repellent used by the samplers), no
nitrogen- or phosphorous-based pesticides commonly used in this area (for example, atrazine,
alachlor, metolachlor, and metribuzin) were detected in the research monitoring wells.
Ground-Water Flow Modeling
The two-dimensional, ground-water flow model FLOWPATH (Franz and Guiguer, 1989) was
used to refine the delineation of the 5-year TOT zone. The area of interest, shown in Figure 2-5
(dashed line), was modeled using the hydrologic boundaries to the east and west (no-flow boundaries)
and constant-flux boundaries to the north and south. The constant-flux boundaries were determined
by using Lippelt and Hennings' (1981) regional water-table map and by calculating the component of
flow perpendicular to the boundary, based on the hydraulic gradient and the hydraulic conductivity.
2-41
-------
ON
f
Zin
1"
ll
-68/S yj
-88/21<
*8/8
a*
8
'cs
1
.Is
"Cn
•a
•^.^
8
I
CO
OJ
2-42
-------
2-43
-------
150
200
a
a
5 10 15 20 25
NITROGEN (NITRATE AND NITRITE) (mg/l)
30
Figure 2-19. Variations in nitrogen concentration with respect to well depth throughout
the research monitoring well network, Stevens Point, Wisconsin.
Variations in bedrock elevation (Figure 2-20) and hydraulic conductivity were included in the
model. Bedrock elevations were based on mapped well log data and geophysical interpretations
(Osborne, 1988). Hydraulic conductivity values derived from aquifer testing were used in the model.
For the buried valley, the hydraulic conductivity was assumed to be 1,240 ft/d. This value assumes that
the hydraulic conductivity derived from the aquifer test at CW6 is an average conductivity for the
saturated thickness of the aquifer in the buried valley. The hydraulic conductivity in areas overlying
bedrock highs was assigned a value of 100 ft/d. Slug-test results from MW13, representative of
the bedrock high areas, indicate significantly lower hydraulic conductivities ranging from 2.1 to 3.5 ft/d
(Table 2-5). The decision to assign a higher value to the bedrock high areas (100 ft/d) was based on
the assumption that the slug test investigated a small portion of the aquifer with lower hydraulic
conductivity. The slug test was performed in the screened interval set in decomposed granite.
Realistically, the average hydraulic conductivity of the bedrock high areas is probably one to two
orders of magnitude higher than the slug-test results indicate, considering the presence of sand above
2-44
-------
X
a
I
I
00
.<£
a
o
"8
w
fc
UJU
—1LU
UJu.
2-45
-------
the decomposed granite layer. The hydraulic conductivity for the remainder of the aquifer was initially
assigned as 320 ft/d based on the average conductivity for the glacial outwash from previous studies
(Holt, 1965; Faustini 1985; Rothchild, 1982).
Surface-water nodes were incorporated to simulate the effects of the Plover River and a small
lake to the west of the Airport wellfield (Figure 2-9). Leakage rates for the river and the lake were
calculated for the model on the basis of the following equation (Franz and Guiguer, 1989, p. 19):
where,
d =
Lr = x R
d
leakage factor in (d)'1
vertical hydraulic conductivity of the river or lake bed in ft/d
thickness of the river or lake bed in feet
correction factor for the ratio of surface areas of the finite-difference block and the
river or lake in the block
For the Plover River, the hydraulic conductivity varied from 0.01 to 0.1 ft/d times the average
hydraulic conductivity of the aquifer (320 ft/d) for calibration purposes, resulting in a K, range of
3.2 to 32 ft/d. This range is effectively one to two orders of magnitude less than the average hydraulic
conductivity of the aquifer. Leakage rates range between values that are considered reasonable for
river and lake deposits (Freeze and Cherry, 1979). Similarly, leakage from the small lake was assumed
to be three to four orders of magnitude less than the average hydraulic conductivity of the aquifer,
0.03 to 0.30 ft/d.
The estimated average annual recharge for the sand plain aquifer, 10 inches, was used as input
to the model (Portage County Planning Department, 1988). The June, 1988, pumping rates were used
as discharge values in the model. The range of pumping rates for June discharges are 1.06 Mgd to
2,48 Mgd for the Airport and Iverson wellfields. Summer discharge rates were chosen for modeling
purposes, as opposed to winter discharges, to simulate maximum stresses on the aquifer.
The model was not calibrated rigorously to specific head measurements because sufficient data
for comparison were lacking. The hydrogeologic mapping performed during this study covers only a
portion of the area of the model. In addition, the hydrogeologic data were obtained during transient
conditions—the aquifer had been stressed by periodic, high-capacity well pumping and was responding
to recent recharge events. The model used for this study, FLOWPATH (Franz and Guiguer, 1989),
assumes steady-state conditions. Regional water-table maps are available. However, these maps are
based on data collected from a variety of sources over a number of years. The regional water-table
maps are useful for determining general ground-water flow directions and regional flow boundaries;
however, they do not represent steady-state hydrogeologic conditions.
2-46
-------
Considering the project duration, model constraints, and data base limitations, the most
realistic representation of the ground-water flow system is obtained by (1) using field-generated values
to establish aquifer parameters; (2) incorporating known heterogeneity, where applicable; and
(3) applying reasonable values for the model input parameters where gaps exist in the data base. The
model was calibrated to represent regional flow paths and water-table elevations. The recharge rate
was varied uniformly over the entire model area. Aquifer and surface-water body hydraulic
conductivity values were varied only in regions for which there were little or no data. Values selected
for hydraulic conductivity are within the range of field measurements for the sand and gravel aquifer.
Recharge values for the model are lower than the average annual precipitation. The known and
unknown input parameters for the ground-water flow model, as well as how these parameter values
were used, or estimated, are summarized hi Table 2-6.
The hydraulic head distribution map and the 5-year TOT zone generated by the model are
shown in Figures 2-21A and 2-21B, respectively. Appendix A contains all input and output data
pertinent to the ground-water flow modeling of the Airport and Iverson wellfields using
FLOWPATH. As expected, the analytically-derived B Zone is larger than the refined 5-year TOT
zone predicted by the numeric model, FLOWPATH. However, the refined 5-year TOT zone extends
farther to the east than previously predicted, particularly in association with the Iverson wellfield
(Figure 2-22). This is due to the spatial variability in hydraulic conductivity, coupled with the
influence of the bedrock highs. As the conceptual model was refined from a homogeneous to a more
heterogeneous aquifer, with an irregularly shaped bottom surface and variable hydraulic conductivity,
the shape of the 5-year TOT zone became increasingly irregular.
2-47
-------
03
•a
S
"8
if
••
2
II
4-1 <4H
O O
o o
»—< »—I
W W
VD
11
Ji'3
C3
o o —
2II
sag
>q o o
U tt] >pj
2-48
-------
(A) HYDRAULIC HEAD DISTRIBUTION
32400-
17200-
2000
0 16000 32000
(CONTOUR INTERVAL IS 10 FEET)
(B) 5-YEAR TIME-OF-TRAVEL CAPTURE ZONE
32400-
17200-
2000-
\
AIRPORT
WELLFIELD
I
N
IVERSON
WELLFIELD
0 16000 32000
(VERTICAL AND HORIZONTAL SCALES IN FEET)
Figure 2-21. (A) Hydraulic head distribution map and (B) 5-year time-of-travel zone for the
Airport and Iverson wellfields, Stevens Point, Wisconsin.
2-49
-------
SCALE IN FEET
Figure 2-22. Comparison of B Zone and 5-year time-of-travel zone areas determined by
analytical and modeling (FLOWPATH) methods, respectively, for the Airport and
Iverson wellfields, Stevens Point, Wisconsin.
2-50
-------
WELLHEAD PROTECTION MONITORING PROGRAM
The purpose of the proposed wellhead protection monitoring program is to preserve the
quality of ground water at the Stevens Point wellfields. The monitoring system will detect changes in
ground-water quality within the WHPA. The proposed program consists of two types of monitoring
systems: one for ambient trend monitoring and one for compliance monitoring. The ambient-trend
monitoring system will detect regional changes in ground-water quality and aid in locating
contaminant sources. This system is targeted to monitor the effects of nonpoint sources and high-
density point sources. The compliance monitoring system is used to (1) assess if monitoring
regulations are met for a source, and (2) provide an early-warning detection system at priority point
sources.
The wellhead protection monitoring program presented here gives priority to protection of the
Airport wellfield. Without the Airport wellfield, the city could not meet the existing water-supply
demand without severe rationing. In addition, manganese concentrations in raw ground water from
CW4 (at the Iversbn wellfield) exceed secondary drinking water standards. This water would require
treatment for use; however, it is sufficiently diluted by addition of higher quality water from the
Airport wells. If the Airport wellfield is contaminated, additional wells would need to be installed, or
a treatment plant established and surface water used to meet the present water demands of the city.
AMBIENT TREND MONITORING
Monitoring Sites
The existing monitoring well system provides adequate coverage for most of the 5-year TOT
zone for the Airport wellfield. Areas with insufficient coverage in this zone include the eastern
portion of the zone and the deeper part of the aquifer to the north and to the west along the 5-year
TOT boundary, especially in the buried valley. All of the existing monitoring wells are located within
the 5-year TOT zone, leaving the city little time to respond to a contamination event.
Additional wells are proposed in the 5- and 10-year TOT zone (well nests N through V) to
offer more response time. These additional wells will serve as early-warning detection indicators of
contamination. Existing and proposed monitoring wells for the wellhead protection monitoring
network are shown in Figure 2-23. In addition, it is recommended that dry wells in the storm-sewer
system and the Plover River be sampled after storms to determine the water quality and the trends in
recharge chemistry due to urban runoff.
2-51
-------
-.*• i
...•A **<>
TXO t-r
•a
03
a I
&0j£
*S i v*
5. a
C!
o
(X
8
2-52
-------
Monitoring Well Construction
It is recommended that all new monitoring sites (N through V) be installed as well nests.
These well nests should be constructed similar to well nest K (Figure 2-10B) and located on public
property to guarantee accessibility. Well nests installed in rights-of-ways should be placed as far as
possible from the road to prevent damage and to reduce the possibility that road construction will
result in removal of the well.
Monitoring Parameters and Frequencies
The ambient-trend monitoring wells and the Plover River should be sampled twice a year.
The first sampling period should occur in April after the ground has begun to thaw and after spring
recharge begins. The second sampling period should begin in September after active land-use
activities, such as pesticide and fertilizer application, are, completed. The ambient-trend well samples
should be analyzed for the indicator parameters: pH, specific conductivity, nitrogen (nitrate and
nitrite), chloride, iron, manganese, nitrogen- and phosphorous-based organics (pesticides), VOCs, and
total petroleum hydrocarbons (TPHs). Sampling should be conducted following standard guidelines
(Wisconsin Department of Natural Resources, 1987), and water levels should be recorded each time a
well is sampled.
Monitoring Data Base
The hydrogeologic and hydrochemical data should be stored on a computerized data base and
updated immediately after the results of each monitoring sample set are received by the city Water
Department. This data base will be used by the Water Department to monitor changes in ground-
water quality and to trigger further investigation if PALs are exceeded. Further investigation may
entail sampling and testing the affected well for other possible contaminants, installing a compliance
monitoring system, and remediation actions. The PALs, as well as primary and secondary drinking
water standards (Wisconsin Department of Natural Resources, 1988), for the ambient-trend indicator
parameters are listed in Table 2-7.
COMPLIANCE MONITORING
Monitoring Sites
There is one existing compliance monitoring network in the monitoring system. This network
of wells was installed to monitor the airport fuel storage tanks. Unfortunately, this network failed to
detect the leak in the UST pumping system in 1988. This incident illustrates the importance of
monitoring system design. In the case of USTs, careful record keeping by the owner may provide the
most reliable, quickest, and least expensive leak-detection system. The first compliance monitoring
wells, or network, should be.installed at the highest-priority point sources (for example, USTs and
large, historical spills in the 10-year TOT zone). Compliance monitoring networks at the remaining
point sources should be added in order of priority, such as feedlots, golf courses, then landfills. New
spills will require immediate evaluation and, possibly, remediation and monitoring.
2-53
-------
TABLE 2-7. PREVENTIVE ACTION LIMITS AND STATE OF WISCONSIN DRINKING
WATER STANDARDS*
Analysis
VOCs
N- and P-based organics
NO2andNO3
a
Mn
Parameters
Benzene
1,1-Dichloroethane
1,2-Dichloroethane
1,2-Dichloroethylene
Ethylbenzene
Tetrachloroethylene
Toluene
1,1,1-Trichloroethane
1,1,2-Trichloroethane
Trichloroethylene
Xylene
Alachlor
Atrazine
EPTC
Simazine
N
a
Mn
PAL Standard
(ngfl)
0.067
85
0.05
0.024
272
0.1
68.6
40
0.06
0.18
124
0.05
0.35
50
215
0.67
850
0.5
0.24
1,360
1
343
200
0.6
1.8
620
0.5
3.5
250
2,150
(mg/1)
2
125
0.025
10
250f
0.05f
Abbreviations: PALs = preventive action limits; VOCs = volatile organic compounds; N = nitrogen;
P « phosphorus; NO2 = nitrite; NO3 = nitrate; Cl = chloride; Mn = manganese.
* Wisconsin Department of Natural Resources, 1988.
t Secondary drinking water standard.
Monitoring Well Construction
The position of monitoring wells around point sources should be determined by careful
investigation of the ground-water flow system in the vicinity of the potential source. The
hydrogeologic mapping done for this study is more regional in nature and does not focus on the local
pattern of ground-water flow at proposed compliance monitoring sites. Important factors influencing
local ground-water flowpaths and contaminant transport are: (1) the effects of variable bedrock
topography, (2) localized changes in the regional gradient due to pumping wells, and (3) the location
and nature of the potential contaminant. Another factor influencing the design of individual
compliance monitoring networks is the accessibility of the site. If monitoring wells cannot be placed
near the source, the density and number of depths monitored should be increased. The exact
placement of the monitoring wells will, necessarily, be site-specific. Generally, the initial monitoring
2-54
-------
system should include wells placed between 10 arid 50 feet downgradient from the potential source and
at least one upgradient well placed 10 to 30 feet from the potential source. The number of
downgradient wells will depend on the size of the source to be monitored; the cost of installation,*"
sampling, and chemical analyses; and the distance from the source. No fewer than three downgradient
wells should be installed (Pfannkuch, 1982; U.S. EPA, 1986).
The primary goal of compliance monitoring is to assess or to ensure that monitoring
regulations are being met for a site or source. Secondly, compliance wells are used for detection
monitoring. Compliance monitoring data should be acquired following strict adherence to quality
assurance and quality control procedures so that data are court admissible. Each compliance
monitoring well should be installed as a single well, somewhat as that depicted in Figure 2-10A and
discussed in the Data Acquisition, Monitoring Wells section. Single wells are preferred to multi-level
wells for compliance monitoring because boreholes are completely separate, thereby reducing the
possibilities of cross-contamination and vertical interconnection. Exceptions to the recommended
single well construction given in Figure 2-10A pertain to the well diameter, well material, well screen,
and gravel pack.
Compliance wells may need to be completed with a minimum, 2-inch-diameter pipe and
screen to accommodate regulation-type sampling and other well entry equipment. The borehole
should be sufficiently wide to accommodate all well materials and tools. The WGNHS and the U.S.
EPA Region V offices provide technical guidance and review of proper well construction, installation,
and development procedures for compliance purposes. The well construction material should be
carefully reviewed regarding its chemical compatibility and reactability to anticipated ground-water
contaminants. In some cases, such as when solvents at high concentrations in ground water are the
target monitoring parameter, stainless steel is recommended over PVC for well construction. All well
components introduced to the borehole should be decontaminated and clean. No solvent-welded
(glued) components should be used in the well construction process; all well pipe and screen should be
flush-threaded.
For compliance purposes, it may be necessary to install acid-resistant, washed and graded silica
sand, rather than native sand backfill for the gravel pack material. The gravel pack should be clean
and free of oil, acid, or other deleterious substances. The well screen slot size should be designed with
respect to the gravel pack sieve size to minimize disruption of normal flow through the well and to
prevent loss of the well depth or screen length (silting-in).
Single monitoring wells may be combined in optimal spacing and depth configurations
(nested) to intercept suspected contaminant flowpaths. If the compliance monitoring well network is
located near a potential contaminant source where dense, nonaqueous-phase liquids (DNAPLs) have
not been used at the source, then shallow monitoring wells should be installed. For DNAPL
contaminant sources that tend to sink, rather than float, within the aquifer, the monitoring wells
should be installed to monitor ground water in deeper portions of the aquifer, potentially to bedrock.
All shallow wells should be screened in the upper 15 feet of the aquifer to access water in the shallow
ground-water flow system and to allow for seasonal changes in the water level.
To fill data gaps in the compliance monitoring network relatively inexpensively, and for
detection purposes only (non-regulatory), multi-level monitoring wells may be considered. The multi-
level well installation method is similar to that at well nests J through M (Figure 2-10B). Casing is
2-55
-------
pushed to the desired depth, cleaned, and a polytube bundle is installed. Then, the casing is extracted,
and the installation is completed when the annular space is filled with collapsed sand, or sand backfill,
and a bentonite seal is installed at the surface.
An economical method for installing multi-level wells (for detection monitoring only) should
be used. For example, hi areas where the water table is very shallow, surface casing with an
expendable drive point can be driven into the aquifer to the desired depth, and a polytube bundle can
be lowered into the casing (Sitets and Chambers, 1990). When the casing is pulled, the drive point
and the polytube bundle are left in place. For deeper installations, a drilling rig can be used to push
the casing, and a sand bailer can be used to clean the casing before installation of the polytube bundle.
Well installation equipment and materials should be decontaminated between each well installation,
using standard procedures to prevent cross contamination within the aquifer (Wisconsin Department
of Natural Resources, 1988).
Monitoring Parameters and Frequencies
Each compliance monitoring network should be sampled for select, potential contaminants or
a characteristic component of the monitored source. For example, if the potential source is a gasoline
storage tank, the target monitoring parameters should be VOCs. The frequency of the sampling is
dependent on the location of the potential source. Sources within the 5-year TOT zone should be
sampled at least twice a year, and sources in the 10-year TOT should be sampled annually, at a
minimum. Potential point sources and recommended monitoring parameters, as well as sampling
frequencies, are listed in Table 2-8.
Monitoring Data Base
The results of the compliance monitoring program need to be kept on file. Any detection of a
contaminant hi a monitoring well should trigger the contingency plan.
CONTINGENCY PLAN
The contingency plan for the Stevens Point water supply system consists of three components.
One component of the plan is reaction to the early-warning monitoring system. When a PAL is
exceeded in an ambient-trend monitoring well, or a specific contaminant is detected in a compliance
monitoring well, then the Water Department should resample the well and begin an investigation into
the source and extent of the contamination. Results of that investigation will determine the nature of
the remedial actions. The procedures outlined by the EPA (U.S. EPA, 1990) can be used, in part, to
assess the site.
The second element of the contingency plan is to react to events that are beyond the
capabilities of the monitoring system (for example, accidental spills in the 5-year TOT zone and leaks
at sources that are not currently monitored). This phase of the contingency plan is currently in
operation and relies on the cooperation of state emergency-response agencies. In the event of an
2-56
-------
TABLE 2-8. COMPLIANCE MONITORING GUIDELINES
Priority Point Sources
Monitoring Parameters
Monitoring Frequency
Airport USTs (old and new)
Gas stations
Highway 66 and Wilshire
Highway 10 and Brilowski
High-density, small USTs
Highway 66 and Airport Rd
Highway 10, near Wildwood
Highway 66, east of
Highway 51
Garages
Highway 66 and Highway 51
Feedlot (FL-A)
Golf courses,
Abandoned landfills
VOCs
VOCs
VOCs
Quarterly
Semiannually
Semiannually
VOCs
N (NO2 and NO3)
N (NO2 and NO3)
VOCs; TPHs; major anions,
cations, and metals
Annually
Semiannually
Semiannually
Annually
Abbreviations: USTs = underground storage tanks; VOCs = volatile organic compounds;
N = nitrogen; NO2 = nitrite; NO3 = nitrate; TPHs = total petroleum hydrocarbons.
accident within the 10-year TOT zone involving the release of hazardous materials, the local and state
emergency-response agencies (for example, the fire department, police, sheriff, highway patrol, and
Department of Natural Resources) will notify Stevens Point Water Department personnel
immediately, and the need for remedial action will be assessed.
If the WHPP fails and a city well is contaminated, the third component of the contingency plan is
implemented: the affected wells will be shut down and other wells will supply the city's needs, the
source of contamination will be investigated, and remedial actions will be taken. New wells would be
required if the contamination is extensive or if the contamination cannot be remediated. One function
of the ambient-trend monitoring system is to assess local ground-water quality to better identify
favorable ground-water development areas. If ground-water contamination is widespread, a treatment
plant may be required.
2-57
-------
-------
CONCLUSIONS
A study was conducted in Stevens Point, Wisconsin, to refine the preliminary wellhead
protection area for the Airport and Iverson wellfields, as well as to design an early-warning detection
ground-water monitoring network. The study entailed source assessment, well installation, aquifer
testing, ground-water sampling, hydrochemical characterization, ground-water flow modeling, wellhead
protection area delineation, and monitoring network design. Conclusions of the study are summarized
as follows:
• Existing point sources within the preliminary wellhead protection area include above-
ground and underground storage tanks (USTs). Nonpoint sources include agricultural
fields, lawns, septic systems, and wetland soils.
• Potential point sources include garages and gas stations, industrial and commercial
facilities, junkyards, landfills, salt and snow storage areas, septic systems, spills, storage
tanks, transformers, and unregulated dump sites. Nonpoint sources include feedlots,
golf courses, lawns, the Plover River, road de-icing runoff, septic-sludge spreading, and
urban runoff.
• Existing contaminant sources were given highest priority for management. Other
sources were ranked based on the source type, quantity, hazard, and location.
Prioritized sources include USTs, irrigated agriculture, high-density, nonsewered
residential areas, landscaped areas, feedlots, spills, landfills, perforated storm drainage
pipes, and dry wells.
• A network of 44 single monitoring wells, 10 well nests, and one multi-level well were
used to measure water levels, to sample ground water, and to conduct aquifer tests in
the shallow, unconfined sand and gravel aquifer. Of the total network, three single
wells and four well nests represent new monitoring points installed for this research.
• Aquifer parameter results from slug, constant discharge, and recovery tests indicate a
range of hydraulic conductivity values for three distinct geologic settings: 820 to 1,700
ft/d for the buried valley, 220 to 240 ft/d for outwash plains, and 2 to 3 ft/d for bedrock
highs.
• Hydrochemical data indicate that nitrogen concentrations, a key indicator of
contamination, have increased over time. Currently, nitrogen concentrations in the
monitoring network range from less than 0.2 to 26.0 mg/l. Nitrogen concentrations
are generally higher in the shallow monitoring wells. Better ground water quality at
depth may indicate a more regional flowpath, more distant agricultural areas, dilution
of recharge water with depth, or a combination of these factors.
2-59
-------
Other indicators of ground-water degradation include iron and manganese from
organic-rich soils located along the Plover River, chloride in proximity to roads where
de-icing occurs, and previous volatile organic compound (VOC) contamination at the
airport and several USTs.
The two-dimensional, ground-water flow model FLOWPATH was used to delineate
the 5- and 10-year time-of-travel (TOT) zones for the Airport and Iverson wellfields.
In comparison, the previous, analytically-derived B Zone is larger; however, the 5-year
TOT zone from FLOWPATH extends farther to the east due to the effects of
pumping at the Iverson wellfield and the presence of bedrock highs.
A long-term ground-water monitoring network is proposed for the Airport and
Iverson wellfields consisting of 34 total existing and proposed wells. Twenty-five
existing wells include 18 single wells and seven well nests. Nine new well locations are
proposed to fill data gaps in the existing network, primarily along the boundaries of
the 5- and 10-year TOT zones.
Wells hi the long-term monitoring network should be sampled twice a year in April
and September for the indicator parameters: pH, specific conductivity, nitrogen,
chloride, iron, manganese, nitrogen- and phosphorous-based organics (pesticides),
VOCs, and total petroleum hydrocarbons (TPHs). Water levels should be recorded
each time a well is sampled.
It is recommended that the Plover River and dry wells in the storm-sewer system be
sampled after storms to determine water quality and trends in recharge chemistry due
to urban runoff.
Compliance monitoring networks should be installed at high-priority point sources
such as USTs, spill sites, feedlots, golf courses, and landfills. No fewer than one
upgradient and three downgradient wells should be installed per network. Single well
construction is recommended for compliance monitoring with some wells combined in
nested configurations to allow detection of parameters at various depths in the aquifer.
Select parameters are chosen for compliance monitoring based on the historical use of
substances at the facility. For example, USTs, gas stations, and garages are specifically
monitored for VOCs; feedlots and golf courses are monitored for nitrogen; and
landfills are monitored for VOCs, TPHs, major anions, major cations, and metals.
The frequency of monitoring for key parameters is site-specific, varying from quarterly
to annually.
The wellhead protection contingency plan consists of three components: (1) reaction
to the early-warning detection system based on preventive action and state drinking
water limits, (2) spill response, and (3) new water-supply development and
implementation.
2-60
-------
RECOMMENDATIONS
The water supply for the city of Stevens Point is currently of high quality. Compared to other
cities, the potential sources of contamination are few and distant. However, the potential for
contamination of the city's water supply exists. The greatest threat to the city's water supply is public
unawareness of aquifer vulnerability. Most citizens are unaware of the fact that what is poured,
spilled, leaked, or dumped on the land surface moves quickly into the underlying ground water. The
sources of greatest concern are those that have already contaminated ground water, including
petroleum from leaking storage tanks, pesticides applied to-crops, nitrogen from overfertilization of
lawns and crops, and nitrogen from human and animal wastes. Other possible sources of
contamination that require careful monitoring include urban runoff, underground tanks that are not
yet leaking, and naturally occurring sources that cause degradation (for example, iron, manganese, and
radioactivity).
The long-term monitoring system proposed in this report is designed to detect potential
contamination events. Its purpose is to provide early warning, to allow time for remediation, and to
prevent contamination of the municipal wells. However, the most efficient method for preventing
contamination is for owners and operators of potential sources to inventory the hazardous and toxic
substances at their facilities and to monitor the correct use of these substances.
The next step in evaluating the potential impact of sources is to model the fate and transport
of contaminants, thereby predicting the effects of sources on ground-water quality over time. Accurate
predictions of contaminant transport require a detailed model of the ground-water flow system.
Evolution of the ground-water flow model from a two-dimensional, steady-state model to a three-
dimensional, transient model is essential to represent the flow system accurately. In addition to
refining the ground-water flow model, the nature of the contaminant source must be understood. The
concentration and release rate of the contaminant from each source should be determined. Field
studies are necessary to verify the validity of the model. Contaminant transport models can be used to
generate more effective land-use restrictions.
Public education of ground-water protection and hazardous waste disposal procedures are two
additional elements of the local WHPP that need to be developed. The new Stevens Point urban area
wellhead protection project, funded by the U.S. Department of Agriculture, provides an appropriate
mechanism to develop these two elements. These are important additions to the city's ground-water
management plan. Following installation of the long-term monitoring well network, other
improvements to the system can be made, including
• Research on contaminant transport through the sand and gravel aquifer
• An increase in the number of monitoring points in the 10-year TOT zone and beyond
• Source assessment beyond the 10-year TOT
• Compliance monitoring at sources classified as a moderate risk
2-61
-------
This project is the first stage of a continuing WHPP and is the core of an historical data base.
As more information is gathered on the ground-water system, the WHPP will be refined and
improved. The TOT zones and the model of the ground-water flow system will become more accurate
and predictive. Ultimately, the ground-water flow system should be simulated three-dimensionally
with a model that can incorporate the variable pumping rate of the city's wells and the transient
nature of recharge. To accomplish this goal, a better understanding of the following hydrogeologic
factors is necessary: (1) the exact location and seasonal variations of the ground-water flow
boundaries, (2) the distribution of hydraulic conductivity (laterally and with depth), and (3) the
variation of recharge rates (with time and location). To model the system accurately, the interaction
between the Plover River and the city wells must be determined. The city's WHPA should continue to
be refined and land-use controls implemented to ensure protection of the city's water supply.
2-62
-------
REFERENCES
Aller, L., T. Bennett, J. H. Lehr, R. J. Petty, and G. Hackett. 1987. DRASTIC: A Standardized
System for Evaluating Ground-water Pollution Potential Using Hydrogeologic Settings. EPA-600/2-
87-035. U.S. Environmental Protection Agency, Office of Research and Development, Robert S. Kerr
Environmental Research Laboratory, Ada, Oklahoma. 455 pp.
Bahr, J. M. 1990. Aquifer Test Results at City Well 6, June 16, 1990. Letter Report to T. Brown,
Project Coordinator of the Stevens Point, Wisconsin, Wellhead Protection Program. University of
Wisconsin-Madison. '
Bouwer, H. and R. Rice. 1976. A Slug Test for Determining Hydraulic Conductivity of Unconfined
Aquifers With Completely or Partially Penetrating Wells. Water Resources Research 16(1): 423-428.
Bowen, B. D. 1987. Potential for Nitrogen Ground-Water Contamination From Animal
Confinement Areas in Central Wisconsin [M.S. Thesis]. University of Wisconsin - Stevens Point.
Chesters, G., M. Anderson, B. Shaw, J. Marker, M. Meyer, E. Rothchild, and R. Manser. 1982.
Aldicarb in Ground-water. Report From the Ground-Water Research Center, University of
Wisconsin, Madison. 38 pp.
Clayton, L. 1986. Pleistocene Geology of Portage County, Wisconsin. Information Circular 56.
Wisconsin Geological and Natural History Survey, Madison, Wisconsin. 19 pp.
Cooper, H. H., Jr. and C. E. Jacob. 1946. A Generalized Graphical Method for Evaluating
Formation Constants and Summarizing Well-Field History. American Geophysical Union
Translations 27(4): 526-534.
Donohue and Associates. 1988. Report on Remediation, Aviation Gasoline Spill, Stevens Point
Municipal Airport. Unpublished Report to the Water Department, City of Stevens Point, Wisconsin
7pp.
Faustini, J. M. 1985. Delineation of Ground-water Flow Patterns in a Portion of the Central Sand
Plain of Wisconsin [M.S. Thesis]. University of Wisconsin - Madison. 117 pp.
Ferris, J., D. Knowles, R. Brown, and R. Stallman. 1962. Theory of Aquifer Tests. U.S. Geological
Survey, Water Supply Paper 1536-E. U.S. Government Printing Office, Washington, D.C. 134 pp.
Finkbeiner, M., B. Moore, T. Brown, G. Disher, and J. Gardner. 1991. Aerial Photographic Source
Assessment for Wellhead Protection, Stevens Point, Wisconsin. U.S. Environmental Protection
2-63
-------
Agency, Office of Research and Development. Environmental Monitoring Systems Laboratory, Las
Vegas, Nevada. 207pp.
Franz, T. and N. Guiguer. 1989. FLOWPATH version 2.0, Two-Dimensional Horizontal Aquifer
Simulation Model. Waterloo Hydrogeologic Software, Ontario, Canada. 75 pp.
Freeze, R. A. and J. A. Cherry. 1979. Ground Water. Prentice-Hall, Inc., Englewood Cliffs, New
Jersey. 604pp.
Good, L. W. and S. W. Madison. 1987. Soils of Portage County and Their Ability to Attenuate
Contaminants. Wisconsin Geological and Natural History Survey, Map 87-8. Madison, Wisconsin.
Harkin, J. G., G. Chesters, P. Kroll, and F. Jones. 1984. Pesticides in Groundwater Beneath the
Central Sand Plain of Wisconsin. Report to North 'Central Region Pesticide Impact Assessment
Program, Columbus, Ohio.
Hickok and Associates. 1965. Ground-Water Investigation at City of Stevens Point, Wisconsin.
Unpublished Report to the Water Department, City of Stevens Point. Minneapolis, Minnesota.
18pp.
Hickok and Associates. 1981. Hydrogeologic Study: Airport Well Field for the City of Stevens Point
Wisconsin. Unpublished Report to the Water Department, City of Stevens Point. Minneapolis,
Minnesota. 23 pp.
Holt, C L. R., Jr. 1965. Geology and Water Resources of Portage County Wisconsin. U.S.
Geological Survey, Water-Supply Paper 1796. U.S. Government Printing Office, Washington, D.C.
77pp.
Horslev, M. J. 1951. Time Lag and Soil Permeability in Ground-Water Observations. Bulletin 36.
U.S. Army Corps of Engineers, Waterways Experimental Station, Vicksburg, Mississippi. 56 pp.
Jones, F. A. 1987. Computer Simulation of Aldicarb Migration and Degradation in the Central Sand
Plain of Wisconsin [Ph.D. Thesis]. University of Wisconsin - Madison. 155 pp.
Lippelt, L D. and R. G. Hennings. 1981. Irrigable Lands Inventory - Phase I, Groundwater and
Related Information. Miscellaneous Report 87-1. Wisconsin Geological and Natural History Survey,
Soil Conservation Service, Madison, Wisconsin. 13 pp.
Neuman, S. P. 1974. Effect of Partial Penetration on Flow in Unconfined Aquifers Considering
Delayed Gravity Response. Water Resources Research 10(2): 303-312.
Osborne, T. J. 1988. Bedrock Elevation Contour Map of the Stevens Point Area. Unpublished
Report Ground-Water Resource Center, University of Wisconsin - Extension, Stevens Point,
Wisconsin.
Pfannkuch, H. O. 1982. Problems of Monitoring Network Design to Detect Unanticipated
Contamination. Ground Water Monitoring Review 2(1): 67-76.
2-64
-------
Portage County. 1986. Unpublished Aldicarb Data for Portage County Wells. Human Services
Department, Division of Community Health Services, Stevens Point, Wisconsin.
Portage County Planning Department. 1988. Portage County Groundwater Management Plan.
Volumes 1 and 2. Stevens Point, Wisconsin. 161 pp.
Renaud, R. K 1987. The Recharge Area and Water Quality of the Stevens Point Municipal Well
Field. Unpublished M.S. thesis. University of Wisconsin - Stevens Point 81 pp.
Rothchild, E. R. 1982. Hydrogeology and Contaminant Transport Modeling of the Central Sand
Plain. M.S. thesis. University of Wisconsin - Madison. 135 pp.
Sitets, W. and L. W. Chambers. 1990. A Method for Installing Miniature Multilevel Sampling Wells.
Ground Water 29(3): 430-432.
U.S. Environmental Protection Agency. 1986. RCRA Ground-Water Monitoring Technical
Enforcement Guidance Document. OSWER-9950.1. U.S. Environmental Protection Agency, Office
of Solid Waste and Emergency Response, Washington, D.C 208 pp.
U.S. Environmental Protection Agency. 1990. Assessing UST Corrective Action Technologies. EPA
600/2-90/027. U.S. Environmental Protection Agency, Office of Research and Development, Risk
Reduction Engineering Laboratory, Cincinnati, Ohio. 124 pp.
U.S. Geological Survey. 1977. National Handbook of Recommended Methods for Water Data
Acquisition. U.S. Department of the Interior, Reston, Virginia. 412 pp.
Weeks, E. P. and H. G. Stangland. 1971. Effects of Stream Flow in the Central Sand Plain of
Wisconsin. U.S. Geological Survey, Madison, Wisconsin. 94 pp.
Wisconsin Department of Natural Resources. 1987. Ground-water Sampling Procedure Guidelines. .
PUBL-WR-153 87. Madison, Wisconsin. 91pp.
Wisconsin Department of Natural Resources. 1988. Wisconsin Administrative Code NR140.
Madison, Wisconsin, pp. 682-683.
2-65
-------
-------
APPENDIX 2-A
FLOWPAIH GROUND-WATER MODEL RESULTS FOR THE
STEVENS POINT, WISCONSIN, AQUIFER
2-67
-------
-------
GO
FLO
CL
O
O
o
at
* 5
o> 5"
m
i
10
O
O
CO
co
1
UJ
CO
_8> DO
iZ co
o
o
• o
g
- o
CM
CO
in
c
_o
O
o
I
CM
CD
TJ
«
a
o
o
o
• o
(O
_0
"5
^;:
_E
CO
o
g
CO
r
1
—r
g
o
o
to
o
o
o
o
o
CM
2-69
-------
•§. 5
1! ?
OS HI
n- ^ a>
a:
-§
O
ru
— 8
ru
CO
2-70
-------
l__
<
Q_
1
L_
-^-*
Q.
0
0
0
»
OJ
CO
o>
:r
.a
T>
as
a>
-2 3
55 DZ
a>
^
LU
u>
C0
cr>
-a
1
DC
O
o
LU
o
0
10
"E :p*
a>
n:
LU
V>
a.
CO
O
M
a>
=5
&
O
o
o
o
o
T
o
5
ru
o
o
• o
o
o
o
ru
CM
O
O
• O
to
O
. O
O
CO
O
O
CM
O
O
o
CM
2-71
-------
« -
£. £
co co
.3
111
co
CO
CM
2-72
-------
OWPATH
i 1
fc ••— 3»
as UJ
e ™
I
8
..
?
X N. X X \
\ \ \ ^ N
S. N. N. \
\
CO
s
\\\\\\\
\ \ '" \% l i \ \ \
8
o
o
• o
?
§
- o
CM
CO
• O
CM
2-73
-------
rf *
o
S 5 3
^ ^ .2
S5 S5 u.
= ft
Ul
ui Si
.. *
& _ _» S
— .£L ' E fe
(M
2-74
-------
ECHO PRINT
FLOBPATH
version 3.0
FLOWPATH was written by Thoias Franz and Hilson Guiguer *
************************************************************
Copyright 1989, 1990
by
Waterloo Hydrogeologic Software
113-106 Seagram Drive
Waterloo, Ontario
H2L 3B8, Canada
ph (519) 746-1798
************************************************************
FLOWPATH logbook for data set : SPBASE
Unit Systesi : English units [ft/gal/d]
***** GRID PARAMETERS *****
Nuuber of x-grid lines : 33
Number of y-grid lines : 34
Grid coordinates (x-grid lines) [ft] :
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
O.OOOOOE+00
2.10357E+03
4.20714E+03
6.31071E+03
8.41429E-KB
9.60326E+03
1.05179E+04
1.12495E+04
1.19812E+04
1.26214E+04
1.29873E+04
1.34446E+04
1.39019E+04
1.43592E+04
1.47250E+04
1.54567E+04
1.60969E+04
1.68286E+04
1.79261E+04
1.89321E+04
1.99382E+04
2.11272E+04
23
24
25
26
27
28
29
30
31
32
33
2.20418E+0
2.32307E-K)
2.40539E-H)
2.52429E+0
2.73464E+0
2.94500E+0
3.15536E+0
3.36571E+0
3.58522E+0
3.79557E+0
4.00000E+0
Grid
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
coordinates (y-grid lines) [ft] :
2.00000E+03425
4.24224E+03426
5.30435E+03427
6.24845E+03428
6.60248E-K)3429
7.54658E+03430
8.37267E+03431
9.55280E+03432
1.03789E+04433
1.13230E+04434
1.19130E+044
1.26211E+04
1.30932E404
1.36832E+04
1.46273E+04
1.56894E+04
1.67516E+04
1.79317E+04
1.87578E+04
1.97019E+044
2.10000E+04
2.18261E+04
23
24
25
26
27
28
29
30
31
32
33
34
2.32422E+04
2.40683E+04
2.52484E+04
2.73727E+04
2.93789E+04
3.16211E+04
3.26832E+04
3.36273E+04
3.46894E+04
3.57516E+04
3.77578E+04
4.00000E+04
2-75
-------
***** WELL PARAMETERS *****
Hmber of wells : 6
).
1
2
3
4
5
6
14
11
11
12
12
10
j
5
10
12
13
14
5
X
[ft]
1.43592E-H)4
1.29873E+04
1.29873E+04
1.34446E-H)4
1.34446E+04
1.26214E-H)4
Y
[ft]
6.60248E+03
1.13230E-H)4
1.26211E+04
1.30932E+04
1.36832E+04
6.60248E+03
well discharge
-1.71300E+06
-1.53300E+06
-2.47900E+06
-1.79900E+06
-1.57200E+06
-1.05600Ei06
***** CONSTRAINED HEAD NODES *****
Hmber of constant head nodes : 1
Ho. i j X Y const, head
[ft] [ft] [ft]
1 14 17 1.43616E+04 1.68023E+04 1.08000E+03
***** ASEAL KECHARGE *****
Hinber of different infiltration/evapotranspiration rates : 1
Ho. infiltration evapotranspiration effective recharge
[L/T] [L/T] [L/T]
1 7.30000E-03 3.50000E-03 3.80000E-03
***** AOJUFER TYPE
Dnconfined aquifer
*****
***** AQUIFER PROPERTIES *****
Hiaber of different aaterial properties : 5
Ho.
1
2
3
4
5
Kxx
[ft/d]
1.00000E-H)2
2.60000E402
1.40000E+03
5.20000E-H)2
5.50000E-HJ2
Kyy
[ft/d]
l.OOOOOE+02
2.60000E-H)2
1.40000E+03
5.20000E-K)2
5.50000E-H)2
Porosity
[-]
3.20000E-01
3.20000E-01
2.50000E-01
3.20000E-01
3.20000E-01
(default)
2-76
-------
********** DISTRIBUTION OF AQUIFER HATERffi, PROPERTIES **********
j 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
* 5
* 5
* 5
* 5
* 5
* 5
* 5
* 5
3 3
3 3
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
5
5
5
5
5
5
5
5
2
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
2
3
3
3
3
3
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2~
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
,3
3
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5 5
5 5
5 5
5 5
5 5
5 5
5 5
5 5
2 2
2 2
2 2
2 2
2 2
2 2
2 2
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
3 3
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5'
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
.1
1
1
2
3
3
3
3
,3
3
3
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
2
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
2
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
'2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
*
*
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
*
*
*
*
*
5
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
* *
* *
* *
* *
* *
* *
* *
* *
* *
* *
2 *
2 *
2 *
2 *
2 2
2 2
' 2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
2 2
12 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
2-77
-------
***** AQUIFER BOTTOM ELEVATIONS *****
Huiber of different aquifer bottoi elevations
No. aquifer bottoi elevation
[ft]
1 1.04000E+03 (default)
2 1.10000E-K)3
3 1.08000E+03
4 1.06000E+03
5 1.02000EW3
6 9.00000E-H)2
7 9.20000E+02
8 9.40000E-K)2
17
9
10
11
12
13
14
15
16
17
9.60000E+02
9.80000E+02
l.OOOOOE+03
1.02000E+03
1.12000BH)3
1.11500E+03
1.03000E+03
1.09000EW3
1.07000E«3
DISIHBOTION OF AQDIFER BOTTOM ELEVATIONS **********
2 3 4 5 6 7 8 910 11121314 15 1617 18 19 20 2122 23 24 25 26 27 28 29 30 3132 33
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
119
9
8
7
6
5
4
3
2
1
2
2
2
2
2
2
2
2
3 3
4 4
1 1
1212
112
4 3
3 3
3 3
16 3
16 3
16 3
16 3
1616
1616
1616
1616
1616
1616
16 3
16 3
3 3
3 3
317
1717
17 4
4 4
2
2
2
2
2
2
2
2
2
3
3
4
1
12
11
11
11
4
4
17
17
3
3
3
3
3
3
17
17
4
4
4
4
4
2
2
2
2
2
2
2
2
2
3
3
4
4
4
1
1
10
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
2
2
2
2
2
2
2
2
2
3
3
3
4
4
1
1
10
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
222222222222222222
222222222222222222
222222222222222222
2 14 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
2 2 14 14 2 2 2 2 2 2 2 2 2 2 2 2 2 2
2 2 14 14 2 2 2 2 2 2 2 2 2 2 2 2 2 2
222222222222222222
222222222222222222
222222222222222222
333333333333333333
333333333333333333
333333344444444444
444444444444444444
444444444444444444
111111144444444444
111111111111111111
10 9 9 9 9 9 9 9 9 9 8 8 8 1 1 1 1 1
1 1 1 1 1 1 11111111111 7 1 1 1 1 1
1 1 1 1 1 1 1 112121212 7 1 1 1 1 1
1 1 1 1 1 1 1 112121212 6 1 1 4 4 4
1 1 1 1 1 1 1 1 12 12 12 12 6 1 1 4 4 4
1 1 1 1 1 1 1 112121212 6 1 1 4 4 4
1 1 1 1 1 1 1 112121212 6 1 1 1 1 1
1 1 1 1 1 1 1 112121212 6 1 1 1 1 1
1 1 1 1 1 1 1 112121212 6 6 6 6 1 1
1 1 1 1 1 1 1 112121212121212 6 1 1
1 1 1 1 1 1 1 112121212121212 6 1 1
1 1 1 1 1 1 1 112121212121212 6 6 6
1 1 1 1 1 1 1 1121212121212121211 6
1 I'l 1 1 1 1 1121212121212121211 6
1 1 1 1 1 1 1 1121212121212121211 6
1 1 1 1 1 1 1 1121212121212121211 £
1111111 112121212121212121111
1 1 1 1 1 1 1 1 15 15 15 15 12 12 12 12 11 11
2
2
2
2
2
2
2
2
2
3
3
4
4
4
4
1
1
1
1
1
1
1
1
1
3
3
1
1
1
1
1
6
11
11
2
2
2
2
2
2
2
2
2
3
3
4
4
4
4
1
1
1
1
1
1
1
1
1
3
3
3
1
1
1
1
6
6
6
2
2
2
2
2
2
2
2
2
3
3
4
4
4
4
1
1
1
1
1
1
1
1
1
3
3
3
1
1
1
1
1
5
5
2
2
2
2
2
2
2
2
2
3
3
4
4
4
4
1
1
1
1
1
1
1
1
1
1
1
1
1
1
5
5
5
5
5
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
1
1
1
1
1
1
5
5
5
5
5
5
5
5
5
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
1
1
1
1
5
5
5
5
5
5
5
5
5
5
5
*
*
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
1
1
1
1
5
5
5
5
5
5
5
5
5
5
5
*
*
*
*
*
3
3
3
3
3
3
4
4
4
4
4
4
4
4
1
1
1
1
1
1
1
1
5
5
5
5
5
5
5
*
*
*
*
*
*
*
*
* *
* *
4 *
4 *
'*
*
4
4
4
4
4
4
4 4
4 4
1 1
1 1
1 1
1 1
1 1
5 5
5 5
5 5
5 5
5 5
5 5
5 5
2-78
-------
************ HYDRAULIC HEAD DISTRIBUTION ************
I 1 23 4 5
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
1.
1.
1.
1.
1,
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
*
*
*
*
*
*
*
*
1224E+03
1221E+03
1212E+03
1200E+03
1190E+03
1182E+03
1172E+03
1164E+03
1158E+03
1153E+03
1151EW3
1151E+03
1152E+03
1152EI03
1150E+03
1146E+03
1139E+03
1130E+03
1115E+03
1097E+03
1079E+03
1064E+03
1048E+03
1026EW3
0992E+03
0958E+03
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.1592E+03
.1520E+03
.1472E+03
.1442E+03
.1417E+03
.1393E+03
.1357E+03
.1292E+03
.1229E+03
.1211EW3
.1195E+03
.1177E+03
.1164E+03
.1151E+03
.1132E+03
.1119E+03
,inmo3
.1109E+03
.1107E+03
.1106E+03
.1103E+03
.1100E+03
.1097E+03
.1093E+03
.1088E+03
.1080E+03
.1069E+03
.1055E+03
.1039E+03
.1022E+03
.1007E+03
.0988E+03
.0952E+03
.0888E-HJ3
1.1535E+03
1.1489E+03
1.1451E+03
1.1425E+03
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1404E+03
1384E+03
1355E+03
1310E+03
1235E+03
1180E+03
1159E+03
1143E+03
1131E+03
1111E+03
1087E+03
1073E+03
1063E+03
1062E+03
1063E+03
1060E+03
1055E+03
1048E+03
1049E+03
1048E+03
1045E-KI3
1039E+03
1029E+03
1016Et03
1003E+03
0990E+03
0978EW3
0959E+03
0919E+03
0834E+03
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.1473E+03
.1443E+03
.1416E+03
.1396E+03
.1381E+03
.1364E+03
.1341E+03
.1304E+03
.1227E+03
.1159E+03
.1137E+03
.1117E+03
.1095E+03
.1063E+03 •
.1041E+03
.1027E+03
.1020E+03
.1017E+03
.1017E+03
.1014EW3
.1010E+03
.1006E+03
.1002E+03
.0999E+03
.0994E+03
.0987E+03
.0979E+03
.0970E+03
.0959E+03
.0949E+03
.0937E+03
.0920E+03
.0883E+03
.0800E+03
1.1414E+03
1.
1.
1.
1390E+03
1370E+03
1360E-H13
1.1350E+03
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1338E+03
1319E+03
1288E+03
1223E+03
1153E+03
1119E+03
1087E+03
1045E+03
1012E+03
1001Et03
0992E+03
0986E+03
0984E+03
0979E+03
0972E+03
0967E+03
0963E-I-03
0959E+03
0955E+03
0949E+03
0943E+03
0935E+03
0926E+03
0917E+03
0908EW3
0898E+03
0883E+03
0853E+03
0803EW3
1.1361E+03
1.1341E+03
1.1325EW3
1.1328E+03
1.1324E+03
1.1317E+03
1.1300E+03
1.1268EW3
1.1205E+03
1.1124E+03
1.1085E-HJ3
1.1046E+03
1.1002E+03
1.0976E+03
1.0969E-t03
1.0960E+03
1.0954E+03
1.0954EI03
1.0947E+03
1.0937E+03
1.0930E+03
1.0926E+03
1.0921E+03
1.0917E+03
1.0911E+03
1.0905E+03
1.0898E+03
1.0891E+03
1.0883E+03
1.0875E+03
1.0867E+03
1.0856E+03
1.0831E+03
1.0792E+03
2-79
-------
10
11
12
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
1.1316E403
1.1299EW3
1.1277E+03
1.1300EW3
1.1307EW3
1.1302BW3
1.1285EHI3
1.1251E+03
1.1187EW3
1.1094E+03
1.1053B03
1.1013EW3
1.Q972BW3
1.0950EW3
1.0943B03
1.0935B03
1.0929EW3
1.0928EW3
1.0918EM3
1.0905EW3
1.Q897EW3
1.0891EHI3
1.0887EM3
U882EW3
1.0877BM3
1.0872EW3
1.0866EW3
1.0859H03
1.0852E+03
1.0846EW3
1.0840EW3
1.0832E+03
1.0813E+03
1.0782E+03
1.1397EW3
1.1366E+03
1.1332B+03
1.1280E+03
1.1291E+03
1.1289E+03
1.1272BW3
1.1233EW3
1.U69BW3
1.1058EW3
1.1017E+03
1.0980E+03
1.0944EI03
1.0928E+03
1.0921E+03
1.0912EW3
1.0907E+03
1.0904E+03
1.0889E+03
1.0B73BW3
1.0862E+03
1.0855BW3
1.0850E+03
1.0846E+03
1.0842E+03
1.0839E+03
1.0834E+03
1.0828E+03
1.0821E+03
1.0815E+03
1.0812E+03
1.0B08EHO
1.0796EM3
1.0770EW3
1.1457E+03
1.1418E+03
1.1379E+03
1.1336E+03
1.1276E+03
1.1277E+03
1.1261E+03
1.1211E+03
1.1153EW3
1.1008E+03
1.0971E+03
1.0942E+03
1.0914EW3
1.0908E+03
1.0900E+03
1.0890E+03
1.0885E+03
1.0877E+03
1.0856E+03
1.0838E+03
1.0824E+03
1.0814E+03
1.0808E+03
1.0805E+03
1.0803E+03
1.0803E+03
1.0801EW3
1.0794W03
1.0787E+03
1.0780E+03
1.0781B03
1.07B2W03
1.0779BW3
1.075BW03
1.
1.
1.
1.
1.
1.
1.
1.
1.
1500E+Q3
1457E+03
1415E+03
1376E+03
1337E+03
1273E+03
1254E+03
11B6E+03
1175EW3
1.0933EW3
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
0910Et03
0893E+03
OS80EI03
0894E+03
0882E+03
0870E+03
0866E+03
0850E+03
0830E+03
0816E+03
0799E+03
0786E+03
0777E+03
0779E+03
0778E+03
0782E+03
0781E+03
0773E+03
0766E+03
0752E+03
0760E+03
1.0766E+03
1.
1.
0766E+03
0748E+03
1.1529EH3
1.1484E+03
1.1442E+03
1.1404E+03
1.1367E+03
1.1318E+03
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1251E+03
1237E+03
1200E+03
1029E+03
0986E+03
0954E+03
0908E+03
0886E+03
0881EI03
0868E+03
0860EI03
0841E+03
0823E+03
0812E+03
0792E+03
0777EW3
0760E+03
0775E+03
0769Ert3
0778E+03
0778E+03
0768E+03
0762E+03
0755E+03
0757EI03
0764E+03
0757E+03
0740E+03
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.1555E+03
.1509E+03
.1466EW3
.1429E+03
.1395E+03
.1355E+03
.1302EI03
.1276E+03
.1225E+03
.1085E+03
.1032E+03
.0996E+03
.0933E+03
.0895E+03
.0882E+03
.0865E+03
.0854E+03
.0830E+03
.0822E+03
.0810E+03
.0785E+03
.0770E+03
.0774E+03
.0779E+03
.0777E+Q3
.0777E+03
.0776E+03
.0779E+03
.0775E+03
.0769E+03
.0769E+03
.0762E+03
.0747E+03
.0732E+03
2-80
-------
13
14
15
16
17
18
34
33
32
31
30
29
28
27
26
25
24
23
22
21
.20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
1.1580E+03
1.1533BW3
1.1490E+03
1.1454BW3
1.1423BW3
1.1388E+03
1.1343E+03
1.1311BW3
1.1251E+03
1.1129E+03
1.1069E+03
1.1026E+03
1.0957E+03
1.0904BW3
1.0884E+03
1.0863E+03
1.0849BHJ3
1.0817E+03
1.0821E+03
1.0815W03
1.0799EW3
1.0790E+03
1.0787EI03
1.0786E+03
1.0785E+03
1.0790E+03
1.0790E+03
1.0791E+03
1.0787E+03
1.0781E+03
1.0782E+03
1.0781E+03
1.0733E+03
1.0725E+03
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1603E+03
1555E+03
1511E+03
1477E+03
1447E+03
1416E+03
1376E+03
1340E+03
1276E+03
1164E+03
1099E+03
1049E+03
0979E+03
0913E+03
0888EI03
0863E+03
0848E+03
0800E+03
0823E+03
0820E+03
0810E+03
0804E+03
0801E+03
0799E+03
1.0798EW3
1.
0800E+03
U801E+03
1.0801E+03
1.0797E+03
1.0786E+03
1.0793EW3
1.0795E+03
1.0776E+03
1.0767BHJ3
1.1627E+03
1.1578E+03
1.1534EHB
1.1501E+03
1.M73EW3
1.1444E+03
1.1408E+03
1.1370E+03
1.1302E+03
1.1197E+03
1
.1130E+03
1.1075E+03
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.1005E-HJ3
.0924E+03
.0891E+Q3
.0865E+03
,0851E-f03
.0824E+03
.0829E+03
.0826E+03
.0819E+03
.0815E+03
.0812E+03
.0810E+03
.0809E+03
.0809E+03
.0810E+03
.0810E+03
.0808E+03
.0804E+03
.0806E-I-03
.0807E+03
.0809E+03
.0805E+03
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1657BW3
1606E403
1562EI03
1529E+03
1503E+03
1476E+03
1443E+03
1403E+03
1334E+03
1234E+03
1165E+03
1109E+03
1046EW3
0966E+03
0907E+03
0868E+03
0858E+03
0843E+03
0839E+03
0834EW3
0829E+03
0826E+03
1.0824E+03
1.0822E+03
1.0821E+03
1.0820E+03
1.0821E+03
1.0822E+03
1.0821E+03
1
0820E+03
1.0821E+03
1
1
1.
0823E+03
0831E+03
0831E+03
1
.1687E+03
1.1634H03
1.1590EW3
1.1558EW3
1.1533E+03
1.1508EW3
1.1477E+03
1.1437E+03
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.1366E+03
.1270E+03
.1202E+03
.1147E+03
.1088EW3
.1018E+03
.0936EW3
.0877E+03
.0866E+03
.0857E+03
.0850E+03
.0844E+03
.0840E+03
.0837E+03
.0835E+03
.0834E+03
.0833Et03
.0831E+03
.0832E+03
.0834E+03
.0835E+03
.0836E+03
.0836E+03
.0837E+03
.0840E+03
.0840E+03
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1721E+03
1665E+03
1620E+03
1589EW3
1565E+03
1541E-HJ3
1511E+03
1471E+03
1400E+03
1308E+03
1242E+03
1188E+03
1133E+03
1067E+03
0975E+03
0892E+03
0875E+03
0865E+03
0857E+03
0850E+03
0845E+03
0843E+03
1.0841E+03
1.0840E+03
1.0840E+03
1.0840E+03
1.0844E+03
1.0849EW3
1.0851E+03
1
0851E+03
1.0852E+03
1.0852E+03
1.0852E+03
1.0851E+03
2-81
-------
19
20
21
22
23
24
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
1.1761E+03
1.1699E+03
1.1653E+03
1.1623E+03
1.1599EW3
1.1577E+03
1.1548EW3
1.150BEt03
1.1439E+03
1.1350EW3
1.1286EM3
1.1236E+03
1.1185EW3
1.1127EI03
1.1053EW3
1.0978E+03
1.0932EI03
1.0907E+03
1.0881E+03
1.0867EW3
1.0860E+03
1.0856E+03
1.0853E+03
1.0852E*03
1.0853EW3
1.0855Et03
1.0857EW3
1.0867E+03
1.0369EW3
1.0B69EI03
1.0869Etfl3
1.0868EW3
1.0866EW3
1.0B63EW3
1.1790EW3
1.1728EW3
1.1684EW3
1.1654E+03
1.1631E+03
1.1609Et03
1.1581E+Q3
1.1542E+03
1.1474E+03
1.1389E+03
1.1329Et03
1.1282E+03
1.1236EW3
1.1185EI03
1.1127E+03
1.1078E+03
1.1044E+03
1.1018E-H13
1.0982EW3
1.0949EW3
1.0930E+03
1.0918E+03
1.0909Et03
1.0898E+03
1.0891E+03
1.0892E+03
1.0888E+03
1.0891E+03
1.0889EI03
1.0887E+03
1.0886E+03
1.0884E+03
1.0880E-H13
1.0875E+03
1.1811E+03
1.1753E+03
1.1710E+03
1.1681E+03
1.1659E+03
1.1638E+03
1.1610EW3
1.1572E+03
1.1507E+03
1.1425E+03
1.1369E+03
1.1325E+03
1.1283EW3
1.1239EW3
1.1192EW3
1J154E+03
1.1127E+03
1.1106E+03
1.1084E+03
1.1051E+03
1.1022E+03
1.0999E+03
1.0980E+03
1.0955E+03
1.0933E-f03
1.0925E+03
1.0912EW3
1.0907E+03
1.0906EM3
1.0904E+03
1.0902E+03
1.0899E+03
1.0894E+03
1.0888E+03
1.1828E+03
1.1775E+03
1.1734EW3
1.1706E+03
1.1684E+03
1.1663E+03
1.1637E+03
1.1600E+03
1.1536EW3
1.1458E+03
1.1405E+03
1.1364E+03
1.1326E+03
1.1286E+03
1.1246E+03
1.1214E+03
1.1192E+03
1.1174E+03
1.1163E+03
1.1140E+03
1.1110E+03
1.1081E+03
1.1055E+03
1.1028E+03
1.0991E+03
1.0960E+03
1.0928E+03
1.0917E+03
1.0918E+03
1.0917E+03
1.0915E+03'
1.0913E+03
1.0906EW3
1.0898E+03
1.1843EW3
1.1793E+03
1.1753E+03
1.1726E+03
1.1705E+03
1.1685E+03
1.1659E+03
1.1623E+03
1.1562E+03
1.1487E+03
1.1437E+03
1.1399E+03
1.1362E+03
1.1326E+03
1.1290E+03
1.1261E+03
1.1240E+03
1.1222E+03
1.1211E+03
1.1192E+03
1.1164E+03
1.1131Et03
1.1103EW3
1.1074E+03
1.1036E-H13
1.0992EW3
1.0944E+03
1.0926E+03
1.0925E+03
1.0924E+03
1.0922E-HJ3
1.0921E+03
1.0915E+03
1.0907E+03
1.1856E+03
1.1808E+03
1.1770E+03
1.1743E+03
1.1723E+03
1.1704E+03
1.1678E+03
1.1644E+03
1.1584E-I-03
1.1513E+03
1.1465E+03
1.1428E+03
1.1394E+03
1.1360E+03
1.1326E+03
1.1299E+03
1.1278E+03
1.1258E+03
1.1240E+03
1.1222E+03
1.1203E+03
1.1179E+03
1.1150EW3
1.1122E-f03
1.1081E+03
1.1007E+03
1.0953E+03
1.0937E+03
1.0934E+03
1.0931E+03
1.0928E+03
1.0926E+03
1.0923E+03
1.0915E+03
2-82
-------
25
26
27
28
29
30 31
32
33
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1867E+03
1821E+03
1784E+03
1758E+03
1738E+03
1719E+03
1695E+03
1661E+03
1604E+03
1535E+03
1489E+03
1454E+03
1421E+03
1389E+03
1356E+03
1329E+03
1308E+Q3
1287E+03
1266E+03
1248EW3
1232E+03
1215E+03
1192E+03
1169E+03
1133E+03
1060E+03
0976E+03
0954E+03
0949EW3
0943E+03
0937E+03
0932E+03
0927EI03
0920EW3
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1879E+03
1835E+03
1799E+03
1775E+03
1755E+03
1737E+03
1714E+03
1681E+03
1627E+03
1561E+03
1517E+03
1484E+03
1452E+OJ
1421E+03
1389E-I-03
1363E+03
1342E+03
1319E+03
1296E+03
1277Et03
1262EW3
1249E+03
1230E+03
1210E+03
1179E+03
1113E+03
1020E+03
0976E+03
1.0968E+03
1.
1.
1.
0960E+03
0953E+03
0943E+03
1.0934E-HJ3
1.
0924E+03
1.1891E+03
1.1850E+03
1.1815EW3
1.1792E+03
1.1774E+03
1.1756E+03
1.1734E+03
1.1704E+03
1.1654E+03
1.1592E+03
1.1549E+03
1.1518E+03
1.1487E+03
1.1457E+03
1.1426E+03
1.1400E+03
1.1377EW3
1.1352E+03
1.1327E+03
1.1303E+03
1.1283E+03
1.1266E+03
1.1250E+03
1.1231E+03
1.1207E+03
1.1174E+03
1.1130E+03
1.1080E+03
1.1048E+03
1.1028E+03
1.1015E+03
1.0999E+03
1.0975E+03
1.0943E+Q3
1.1901E+03
1.1860E+03
1.1825E+03
1.1802E+03
1.1784E+03
1.1768E+03
1.1747E+03
1.1720E+03
1.1675E+03
1.1618E-HI3
1.1577E-HJ3
1.1546E+03
1.1517E+03
1.1489E+03
1.1460E+03
1.1434E+03
1.1409E-HJ3
1.1381E+03
1.1352E+03
1.1326E+03
1.1304E+03
1.1288E+03
1.1272E+03
1.1254E+03
1.1232E+03
1.1207E+03
1.U79E+03
1.1148E+03
1.1121E+03
1.1099E+03
1.1082E+03
1.1062Et03
1.1029E+03
1.0981E+03
1.1916E+03
1.1866E+03
1.1824E+03
1.1802E+03
1.1786E+03
1.1771E+03
1.1752EW3
1.1727Et03
1.1686E+03
1.1633E+03
1.1593E+03
1.1564E+03
1.1536E+03
1.1509E+03
1.1482E+03
1.1457E+03
1.1432E+03
1.1404E+03
1.1374E+03
1.1347E+03
1.1325E-W3
1.1308E+03
1.1293E+03
1.1277E+03
1.1259EW3
1.1238E+03
1.1213E+03
1.1187E+03
1.1163E+03
1.1143E+03
1.1127E+03
1.1106E+03
1.1072E+03
1.1021E+03
*
4
1.1803E+03
1.1792E+03
1.1780E+03
1.1766E+03
1.1751E+03
1.1729E+03
1.1691E+03
1.1640E+03
1.1601E+03
1.1573E+03
1.1546E+03
1.1520E+03
1.1494E+03
1.1470E+03
1.1447E+03
1.1420E+03
1.1390EW3
1.1364E+03
1.1343E+03
1.1326E+03
1.1312E+03
1.1296E+03
1.1280E+Q3
1.1261E+03
1.1238E+03
1.1214E+03
1.1192E+03
1.1173E+03
1.1158E+03
1.1138E+03
1.1105E+03
1.1054E+03
*
*
i
'*
t
1.1749E+03
1.1743E+03
1.1727E+03
1.1692E+03
1.1640E+03
1.1597E+03
1.1570E+03
1.1546E+03
1.1522E+03
1.1498EW3
1.1476E+03
1.1454E+03
1.1429E+03
1.1402E+03
1.1377E+03
1.1357E+03
1.1342E+03
1.1328E+03
1.1314E+03
1.1297EW3
1.1278EW3
1.1255E+03
1.1232E+03
1.1212E+03
1.1195E+03
1.1180E+03
1.1161E+03
1.1128E+03
1.1074E+03
i
4
*
*
*- '
1.1563E+03
L1554E+03
1.1538E+03
1.1517E+03
1.1493E+03
1.1473E+03
1.1454E+03
1.1432E+03
1.1409E+03
1.1385E+03
1.1364EW3
1.1348E+03
1.1336E+03
1.1323E+03
1.1307E+03
1.1289E+03
1.1267E+03
1.1245E+03
1.1225E+03
1.1210E+03
1.1196E+03
1.1178EW3
1.1149E+03
1.1102E+03
*
i
*
i
1.1469E+03
1.1460E+03
1.1448E+03
1.1431EM3
1.1410E+03
1.1388E+03
1.1368E+03
1.1352E+03
1.1340E+03
1.1327E+03
1.1312E+03
1.1294EW3
1.1273E+03
1.1251E+03
1.1233E+03
1.1218E+03
1.1206E+03
1.1190E+03
1.1166E+03
1.1137E+03
***** PATHLINE 4 PAETICLE TMCKIHG DATA *****
Nuiber of forward particles : 0
Huiber of reverse particles : 0
Particles released at wells :
Well-No. Particles released
1 10
2 10
3 10
4 10
5 10
6 10
2-83
-------
-------
EPA-600/R-93/
APRIL 1993
CHAPTERS
MONITORING SYSTEM DESIGN FOR WELLHEAD PROTECTION
LITTLETON, MASSACHUSETTS
by
Alan H. Cathcart and Savas C. Danos
Littleton Water Department
Littleton, Massachusetts 01460
Thomas Franz and Nilson Guiguer
Waterloo Hydrogeologic Software
Waterloo, Ontario, Canada N2L-5Y9
Beth A Moore
Lockheed Environmental Systems & Technologies Company
Las Vegas, Nevada 89119
April 1993
Contract Number CR-816199-01-1
Project Officer
Steven P. Gardner
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas, Nevada 89193-3478
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89193-3478
-------
NOTICE
This report is the result of research supported by the U.S. Environmental Protection Agency's
Environmental Monitoring Systems Laboratory in Las Vegas, Nevada, as part of its efforts to provide
technical assistance to state, tribal, and local governments on the implementation of the Wellhead
Protection Program. The specific methods and approaches contained in this document have
undergone peer review but do not constitute official Agency endorsement or policy recommendations.
The Office of Research and Development provides this information to help solve complex technical
problems related to refined delineation and ground-water monitoring of wellhead protection areas in
various hydrogeologic settings. Further assistance is available from the Environmental Monitoring
Systems Laboratory hi Las Vegas, the Office of Ground Water and Drinking Water in Washington,
D.C, and the ground-water offices in the ten U.S. Environmental Protection Agency regions.
3-ii
-------
ABSTRACT
As a management approach for wellhead protection of Production Well Number 5, PW-5, the
Littleton Water Department proposes source-assessment and early-warning detection monitoring.
Techniques for refined delineation of the PW-5 wellhead protection area and long-term monitoring
approaches are discussed in this report. Production Well Number 5 is completed at a depth of 167
feet within saturated, stratified valley-fill deposits. The aquifer is unconfined and it receives significant
surface-water recharge (20% to 25%) from nearby Spectacle Pond and Bennetts Brook.
Land-use activities within the wellhead protection area cover a broad range of commercial,
industrial, and to a lesser degree, agricultural operations. Collectively, these land-use activities pose
potential contamination threats to the aquifer, including heavy metals, volatile organic compounds,
pesticides, and nutrients. Baseline monitoring results indicate that, to date, ground-water quality
within the capture zone of PW-5 is virtually unaffected by source operations. Sodium is the only
exception. Slightly elevated levels of sodium in surface water and the shallow aquifer are attributed to
roadway de-icing activities. Manganese and iron concentrations are elevated throughout the recharge
area of PW-5, primarily because of their occurrence in wetland sediments and glacial deposits. The
levels of these parameters have increased at PW-5 for several years and may warrant treatment in the
future.
The PW-5 wellhead protection area consists of three protection zones delineated using a
combination of numerical ground-water flow modeling (FLOWPATH, FLOWCAD, and
MODFLOW) and hydrogeologic mapping. Zone I is the 400-foot sanitary protective radius
mandated by the State of Massachusetts. Zone II is the most critical management area and was
delineated conservatively as the union of three numerical capture zone solutions. These numerical
solutions incorporate two- and three-dimensional flow, as well as steady-state and transient flow
conditions. Local and regional ground-water flow simulations are based on the results of short- and
long-term aquifer testing.
Zone II generally represents the steady-state capture zone for PW-5 that corresponds
approximately to the 400-day travel-time contour. Flowpath simulations indicate that Zone II extends
to the bottom of the aquifer and is constrained by bedrock and glacial till. Within Zone II, three
existing wells and two new wells are proposed for inclusion in the long-term monitoring network for
early-warning detection and source assessment purposes. These wells lie along either the 150-day or
the 300-day travel-time contours. Screened intervals for the new monitoring wells were chosen based
on results from MODPATH computer flow simulations. Monitoring parameter groups for these wells
include general water quality parameters (cations, anions, alkalinity, etc.), site-specific parameters
(total coliform, heavy metals, volatile organic compounds, etc.), and physical parameters (water level,
pH, temperature, etc.). Recommended monitoring frequencies for these parameter groups vary from
quarterly to annually, depending on the travel-time distance from the monitoring well to PW-5 and
the monitoring well depth.
3-iii
-------
Zone in is defined as the upgradient area of the aquifer that contributes to Zone II and
extends to the watershed boundary. Zone III is monitored at two surface-water locations, one at the
inflow and one at the outflow of Spectacle Pond. In addition, Zone III is monitored biannually at
existing compliance networks around waste management and industrial sites. Monitoring parameters
for the compliance wells include general water quality, site-specific, and physical parameter groups.
Monitoring results ultimately support land-use planning, risk-assessment screening of new developers,
existing development compliance, and general Wellhead Protection Program management.
The Littleton Wellhead Protection Program incorporates contingency planning. Catastrophic
releases initiate a spill-response plan that involves many departments and agencies. In the event of
contamination of PW-5 or another production well, Littleton has sited a new production Well. The
proposed well site is approved by the State, and protection Zones I, II, and III are delineated. The
adjacent town of Boxborough shares the recharge area to the proposed well. Boxborough has
adopted complementary strategies with Littleton to ensure its water qualify protection.
3-iv
-------
CONTENTS
Abstract iii
Figures vii
Tables xi
Abbreviations, Symbols, and Conversion Factors xiii
Acknowledgements xvii
Background for the Littleton Case Study 3-1
Introduction 3-1
Wellhead Protection Program Overview 3-1
Hydrogeologic Setting 3.4
Preliminary Wellhead Protection Area . 3-5
Source Assessment 3-10
Research Monitoring Program 3-14
Data Base Limitations 3-14
Research Objectives 3-15
Research Tasks 3-15
Data Acquisition 3-17
Hydrogeologic Data Review 3-17
Ground-Water Monitoring Stations 3-17
Aquifer Testing 3-20
Peat-Probing Exercise 3-27
Ground-Water Sampling and Analysis . 3-27
Data Interpretation . 3.34
Conceptual Model 3-34
Aquifer Testing and Analyses 3-36
Ground-Water Flow Modeling 3-44
Water Quality Monitoring 3-66
Refined Wellhead Protection Areas 3-74
Wellhead Protection Monitoring Program 3-77
Monitoring Objectives 3-77
Monitoring Sites 3-77
Ground-Water Stations .... '. 3-77
Surface-Water Stations .... .T 3-80
Monitoring Parameters and Frequencies 3-80
Ground-Water Stations 3-81
Surface-Water Stations 3-83
Quality Assurance and Quality Control 3-84
Monitoring Data Base Storage, Update, and Retrieval Systems 3-84
Contingency Planning 3-85
3-v
-------
CONTENTS, Continued
Conclusions
References
Additional Bibliography
3-89
3-93
3-97
3-vi
-------
FIGURES
Number Page
3-1 Regional location of the town of Littleton in northeastern Massachusetts 3-2
3-2 Location of PW-5, Spectacle Pond, and the distribution of valley-fill
deposits within the Bennetts Brook watershed, Littleton, Massachusetts 3-3
3-3 Location of Spectacle Pond and PW-5 with regional topography, Littleton,
Massachusetts 3-6
3-4 Construction diagram for PW-5, Littleton, Massachusetts 3-7
3-5 Aquifer and Water Resource Overlay Districts within the Bennetts Brook
and Beaver Brook watersheds, Littleton, Massachusetts 3-9
3-6 Location of existing land-use activities within the Bennetts Brook watershed,
Littleton, Massachusetts 3-12
3-7 Ground-water monitoring device construction diagrams for (A) single-drilled
well; (B) MicroWell; and (C) piezometer 3-18
3-8A Ground-water elevation stations for near-field studies, Bennetts Brook
watershed, Littleton, Massachusetts : 3-21
3-8B Ground-water elevation stations for far-field studies, Bennetts Brook
watershed, Littleton, Massachusetts 3-24
3-9 Near-field, water-level, observation and water quality monitoring network,
Bennetts Brook watershed, Littleton, Massachusetts 3-25
3-10 Peat-probing transects at the eastern lobe of Spectacle Pond, Bennetts Brook
watershed, Littleton, Massachusetts 3-28
3-11 Watersheds and hydrogeologic divides near Bennetts Brook, Littleton,
Massachusetts 3-35
3-12A A location map of stratigraphic sections across Spectacle Pond, Bennetts
„ Brook watershed, Littleton, Massachusetts 3-37
3-vii
-------
FIGURES, Continued
Number Page
3-12B Stratigraphic sections across Spectacle Pond, Bennetts Brook watershed
Littleton, Massachusetts 3-38
3-13 Neuman type-curve analysis of piezometer P-20 recovery data, short-term
test, Bennetts Brook watershed, Littleton, Massachusetts 3-39
3-14 Regional two-dimensional model boundary and grid domain for the Spectacle
Pond aquifer, Bennetts Brook watershed, Littleton, Massachusetts 3-46
3-15 Regional bedrock elevation contour map, Bennetts Brook watershed, Littleton,
Massachusetts < 3-47
3-16 Regional hydraulic head elevations for the Spectacle Pond aquifer, Bennetts
Brook watershed, Littleton, Massachusetts 3-50
3-17 Hydraulic head elevations in the vicinity of PW-5, Bennetts Brook watershed,
Littleton, Massachusetts 3-51
3-18 Comparison of the zone of contribution and zone of influence for PW-5 in the
presence of a regional hydraulic gradient, Bennetts Brook watershed,
Littleton, Massachusetts 3-52
3-19 Time-related capture zones determined by two-dimensional modeling in the
vicinity of PW-5, Bennetts Brook watershed, Littleton, Massachusetts 3-53
3-20 Three-dimensional model boundary, grid domain, and cross section locations
in the vicinity of PW-5, Bennetts Brook watershed, Littleton,
Massachusetts » 3-56
3-21 Multi-layer grid for the three-dimensional model in the vicinity of PW-5
Bennetts Brook watershed, Littleton, Massachusetts 3-57
3-22 Cross sections of hydraulic head distribution, E-E' and. D-D', determined by
MODFLOW and MODPATH 3-59
3-23 Plan view of hydraulic head distributions for each layer of the three-
dimensional model, Bennetts Brook watershed, Littleton, Massachusetts 3-60
3-viii
-------
FIGURES, Continued
Number
3-24 Time-related capture zones for PW-5 for each layer of the three-dimensional
model, Bennetts Brook watershed, Littleton, Massachusetts , 3-61
3-25 Capture zone for PW-5 determined by FLOWCAD using the 1991 Massachusetts
Department of Environmental Protection delineation criteria, Bennetts Brook
watershed, Littleton, Massachusetts 3-64
3-26 Comparison of the capture zones for PW-5 generated by the 3-D steady state,
2-D steady-state, and 2-D transient flow models, Bennetts Brook watershed,
Littleton, Massachusetts 3-65
3-27 Piper diagram comparison of average water quality in shallow and deep ground-
water zones near PW-5, Bennetts Brook watershed, Littleton, Massachusetts .... 3-67
3-28 Piper diagram comparison of ambient water quality changes during the long-term
aquifer test in the (A) shallow and (B) deep aquifer systems, Bennetts
Brook watershed, Littleton, Massachusetts 3-68
3-29 Increasing concentrations of iron (Fe) and manganese (Mn) in PW-5 from 1984
through 1990, Bennetts Brook watershed, Littleton, Massachusetts 3-70
3-30A Spatial distribution of iron concentrations near PW-5, Bennetts Brook
watershed, Littleton, Massachusetts 3-71
3-30B Spatial distribution of manganese concentrations near PW-5, Bennetts Brook
watershed, Littleton, Massachusetts 3-72
3-31 Water temperature changes observed at depths greater than 25 feet near PW-5
during the long-term aquifer test, Bennetts Brook watershed, Littleton,
Massachusetts 3-73
3-32 Comparison of temperatures at deep and shallow cluster wells exhibiting
warming trends during the long-term aquifer test, Bennetts Brook
watershed, Littleton, Massachusetts 3-73
3-33 Proposed wellhead protection zones for PW-5, Bennetts Brook watershed,
Littleton, Massachusetts 3-75
3-ix
-------
FIGURES, Continued
Number
Page
3-34 Optimized wellhead protection monitoring stations for PW-5, Bennetts Brook
watershed, Littleton, Massachusetts 3-79
3-35 Spill-response hierarchy chart, Littleton, Massachusetts 3-86
3-36 Proposed well site and protection zones, Bennetts Brook watershed,
Littleton, Massachusetts 3-87
3-x
-------
TABLES
Number
3-1 Potential Sources of Contamination Within the Bennetts Brook Watershed,
Littleton, Massachusetts
3-2 Distances from PW-5 to Monitoring Stations, Station Depths, and Screen
Lengths for the Short-Term Aquifer Test, Bennetts Brook Watershed,
Littleton, Massachusetts
3-3 Distances from PW-5 to Stations, Station Depths, and Screen
Lengths for the Long-Term Aquifer Test, Bennetts Brook Watershed,
Littleton, Massachusetts
3-4 Water Quality Data for the Monitoring Network near PW-5, Littleton,
Massachusetts
Page
3-11
3-22
3-26
3-29
3-5 Aquifer Parameter Results for the Short-Term Aquifer Test at PW-5, Littleton,
Massachusetts
3-6 Aquifer Parameter Results for the Long-Tenn Aquifer Test at PW-5, Littleton,
Massachusetts
3-40
3-43
3-7 List of Chemical and Physical Monitoring Parameters for the PW-5
Wellhead Protection Area, Littleton, Massachusetts
3-8 Long-Term Monitoring Program for the PW-5 Wellhead Protection Area, Bennetts
Brook Watershed, Littleton, Massachusetts
3-81
3-82
3-xi
-------
-------
ABBREVIATIONS, SYMBOLS, AND CONVERSION FACTORS
ABBREVIATIONS
2-D
3-D
A
amsl
BA
BP
C
cfs
Cond
d
DEP
DNAPL
elev
EPA
ft
ft/d
ft^/d
ftVmin
GIS
gpd
gpd/ft
gpf
gpm
GWQ
H
Hard
Herb
HM
in
in/yr
1
LF1
LNAPL
(ig/1
M
mg
two-dimensional
three-dimensional
annually
above mean sea level
biannually
before pumping
Cooper-Jacob corrected drawdown analysis
cubic foot per second
specific conductivity
day
(Massachusetts) Department of Environmental Protection
dense, nonaqueous phase liquid
elevation
U.S. Environmenal Protection Agency
foot
foot per day
square foot per day
square foot per minute
Geographic Information System
gallon per day
gallon per day per foot
gallon per foot
gallon per minute
general water quality parameters
Hantush analysis with aquitard storage
hardness
herbicide
heavy metal parameters
inch
inch per year
liter
miscellaneous monitoring station number 1
light, nonaqueous phase liquid
microgram per liter
microSieman per centimeter
monthly
milligram
3-xiii
-------
mg/l milligram per liter
Mgd millon gallon per day
min minute
MPA microscopic participate analysis
MW monitoring well
MW-25d monitoring well, 25 feet from PW-5, deep well
MW-25s monitoring well, 25 feet from PW-5, shallow depth well
MW-lOOm monitoring well, 100 feet from PW-5, moderate depth well
MW-C compliance monitoring well
MW-E existing monitoring well
MW-P proposed monitoring well
N Neuman analysis
N/A not applicable
ND not detected
MR no response
NRC National Response Center
OD outside diameter
OW observation well
P piezometer
PCB polychlorinated biphenyl
Pest pesticide
Pt pumping test
PVC polyvinyl chloride
PW production well
PW-5 Production Well Number 5
Q quarterly
QA quality assurance
QC quality control
Rt recovery test
SI staff guage number 1
SCH schedule
SDWA Safe Drinking Water Act
SS stainless steel
SW surface-water monitoring station
TC total coliform
TDD Theis drawdown versus distance analysis
TDS total dissolved solid
Temp temperature
Th Theis corrected drawdown analysis
TOT time of travel
USGS U.S. Geological Survey
UST underground storage tank
VOC volatile organic compound
WHPA wellhead protection area
WHPP Wellhead Protection Program
WL water level
3-xiv
-------
zoc
ZOI
year
zone of contribution
zone of influence
SYMBOLS
Ag
As
Ba
Ca
Cd
a
Cr
Cu
oF
Fe
HC03
Hg
K
Mg
Mn
Na
NH3
NO2
NO3
Pb
Se
SO4
silver
arsenic
barium
calcium
cadmium
chlorine
chromium
copper
degree Fahrenheit
iron
bicarbonate
mercury
potassium
magnesium
manganese
sodium
ammonia
nitrite-nitrogen
nitratre-nitrogen
lead
selenium
sulfate
P
b
K
Kr
Kz
KJr/Kz
n
r
R
S
Sy
T
Neuman analysis-fitting parameter
saturated aquifer thickness
hydraulic conductivity
horizontal hydraulic conductivity
vertical hydraulic conductivity
anisotropy ratio
porosity
radius
distance from Production Well Number 5 to monitoring station
storativity
specific yield
transmissivity
3-xv
-------
CONVERSION FACTORS
Multiply
acre-foot
cubic foot per second
foot
foot per second
gallon
gallon
gallon
gallon per day
gallon per day per foot
gallon per day per square foot
inch
inch per year
mile
million gallons per day
square foot per minute
square foot per second
square mile
By
1230
0.0283
0.3048
0.3048
3.785
0.134
0.00379
0.000003528
0.000207
0.0408
0.0254
25.4
1.609
2.629
0.0929
0.0929
2.59
To Obtain
cubic meter
cubic meter per second
meter
meter per second
liter
cubic foot
cubic meter
cubic foot per second
square meter per day
meter per day
meter
millimeter per year
kilometer.
cubic meter per minute
square meter per minute
square meter per second
square kilometer
co +32
3-xvi
-------
ACKNOWLEDGEMENTS
This document was prepared for the U.S. Environmental Protection Agency (EPA), Office of
Water and Drinking Water under Contract Number CR-816199-01-0 among the U.S. Environmental
Protection Agency, Environmental Monitoring Systems Laboratory in Las Vegas, Nevada (EMSL-
LV) and the Littleton Water Department, Littleton, Massachusetts. We gratefully acknowledge the
support of Steven Gardner (EMSL-LV) who served as the EPA Project Officer. The authors are
indebted to the following individuals who provided critical review comments: Dr. Charles Kreitler
(University of Arizona), Douglas Heath (U.S. EPA Region I), Dr. Philip Berger (U.S. EPA Office of
Water), Theodore Morine (D.L. Maher, Co.), and Dr. Robert Cleary (Robert Cleary & Assoc.).
Special thanks are extended to the many people at the Littleton Water Department whose
support has been invaluable, including General Manager Curtis J. Lanciani and the Board of Water
Commissioners; Deborah Rickley, who assisted in data acquisition; Kay Johnson, who provided
administrative support; and the Water Operations staff and employees for then: assistance in field
work and data acquisition. John Nicholson and Carolyn Cameron of Lockheed Environmental
Systems & Technologies Company (LESAT) provided technical writing and editing support for
preparation of the manuscript Steve Garcia (LESAT) and Richard Pearce (Waterloo Hydrogeologic
Software) contributed excellent graphical products to the report. Shalena Fendzlau (LESAT) is
graciously acknowledged for her patience and expertise in word processing support.
3-xvii
-------
-------
BACKGROUND FOR THE LITTLETON CASE STUDY
INTRODUCTION
Five years before enactment of the U.S. Environmental Protection Agency's (EPA) Wellhead
Protection Program (WHPP), the town of Littleton, Massachusetts, researched and adopted a
comprehensive aquifer and watershed protection program that has served as a model for state and
local governments. Littleton is a semi-rural community that has experienced significant commercial
and industrial development over the past 10 years. The town is located in northeast Massachusetts,
approximately 35 miles northwest of Boston, at the intersection of two major transportation routes,
1-495 and Route 2 (Figure 3-1). Development of the Littleton WHPP is briefly discussed in this
paper, together with recent refinements in wellhead protection area (WHPA) delineation and
ground-water monitoring in proximity to Production Well Number 5 (PW-5), located within the
Bennetts Brook watershed (Figure 3-2).
Since the inception of Littleton's aquifer and water resource protection plan, ground-water
monitoring activities have played an integral role in the overall success of the program. Because of
the formidable time and expense required to conduct these monitoring activities, it became necessary
to establish an optimized monitoring program. The benefits of such a program include more reliable
ground-water monitoring, reduced analytical costs (otherwise misappropriated at improperly sited
monitoring stations), and increased efficiency for related WHPP efforts. Optimized monitoring
requires accurate WHPA delineation as a prerequisite for effective land-use planning and management
policy development.
At present, the Spectacle Pond aquifer (Figure 3-2) is a primary water resource for the town
of Littleton. Because of its high capacity, good water quality, and potential vulnerability to future
development, this aquifer was selected for reevaluation using recently developed technologies.
Specifically, computer simulation techniques were employed to delineate the WHPA with associated
ground-water flowpaths and travel times accurately. The goal of this research was to create a water
quality monitoring program within the refined WHPA established for PW-5. These activities will
ensure the protection of this water supply in the future.
WELLHEAD PROTECTION PROGRAM OVERVIEW
Littleton's boundaries encompass 16 square miles, and'the population of the town is
approximately 7,300. Over two-thirds of the community are serviced by the public water supply
system. The daily water demand varies from 800,000 to 1,500,000 gallons per day (gpd) and is met by
four ground-water production wells located in two separate watersheds (Bennetts and Beaver Brook
watersheds, Figure 3-2). The total yield of these wells is approximately 2,800,000 gpd. Residents
outside of the town's distribution system generally depend on low-yielding bedrock wells to meet their
3-1
-------
MASSACHUSETTS
Figure 3-1. Regional location of the town of Littleton in northeastern Massachusetts.
(Moore et al, 1990, p. 184)
water needs. At present, there is no public sewage collection system, and sanitary wastes are
discharged into individual septic systems. Some industries discharge waste in excess of 15,000 gpd, and
these industries are serviced by independent secondary or tertiary treatment facilities.
In 1979, the citizens of Littleton formed an Ad Hoc Committee for Ground-Water Protection
to administer a town-wide hydrogeologic evaluation. The study was conducted in response to an
estimated 70% loss of ground-water supply capacity by the town of Acton, which borders Littleton on
the southeast (Figure 3-2). The loss in capacity was attributed to organic contamination from a
nearby industrial development. Littleton's citizens appropriated $70,000 to hire a consultant to
complete a hydrogeologic investigation, to site new ground-water development areas, and to
recommend protection strategies for existing and proposed supplies. In 1981, the consultants made
recommendations, which served as the foundation for the creation of zoning overlay districts, the
adoption of a hazardous materials bylaw, and the adoption of other local health and wetlands
protection ordinances.
Most preexisting industrial and commercial facilities that pose a contamination risk and are
located within overlay districts willingly cooperate with regulatory agencies in protecting supply wells.
Proposed developments within overlay districts are subject to a risk-assessment screening process. If
low to moderate contaminant risk is determined and a development permit is issued, then the
developer must demonstrate nondegradation of ground-water quality at the property perimeter.
Compliance is provided by certified laboratory analyses and maintained through scheduled sampling of
onsite monitoring wells.
3-2
-------
LITTLETON
ROUTE 2
1-495
BOXBOROUGH
•Spectacle Pond
LEGEND
- TOWN BOUNDARY
- STREAM
POND
WATERSHED BOUNDARY
%%% BEDROCK AND GLACIAL TILL
I VALLEY-FILL DEPOSITS
PRODUCTION WELL (PW)
ACTON
SCALE IN MILES
Figure 3-2. Location of PW-5, Spectacle Pond, and the distribution of valley-fill deposits within the
Bennetts Brook watershed, Littleton, Massachusetts. (Modified after Moore et al., 1990, p. 185)
3-3
-------
Town governing boards such as the Planning Board, the Board of Health, and the
Conservation Commission participate in regulation of the WHPP. Daily implementation of
ground-water monitoring, data collection, source assessment inspections, and overall management is
performed by the Littleton Water Department. Hydrogeologic and engineering consultants are hired
as needed. Development of the WHPP in Littleton is guided by the Board of Water Commissioners
and the General Manager of the Water Department, who have been instrumental in the evolution of
the water resource protection strategy and have encouraged its progression.
HYDROGEOLOGIC SETTING
Littleton is situated in a broad lowland belt where Pleistocene-age glacial and postglacial
deposits form a discontinuous mantle over igneous and metamorphic rocks of Paleozoic age (Jahns,
1953). Bedrock underlying the Spectacle Pond study area is identified as the Lower Silurian
Chelmsford fades, composed of a porphyritic biotite granite. Topographically prominent features owe
their expression to resistive schists and gneisses, whereas other elongated highs are composed of
glacial drift. The northeast-trending orientation of bedrock eminences reflect structural trends
associated with a high-angle thrust fault, regionally known as the Clinton-Newbury fault system, and
Tertiary Period drainage patterns (Zen, 1983). Drumlin, striae, and groove orientations indicate a
southeast advancement and depositional environment attributed to the Wisconsin ice sheet (Jahns,
1953).
The hydrogeologic setting of public ground-water resources in Littleton is characterized as an
unconfined to semi-confined, buried valley system. Surface water and ground water generally flow
northeast, parallel to the strike of the fault system within the Bennetts Brook and Beaver Brook
watersheds (Figure 3-2). Typically, these valley-fill deposits include outwash sands and gravels;
glaciolacustrine deposits with sequences of gravel, sand, silt, and carved clay; ice contact features of
boulder to sand-size fractions; and postglacial peat, muck, silt, and fine sands indicative of wetland
environments.
To simplify the local hydrogeologic description, the areal extent of relatively impermeable
bedrock and glacial till deposits is represented by the shaded areas in Figure 3-2. Conversely, the
unshaded areas consist of highly-permeable outwash sands and gravels, as well as moderately-
permeable stratified drift and recent deposits. Figure 3-2 shows the predominance of low-yield
bedrock and glacial tills in the southern portion of town.
North-to-south trending gravity profiles completed across the Beaver Brook watershed and
test-well depth data indicate that bedrock depths range from 0 to 140 feet (Kick, 1980). Similar
profiles and test wells completed within the Bennetts Brook watershed indicate that bedrock depths
range from 0 to 105 feet Depths to water typically range from 0 to 25 feet and 0 to 35 feet within
the Beaver Brook and Bennetts Brook watersheds, respectively. Other significant aquifer property
ranges include hydraulic conductivities between 1,200 and 3,100 gallons per day per foot (gpd/ft);
porosities between 25% and 35%; and storativities between 0.05 and 0.0003 (Goldberg-Zoino &
Associates Inc., 1987; D. L. Maher Co., 1985). Aquifer studies indicate that semi-confined conditions
exist in areas where peat and muck deposits are significant, particularly in the wetlands.
3-4
-------
The Spectacle Pond study area is located in the northwest corner of Littleton within the
Bennetts Brook watershed, a sub-basin of the Stony Brook watershed that feeds into the Merrimack
River (Figure 3-3). The pond is one of the more prominent hydrologic features within the study area,
covering more than 70 acres. Spectacle Pond consists of two interconnected lobes that are oriented
southwest to northeast (Figure 3-3). The southwest lobe has a maximum depth of 33 feet, and the
northeast lobe, adjacent to PW-5, has a maximum depth of 14 feet. Local topographic relief is
characteristic of widespread glacial ice-contact deposits found throughout New England.
Production Well Number 5 is located on the southeast shore of the pond. This 28- X 48-inch,
gravel-packed well was constructed in 1983 to a depth of 50.5 feet within the valley-fill deposits. The
bottom 15 feet of the well consists of a 24-inch screen (Figure 3-4), which draws water from the
unconfined aquifer. At the time of installation, the calculated safe yield of the well was 1,100 gallons
per minute (gpm), with a specific yield of approximately 1,100 gallons per foot (gpf). At present, the
well delivers an average of 650 gpm into the Littleton distribution system. On the northwestern shore
of the pond, the bordering town of Ayer (Figure 3-2) has developed two production wells, which draw
approximately 1,050 gpm from the same aquifer.
PRELIMINARY WELLHEAD PROTECTION AREA
Prior to 1981, the Massachusetts Department of Environmental Protection (DEP) established
siting criteria requiring municipalities to procure and protect a 400-foot radius around public wells.
This regulation was enacted to reduce the potential deleterious effects attributed to pathogenic
microorganisms resulting from sewage discharge in the immediate area of a municipal production well.
At a time when further water quality protection guidance was unavailable, Littleton's ground-water
consultants developed a process for selecting favorable development areas for ground-water supplies
and delineated WHPA's for land-use regulation enactment.
During the initial hydrogeologic data search, the consultants identified gross deficiencies in the
data. Therefore, geophysical techniques, including gravity profiling and electrical resistivity, were
utilized to estimate depths to bedrock and the saturated thickness of outwash deposits within proposed
regulatory districts. Depth and thickness estimates were verified and calibrated through test borings,
providing more detail of sediment types and stratigraphy.
Protection districts were delineated primarily based on the radius of influence of the
production well determined by aquifer testing. This area was then extended to include the zones
contributing to ground-water and surface-water recharge. The boundaries of protection districts were
justified primarily on the basis of local hydrogeologic conditions, but were extended to include
geopolitical boundaries as well. Other considerations in the delineation process included potential
areas of development; existing potential sources such as road salting activities, an active landfill, service
stations, and industrial development; and zoning issues relating to property boundaries.
The final recommendations for the Aquifer and Water Resources Overlay Districts were
adopted as a zoning bylaw in 1981 (Town of Littleton Planning Board, 1988). The Aquifer Overlay
Districts include the zone of influence (ZOI) and the upgradient, ground-water recharge area for each
3-5
-------
BENNETTS BROOK
WATERSHED
RECHARGE AREA
TO PW-5
< 2,000 t 4.000 , 6,000
SCALE IN FEET
Figure 3-3. Location of Spectacle Pond and PW-5 with regional topography, Littleton, Massachusetts.
3-6
-------
(2.5-IN OD DBS WELL)
0-r
2-
4
6-
8
10-
12
14-
16-
18-
20-
22-
24-
26
28-
30-
32-
PUMP HEAD BASE PLATE
5-FT DBS WELL SCREEN
(2.5-IN OD, 0.01-IN SLOT)
^^j;
>;
"I
n
I. I
j k
3 t
«»
DREEN
SLOT)
£=
••
'
i
•=
.. .
-i
r^
i-1
1
J
^^
~:
t
.
^
1
is
(197-FT ELEV)
1*1 FT wn i ^pnrrM ^T FT nn n 11 IN ^i HT^
Figure 3-4. Construction diagram for PW-5, Littleton, Massachusetts.
(Modified after Haley and Ward, Inc., 1991)
3-7
-------
municipal well. Water Resources Overlay Districts were created as a less critical zone of protection
and include the watershed boundaries identified within the town of Littleton (Figure 3-5). The overlay
districts originally encompassed developed and undeveloped parcels of land, posing serious regulatory
issues for the town. Because the town could not afford to purchase multi-million dollar land parcels
and preexisting industries, performance criteria were established within the new zoning bylaw. This
enabled land development to progress within these districts, while placing the responsibility of water
resource protection on individual land owners. The bylaw was written to make individual owners and
developers (existing and future) responsible for adverse environmental impacts within their property.
Recently, the DEP adopted new WHPA delineation and management requirements, outlined
within the "Guidelines and Policies for Public Water Systems" (Massachusetts Department of
Environmental Protection, 1991). To facilitate land-use management activities, the protection area is
subdivided and prioritized using the following WHPA zones:
• Zone I: The protective radius required around a public water supply well or wellfield.
• Zone II: The area of an aquifer that contributes water to a well under the most severe
pumping and recharge conditions that can be realistically anticipated (180 days
pumping with no recharge from precipitation), extended upgradient to prevailing
hydrogeologic boundaries.
• Zone III: The land area beyond the Zone II area from which surface water and
ground water drain.
For practical purposes, the current state WHPA guidelines closely resemble the Littleton
Overlay Districts established for the Spectacle Pond aquifer. Zone I, the 400-foot protective radius
mandated by the state, is equivalent to the 14 acres surrounding PW-5 owned and controlled by the
Littleton Water Department. Zones II and III, which differentiate the primary recharge area to PW-5
from farther upgradient recharge areas, are equivalent to the Aquifer and Water Resources Districts
established by the town. The Aquifer District actually covers a larger area than Zone II, extending
into the sands and gravels located upgradient of PW-5, but terminating at the less permeable till and
bedrock contacts rather than at other recharge boundaries such as streams or ponds.
Proposed commercial and industrial developments within the overlay districts are subject to
risk evaluation through the special-permit granting authority of the Littleton Planning Board. High-
risk facilities, such as gasoline stations or landfills, cannot be developed within the more sensitive
aquifer district. Moderate- to low-risk facilities are systematically evaluated for hazardous materials
storage, handling, and disposal practices. Other considerations in this screening process include the
percent of impermeable lot coverage (parking lots, buildings, etc.), nutrient loading, proximity to
production wells, and other risks identified during the review process.
In addition to the Aquifer and Water Resources Overlay District Bylaw, Littleton adopted a
Toxic and Hazardous Materials Bylaw (Town of Littleton Planning Board, 1981). This bylaw outlines
requirements for mandatory registration of toxic and hazardous substances used or stored within all
local businesses. Along with current inventories, proper containment for these substances is
mandatory. Regulations regarding underground storage tanks (USTs) are also addressed within this
bylaw. Because these provisions are unrelated to zoning regulations, facilities cannot be
"grandfathered" or waive the onus of compliance.
3-8
-------
LITTLETON
•Spectacle Pond
\ s-
ROUTE 2—3!
1-495
BOXBOROUGH
LEGEND
TOWN BOUNDARY
- STREAM
POND
AQUIFER OVERLAY
DISTRICTS
WATER RESOURCES
OVERLAY DISTRICTS
PRODUCTION WELL
(PW)
ACTON
SPECTACLE
POND
AQUIFER
OVERLAY
DISTRICT
SCALE IN MILES
Figure 3-5. Aquifer and Water Resource Overlay Districts within the Bennetts Brook and Beaver
Brook watersheds, Littleton, Massachusetts. (Modified after Moore et al, 1990, p. 188.)
3-9
-------
Within Littleton, the greatest risks for ground-water contamination are posed by smaller, less
accountable operations that were established years before water resource protection issues were
realized. Larger businesses that handle significant volumes of toxic or hazardous substances on a
day-to-day basis have independently learned of the environmental risk posed by their activities. To
reduce their liability, many of these companies have developed substantial risk reduction programs. In
many instances, the smaller operations posing the greatest risks tend to be overlooked within the
regulatory framework. For this reason, Littleton has implemented very low reporting quantities of
substances under the Toxic and Hazardous Materials Bylaw. The monitoring and documentation
process has become a valuable educational exercise for these smaller industries.
The bylaw is administered locally through the Board of Health, the Building Inspector, and
the Fire Chief. Daily compliance is monitored through scheduled audits conducted by the Water
Department. These audits provide a method by which to identify potential ground-water
contamination sources and to update existing records. Audits also provide an opportunity to educate
commercial and industrial customers about cooperative wellhead protection practices. For some
industries, potential liabilities have been minimized by source reduction practices.
SOURCE ASSESSMENT
Comprehensive source assessment is a necessary component of any" effective ground-water
monitoring program. Shallow sand and gravel aquifers are particularly susceptible to contamination
from wastes generated by overlying impoundments (Delaney and Maevsky, 1980). Hence, proper
identification, evaluation, and education of operations that pose threats of contamination within
protection districts helps to reduce deleterious impacts from inattentive businesses. In Littleton, risks
are minimized by regulating the storage, handling, and disposal activities associated with hazardous
materials. Source reduction is the preferred method of minimizing risks, but other management
strategies are employed if this option is not feasible.
Since 1981, the special permitting process (addressed within the site plan review) has enabled
Littleton's Planning Board to screen and identify moderate- and high-risk operations proposed within
the Aquifer and Water Resources Districts. Preexisting facilities are subject to screening if they plan
an increase of onsite storage of hazardous materials or an increase in lot coverage (above 20% total
property area), or when there is a change in proprietorship. Collateral risk-assessment activities are
conducted by the Littleton Board of Health by means of the Toxic and Hazardous Materials Bylaw.
Source assessment activities are supplemented with site inspections by the Water Department.
Initially, audits were performed at larger industries known to use or suspected of using significant
quantities of toxic or hazardous materials. The priority of site audits was determined by the proximity
of the facility to a municipal production well and the general outward appearance of the facility. A
comprehensive list of potential point and nonpoint sources within the Bennetts Brook watershed is
presented in Table 3-1.
3-10
-------
TABLE 3-1. POTENTIAL SOURCES OF CONTAMINATION WITHIN THE BENNETTS
BROOK WATERSHED, LITTLETON, MASSACHUSETTS
Agricultural
Commercial
Industrial
Animal burial areas
Animal feedlots
Chemical application
Chemical storage
Irrigation
Manure spreading and pits
Auto repair shops
Car washes
Fuel storage tanks
Gasoline stations
Metal plating operations
Paint shops
Railroad tracks
Road de-icing activities
Scrap and junkyards
Pool supply outlets
Fuel Storage Tanks
Machine and metalworking
shops
Toxic and hazardous spills
Warehousing and distribution
activities
Wells (operating and
abandoned)
Residential
Waste Management
Naturally Occuring
Fuel storage tanks
Household chemical and
hazardous waste products
Fertilizer and pesticide
application
Sanitary septic systems
Sewer lines
Swimming pools (chlorine)
Unregulated agricultural and
farmland disposal (farm
dumps)
Municipal landfills
Recycling reduction facilities
Salt storage facilities
Stormwater drains and
retention basins
Transfer stations
Iron
Manganese
Generally, land-use activities identified during source assessments covered a broad range of
commercial and industrial operations. Those identified throughout the Littleton study area are
depicted in Figure 3-6. Fortunately, while no major hazardous releases have been reported to date,
gasoline stations, metal plating shops, landfills, industrial developments, and residential subdivisions
already exist within the watershed. Collectively, these land-use activities pose potential risks of
nutrient, solvent, and heavy metal contamination to the aquifer. In addition to these anthropogenic
sources, naturally occurring contamination sources are listed. Manganese and iron are detected at
elevated levels in the recharge area of PW-5 and occur in wetland sediment and glacial deposits. High
levels of manganese and iron in PW-5 may warrant water treatment in the future.
To better prioritize sites according to the actual ground-water contamination risks posed, site-
specific information was categorized and reviewed. Information collected during site inspections was
combined with information collected through bylaw-mandated reporting requirements and
incorporated into a computerized source data base. The data base contains information regarding the
category and quantities of hazardous materials identified for each operation and information about
containment practices. The data base also includes information compiled from DEP and EPA offices,
3-11
-------
forge
.Pond
Spectacle ..••"
Pond
I
I
I
SALT
STORAGE
AGRICULTURAL
V/////A INDUSTRIAL & COMMERCIAL
RESIDENTIAL & OPEN SPACE
WASTE MANAGEMENT
TOWN BOUNDARY
WATERSHED BOUNDARY
ROADWAY
»-H RAILROAD TRACK
BROOK
PRODUCTION WELL (PW)
SCALE IN FEET
Figure 3-6. Location of existing land-use activities within the Bennetts Brook watershed,
Littleton, Massachusetts.
3-12
-------
such as hazardous waste generator identification numbers, state audit sites, disposal sites, and
incidence response records. This information can be used to cross-reference information with state and
federal regulatory agencies.
Ground-water contamination is often the legacy of historic land-use activities. In Littleton,
economic growth between the early 1970s and the late 1980s transformed the community from a rural,
agricultural base to a residential and commercial base. While existing clean developments could
conceal potential contamination, it was important to identify past land-use practices, at least in a
general sense. Former agricultural activities were identified throughout the community. Potential
sources associated with agricultural land use include improper pesticide storage and application, rural
dumps, and nitrate loading from seasonal fertilization and manure stockpiles.
For potential contamination sources identified within the WHPA, a general land-use
classification scheme was adopted for prioritizing each site. Because land-use classification is
qualitative and, therefore, imprecise as an assessment technique, shortcomings were anticipated.
A superior ranking system would utilize more accurate, site-specific, chemical inventories and
containment procedures. This combined quantitative and qualitative approach would compensate
for inconsistencies observed from one industry to the next.
In addition to the audit program, ground-water monitoring networks have been installed
around industries that pose the greatest risk of ground-water contamination. Biannual sampling and
analyses for general water quality parameters, heavy metals, and volatile organic compounds (VOCs)
are performed by the Littleton Water Department. Results from these sampling activities verify if "at
risk" industries are conducting operations in accordance with local bylaws, which ban the release of
toxic or hazardous substances in any water resources. These historic water quality records further
support our assessment of regional water quality issues within the Spectacle Pond aquifer.
3-13
-------
RESEARCH MONITORING PROGRAM
DATA BASE LIMITATIONS
For the Spectacle Pond Aquifer and Water Resources Districts, the Littleton Water
Department has compiled a data base that includes point-source water quality information, historical
water-level measurements, aquifer parameters, well construction logs, borehole information,
geophysical surveys, site assessment records, aerial photographs, and technical reports detailing
regional investigations. Although the existing data base is extensive, aquifer testing conducted in 1981
did not provide enough detailed information for accurate development of an effective ground-water
monitoring program.
In 1981, a water resource consultant conducted aquifer tests and concluded that the aquifer
within the Spectacle Pond study area is hydrologically isolated from the surface-water system. This
interpretation is disputable based on deficiencies of the aquifer-test methodology and subsequent
analyses, as follows:
• An insufficient number of observation points were used to quantify induced infiltration
from Spectacle Pond with accuracy.
• Pumped ground water was discharged 50 feet from PW-5 to a swale; therefore,
recirculation of discharge water may have caused incorrect calculations of the area of
influence.
• The aquifer test was conducted for 5 days, and steady-state conditions may not have
been attained.
* The capture zone, ground-water flawpaths, and time of travel (TOT) to PW-5 were
not determined.
In 1981, the town planners extended the protection zone to the upgradient hydrogeologic
recharge boundaries determined as the glacial till and bedrock uplands, thereby somewhat
compensating for these deficiencies. This delineation approach, conservative by design, resulted in
overprotection. Because of escalating land values, this approach is generally unpopular in many
communities. Opposition by land owners and developers and the labor and expense required to
monitor these districts are only a few of the more obvious shortcomings of overprotection.
Considering management issues alone, an optimum protection zone should encompass only those
areas in which a contamination release will realistically pose a threat to the water supply in question.
344
-------
Other deficiencies of data, which impaired the ability to design an optimum monitoring
program for the study area, included:
• Lack of detailed horizontal and vertical hydrogeologic characterization
• Insufficient surface-and ground-water flow information within a one-half-mile radius
of PW-5 ("near-field" scale) and to the boundaries of the aquifer (5-mile radius, "far-
field" scale)
• Lack of near-field water quality data
• Inadequate aquifer testing
RESEARCH OBJECTIVES
To address deficiencies in the data base, a comprehensive investigation of the hydrogeologic
properties within the Bennetts Brook watershed was conducted. The investigation included several
tasks to determine aquifer properties, surface- and ground-water flow system interactions, and ambient
and contaminant water quality. These tasks, coupled with source inventory and prioritization, led to
the development of an optimum monitoring program for the Spectacle Pond aquifer. The monitoring
program was designed to provide a relatively simple and cost-effective means of ensuring protection of
PW-5. It called for monitoring sites to be installed at strategic locations based on flowpath
identification; sampling frequencies to be instituted using TOT criteria; and analytical parameters to
be defined by source assessment activities, including land-use identification, site inspections, and
near-field water quality analyses.
RESEARCH TASKS
The most critical task performed in any hydrogeologic study is the development of a
conceptual model that realistically describes the system of interest. To better understand the
hydrogeologic conceptual model for the Spectacle Pond aquifer, a series of tasks was designed to
qualitatively determine aquifer properties and flow system dynamics. Confirmation and modifications
to the conceptual model were accomplished by a review of hydrogeologic data, installation of
ground-water monitoring devices, aquifer testing, ground-water flow modeling, and comprehensive
water quality analyses.
Over the years, a significant amount of hydrogeologic information was generated within the
Bennetts Brook watershed through independent studies. The historic data review process helped
identify and integrate hydrogeologic characterizations, aquifer testing results, and site assessment
findings detailed in these previous studies/ The review provided a means to collect background
information and to highlight data deficiencies within the study area. Federal and state hydrogeologic
investigations, as well as private site assessments, were sought during this review process.
Ground-water monitoring devices were installed to acquire hydraulic head data, stratigraphic
information, and water quality characterization. Because of the range in costs and utility for various
types of monitoring systems, the selection of each device for this study depended on the location of the
drilling site, local hydrogeology conditions, types of pollutant(s) expected, and overall installation
expense (Scalf et al., 1981). In most studies, cost is a limiting factor and often dictates the selected
3-15
-------
number and design of monitoring devices used. Where appropriate, preexisting and newly installed
ground-water monitoring wells, observation wells, "MicroWells" (Cadwgan and Swallow, 1991), and
small-diameter piezometers were used for this study.
To simulate ground-water flow conditions for the near-field and far-field systems, a phased
approach of aquifer testing and modeling was outlined. The approach was iterative so that computer
simulations could be calibrated to generate an accurate numerical flow model of the Spectacle Pond
aquifer. An initial, 3-day aquifer test, referred to as the "short-term test," was performed in May 1990.
Another aquifer test, lasting 12 days and referred to as the long-term test," was performed in August
1990.
Monitoring stations were strategically located to acquire spatial aquifer parameter. Drawdown
and recovery data were interpreted using four different analytical methods for distinct aquifer
conditions. In addition to providing best estimates for aquifer parameters, separate analyses allowed
for evaluation of ground-water flow systems in different parts of the aquifer.
Steady-state, numerical, ground-water flow models were used to simulate far-field and near-
field aquifer responses under pumping and nonpumping conditions. To simulate normal
hydrogeologic flow conditions, necessary for siting appropriate monitoring stations within the capture
zone of PW-5, average long-term recharge and discharge input parameters were compiled. A two-
dimensional (2-D), horizontal plane model, FLOWPATH (Franz and Guiguer, 1989), was used (1) to
simulate far-field ground-water flow, surface- and ground-water interactions, infiltration events, and
geologic controls; and (2) to determine time-related capture zones. The three-dimensional (3-D) flow
model, MODFLOW (McDonald and Harbaugh, 1988), simulated vertical flow components in
proximity to Spectacle Pond and PW-5. This near-field model generated capture zones, flowpaths,
and travel times associated with PW-5. Near-field and far-field steady-state model results were
evaluated in designing the long-term monitoring network and in determining protection zones within
the WHPA.
In contrast to steady-state modeling, transient-flow modeling permits simulation of
time-related responses in a hydrogeologic system with various pumping and recharge stresses. Current
WHPA delineation criteria for the State of Massachusetts (Massachusetts Department of
Environmental Protection, 1991) require the development of a such a model. Stress conditions of
continual pumping for 180 days at a safe yield with no recharge from precipitation were imposed on
the aquifer's natural, nonpumping, hydraulic head conditions. By design, this drought simulation will
expand the downgradient boundary of the resulting WHPA. Using these criteria, the Spectacle Pond
aquifer system was simulated with a transient-state, 2-D, numerical flow model, FLOWCAD (Franz
and Guiguer, 1991). The final hydraulic head distribution was then used to delineate the capture zone
for these extreme hydrologic conditions.
In addition to the aquifer testing and modeling tasks, sampling and hydrochemical analyses
were performed for the following reasons:
• To characterize near-field water quality
• To corroborate or refute ground-water modeling interpretations
• To establish ambient water quality data within the Bennetts Brook watershed
3-16
-------
DATA ACQUISITION
Hydrogeologic Data Review
Reference site-specific and regional characteristics were examined during development of the
initial hydrogeologic conceptual model for the Bennetts Brook watershed. Standard references, such
as U.S. Geological Survey (USGS) surficial geology and topographic quadrangle maps and the USGS
Hydrological Atlas, were reviewed. Site-specific reports detailing original aquifer test analyses, soil
boring logs, geophysical surveys, and historical ground-water quality data were also reviewed.
Following the literature review, a site visit was conducted by all principal investigators. This
exercise was designed to orient participants involved in the study, through a realistic overview of the
local geomorphology and a review of regional assumptions and interpretations. Reconnaissance site
visits, such as the one conducted by the principal investigators, highlight site conditions that may not
be evident from the literature review and may be significant for conceptualization by both computer
modelers and field technicians.
Ground-Water Monitoring Stations
Ground-water monitoring wells, observation wells, Micro Wells, and piezometers are found
throughout Jhe Bennetts Brook watershed. Each device was installed for a specific purpose, but these
devices were used collectively for regional water-table mapping. Every device was field-tested to
confirm whether it maintained a suitable hydraulic connection with the underlaying aquifer. The
integrity of each well was confirmed with a modified falling-head recovery test. If upon adding a
known quantity of water, the original water-level within the well recovered quickly, the hydraulic
connection was considered acceptable. If the response was delayed, the well was redeveloped and
retested. If it failed this second test, the well was not used for future monitoring activities.
Monitoring Wells-
Monitoring wells were installed where chemical and physical properties of an underlying
aquifer needed to be identified. The drilling method and monitoring well design employed is
dependent on the hydrogeologic environment encountered. Within the Bennetts Brook watershed,
hollow-stem augers or drive-and-wash drilling methods were used successfully. After the soil borings
were advanced, monitoring well casings and well screens were inserted into the boring. These
materials were typically constructed of polyvinyl chloride (PVC) or stainless steel. Once the
monitoring well was emplaced, the original driving casing was pulled, which allowed for representative
ground-water samples to be collected from the aquifer.
Monitoring wells were advanced using a 8.25- x 4.25-inch hollow-stem auger drilling rig. The
final well construction consisted of 2-inch-diameter, schedule 40, flush-joint threaded PVC pipe,
coupled to 5- or 10-foot lengths of 0.010-inch slotted well screen (Figure 3-7A). Screen lengths were
determined by the depth and extent of aquifer isolation required. For example, most cluster wells,
installed at multiple depths within the same general area were equipped with 5-foot screen lengths.
Singular wells were equipped with 10-foot screens to provide a greater depth range for monitoring.
3-17
-------
cc
UJ
UJ
O
N
UJ
CL
m
i^
o
DC
UJ
o
CO
8
z
t^ •*>
K^ I"*™
K£ o
i
LU Oxn
LU LUt
K-m OS
co-2 2r
ZD CC
CO Q
\/
^ i
Js
•o
^
5g
UJ-J
i
Q
Z
d
-T^J
y
i—
UJ
UJ
O
1\ D
^^
1 ol
[7
BBB..-*1.^1-.-
ffilna-v V. •'».••;
oO
•^•z
xco
"^
coo
zo
CVIQ-
-"• •.'. :.\/
•..-o.VvA
• •••".••• o
f?.'.;.-..V
_i
LL.
*£
Q
<
ffl
z:^
<>
U.IH:
_l<
\
»".v:-'.C-»V-:;:::
>•;•;%•:-••:«
<
UJ f—
CO _
LLJ C_5
Z OC
\*
i'B'''\'''°':\^
•"•'"•' *:.v:V^
-
z
of*
wB
1 1
^z
\
\
Illllf
••
Qi^
^~CJ
COCL
-------
All screens were filter-packed with clean Ottowa sand and capped with a 6-inch bentonite clay seal to
minimize effects from vertical leakage. At the surface, each well was protected with a locking case and
cap and sealed with 2-foot-thick cement grout. All wells were developed for several hours until a clear
discharge water was observed.
Split-spoon sampling was performed at 5-foot increments to identify local stratigraphy (U.S.
EPA Region 1, 1989). In areas where the larger, truck-mounted auger rig could not be taken, a
portable drive-and-wash, cable tool drilling rig was employed. With the exception of the smaller-
diameter PVC casing (1.5-inch), the final design of these wells is similar to the monitoring wells
described above.
Observation Wells--
Observation wells were installed to determine local stratigraphy and ground-water flow
conditions, without regard to ground-water quality. These wells were installed using the same
drive-and-wash techniques used to drill the monitoring wells. Observation wells differ from
monitoring wells in that the galvanized-steel casing used to advance each boring is left in place.
Generally, the well screen, placed at the bottom of the casing, is also constructed with galvanized steel.
Because this steel is not replaced with more chemically inert materials such as PVC or stainless steel,
it is not possible to collect truly representative ground-water samples from these wells. However, the
simplified drilling process decreases the amount of time and overall expense required for each
installation. Where ground-water quality is not required, observation wells provide a cost-effective
means of obtaining hydrogeologic properties.
MicroWells—
The Micro Well installation process employs a high-frequency vibratory hammer mounted on a
VibraDrill all-terrain drilling machine, which can advance 0.5-inch, steam-cleaned, steel casing to
depths of over 100 feet (vibracore process) (Cadwgan and Swallow, 1991). The bottom 10-foot
section of the steel casing is machined with a double row of longitudinal slots 0.015-inch wide (Figure
3-7B). After the casing is driven to the desired depth, the well is naturally sealed by the surrounding
aquifer and developed using a peristaltic pump attached to 0.5-inch-diameter polyethylene tubing. The
well is purged until clear discharge water is observed. Although Micro Wells can be used to obtain
hydraulic head data and water quality samples, they are used hi this study for water-level monitoring,
exclusively. While the Micro Well drilling technique will reduce the time and costs associated with
more traditional drilling methods, the vibracore process is effective only in sand and gravel
environments. At the time of this study, the vibrocore process did not incorporate stratigraphic
sampling; however, now it does.
Piezometers--
Piezometers are installed for hydraulic head measurements exclusively. Each is composed of
0.75-inch, schedule 80, mild-steel casing machined to 7- and 14-foot lengths (Figure 3-7C). Manually
driven piezometers seldom reach depths greater than 30 feet but provide a useful method for collecting
water-level measurements hi shallow sand and gravel aquifers without significant equipment or cost.
These devices can be advanced only in soft materials that are relatively free of cobbles and boulders
(Roy et aL, 1984). For this study, each point is hand-driven through the upper soil horizons and set
into the underlying sands and gravels. The drive points are perforated with multiple small-diameter
holes to provide a hydraulic connection with the aquifer. Fine-grained sediment and organic materials
entrapped during the installation process are removed by air-jetting techniques.
3-19
-------
Aquifer Testing
Hydrogeologic systems can be defined using fundamental aquifer properties such as saturated
thickness (b), porosity (n), hydraulic conductivity (K), transmissivity (T), storativity (S), and specific
yield (Sy). Saturated thickness is measured as the aquifer thickness saturated with ground water. The
volume of water stored within the aquifer is related to porosity, defined as the ratio of aquifer void
space to total aquifer volume. Ground-water flow through this void space is controlled by grain size
distribution and sorting characteristics of the matrix. Hydraulic conductivity is a coefficient of
proportionality, which describes the flow velocity and direction of ground-water moving through an
aquifer. Often, hydraulic conductivity is separated into its horizontal and vertical flow components,
Kr and Kz, respectively. The amount of water transmitted through a unit width of aquifer under a
unit gradient, transmissivity, is calculated by multiplying saturated thickness by hydraulic conductivity.
Finally, for water supply studies, the volume of water stored or expelled per unit area of the aquifer
per unit change in pressure, storativity, is particularly important. Generally, in an unconfined system,
storativity is equivalent to specific yield. Collectively, these properties, which are used to characterize
an aquifer system, can be used to predict hydrologic responses to natural (recharge) or induced
(pumping) stresses. The qualitative measurement for each property is determined using standard
aquifer testing methods.
The short-term aquifer test was conducted at minimum cost by using existing monitoring sites
in proximity to PW-5 (Figure 3-8A). To determine ground-water responses in areas where no
monitoring stations exist, inexpensive piezometers were installed. Discharge water during testing was
diverted from the near-field aquifer area through the existing water supply distribution network. The
information obtained during the short-term aquifer test, although somewhat incomplete, proved useful
in the siting of additional wells for the long-term aquifer test. The long-term aquifer test was
scheduled to obtain accurate values of hydrogeologic properties within the near-field capture zone of
PW-5 (Figure 3-8A).
Drawdown and recovery data obtained during the short- and long-term aquifer tests were
analyzed using the following graphical methods:
• Neuman (1974) method for horizontal and vertical movement in an unconfined
aquifer, modified for a partially penetrating well
• Theis (1935) nonequilibrium formula for a non-leaky, confined aquifer, modified by
Kruseman and de Ridder (1990) for unconfined aquifers
• Cooper-Jacob (1946) method for graphical solution of the modified non-leaky
confined formula
• Hantush (1964) modified method for a leaky, semi-confined aquifer, considering
partial penetration of the well and storage in the aquitards
3-20
-------
3-21
-------
Short-term Aquifer Test-
In addition to PW-5, a network of water-level monitoring stations was employed in the
short-term test; these included four observation wells (OW), nine piezometers (P) and one monitoring
well (MW). These observation stations were located at distances ranging from 2 to 1,393 feet from
the pumping well, PW-5 (Figure 3-8A). Information regarding the distance from each monitoring
station to PW-5, station depths, and screen lengths is given in Table 3-2. Prior to the short-term test,
PW-5 was shut down for a period of 96 hours to determine the static, prepumping water-level
condition. At 12:00 P.M. on May 22,1990, PW-5 was turned on, and a constant pumping rate of 650
gpm was established for the duration of the test. Water-level responses were monitored in the
near-field network. After 73 hours, at 1:00 P.M. on May 25,1990, PW-5 was shut off and water-level
recovery measurements were collected for the next 72 hours.
TABLE 3-2. DISTANCES FROM PW-5 TO MONITORING STATIONS, STATION DEPTHS,
AND SCREEN LENGTHS FOR THE SHORT-TERM AQUIFER TEST, BENNETTS
BROOK WATERSHED, LITTLETON, MASSACHUSETTS
Monitoring
Station
PW-5
OW-2
OW-100
OW-150
OW-400
P-A
P-Cd
P-Cs
P-Gd
P-Gs
P-H
P-I
P-J
P-L
MW-T
Distance from PW-5
(ft)
0
2
100
150
400
20
50
50
441
441
510
993
1,393
655
800
Station Depth
(ft)
50
40
40
40
40
40
40
14
20
7
7
7
7
7
35
Screen Length*
(ft)
15
10
10
10
10
10
10
3
5
3
3
3
3
3
10
Abbreviations: PW-5 = Production Well Number 5; OW = observation well; P = piezometer;
d — deep wellj s = shallow depth well; MW = monitoring well.
* All screens are located at the bottom of each monitoring station.
3-22
-------
Hydraulic head data at the 15 observation stations were measured with hand-held
electronic water-level indicators to an accuracy of ± 0.01 foot. To address effects from regional,
ambient water-level fluctuations during the aquifer test, two observation wells (OW-la and OW-lb)
located far outside the zone of influence of PW-5 were monitored (Figure 3-8B). Surface-water
elevations were collected at a staff gauge, SI, located near the eastern shore of Spectacle Pond (Figure
3-8A). Other physical parameters recorded during the aquifer test included barometric pressure,
rainfall, and temperature.
During analyses of the short-term test results, it was noted that several newly-installed
piezometers did not respond reliably, especially those observation stations located farthest from PW-5
(P-N, P-J, and P-I). It was also decided that the number of water-level stations across the pond was
insufficient. Specifically, water-level data on the other side of the pond were required to evaluate
connectivity between the pond and the aquifer and to determine the recharge contribution from the
pond to PW-5.
Long-Term Aquifer Test—
From the values obtained during the short-term test, a sensitivity analysis was performed using
FLOWPATH varying transmissivity, leakage through the bottom of the pond, specific yield, and
storage. The results of the sensitivity analyses demonstrated that the aquifer would need to be stressed
for a period of at least 10 days for a water-level response to be detected in wells located on the
northern shore of the pond. Based on these findings, the long-term aquifer test was conducted over a
period of 287 hours (approximately 12 days).
The location of observation stations used during the long-term aquifer test are shown in
Figure 3-9. Information regarding distances to monitoring stations from PW-5, station depths, and
screen lengths are contained in Table 3-3. Four observation wells, OW-100, -150, -400, and -800
(formerly identified as MW-T), and two piezometers, P-20 and -50, from the short-term test were
included for use in the long-term test. Observation Well-1200 and seven newly-installed monitoring
stations [MW-25 (shallow and deep), MW-100 (shallow and deep), MW-150, MW-375 (shallow and
, deep), MW-425, MW-660 (shallow and deep), and MW-800] were also integrated into the long-term
test network.
Well PW-5 was shut down for 104 hours prior to the start of the long-term test to allow for
water-level recovery. Beginning on August 16, 1990, PW-5 was pumped at a constant rate of 650 gpm.
After 72 hours, PW-5 was shut down and water-level recoveries were monitored for an additional 48
hours, ending on August 30, 1990.
Water-level elevations for most of the 20 monitoring stations were recorded with electronic
hand-held indicators. Water-level data recorded at three cluster wells, MW-25, -100, and -375, and two
single wells, MW-150 and -425, were measured using an automatic datalogger system. An automatic
datalogger provides a means of recording rapid changes in hydraulic heads and temperatures
anticipated at those stations suspected of exhibiting maximal response to pumping. Dramatic
temperature changes provide a positive indication of surface-water and ground-water interconnection.
3-23
-------
3-24
-------
3-25
-------
TABLE 3-3. DISTANCES FROM PW-5 TO MONITORING STATIONS, STATION DEPTHS,
AND SCREEN LENGTHS FOR THE LONG-TERM AQUIFER TEST, BENNETTS
BROOK WATERSHED, LITTLETON, MASSACHUSETTS
Monitoring
Point
PW-5
P-20
MW-25s
MW-25d
P-50
MW-lOOs
MW-lOOm
OW-lOOd
OW-150
MW-150
MW-375s
MW-375m
MW-375d
OW-400
MW-425
MW-660s
MW-660d
MW-800
OW-800 (MW-T)
OW-1200
Distance from PW-5
(ft)
0
20
25
25
50
100
100
,100
150
150
375
375
375
400
425
660
660
800
800
1,200
Station Depth
(ft)
48.0
40.0
18.0
34.0
40.0
18.0
35.0
40.0
40.0
70.0
10.0
22.0
40.0
40.0
73.0
37.0
63.5
65.0
35.0
34.0
Screen Length*
(ft)
15
10
5
5
10
5
5
10
10
10
5
5
5
10
10
10
10
10
10
10
Abbreviations: PW-5 = Production Well Number 5; P = piezometer; MW = monitoring well;
s = shallow depth well; d = deep well; m = moderate depth well; OW = observation well.
* All screens are located at the bottom of each monitoring station.
3-26
-------
As in the short-term test, ambient ground-water levels were recorded, at OW-la and -Ib.
Surface-water elevations from SI, located within Spectacle Pond, were also monitored. In addition,
regional barometric pressure, rainfall, and temperature, starting 5 days prior to and during the test,
were recorded.
The water-levels recorded during the long-term aquifer test were analyzed graphically with
type curves and numerically with least-square fitting for different aquifer settings. Hydraulic responses
were compared to the Neuman analysis for unconfined aquifers with delayed yield, the Theis and
Cooper-Jacob methods corrected for unconfined aquifers, and Hantush's solution for leaky aquifers
with storage in the aquitard. Matching solutions provided estimates of hydraulic conductivity and
storativity and indicated prevailing aquifer conditions at each observation station.
Peat-Probing Exercise
The areal extent and thickness of peat deposits that underlie the eastern lobe of Spectacle
Pond basin may significantly affect induced flow from the pond to PW-5. To determine the
configuration of the peat layer, field measurements were collected on February 6, 1990, when the ice
cover on the pond supported field technicians and equipment. Peat layer thicknesses were determined
by coring through the pond ice using a hand-held ice auger, then driving 10-foot-lengths of small-
diameter (0.75-inch), threaded rod through the peat until refusal was encountered. The interface
between peat and the sand and gravel was easily confirmed by the grinding noise made when the metal
probe struck the sand and gravel layer. Peat depths and thicknesses were determined at 50-foot
intervals along several transects at the eastern lobe of the pond (Figure 3-10). Peat layer thickness
ranged from less than 1 foot to 35 feet.
Ground-Water Sampling and Analysis
Ground-water monitoring activities conducted within the Spectacle Pond aquifer and the
capture zone of PW-5 were designed to characterize water quality for long-term monitoring planning.
The scope of these monitoring activities are broadly defined by Todd et al. (1976) and include
(1) ambient trend monitoring (temporal and spatial parameter trends within the watershed), and
(2) source assessment monitoring (measurement of key indicators of contamination to determine the
impact of a potential or known source). Sample collection is also discussed in this section.
The water quality parameters grouped in Table 3-4 were differentiated according to specific
monitoring objectives: general criteria, site-specific criteria, and combined general and site-specific
criteria. General water quality criteria include those ground-water constituents that are typically
identified in undeveloped, pristine watersheds where anthropogenic stresses are minimal. Site-specific
criteria include chemical constituents that originate from man-made stresses including point source
(isolated) or nonpoint source (regional) contamination. Parameters attributed to both natural and
artificial sources are listed in the combined general and site-specific category.
3-27
-------
3-28
-------
50
CHUSETI
2
«i
4_
o ^
E "§>
n «
VT •!-»
o *•*
fsl
gj *-*
O -a
II
t-t ^
EH -2
II
FA FOR T
ations PW-
II
D tJ
O CO
ct
III
i
i
I
o
h
1
*
0)
O
1
f
CO
1
J
O
1
O
U
•§
6
e
£
s
*
0
ffi
o
!
CO
p
3
„
o"
"2
3
0.— «
sj?
= !
p" *•
S jjj
^f Ov OO C*«
00 vo C~ vo
1-1 1-t T-l 1-1
S 2 3 S
dodo
S32 3 2
d d d d
co CM CM «n
t~ vo t- o\
VO vo vo VO
CM O t- •*
*-< CM CO t—
CM CM CM 1-1
O O O CM
S3S33"
oo vo f co
CM CM CM CM
VJ VJ CO O
CO CO CO CO
oq 'ar i-; vq
gsss
^t QO OO t^
O\ f^ Tl- ^H
01 n so VD
i
oq tn tn QO
^^ CO ^^ ^5
r* «s cs cs
OO ^i ^H 00
o o o o
o o o o
QO vo vo ^5
GO cs cs en
co en cs ^D V\
»-5 oo l> «n
TT «0 i-« CS
p in p p
ov cs t^ oq
vo in cs o
S
CO CO CO CO
t-; "* CO 00
vo in ^j- co
rr p oq oq
Ov VO CM i-l
in vo in vo
p p p co
CO CM QO "^
in m v8 in
•o
aaaa
ssss
r-5 i-5 d d
IIS5
SSSit
t^ t^ VO t^
VO VO VD \O
^3 C3 t*1" CO
T-* oo o tn
CS ** CS T^
cs cs cs oq
CS ^3 CS CT\
35S3
"3 o vo m
JS} 00 CO CO
oq p p vj
^5 ^f VO "^
^r ^f 01 i— *
p VJ p CM
VJ 00 O\ l^;
t^ CM Ov OV
S in
BBBB
TH
co oo cs m
S S » 3
f~ m CM 1-1
dodo
dodo
CO CO CM S
•^- i-< CO CM
t- t- t- t-
CS ^5 t** CO
CM 0 0 p
S VI S O
in in CM r-<
CM CM CM CM
»n 1-1 r— co
VO Tf CO CO
p vq p vq
° vo vS'vo
CM. 0 vj oq
O» CM t— 00
t- r~ vo in
oq i-; CM vj
co oo in co
CO CM CM CM
S CM 00 CM
f^. \o r*^
r-t
i
O T-H O 0\
to vo oo
000
e§s>
cR c^ CM S
in o in t~
vo t~ vo vo
c^ o oo co
1-1 in o ^r
CM CM CM TH
p p oq oq
vo o ^f in
CM m CM oo
co co in vo
CM co i-5 i-5
oo i-« CM •*
CO Tt CO CM
t- 0 00 0
1-1 O O **•
CM ^ m CM
CM t~; i-; co
'S' t^ CO CM
i— J O\ CO (O
CO ^3 OS 00
CKfci:
MW-375S
II. 8
co vo oo
o o o
i— r- p
So ! 3
co co i »n
P ^ i *";
OO t"- * t**
O Tf ( O
0 Tl- 0
s vi • s
as ' s
co vq . ox
*3" Tf 'CO
t-; oq oq
Sco ; TI-
vo ' r-
VJ 0 CM
in T-I ' o\
t< 00 ' VO
p »H t~;
t** vi ' t~-
CO CO ' CM
o\ t- i >n
m r- ' t-
s
3-29
-------
g
a
3
S
*™s
£
•a
o
1
1
CM
1
S
f
1
i
1
2>
»
«j
"
C
i
CM
^
«s
ff
JO
0-
5
u
S
A
9,
g
i
i"
o"
s3
n
Ioaiupic
Location
i i
* »
.
* *
* *
* »
» *
* »
o o
d o
O O lq O
TT C> oo ^r
Q Q Q Q
Z Z Z Z
gggg
§Q Q Q
Z Z Z
OH
O 0 °. 0
M C> § CJ
Q Q Q Q
»7 trr C5 i^
X-i ^H XH X-i
D Q Q Q
iiaa
MW-25s
ssss
gggg
^-« ^M <*-« XH
gggg
^ Q o f*
PQ T*1
MW-25d
°- ° 13 °-
o Q Q Q
1-1 Z Z Z
BSBB
cn =3 P
>n o m t~
O Q Q Q
Z Z Z Z
BiBB
oo Q oo Q
** Z « Z
1
P|g °-
Q Q <= 0
§QQ9
Z Z Z
VO Tf Q O
T-H 1-H ^ CS
MW-375s
P £ ! =?
BB-B
Q Q , Q
Z Z ! Z
*H 2 'IS
MW-375d
3-30
-------
W
CO
D
1
<
C/5
•
^^ O
O 03
W w)
C o
3 p
«? 1
5^ Bl
EH pt
b4 •§
p^ S
O 2
H ^
o (2
11
HH ™
5
Z OT
s -a
s 1
A FORTH
MW-425 t
Si
fcl
^ •*
S-s
w *2
1
a
i
^-K
*p
S
|
VvS
e
R
£
$
ca
6
1
£!•
3
n
b
1
-^
•a
md Site-Sp
i
.§•
i
o
1
o
•
?!
S
I'
^
e.
S
a
S
£
0
PS
CL,
1
<»
P
5T
U!
f
•2
«
ffi
0
e'er
S5-
S a
ft
-1 a.
|1
% J
VO 00 00 0)
^ i/i Q 0!
c!i ?4 1*1 cs
M CS 0 •*
Tt Tl- 1-1 »-l
<3 C3 tH 0
Q «S M O\
® •* IS O
<-i O CJ O
9 55S
m oo m "*•
\o \d vo vd
o \A vo ^h
c^ c5 ^5 **)
«s en en t^ TH p
CO CO CO CO
oo a\ m oo
i/i iri •* co'
co oo o\ o w-»
M ^ « ^
MW-425
n- m cs o\
CM co O n »n ^t >n
O O O O
0030
OO •fl" CT\ O
\d vo O >n
T-H CO OQ P-
co co oi vo
co t- •* in
t~ vo vo vo
o o r~ i-
TT CS <7\ in
CM CS TH T-l
O CM M n-
5 S 38
\O n en
00 ^ CO Tf
l-H l-H TH l-t
cs Tr o vo
?2g §8S
p o\ in in
TH *H CS VO
OO t- i-c OO
t ^ -^ «^
S RiSS
!n K Sin
n ^ ^^
|
vo co p oo
S S 3 2
8 » S S
0 0 0 C3
8 co f- 5
•* cs o
CO O O O
S R 38
•^ CS CS vo
VO vo vo vo
oo o in >n
CS TH O\ rH
CS CS rH r-c
oo p p p
\O VO t*> ^N
«n CN r-l \O
m t~ o CTV
CO CS CN TH
vo
-------
•ff
•§>
1
o
1
u-l
I
i
U3
V) V* O VI
O O V) p
poop
in
3-32
s|
I
,-s
s
s
5
g
is:
Q Q
Q Q
O Q
Z Z
•a o
If
I
Q Q Q
"1
Q Q
Z Z
t-; Q
o» E
ffi ^
fo,
w.
•O II
3
CO
1
JJ §
II B
-------
Ambient Trend Monitoring— .
Ambient-trend monitoring parameters are typically influenced by the native hydrogeochemical
environment. Ground water derived from unconfined aquifers in Massachusetts is typically soft to
moderately hard, slightly acidic, and low in total dissolved solids (TDS) (Delaney and Maevsky, 1980).
Most of the higher-yielding aquifers are found within the sand and gravel, glacial meltwater deposits
where predominate constituents include sodium, calcium, magnesium, potassium, sulfate, chloride, and
bicarbonate.
To evaluate spatial water quality variations within the near-field pumping area, samples were
collected on August 15,1990, throughout the monitoring well network shown in Figure 3-9 and
detailed in Table 3-4. These samples were collected under static water-level conditions created by
shutting down PW-5 48 hours prior to the sampling event. Temporal variations in water quality were
evaluated by comparing successive samples collected from the same monitoring station over the
duration of the long-term aquifer test. These sampling events were conducted on August 17, August
22, and August 27 and represent days 1, 6, and 11 of the long-term test, respectively.
Water quality data obtained during these sampling events are shown in Table 3-4. Several
graphical techniques were used to identify spatial and temporal water quality variations. Particular
attention was focused on hydrochemical changes within the near-field capture zone. In general, these
trends are best illustrated by comparing the samples collected prior to pumping versus 11 days after
pumping.
Source Assessment Monitoring—
Source assessment monitoring was conducted to identify ground-water contamination threats
within the Bennetts Brook watershed. Review of source assessment records indicate that the most
probable ground-water contamination sources are posed by indiscriminate dumping activities within
industrial and commercial developments (U.S. EPA, 1990a). Disposal of toxic substances used within
residential developments and chemical applications used for agricultural operations (historical and
existing), illegal dumping, leaking USTs, and road salting operations were also considered. Naturally
occurring constituents such as iron and manganese, were also evaluated. At elevated concentrations,
iron and manganese will create significant aesthetic problems in water supply systems (Silvey and
Johnston, 1977).
Monitoring of all the potential contaminants associated with the identified land-uses listed in
Table 3-1 is beyond the scope and budget of this investigation. To maximize the amount of
information obtained during the sampling program, while minimizing the analytical costs, target
chemicals were identified. The "site-specific" parameters in Table 3-4 represent the more ubiquitous
compounds identified in past ground-water contamination studies. In addition, monitoring wells,
which were sampled for VOCs (EPA method 524.2), heavy metals (the Safe Drinking Water Act),
pesticides (EPA method 608), and herbicides (EPA method 8150), were carefully selected based on
upgradient source assessment activities. Although semi-volatile constituents are often included in
ground-water contamination investigations, they were not included in this study as initial screening
parameters because of their high analytical cost. Studies have demonstrated that, in the absence of ^
VOCs, these compounds are seldom detected in ground-water (Rosenfeld, 1990). Analysis for VOCs
is a screening technique; if detected, their presence will justify the need for more complete compound
analysis.
3-33
-------
Source-assessment monitoring stations were chosen based on the conceptual understanding of
ground-water flow and with consideration of existing or historical upgradient threats. With these
factors in mind, PW-5, MW-425, MW-660 (deep and shallow wells), and MW-800 were included for
source assessment monitoring. In general, the Water Department attempts to site source assessment
monitoring wells in groups of 3 to 4. At a minimum, one well monitors the upgradient source area,
while 2 to 3 wells monitor the downgradient source area.
Sample Collection-
Representative water samples were obtained by purging each well a minimum of three times
the borehole volume of water, or until field-measured temperatures and conductivities had reached
stabilization. To eliminate possible errors introduced by cross contamination, each well was evacuated
and sampled by a dedicated, inertial pump composed of small-diameter [0.625 inch OD] polyethylene
flexible-tubing coupled to a Teflon foot valve. The effectiveness of these sampling devices for
development, purging, and sampling is documented by Rannie and Nadon (1988).
To minimize quantitative errors attributed to sample collection and analysis, a water quality
assurance (QA) and quality control (QC) plan was developed (Huibregtse and Moser, 1976). Storage,
handling, and analytical procedures were conducted in accordance with standard EPA methods. All
analyses were performed at state-certified analytical laboratories, which adhere to standard analytical
protocol including chain-of-custody documentation, percent recovery and surrogate spike analysis, and
use of trip and laboratory blanks. As an additional QA/QC check, only analytical results with a
cation-anion balance within 20% error or less were considered acceptable for interpretation.
DATA INTERPRETATION
Conceptual Model
The Bennetts Brook watershed is composed of approximately 3.1 square miles of glacial drift,
1.1 square miles of wetlands, and 3.2 square miles of glacial till and bedrock with an estimated annual
ground-water recharge rate of 3.1 million gallons per day (Mgd) (Metcalf and Eddy, Inc., 1981). The
watershed is bound to the west by a ground-water divide and to the north, south, and east by glacial
till and bedrock prominences. Permeable glacial outwash deposits appear to pinch out along the till
and bedrock uplands. Regional topographic relief varies from 190 feet above mean sea level (amsl) at
the pond to 350 feet amsl along the drumlins to the north.
Three distinct aquifers border the study area (Figure 3-11). Ground water within the Spectacle
Pond aquifer originates at the headwaters of the Stony Brook subbasin and feeds into the Merrimack
River basin. The Cow Brook subbasin envelopes the aquifer due north and flows northeast toward
the Merrimack River. These two aquifers are separated by a surface-water divide formed by bedrock
prominences. The third aquifer located west of Bennetts Brook drains into the Nashua River basin.
A ground-water divide, located adjacent to the southeastern shore of Sandy Pond, separates the
Nashua River basin from the Stony Brook subbasin.
3-34
-------
CO
3-35
-------
Regional geomorphology and aquifer-test interpretations indicate that these aquifers are
predominantly unconfined. The regional ground-water table is shallow and varies in depth between
0 and 25 feet In the immediate vicinity of PW-5, unconfined conditions exist to the east within the
sands and gravels. Beneath the wetlands that extend to the west and under Spectacle Pond, the aquifer
is semi-confined.
In the areas of Spectacle Pond and Forge Pond, the glacial outwash deposits have high
transmissivity values of greater than 4,000 square feet per day (ft2/d) and high potential yields of
greater than 300 gpm. Zones of lower transmissivity (less than 1,350 fWd) are found along the
perimeter of buried valleys where the outwash materials are poorly sorted and the aquifer thickness
pinches out into till and bedrock uplands (USGS, 1985).
Spectacle Pond appears to be dimictic (subject to spring and fall lake turnover) in the deeper
basins but polymictic (continuous mixing) throughout the rest of the pond, with gradually sloping
banks extending into the two basins. Landforms adjacent to the pond display characteristic
ice-contact relief. The present shoreline may reflect impacts from a man-made weir near the outflow
channel by Route 119. Seasonal vegetative growth and decay cycles contribute to the formation of
peat deposits within each basin. Surface-water flows northeast and is restricted at the outflow channel
by the weir. The direct correlation between drainage area size and mean annual discharge for
watersheds within north-central Massachusetts show that the mean annual discharge from this
watershed is approximately 12.09 cubic feet per second (cfe).
Ground-water and surface-water flow within the aquifer is influenced by constant municipal
well withdrawal along the northwest and southeast shores of the pond. These production wells
exaggerate the hydraulic connection between the surface- and ground-water systems. The extent of the
capture zones is influenced by the rate of induced infiltration, which in turn is influenced by the areal
extent, thickness, and type of peat deposits present.
Extensive organic deposits were identified in the eastern lobe of Spectacle Pond during peat-
probing activities. Measured thicknesses vary from less than 1 foot to greater than 34 feet (Figures
3-12A and 3-12B). The greatest peat thicknesses were encountered near the existing outflow where
the outflow stream channel appears to be sink for organic debris carried through the pond
(stratigraphic sections C-C and D-D'). The peat encountered during field probing was loosely
packed and fibrous, with variable thickness across the basin. A large, irregularly-contoured
prominence, possibly a submerged ice-contact feature, was identified within the eastern basin, which
may allow a hydraulic connection between the pond and the aquifer (stratigraphic sections B-B' and
C-C).
Aquifer Testing and Analyses
Short-Term Aquifer Test-
The drawdown and recovery data for each monitoring station used during the short-term
aquifer test were analyzed using four different methods to simulate distinct aquifer conditions. The
computer program AQTESOLV (Duffield and Rumbaugh, 1989) was used to perform these analyses.
AQTESOLV permits aquifer-test analysis using both type-curve and automatic-fitting procedures. A
typical printout from AQTESOLV, which depicts the fit of the recovery data for piezometer P-20
using the Neuman analysis, is shown in Figure 3-13.
3-36
-------
SPECTACLE
'POND
PEAT
BEDROCK
SAND AND
GRAVEL
Figure 3-12A A location map of stratigraphic sections across Spectacle Pond, Bennetts Brook
watershed, Littleton, Massachusetts.
Ancillary water-level effects associated with changes hi atmospheric pressure and recharge
from rainfall were neglected in the data interpretation. Walton (1987) noted that water-level changes
induced by fluctuations in barometric pressure are commonly negligible for unconfined aquifers. A
very small variation in atmospheric pressure was measured during the aquifer test (0.4 foot of water);
therefore, data were not corrected for this effect. Data were not corrected for precipitation because
neither the water levels in the piezometers outside the area-of-influence of the pumping well, nor the
water levels in Spectacle Pond showed a clear correlation to precipitation during the test. The
municipal wells in the town of Ayer, located on the western shore of Spectacle Pond (Figure 3-2),
were run at a constant rate throughout the pumping and recovery portions of the test. Therefore, no
transient effects were caused by the operation of these wells.
3-37
-------
CO
^ '• • . I'S'
##%&
* » * * * r v /v X
I
/x-'Vy''
ffi
(U
5 i"
§ £
(|SU1B U)
NOI1VA313
O
O
I
I
-'•/..">/N
(ISOIB »)
NOI1VA313
o in
S t
(ISLUE «)
NOI1VA313
LU
3-38
-------
I
Q
I
cr
Q
1,000
100
10
I I I|HH[ - 1 llllllll| - 1 1 llllll|
TRANSMISSIVITY, T = 8.13 ft2/min
TTT
I I I Illlll
STORATIVITY, S = 0.020
SPECIFIC YIELD, Sy = 0.066
NEUMAN ANALYSIS-FITTING PARAMETER, #= 0.03
-TYPE CURVE
i i mini i i i mill i i i iiml I i I Illlll 1 I I HIM
0.1 1 10 100 1,000 10,000
TIME (min)
Figure 3-13. Neuman type-curve analysis of piezometer P-20 recovery data, short-term test,
Bennetts Brook watershed, Littleton, Massachusetts.
The analytical results of the short-term test are shown in Table 3-5. The method that
provided the best fit to the observed data is marked with an asterisk. Generally, the best fit for the
drawdown and recovery curves for observation stations were obtained using the Neuman type curves
for unconfined aquifers. The response for piezometer P-L, located on a near-shore island in Spectacle
Pond (Figure 3-8A), resembled the Hantush type curve for leaky aquifers. This fit may reflect the
semi-confining characteristics of the peat layer at the bottom of the pond.
Upon completion of the short-term aquifer test analyses, horizontal hydraulic conductivity
values were found to range from 127 to 2,220 ft/d, with an average value of 560 ft/d. The
representative horizontal conductivity value was estimated to be approximately 50 times greater than
the vertical conductivity value.
The pumping time was too short for reliable estimates to be obtained for delayed yield at
points located farther than 600 feet from PW-5. Therefore, the results from OW-800 (Figure 3-9)
across the pond may be masked by this effect Data from piezometers P-H, P-I, P-J, and P-N were
3-39
-------
TABLE 3-5. AQUIFER PARAMETER RESULTS FOR THE SHORT-TERM AQUIFER
TEST AT PW-5, LITTLETON, MASSACHUSETTS
Aquifer Parameters
Monitoring R
Station (ft)
PW-5 0
OW-2 2
P-20 20
P-50 50
P-Cs 50
OW-100 100
OW-150 150
P-150 150
Method
Pt-Th
Pt-C
Pt-N
Rt-N*
Pt-Th
Rt-Th
Pt-C
Pt-N
Rt-N*
Pt-Th
Rt-Th
Pt-C
Rt-C
Pt-N*
Rt-N
Pt-Th
Rt-Th
Pt-C
Rt-C
Pt-Th
Pt-C*
Pt-N*
Rt-N
Pt-Th
Rt-Th
Rt-C
Pt-N
Rt-N*
Rt-Th
Rt-C
Pt-Th
Pt-C*
T
(ftVmin)
24
24
5
32
17
26
25
13
8
17
23
19
22
21
23
21
24
21
24
18
18
21
21
27
33
29
18
27
39
35
21
21
Kr
(ft/d)
576
576
127
775
410
640
604
316
194
412
573
477
532 '
523
564
513
576
508
588
451
451
511
518
664
796
715
444
648
938
840
760
508
S Sy
orS
0.01400
0.01100 0.0810
0.02000 0.0660
0.0150
0.0028
0.0052
0.0032
0.00400 0.0300
0.01600 0.0260
0.0220
0.0250
0.0200
0.0230
0.0550
0.0540
0.00310 0.0160
0.00150 0.0110
0.0030
0.0017
0.0029
0.00300 0.0350
0.00630 0.0150
0.0037
0.0054
0.0041
0.0041
P
0.001
0.030
0.004
0.030
0.010
0.010
0.180
0.060
0.100
0.200
Kr/Kz
1.00
0.03
28.00
4.00
69.00
69.00
21.00
63.00
28.00
14.00
(continued)
3-40
-------
TABLE 3-5. AQUIFER PARAMETER RESULTS FOR THE SHORT-TERM AQUIFER
TEST AT PW-5, LITTLETON, MASSACHUSETTS (continued)
Aquifer Parameters
Monitoring
Station
OW-400
P-Gd
P-Gs
R
(ft)
400
441
441
Method
Pt-N*
Pt-Th
Rt-Th
Pt-C
Pt-N*
Pt-Th
Pt-C
Pt-N"
Pt-Th
T
(ft^min)
45
48
92
41
22
26
28
27
26
Kr S
(ft/d)
1,101 0.01300
1,171
2,220
991
544 0.00041
631
672
664 0.00170
624
sy
orS
0.0670
0.0620
0.0660
0.0810
0.0270
0.0210
0.0180
0.0400
0.0350
P
1.500
0.490
0.400
Kr/Kz
30.00
109.00
135.00
P-H
510
P-L
P-I
P-J
MW-T
All Wells
655
993
1,393
800
N/A
Pt-Th
Pt-C
Rt-H*
Pt-N*
Pt-Th
Pt-C
Pt-Th
Pt-C*
Pt-N*
Pt-Th
Pt-C
TDD
35
36
11
33
42
42
89
93
59
59
62
31
842
862
263
813
1,023
1,010
2,150
2,236
1,435 0.02700
1,418
1,495
746
0.1200
0.1000
0.0086
0.0140
0.0083
0.0077
0.1000
0.0780
0.0270
0.0280
0.0190
0.0340
r/p= 1.800
6.000 45.00
5.000 83.00
Abbreviations: PW-5 = Production Well Number 5; R = distance from PW-5 to monitoring station;
T = transmissivity; Kr = horizontal hydraulic conductivity; S = storativity; Sy = specific yield;
P = Neuman analysis-fitting parameter [(Kz/KrXr/b)2]; Kr = horizontal hydraulic conductivity;
Kz = vertical hydraulic conductivity; Kr/Kz = anisotropy ratio; Pt = pumping test; Th = Theis
corrected drawdown analysis; C = Cooper-Jacob corrected drawdown analysis; OW = observation
well; N = Neuman analysis; Rt = recovery test; P = piezometer; s = shallow depth well; d = deep
well; H = Hantush analysis with aquitard storage; MW = monitoring well; N/A = not applicable;
TDD = Theis drawdown versus distance analysis for all wells.
* = Best-fit method.
3-41
-------
not reliable enough to be analyzed. It is possible that these piezometers were clogged by native
materials. In conclusion, it was determined that additional monitoring points across the pond and a
longer pumping period would be necessary in the design of the long-term aquifer test.
Long-Term Aquifer Test-
Data from the long-term aquifer test were interpreted in the same way as the short-term test
data. For reasons discussed earlier, barometric pressure and pumping effects from the town of Ayer's
water supply wells were not corrected in the analyses. However, rainfall data showed a clear
correlation to pond water levels. Water-level data from observation wells near the pond were
corrected in a linear fashion for temporal variations in the pond.
Interpretations of the long-term test data are summarized in Table 3-6. There were no water-
level responses in OW-1200 located on the far side of the pond (Figure 3-9), indicating that
contribution from the northwestern portion of the aquifer is drawn entirely from surface water
through the bottom of the pond. Investigations of peat thickness in the area between PW-5 and
OW-1200 showed the existence of a sandy ridge along the bottom of the pond, halfway between these
two stations (stratigraphic sections B-B' and C-C, Figure 3-12B). This ridge provides direct
hydraulic connection between the surface water and ground water. On the other hand, OW-800,
located on the opposite side of the pond but farther .east than OW-1200, showed measurable
drawdown. Hence, water entering PW-5 from the north is induced flow from the pond and
ground-water flow beyond the pond.
Water-level monitoring at cluster wells MW-375s, -375m, and -375d (designating shallow,
moderate, and deep well depths, respectively), located on the near-shore island (Figure 3-9), reveal a
downward hydraulic gradient during testing which increased hi time before stabilization. This vertical
gradient indicates that water from the pond is recharging the aquifer when PW-5 is operating.
Immediately following the start-up of the pump, most of the flow is horizontal. Thereafter, water is
withdrawn from the pond until equilibrium is established. The hydraulic conductivity values estimated
from these monitoring stations range from 2,000 to 3,000 ft/d, well above the average value calculated
for the other stations (approximately 520 ft/d). In conclusion, this section of the aquifer, receives
approximately five to six times more water from surface water than from ground water. By averaging
all ground-water fluxes reaching the well, approximately 20% to 25% of the water recharging PW-5 is
believed to be derived from Spectacle Pond.
At all other monitoring stations, responses to pumping were similar to those observed during
the short-term test. Again, hydraulic conductivity values ranged from about 300 to 3,600 ft/d with an
average value of 520 ft/d. These values are typical for well-sorted sand and gravel outwash deposits.
The Neuman analysis-fitting parameter (p), used to calculate the anisotropy ratio of Kr/Kz, indicates
that regional values of vertical hydraulic conductivity (Kz) are relatively small. This suggests that the
prevailing ground-water flow component is horizontal (Kr). The anisotropy ratio, determined using
data collected during the short- and long-term aquifer tests, is approximately 35:1. The average
specific yield, which was calculated using the best-fit analytical solution, was approximately 0.1054.
Typically, specific yield is representative of porosity values within sand and gravel aquifers. Pond
effects may have created overestimates of specific yield at MW-25s and MW-25d (Figure 3-9). The
average storativity value of 0.0216 is one order of magnitude lower than the specific yield and is
characteristic of an unconfined aquifer.
3-42
-------
TABLE 3-6. AQUIFER PARAMETER RESULTS FOR THE LONG-TERM AQUIFER
TEST AT PW-5, LITTLETON, MASSACHUSETTS
Aquifer Parameters
Monitoring
Station
P-20
MW-25s
MW-25d
P-50
P-lOOs
OW-150
MW-375s
MW-375d
OW-400
MW-425
MW-800
R
(ft)
20
25
25
50
100
150
375
375
400
425
800
Method
Pt-Th*
Rt-Th
Pt-C
Pt-N
Rt-N*
Pt-Th
Pt-N*
Rt-N
Pt-N
Rt-N*
Pt-Th
Pt-C
Pt-N*
Rt-N
Pt-N*
Rt-N
Rt-C
Pt-H*
Pt-Th
Pt-Th
Rt-Th*
Pt-C*
Pt-Th
Pt-N*
Rt-N
Pt-C
Pt-Th
T
(ftVmin)
11
13
21
19
24
20
16
19
16
15
20
19
18
19
18
20
24
96
110
110
136
37
31
16
18
20
15
Kr
(ft/d)
270
315
507
463
576
484
393
458
399
361
487
473
438
471
446
490
589
2,312
2,640
2,654
3,267
888
744
404
449
486
367
S
0.0410
0.0560
0.0340
0.0450
0.0056
0.0041
0.0044
0.0037
0.0057
0.0120
0.0420
0.0370
0.0031
0.0023
sy
orS
0.100
0.200
0.340
0.440
0.410
0.230
0.480
0.370
0.051
0.061
0.026
0.022
0.029
0.014
0.130
0.150
0.180
0.014
0.014
0.030
0.012
0.120
0.087
0.061
0.085
P
0.20
0.20
0.10
0.10
0.06
0.06
0.20
0.30
1.50
1.50
7.00
7.00
Kr/Kz
0.9
0.9
1.7
1.7
12.0
12.0
14.0
9.0
4.0
4.0
7.0
7.0
(continued)
3-43
-------
TABLE 3-6. AQUIFER PARAMETER RESULTS FOR THE LONG-TERM AQUIFER
TEST AT PW-5, LITTLETON, MASSACHUSETTS (continued)
Aquifer Parameters
Monitoring
Well
OW-800
OW-1200
All Wells
R
(ft)
800
1,200
N/A
Method
Pt-C
Rt-Th
TDD
T
(ftVmin)
53
42
NR
36
Kr
(ft/d)
1,273
1,016
NR
801
S Sy
orS
0.130
0.034
p Kr/Kz
0.40 444.0
Abbreviations: PW-5 = Production Well Number 5; R = distance from PW-5 to monitoring station;
T = transmissivity; Kr = horizontal hydraulic conductivity; S = storativity; Sy = specific yield;
P - Neuman analysis-fitting parameter [(Kz/Kr)(r/b)2)]; Kr = horizontal hydraulic conductivity;
Kz «- vertical hydraulic conductivity; Kr/Kz = anisotropy ratio; P = piezometer; Pt = pumping test;
Th — Theis corrected drawdown analysis; Rt = recovery test; MW = monitoring well; s = shallow
depth well; C — Cooper-Jacob corrected drawdown analysis; N = Neuman analysis; d = deep well;
OW = observation well; H = Hantush analysis with aquitard storage; NR = no response; N/A = not
applicable; TDD = Theis drawdown versus distance analysis for all wells.
* « Best-fit method.
Ground-Water Flow Modeling
Two-Dimensional, Steady-State, Regional Flow Model-
Model Description-FLOWPATH is a steady-state, 2-D, horizontal plane, ground-water flow
and pathline analysis model (Franz and Guiguer, 1989). FLOWPATH is based on the block-centered,
finite difference and the particle tracking methods. The model can simulate horizontal ground-water
flow in heterogeneous aquifers, anisotropic aquifers, and in confined, unconfined, and leaky aquifers.
It can simulate withdrawal and injection of water at multiple wells.
Model Assumptions and Limitations—The implementation of the 2-D, finite difference method
yields a numerical approximation of the modeled system. Upon discretization of the natural system
into a large number of rectangular cells, a differential water balance is written for each cell. In the
block-centered, finite difference formulation, a system of equations is solved for the hydraulic head at
the center of each cell. All flow parameters, including the calculated hydraulic head value, represent
an average value over the entire cell volume, neglecting any small-scale variations or vertical gradients.
Good approximations are usually obtained for regional-scale ground-water flow, where the horizontal
extent of the aquifer is much greater than its vertical dimension.
The 2-D formulation does, however, generate inaccuracies around the partially-penetrating
production well and close to surface-water bodies where 3-D flow prevails. In the context of
regional-scale modeling, and especially for the purpose of siting monitoring wells at a significant
3-44
-------
distance from the pumping well, these inaccuracies are considered minimal and are not expected to
affect the validity of the modeling results. A steady-state, 2-D simulation of regional ground-water
flow was conducted to provide conservative, long-term estimates of the capture zone for PW-5.
Grid Design and Boundary Conditions—The model grid design satisfies two main
requirements: (1) the hydraulic gradients induced by pumping the municipal wells in Ayer and
Littleton are properly represented, and (2) the variable topographic relief in the Bennetts Brook
watershed, with its steep slopes, is properly represented in the numerical description of unconfined
flow. Some difficulties were encountered initially in meeting the second requirement. Specifically, the
rapid variation in elevation head causes large gradients in the total head, while the water depth
remains small. This condition can be simulated in a 2-D model by using a very fine model grid. For
these reasons, and because of the irregular geometry of the bedrock and glacial till outcrops (model
boundaries), a largely uniform grid was designed (Figure 3-14). The grid is comprised of 5,418 cells,
the majority of which have dimensions of approximately 150 X 100 feet. The model grid covers a
total area of approximately 5 square miles (approximately 3X2 miles).
The domain boundary primarily reflects the bedrock and glacial till outcrops, which generally
occur at an elevation of approximately 250 feet (Figure 3-15). For the regional model, it was assumed
that the water table coincides with the sediment pinch-out (at outcrops) and was modeled as a fixed-
head boundary. Alternatively, a flux boundary could have been chosen; however, because site-specific
information about the rate of infiltration along these contacts was lacking, a fixed-head boundary was
preferred. East and west of the Bennetts Brook watershed, Sandy Pond and Forge Pond were
represented as constant-head (flow) boundaries. Short sections of the domain boundary adjacent to
these ponds were assumed to coincide with streamlines and, therefore, were modeled as fixed-head
boundary segments.
Bedrock Elevation—In an unconfined aquifer model, because the aquifer transmissiviry equals
the product of hydraulic conductivity and saturated thickness, it is important to quantify the elevation
difference between the water table and the aquifer bottom. For this reason, the relationship between
hydraulic head and transmissiviry requires special attention because of complications caused by the
variation in underlying bedrock topography within the model domain.
The base of the aquifer consists of relatively impermeable bedrock. Bedrock outcrops
reported by the USGS (1985), gravimetry mapping along several profiles (Metcalf & Eddy, Inc., 1981),
and borehole information were used to interpret the bedrock depths and configuration. The bedrock
surface was generated using an interpolation algorithm commonly known as Kriging (de Marsily,
1986). This provided a nodal representation of the bedrock surface, which is considered sufficiently
accurate, and is shown in Figure 3-15.
Surface-Water and Ground-Water Recharge—The interaction between surface water and
ground water in the area of PW-5 was one of the major processes to be determined in the regional
modeling study. Data on the bathymetry of Spectacle Pond were obtained from a previous study
(GHR Engineering Associates, Inc., 1989) and supplemented by the peat-probing investigation. The
peat-probing and aquifer-testing exercises indicate that where peat exists, the peat acts as a relatively
impermeable layer that isolates ground water from the pond. Where the peat layer pinches out,
approximately in the middle of the eastern lobe of Spectacle Pond, significant hydrologic connection
between the pond and ground water was observed.
3-45
-------
3-46
-------
g
-------
In the FLOWPATH simulation model, the interaction of ground water and surface water is
modeled using the principle of a leaky aquifer. With this approach, it is possible to account for the
fluid flux between the pond and the aquifer, obtained from Darcy's Law. If the water-table elevation is
greater than the bottom elevation of the pond, the leakage flux is proportional to the head difference
in the pond and the aquifer. Otherwise, the leakage flux is proportional to the depth of water in the
pond. The leakage flux is also proportional to the hydraulic conductivity of the sediments lining the
surface-water body, and inversely proportional to the thickness of these sediments. The leakage factor
used within the modeling exercise is based on site-specific values for peat thickness and hydraulic
conductivity. For Spectacle Pond, the leakage factor was estimated to be 0.454 gpd, representing a
peat layer with an average thickness of approximately 5 feet and a hydraulic conductivity of 10 ft/d.
The hydraulic conductivity estimate was provided by William Nichols (USGS, Augusta, Maine,
personal communication, 1990) and is based on field tests conducted in similar peat environments
studied in Maine.
Flow data on other surface-water features (tributaries to Spectacle Pond, including Bennetts
Brook) in the study area were limited. At the weir located at the outflow of Spectacle Pond, the water
level of Gilson Brook drops approximately 8 feet. For all streams, a leakage factor of 1.32 gpd and a
water depth of 2 feet were used. Monthly precipitation data from January 1975 to July 1989 were
evaluated from records kept by the Littleton Water Department. The average annual precipitation
rate during this time period was 48.7 inches per year (in/yr). The resulting annual ground-water
recharge rate is approximately 12.2 in/yr, which represents 25% of the total precipitation rate
(Frimpter et al., 1988). The remaining 75% is accounted for by evapotranspiration (50%) and
runoff (25%).
The model recharge rates were chosen to be spatially variable. In areas of higher hydraulic
conductivity, it was assumed that the infiltration rate is higher than average. Recharge rates in aquifer
zones with lower hydraulic conductivity values were assumed to be somewhat lower than the estimated
average. The model net recharge rates (infiltration minus evapotranspiration) ranged from 8.6 in/yr to
35.8 in/yr, with an average of 14.3 in/yr. Because of the uncertainties and seasonal variations inherent
in the surface-water parameters, the leakage factors and recharge rates used to describe the
ground-water and surface-water interaction were studied in a sensitivity analysis. The results of the
sensitivity analysis are discussed along with the 2-D ground-water flow modeling results.
Hydraulic Conductivity Distribution—U.S. Geological Survey ground-water availability
mapping (1985) indicates variations in transmissivity ranging from 0 to 4,000 ft2/d in the Bennetts
Brook watershed. The highest transmissivities are found around Spectacle Pond, where highly
permeable outwash deposits occur. Transmissivities decrease toward the outcrop areas where these
valley-fill deposits pinch-out. Somewhat lower transmissivities are found in the eastern portion of the
model domain where bedrock is reported to occur at shallow depths (GHR Engineering Associates,
Inc., 1989).
The hydraulic conductivity distribution was the most important calibration parameter
considered during the modeling exercise. Hydraulic conductivity was assumed to be isotropic
throughout the model domain, an accepted practice for horizontal, regional-scale modeling. The
described transmissivity distribution was implemented qualitatively in the model hydraulic conductivity
3-48
-------
distribution. Calibrated hydraulic conductivities range from 26 to 230 ft/d, with the lowest values
occurring in the topographically-high areas within the model domain. The greatest values occur
around Spectacle Pond and Forge Pond and along the Bennetts Brook valley. A hydraulic
conductivity value of 90 ft/d was assigned to most of the model area.
Observed Versus Calculated Hydraulic Head Distribution-In order to calibrate the regional
model, an even distribution of monitoring stations for ground-water level measurements throughout
the domain was desired (Figure 3-8B). A complete survey of ground-water level measurements was
conducted on May 21,1990, after PW-5 was shut-down for 3 days. The field-measured hydraulic
heads (accurate to ± 0.01 foot) are shown in Figure 3-16. Also included in Figure 3-16 is the
calibrated, model-synthesized hydraulic head distribution displayed in 5-foot contour intervals.
Measured and calculated hydraulic heads from observation stations located within the near-field region
of PW-5 are shown in greater detail in Figure 3-17. The comparison between measured and calculated
heads show reasonably good agreement in the vicinity of PW-5. Larger discrepancies occur in areas
close to the domain boundary where the fixed-head boundary condition strongly constrained the
model results. Despite these deficiencies, the model produced an acceptable simulation of the
regional-scale ground-water flow regime, particularly in the area of greatest interest, PW-5.
Model Results-Upon calibration, FLOWPATH was used to delineate the capture zone for
PW-5. The model was run with a pumping rate of 650 gpm at PW-5 to determine the hydraulic head
distribution under long-term, steady-state pumping conditions. For the determination of the capture
zone, the ground-water travel-time criterion was employed. The time-related criterion defines the
WHPA as the aquifer volume through which ground water moves toward the well, arriving at the well
within a specified time. The affected aquifer volume is called the zone of contribution (ZOC) and is
also referred to as the time-related capture zone.
A second, less sophisticated criterion for WHPA delineation is drawdown. For the
hydrogeologic setting of the study area, model test results from FLOWPATH indicate that the
drawdown criterion produced inappropriate protection results (Figure 3-18). In other words, the ZOI,
which is uncorrected for the regional gradient effect, should not be used to define the WHPA in a
setting with a significant regional ground-water gradient because the area will be overprotective in the
downgradient direction. The regional gradient should be subtracted from the ZOI to generate the
ZOC, which should serve as the basis for the WHPA (U.S. EPA, 1987). For these reasons, and
because of the more quantitative nature of the time-related criterion such as the ability to better
predict contaminant transport rates, the time-related criterion is generally preferred over the
drawdown criteria for delineating a WHPA
Time-related capture zones for various intervals (50, 100,150, 200, 300, and 400 days) are
shown in Figure 3-19. The capture zones are narrow and preferentially extend toward the south. The
capture zones also extend to the north across the pond, which indicates ground-water recharge from
the northern portion of the aquifer to the northwest from Spectacle Pond. The effect of surface-water
influences on the capture zone has major implications for wellhead protection strategies. Any
pollutant entering Spectacle Pond by way of the surface water or ground water could potentially
contaminate the well. Therefore, the entire Spectacle Pond watershed requires some degree of
protection, with greater attention focused on the capture zone for PW-5.
3-49
-------
3-50
-------
•S3
1
a
I
o
I
-------
Q
O
O O O CE
Z O < =!
O en O <
Q. m a: o:
3-52
-------
3-53
-------
The quantitative relationship between the well discharge and the pond leakage is addressed in
the 3-D modeling study. The time values shown on the capture zones indicate the approximate time,
in days, required for fluid particles located on the time contours to reach PW-5. The area of the
capture zone did not increase for times greater than 400 days. This condition indicates that the
amount of water withdrawn by the production well and the amount of recharge (precipitation plus
leakage) occurring within the 400-day contour are in equilibrium. The 400-day capture zone,
therefore, represents steady-state conditions for a discharge rate of 650 gpm at PW-5. The recharge
zone would increase if the well were pumped at a higher rate.
Model Sensitivity— A sensitivity analysis was performed to determine the effects of pond
leakage, ground-water recharge, and aquifer porosity on the delineation of the WHPA for Spectacle
Pond. The purpose of this analysis was to investigate values of these parameters within their
uncertainty ranges. The leakage factor is given by the ratio of hydraulic conductivity and thickness of
the pond bed. An increase in the leakage factor in the pond resulted in a decrease in hydraulic head
in the pond area. For very high leakage factors, the ground-water level in the aquifer around the pond
approached the pond water-level elevation. For low leakage factors, which correspond to low
hydraulic conductivities of the pond bed, the hydraulic heads increased. This behavior indicates that
the pond is recharged by ground water flowing from the topographically-higher areas through the
pond bottom. Greater recharge occurs in areas of higher hydraulic conductivity, where a good
connection between the ground water and the pond exists. For increased leakage factors, the
production well derives more water from the pond and the upgradient areas located to the south,
adjacent to the pond. For reduced leakage factors, the well derives most of its water from the south
and a reduced amount from the pond.
Increased ground-water recharge causes a general reduction in the size of the time-related
capture curves. The recharge area extends to the aquifer boundary, as in the original calibrated
simulation. Ground water and contaminants originating at the boundary, however, reach the well
within a shorter period of time. A reduction of the recharge rate has the opposite effect.
Porosity affects the magnitude of the average linear ground-water velocities, but not their
directions. Travel times, therefore, decrease or increase for decreasing or increasing porosities,
respectively. The shape of the time-related capture curves with the adjusted time levels remains the
same. For example, the 200-day contour for the simulation shown in Figure 3-19, with a porosity of
30%, would change to a 100-day contour for a porosity of 15%.
Implications of the Regional Approach—To some degree, the limitations of a 2-D model will
bias the size and shape of the predicted WHPA The 2-D formulation neglects vertical flow gradients;
therefore, it appropriately simulates regional-scale systems in which flow occurs predominantly in the
horizontal plane. Three-dimensional flow conditions with significant vertical components exist in the
immediate vicinity of partially-penetrating wells and shallow surface-water bodies and in strongly
variable topography.
A 2-D horizontal model cannot resolve vertical layering within the aquifer. Vertical variations
in aquifer parameters such as hydraulic conductivity and porosity are averaged over the aquifer
thickness. In addition, the withdrawal rates of partially-penetrating wells are evenly distributed over
3-54
-------
the entire aquifer thickness. This results in lower stress on the aquifer over a larger area than the
partially-penetrating well would exert Withdrawal from a partially penetrating well may exert very
high stress over a shorter screen length and very small stress over the rest of the well bore.
These conditions all apply to PW-5. The Bennetts Brook watershed has variable topography,
and the partially-penetrating production well, PW-5, is located close to a shallow pond with a
semi-pervious peat layer at the pond bottom. The fine grid and the relatively extensive data base of
the 2-D model suggest that the regional and local flow systems were described appropriately by
FLOWPATH. For this study, a comparison of the 2-D versus the 3-D model simulations is useful to
address all of the limitations of the 2-D simulation.
Three-Dimensional, Steady-State, Local Flow Model—
Modeling Purpose—One of the essential assumptions inherent in the use of the 2-D model is
that the predominant direction of flow in the aquifer is horizontal. For most thin stratified-drift
aquifers, this is generally true except in recharge or discharge areas. In the vicinity of the partially-
penetrating PW-5, vertical flow will be greatest near the leaky stream bed of Spectacle Pond. A 3-D
flow model of this section of the aquifer was constructed (1) to obtain better definition of flow
patterns beneath and around the pond, and (2) to determine the effects of the partial well penetration
and of vertical anisotropy. Another model, which is basically a post-processor of the flow model, was
used to determine pathlines and times of travel. The numerical model used to simulate 3-D flow of
ground water in this area is MODFLOW (McDonald and Harbaugh, 1988).
Model Description—MODFLOW utilizes a finite-difference numerical method to solve the
3-D flow equations. In addition to the options available in the 2-D model, the 3-D code allows
simulation of vertical flow components between adjacent aquifer units. The units may or may not be
separated by confining layers. Furthermore, the model simulates leakage to or from the aquifer
through a leaky stream bed in any model layer. Aquifer layers may be confined, unconfined, or a
combination of each, and recharge may be applied to any model layer. The post-processing of the
hydraulic heads simulated by MODFLOW was completed with MODPATH (Pollock, 1989).
Model Exercise—The 3-D model for the area in the vicinity of the production well is based on
the same grid used in the 2-D model. The 3-D grid uses essentially the same distribution of hydraulic
properties, recharge, and discharge, except that it is divided vertically into four layers. The finite
difference grid used in the 3-D simulation is shown in Figure 3-20. Layer 1 (top layer) of the model
has the same areal extent as the near-field, 2-D model; however, the remaining layers are of lesser
areal extent to more accurately simulate the bedrock topography (Figure 3-21). The horizontal grid
spacing is approximately 300 feet, except near the well where it is 50 feet. The vertical grid spacings of
45, 15, 15, and 30 feet correspond to Layers 1 through 4, respectively (Figure 3-20).
Layer 1 in the model is simulated for unconfined conditions. Transmissivity is computed as a
function of saturated thickness and hydraulic conductivity after each model iteration. Nodes where the
leaky stream bed is simulated are also located in the top layer. Constant-head boundary conditions
were assigned for all four sides in the top layer, using hydraulic heads generated by the steady-state
2-D simulations. The boundary conditions for the bottom three layers of the model are simulated as
constant-head where there is no bedrock intercepting the layer, and otherwise, as no-flux boundaries.
The bottom three layers are interactive with the layers immediately above or below.
3-55
-------
SPECTACLE.
POND
Figure 3-20. Three-dimensional model boundaiy, grid domain, and cross section locations in the
vicinity of PW-5, Bennetts Brook watershed, Littleton, Massachusetts.
3-56
-------
\
LAYER 1
LAYER 2
LAYER 3
LAYER 4
Figure 3-21. Multi-layer grid for the three-dimensional model in the vicinity of PW-5 (small black
square), Bennetts Brook watershed, Littleton, Massachusetts, (Large black areas represent bedrock)
3-57
-------
Horizontal hydraulic conductivities for the 3-D model are the same as those used in the 2-D
model. The vertical hydraulic conductivity was determined as 1/35 of the horizontal values (the
average ratio obtained from long-term aquifer test). A stratigraphic section was drawn for each model
column to define material properties. Four examples; of stratigraphic sections with the finite difference
grid are shown in Figure 3-12. In finite difference blocks where two different lithologic materials were
present (for example, aquifer and peat or aquifer and bedrock) the harmonic mean of the respective
hydraulic conductivities was used. Recharge to the model was applied to Layer 1 in the same
proportions as that used for the 2-D simulation. The production well is present in Layer 3 (Figure
3-20), thus simulating the 15-foot screen length located between the depths of 35 and 50 feet.
Model Results—As in the 2-D modeling exercise, simulations were run to steady-state
conditions so that the maximum effects of pumping could be observed. At steady state, the sources of
water to the production well are (1) induced infiltration from the pond, and (2) captured ground-water
discharge from areal precipitation and runoff from upland till and bedrock outcrops.
Simulated hydraulic head distributions under pumping conditions at a rate of 650 gpm are
shown in Figures 3-22 and 3-23. Cross section E-E' (Figure 3-22) indicates that the pond is receiving
ground water from the uplands through the bottom and is, therefore, effluent in areas far from the
well. In contrast, cross section D-D' indicates that in areas adjacent to the well, part of the
production water is withdrawn from Spectacle Pond. Both cross sections in Figure 3-22 show
downward vertical gradients from the valley uplands toward the center of the flow system.
Hydraulic head distributions for the four layers in the model are presented in Figure 3-23 in
the plan view. Maximum drawdown occurs in Layer 3, the layer in which the well is screened. The
horizontal, 2-D simulation represents an average of the heads in all layers; therefore, no direct
comparison is possible. However, the layer that best matches the 2-D model simulation is Layer 2.
The surface-water contribution to the discharge of PW-5 is determined by the hydraulic
conductivity and saturated thickness of the peat layer, and by the horizontal versus the vertical
hydraulic conductivities of the aquifer. For average conditions, a water-budget calculation from the
model indicates that 19% of the production water (approximately 120 gpm of the 650 gpm total) is
induced from surface water. The value of 19% is very near the 20% to 25% range estimated from the
aquifer-test analysis.
On the basis of the 3-D head distribution data, time-related capture zones were calculated for
50,100, 200, and 400 days. The 400-day capture zone is equal to the steady-state capture zone. In
3-D, the outline of a given capture zone represents an irregularly-shaped surface. The intercept of the
computed surface with the four model layers is shown in Figure 3-24. In all layers, the capture zones
extend to the aquifer limits, both north and south. In Layer 3, where the well screen is completed,
there is a stronger horizontal flow component than in the other layers. Therefore, the capture zone in
Layer 3 extends farther to the east and west. This east-west extension is not as pronounced in Layer 1
because of the predominant vertical flow components close to the pond. The capture zone in Layer 1
represents the surficial expression of the area that should be prioritized for protection against
potential contamination threats. The capture zones generated for the other layers will prove useful in
designing multi-level monitoring wells for early-warning detection of ground-water contamination.
3-58
-------
E'
250
~5> 225.
-------
LAYER 1
LAYER 2
LAYER 3
LAYER 4
LEGEND
HYDRAULIC HEAD CONTOUR (feet amsl)
CONE OF DEPRESSION
PRODUCTION WELL (PW5)
BROOK
• ROAD
• RAILROAD TRACK
Figure 3-23. Plan view of hydraulic head distributions for each layer of the three-dimensional model,
Bennetts Brook watershed, Littleton, Massachusetts.
3-60
-------
LAYER 1
LAYER 2
LAYER 3
LAYER 4
LEGEND
- TIME-RELATED CAPTURE ZONE (DAYS)
(MODFLOW; 3-D. STEADY-STATE MODEL)
» - PRODUCTION WELL
— - BROOK
ROAD
~-~-~ - RAILROAD TRACK
Figure 3-24. Time-related capture zones for PW-5 for each layer of the three-dimensional model,
Bennetts Brook watershed, Littleton, Massachusetts.
3-61
-------
Two-Dimensional, Transient, Regional Flow Model--
Modeling Purpose-The 2-D, transient model, FLOWCAD (Franz and Guiguer, 1992), was
used to delineate capture zones for extreme hydrologic conditions given by the Massachusetts
Department of Environmental Protection (1991). Under the DEP criteria, the hydrogeologic
simulation is controlled by stressed recharge conditions (no precipitation) for a specific time period
(180 days). These delineation criteria result in conservative capture zones and associated Water
Resource Protection Areas. Zone I is set by the State at a fixed radius of 400 feet; Zone II is
determined using the following procedure:
1. Construct a water-table contour map representative of the predeveloped, long-term
average conditions in the aquifer.
2. Predict drawdowns by imposing Zone II criteria (180 days of pumping at safe yield
rates with no recharge from precipitation), using an appropriate analytical or
numerical model.
3. Determine the Zone II water-table contours by subtracting the predicted drawdowns
from the long-term average water-table contours.
4. Construct a flow net based on the resulting Zone II water-table contours.
5. Identify the ground-water divide, induced by pumping, which separates the area of
contributing water to the well (Zone II) for the aquifer outside of Zone II.
6. Extend the ground-water divide upgradient to its point of intersection with prevailing
hydrogeologic boundaries.
7. Delineate Zone II for the production well as the area determined by this procedure.
Zone II was delineated for PW-5 using FLOWCAD to meet DEP criteria. The results of the
transient-state modeling are compared to the steady-state, 2-D modeling and the 3-D, near-field model
in the following section, titled, "Comparison of Model Results."
Model Description-FLOWCAD is an extension of FLOWPATH (2-D, numerical model),
except that it performs transient-state modeling. Specifically, FLOWCAD simulates time-related
responses of the hydraulic head distribution to recharge and discharge stresses.
Model Exercise-The calibrated hydraulic heads simulated with FLOWPATH, representing
average recharge conditions under nonpumping conditions, were used to construct the regional water-
table contour map (Procedure 1). FLOWCAD was used to predict production well drawdowns
(Procedure 2). The same model grid, bedrock elevations, boundary conditions, and aquifer property
distributions used in the steady-state, regional FLOWPATH simulation were used for the FLOWCAD
simulation. All model assumptions and limitations discussed in the regional FLOWPATH simulation
apply to the transient exercise.
3-62
-------
The regional water-table contour elevations previously determined were used for the initial
hydraulic head distribution. The production well was then pumped at a rate of 650 gpm, and recharge
from precipitation was assigned a zero value. The model simulated 180 days of continuous pumping,
and the resulting hydraulic head distribution was used to predict the Zone II water-table contour.
Using this hydraulic head distribution, a pathline analysis was performed to determine the ZOC for
PW-5.
Model Results—The resulting capture zone is shown in Figure 3-25. Because the capture zone
is controlled primarily by the limited recharge condition rather than the time-related constraint, it
extends to the hydrogeologic boundaries, thus satisfying the Zone II criteria outlined earlier
(Procedure 6). While hydrologic boundary conditions were reached to the north and south (bedrock
and till), and to the west (Spectacle Pond), the transient-state capture zone expands toward the east
and west to compensate for the loss of recharge from precipitation. Generally, as recharge from
precipitation decreases, the areal extent of the ZOC increases, if hydrogeologic boundaries are not
encountered.
Comparison of Model Results—
A graphical comparison of the steady-state 2-D (regional scale), steady-state 3-D (local scale),
and transient 2-D (regional scale) capture zones is shown in Figure 3-26. The variations in size and
shape of the capture zones are directly related to the assumptions and limitations of the initial model.
Spatial factors in model design are best exemplified by comparing the 2-D and 3-D steady-state
solutions. Effects of temporal and recharge conditions are best represented when comparing the 2-D
steady-state results with the 2-D transient results. Depending on the predetermined modeling and
wellhead protection objectives, each can be used effectively to delineate a WHPA.
The 2-D and 3-D steady-state models generated slightly different capture zones. Compared to
the 3-D model, the 2-D solution underestimates the eastern and western directions and overestimates
the contributing areas to the north and south. The reason for these differences is the 3-D model,
which accounts for variations in aquifer depth and the resultant vertical flow field, making generation
of a more realistic 3-D ZOC possible. For the relatively thin aquifer considered in these simulations,
the 3-D ZOC extends throughout the entire depth of the aquifer. The enveloping surface of the
contributing area is close to vertical everywhere, unless constrained by the bedrock. In general, the
3-D ZOC is egg-shaped at depth (Figure 3-24); the deeper boundaries (Layers 3 and 4) are
constrained by the bedrock valley.
Neither the 2-D steady-state nor the 2-D transient model accounts for the vertical flow field
and variations in aquifer depth. Consequently, these model limitations cannot be a factor in the
comparison of the simulated capture zones. Interestingly, their respective capture zones are similar in
shape. The capture zone created by the transient flow model extends well beyond the downgradient
(eastern) boundary of the steady-state equivalent and slightly beyond the western boundary. In both
cases, the aquifer boundaries are reached to the north and south. The wider capture zone generated
by the 2-D transient model underscores the influence of limited recharge during the 180-day
simulation. The effects of temporal variations, steady-state versus 180 days of pumping, are not as
significant as the recharge (precipitation) effect
3-63
-------
3-64
-------
3-65
-------
Water Quality Monitoring
Ambient Trend Monitoring—
The chemical composition of ground water within glaciofluvial deposits can be classified into
several different water quality types depending on the relative abundance of anions and cations. Using
the classification system described by Freeze and Cherry (1979), the water quality within the Bennetts
Brook watershed has been characterized as Type I Water. This is defined as slightly acidic and very
fresh [total dissolved solids (TDS) <100 mg/1], in which sodium, calcium, or magnesium are the
predominant cations, and bicarbonate is the predominant anion (Table 3-4). More detailed chemical
analyses may be conducted graphically using various techniques such as Piper, Stiff, or radial plotting
(Hem, 1989). These analyses are created by plotting the percent milliequivalents of the major anions
against the percent milliequivalents of major cations. For water quality data collected within the
Spectacle Pond aquifer, trilinear Piper diagrams were used to evaluate these chemical relationships.
This graphical representation helped identify two distinct hydrochemical fades that exist within the
study area (Figure 3-27). A sodium-bicarbonate water type was identified above 25 feet in depth, and
a calcium-bicarbonate water type was identified below 25 feet in depth.
To evaluate ambient water quality changes over the duration of the long-term aquifer test, a
series of samples were collected from the near-field monitoring well network. The results of these
tests indicate that the water quality at all stations, including the shallow and deep cluster wells as well
as PW-5, shifted toward more saline (sodium chloride) conditions as the aquifer test progressed
(Figure 3-28). Although the concentration of sodium is low, its increase in concentration is
proportionately higher than the other cations measured (such as calcium and magnesium). It appears
that the proportionately higher sodium concentrations observed within the surface and shallow ground
waters are migrating downward within the area-of-influence of PW-5, thereby changing the water
chemistry throughout the monitoring well network. This temporal trend in chemistry is supported by
modeling results, which indicate a downward flow gradient within the capture zone under pumping
conditions.
Source Assessment Monitoring—
Chemical Parameters—Target chemical parameters were analyzed to evaluate anthropogenic
and natural impacts on water quality. Anthropogenic impacts are traced to specific land-use activities.
In the Spectacle Pond study area, potential sources of heavy metals, VOCs, pesticides, herbicides, and
nutrients were identified. Source-assessment monitoring locations were carefully selected based on
their downgradient position with respect to existing and potential threats. If the source is believed to
issue chronic releases (steadily over time with little variation in concentration), then only one sample
was collected to provide an accurate representation of ground-water quality (Achinger and Shigehara,
1968). This procedure was adopted to establish baseline water quality characterization within the
capture zone. As a QC measure, and to monitor possible errors incurred during the initial sampling
event, the monitoring network was sampled on two separate occasions, once on August 15, 1990, and
again on August 27,1990. It is important to emphasize that for compliance monitoring or for more
stringent Q A/QC objectives, additional data would be required.
3-66
-------
CATIONS
PERCENT OF TOTAL
(MIUJEQUIVALEUTS PHI LITER)
AN1ONS
LEGEND
X - MONITORING STATIONS
<25 FEET DEEP
O - MONITORING STATIONS
>25 FEET DEEP
Figure 3-27. Piper diagram comparison of average water quality in shallow and deep ground-water
zones near PW-5, Bennetts Brook watershed, Littleton, Massachusetts.
3-67
-------
(A)
reHOKT OF TOTAL
(MIUlEOUtVAUKTS P£H UTEJ!)
(B)
LEGEND
X - AVERAGE WATER QUALITY AND
DISTRIBUTION ENVELOPE BEFORE
PUMPING
O - AVERAGE WATER QUALITY AND
DISTRIBUTION ENVELOPE AFTER
11 DAYS OF PUMPING
KXCEHT OF TOTAL
( UUJEOUIVAUHTS PEH LtTBI)
Figure 3-28. Piper diagram comparison of ambient water quality changes during the long-term aquifer
test in the (A) shallow and (B) deep aquifer systems, Bennetts
Brook watershed, Littleton, Massachusetts.
3-68
-------
Results from these analyses, detailed in Table 3-4, show that, with the exception of sodium,
ground-water quality within the capture zone of PW-5 is unaffected by land-use activities. The higher
levels of sodium detected within the surface water and the shallow aquifer may result from roadway
de-icing activities.
Naturally occurring iron and manganese have been detected at elevated concentrations within
the Spectacle Pond watershed (Bray et al., 1990). Over the past several years, iron and manganese
levels within the PW-5 have been gradually increasing (Figure 3-29). Locally, total iron and total
manganese concentrations are elevated throughout the recharge area of PW-5 (Figures 3-30A and
3-30B, respectively). The average concentrations of iron and manganese were 2.03 mg/1 (range, 0.13
to 6.9 mg/1) and 0.48 mg/1 (range, 0.06 to 1.07 mg/1), respectively. These ground-water concentrations
were significantly higher than the levels detected in either the surface water or PW-5 (0.18 mg/1 and
0.08 mg/1 for iron and manganese, respectively). This discrepancy may be caused, in part, by the high
amounts of suspended clay particulates found within the monitoring well network during the sampling
events. Negatively charged clay particulates can preferentially attract iron and manganese cations.
Hem (1989) has suggested that colloids smaller than 0.45 micron can pass through most standard filter
membranes. If the water sample contains a high percentage of these colloids, total iron and
manganese analyses performed on the resulting filtrate may result in erroneously elevated values.
A temporal trend analysis, generated by plotting the concentrations of iron and manganese
over the duration of the long-term aquifer test, was also performed. Other than a very slight decrease
in manganese concentration, no significant trends were observed (Table 3-4).
Physical Parameters—To verify hydraulic connection between the surface-water and
ground-water systems, physical parameters were measured and compared between the pond and the
monitoring network. The spatial variation in temperature measurements was assessed horizontally
and vertically, under nonpumping water-table conditions. Temporal changes were measured
throughout the long-term aquifer test for 11 days.
On August 15,1990, the day before the long-term aquifer test began, baseline temperatures
were measured throughout the monitoring network. The overall temperature range was 52°F to 77°F.
These values are typical for late-summer conditions within this region (Wetzel, 1975). The pond was
slightly stratified, as indicated by a temperature of 77°F near the surface (epilimnion) and 73°F near
the bottom of the pond (hypolimnion). The greatest vertical temperature gradient was observed
within cluster well MW-375, where a 10°F difference was measured between the shallow well (8 feet)
and the deep well (47 feet). Temperature ranges measured in wells, set at a minimum depth of 20
feet, indicate a less pronounced, lateral range of 59°F at MW:375d to 54°F at MW-800.
Automatic temperature probes installed in several of the near-field monitoring wells were used
to record rapid, early-time and late-time temperature changes during the startup and shutdown stages
of the long-term aquifer test. These recordings, in addition to periodic measurements taken as part of
the water quality sampling events, highlighted two significant trends. First, monitoring stations
located between Spectacle Pond and the production well (MW-25, MW-100, and MW-375; Figure 3-9)
showed a noticeable warming trend as the aquifer test progressed. Other monitoring stations
(MW-425, MW-660, and MW-800; Figure 3-9) located downgradient of PW-5 showed a less
3-69
-------
1984
1985
1986
1987
TIME (yr)
1988
1989
1990
Figure 3-29. Increasing concentrations of iron (Fe) and manganese (Mn) in PW-5 from 1984 through
1990, Bennetts Brook watershed, Littleton, Massachusetts.
pronounced cooling trend (Figure 3-31). Secondly, at cluster wells where warming trends were
observed, both the shallow and deep wells responded similarly to overall temperature changes
(Figure 3-32).
In general, ground-water temperatures increase, both vertically and laterally, toward Spectacle
Pond. As the aquifer was stressed during pumping, ground water between the pond and production
well displayed a significant warming trend at all depths. These measurements demonstrate the
significant hydraulic connection that exists between warmer surface water and cooler ground water
within the capture zone of PW-5.
Using a technique described by Lapham (1989), quantitative estimates were made for
downward flow velocities at MW-600s and -600d, MW-25s and -25d, and MW-375s and -375d. This
technique establishes one relationship between vertical temperature profiles in sediment beneath a
surface-water body and vertical ground-water flow. Then, a secondary relationship is established
between these factors and the effective vertical hydraulic conductivity of the sediment. Using this
relationship, vertical hydraulic conductivities were measured from 0.01 ft/d, under nonpumping
3-70
-------
Q E Q O Q OC
O g Z O < =!
-------
=
cc O o tr O <
Q. S D. m oc or
rf
1
52
o
a
i
§
-------
4 6 8
TIME AFTER PUMPING (d)
12
Figure 3-31. Water temperature changes observed at depths greater than 25 feet near PW-5 during
the long-term aquifer test, Bennetts Brook watershed, Littleton, Massachusetts.
80
-70
UJ
IT
I
or
UJ
Q-
UJ
60
50
40
MW-25
SHALLOW
WELL
DEEP
WELL
MW-100
MW-375
11
11
Figure
0 11 0
TIME AFTER PUMPING (d)
3-32. Comparison of temperatures at deep and shallow cluster wells exhibiting warming trends
during the long-term aquifer test, Bennetts Brook watershed, Littleton, Massachusetts.
3-73
-------
conditions, to greater than 1.0 ft/d under pumping conditions. These values, although somewhat lower
than those obtained during the short- and long-term aquifer tests, provide additional evidence of the
downward hydraulic gradient detected within the cap ture zone of PW-5.
REFINED WELLHEAD PROTECTION AREAS
To facilitate ground-water monitoring activities and land-use management efforts, the
Spectacle Pond WHPA has been sub-divided into three protection zones: Zones I, II, and III
(Figure 3-33). Each zone encompasses an area where contamination threats represent a specified
level of risk to the water quality at PW-5. Consequently, each zone requires different levels of risk
abatement as the potential for drinking water contamination decreases from Zone I to Zone III.
Establishing protection zones within the WHPA provides WHPP managers with a method of
determining variable land-use control measures and water quality monitoring activities.
Zone I is defined as the land surrounding PW-5 that is currently owned and controlled by the
Littleton Water Department. This 14-acre area encompasses the 400-foot protective radius mandated
by the Massachusetts DEP and applies to all public water supplies with planned yields of 100,000 gpd
or greater. The fixed-radius criteria of Zone I is designed to reduce potential sources of pathogenic
contamination introduced by septic discharge. Septic effluent is one of the more common source(s) of
pathogenic contamination. Although the threat is greatest in surface-water systems subject to direct
contamination from point-source sewage discharge or abundant native fauna, ground-water supplies
can be contaminated indirectly by induced surface-water contributions. Pathogenic organisms that are
subject to tortuose ground-water transport, especially where the travel times are 30 to 90 days or more,
have a fairly poor survival rate. While this process provides some measure of protection, it is not
entirely effective. Because of the proximity and short travel times calculated between Spectacle Pond
and PW-5 (approximately 50 days), land-use controls within Zone I may not be completely effective in
eliminating pathogenic contamination at PW-5.
Using the most conservative approach, Zone II was delineated as the union of the capture
zones, which were established using the 3-D, steady-state criteria (MODFLOW simulation) and the
2-D, transient criteria (FLOCAD simulation) employed by the Massachusetts DEP (1991). The 2-D,
steady-state ZOC (FLOWPATH simulation) is encompassed within this union. Ground-water
flowpaths and TOTs, obtained during the 3-D, steady state modeling exercise, were used to develop
the long-term optimized monitoring program for Zone II. The ground-water monitoring program
will provide detailed, baseline information regarding local water quality impacts.
Zone III is defined as the upgradient area of the Bennetts Brook watershed, which recharges
Zone II. Because regional impacts on water quality at PW-5 are buffered by Spectacle Pond (dilution
effect), land-use regulations and continued source assessment monitoring will provide sufficient water
resource protection within Zone III.
Zones I, II, and III are created to facilitate cost-effective and successful WHPP management.
Each zone requires a specific level of protection. Zone I must be owned and controlled by the local
water department or town. Strict land-use controls and ground-water monitoring activities are
3-74
-------
3-75
-------
required within Zone II. Zone HI necessitates practical, yet comprehensive, land-use regulations with
additional ground-water monitoring activities performed around high-risk, land-use activities. Without
differential zoning of the WHPA, administrators may not have sufficient information regarding the
ground-water flow dynamics to properly allocate monies and management efforts needed to maintain
WHPP goals.
3-76
-------
WELLHEAD PROTECTION MONITORING PROGRAM
MONITORING OBJECTIVES
The existing ground-water monitoring program for PW-5 incorporates water quality testing to
satisfy state and federal drinking water regulations and source assessment monitoring to satisfy local
land-use requirements. Ground-water flowpaths and TOTs recently identified as a result of this
research will be used to develop a more comprehensive, long-term program. While monitoring
program design within Zone II of PW-5 is emphasized, the upgradient recharge zone, Zone III, will
also be evaluated. To meet long-term monitoring objectives, existing and new monitoring wells
(identified as MWs) and surface-water stations (identified as SWs) are proposed. Each station is sited
for a specific monitoring purpose. Monitoring parameters and sampling frequencies are also
proposed for each station. Specific chemical and physical parameters were selected based on baseline
water quality information and potential contamination sources. Monitoring frequencies were
determined from hydrogeologic properties and associated ground-water travel times.
Ultimately, the success and reliability of the monitoring program is dependent on the accuracy
of the capture zone delineation, ground-water flow paths and travel times, and baseline water quality
determination and source assessment activities. Source audits will continue to be performed to identify
and prioritize land-use risks because Zones II and III are subject to regional development. Dramatic
changes in land-use activities may result in the addition of new monitoring stations.
MONITORING SITES
Although dilution effects from Spectacle Pond may reduce contamination burdens originating
in Zone III, ground-water compliance monitoring activities will continue to be conducted for source-
assessment purposes. Within Zone II, a monitoring well network will be established to evaluate water
quality within the capture zone of PW-5. In addition to ground-water monitoring wells, surface-water
stations will be sampled to evaluate effects from induced recharge identified between Spectacle Pond
and PW-5.
Ground-Water Stations
All ground-water monitoring stations will conform to accepted EPA structural designs, as
described in the Ground-Water Monitoring Stations section. Within the unconsolidated sediments of
Zones I, n, and III, all monitoring wells will be constructed with 2-inch-diameter, schedule 40,
flush-threaded PVC casing and 0.01-foot slotted well screens. The annular space between the drill
boring and PVC screen will be backfilled with clean silica sand, and sealed 1 foot above the screen
with bentonite clay. Protective casing and security locks will be placed over exposed risers and
3-77
-------
cemented in place 2 feet below ground level (Figure 3-7A). Sampling protocol for surface- and
ground-water monitoring stations will conform to accepted EPA methodologies, as previously
described and referenced.
Existing Ground-Water Stations-
Compliance monitoring wells, which were installed around the perimeter of existing
ground-water sources, will continue to be used for source-assessment water quality evaluations.
Compliance monitoring stations (abbreviated as MW-C; shown as a well symbol in Figure 3-34) are
sited around the municipal landfill and transfer station and a large industrial and commercial complex
within Zone III.
By designation, Zones I and II are more stringently regulated than Zone III; therefore, all
land-use activities identified within these regions will be strictly monitored. Development is prohibited
within Zone I. Zone II contains residential and commercial developments, which pose direct
ground-water contamination threats to PW-5. To meet the monitoring needs within this area, existing
monitoring wells (identified as MW-Es) and proposed monitoring wells (identified as MW-Ps) were
incorporated into the optimized monitoring network (Figure 3-34). Flowpath and travel-time results
were consulted in determining well sites.
Monitoring well 660s and MW-660d (renamed MW-Els and MW-Eld, respectively) will be
used to evaluate impacts from an agricultural supply warehouse located within the northeast quadrant
of Zone H. Pesticides and solvents are sold at this facility, and past storage practices are not well
documented. These two wells will also be used to monitor Route 119 and railroad activities
immediately upgradient The cluster arrangement of deep and shallow well screens will permit
monitoring of both light, nonaqueous-phase liquids (LNAPLs such as petroleum distillates) and
dense, nonaqueous-phase liquids (DNAPLs such as industrial-strength solvents and pesticides). The
locations of MW-Els and MW-Eld correspond to the 150-day and 300-day travel-time contours,
respectively (Figure 3-34). Assuming contaminant transport through adhesive processes, this well pair
is expected to provide between 150 to 300 days of early warning detection.
Monitoring well 800 (renamed MW-E2 in Figure 3-34) is located along the eastern edge of
Zone n near a dense commercial district. This well will be used to monitor the commercial
development as a potential source. It will be monitored for site-specific parameters in the event of a
severe regional drought and for regional water quality parameters. Located along the Zone II
boundary, MW-E2 corresponds to approximately the 400-day travel-time contour. However, this
aspect of the well location is not as critical because MW-E2 is a source-assessment monitoring well
rather than one for early warning detection.
Recommended Ground-Water Stations—
To complete the ground-water monitoring network within Zone II, two proposed monitoring
wells (identified as MW-P1 and MW-P2) were sited in the northern and southern portions of the
zone, respectively. These wells will be installed for early warning monitoring. Because the aquifer
pinches-out to the north and south, the well screens will be set at shallow depths of less than 30 feet.
Monitoring well PI and MW-P2 are situated to monitor nonpoint source activities attributed to the
residential development. Both wells will be fully screened to allow for the detection of both LNAPLs
and DNAPLs. The locations of MW-P1 and MW-P2 correspond to the 300-day travel-time contour
(Figure 3-34). These wells are expected to provide approximately 300 days of early warning detection.
3-78
-------
e
CM
2tSs
uT a 5
•s s s.
E 1. 2
d li 3 I
UI UJ 111 CCS
5 5 S tfw
O <9
0--O.Z zoz 3z 5 3=
-------
Although the intent of the long-term monitoring program is to maintain the clean ground-
water resources available to PW-5, samples will still be collected from PW-5 to comply with the
SDWA. These samples will also demonstrate the effectiveness of the wellhead protection monitoring
program. If water quality problems are encountered at PW-5 and not at other monitoring stations in
Zone n, then an investigation, which will utilize all available monitoring stations within Zone II, will
follow to determine the origin of the contamination.
As additional wells are incorporated in the long-term monitoring network over time; criteria
should conform to standard installation and design protocols and guidelines. Well placement and
design specifications will be determined by the type of contamination identified and local flowpath and
travel-time characteristics. At present, PVC and stainless steel are acceptable materials for routine
ground-water quality monitoring stations. If ground-water contamination is identified, Micro Wells
may be employed for plume delineation.
Surface-Water Stations
Spectacle Pond is a discharge area for regional ground-water flow and it receives surface water
from the upstream reaches within the Bennetts Brook watershed. For these reasons, the water quality
within Spectacle Pond reflects that of the entire watershed. Spectacle Pond contributes 20% to 25%
of the total recharge to PW-5; therefore, the surface-water quality of the pond must be considered in
the long-term monitoring program. The water quality evaluation of Spectacle Pond provides a
method of determining regional land-use impacts from the towns of Littleton and Ayer. Ayer has not
yet adopted a comprehensive WHPP.
Even though the travel time between Spectacle Pond and PW-5 is approximately 50 days,
significant changes in surface-water quality associated with land-use activities will occur. It is
anticipated that these changes will be gradual because of ground-water mixing and surface-water
dilution. To evaluate surface-water quality changes, the inflow (SW-1) and outflow (SW-2) regions of
Spectacle Pond will be monitored regularly (Figure 3-34). To evaluate impacts from historic and
potentially chronic sources, sediment samples will be collected from the bottom of each lobe of
Spectacle Pond. The parameters of interest for analysis of the sediment samples are heavy metals and
PCBs.
MONITORING PARAMETERS AND FREQUENCIES
Ambient-trend, source-assessment, and early-warning monitoring objectives were integrated in
the long-term monitoring strategy. Ambient trend monitoring is done to determine regional water
quality variations and changes over time. Source assessment monitoring is done to evaluate
source-specific water quality degradation. Early warning monitoring is done to decrease potential
risks for public exposure to ground-water contamination within Zone II. Each goal will be met by
analyzing samples from monitoring stations for target parameters at scheduled frequencies.
A complete list of chemical and physical monitoring parameters for the WHPA is given in
Table 3-7. Chemical parameters are divided into general water quality (GWQ) and site-specific
parameters. General water quality parameters are included to monitor long-term, basin-wide
fluctuations in the hydrochemistry. Site-specifc parameters constitute a short list of indicator
3-80
-------
TABLE 3-7. LIST OF CHEMICAL AND PHYSICAL MONITORING PARAMETERS FOR
THE PW-5 WELLHEAD PROTECTION AREA, LITTLETON, MASSACHUSETTS.
Chemical Parameters
General Water Quality
Alkalinity (total)
Ammonia
Bicarbonate
Calcium
Chloride
Hardness
Iron
Magnesium
Manganese
Nitrate-nitrogen
Nitrite-nitrogen
Potassium
Sodium
Total dissolved solids (TDS)
Site Specific
Microscopic particulate analysis
(MPA)
Total coliform (TC)
Heavy metals (HM)
Volatile organic compounds
(VOCs)
Pesticides (Pest)
Herbicides (Herb)
Heavy Metals
Arsenic
Barium
Cadmium
Chromium
Copper
Lead
Mercury
Selenium
Silver
Physical Parameters
Water Level (WL)
Specific Conductivity (Cond)
Temperature (Temp)
Abbreviation: PW-5 = Production Well Number 5.
chemicals that are likely to be associated with prioritized sources. Physical tests will be conducted at
all monitoring stations for water quality and flow-field characterization. Details of the long-term
monitoring program for the PW-5 WHPA are presented in Table 3-8. Sample frequencies determined
by perceived risks of contamination from upgradient land uses and travel times are also listed in
Table 3-8.
Ground-Water Stations
Compliance monitoring within Zone III will continue to be performed biannually (BA).
Samples are collected in the spring and fall when the regional water table is typically highest, and
lowest, respectively. Because the compliance monitoring wells are located around known or suspected
contamination sources (Figure 3-34), site-specific criteria and general water quality testing criteria are
required. The number and location of compliance monitoring wells are directly related to land-use
activities within Zone III.
3-81
-------
•a
6
o
8
I
1
S.
1
3
3
3
ec
3
3
3
3
n
cm
as
1
I
o
§
S
va •
M
r
|S
O *•*
I"
«
E
M
E2
s
o
3
3
3
3
P2
3
3
3
3
3
3
3
l§ I
!•§»
3-82
£ If <
ila
AT °*
—< ^
n II II
I-" 8, a
*T3 !J- I
2 .g
ill
^
•1
-------
The primary objective for all monitoring activities in Zone II is early warning detection of
ground-water and surface-water contamination. Source-assessment and ambient-trend monitoring
goals are also met by sampling ground water. The period of advanced warning for each monitoring
station is dependent on travel times based on modeling results. Although MW-Els and MW-Eld are
equidistant from PW-5, ground-water travel times from each well to PW-5 are 150 days and 300 days,
respectively (Figure 3-34). The variation in ground-water velocity at depth indicates that shallow
ground-water contamination will arrive at PW-5 in less time than along deeper ground-water
flowpaths. Contaminant transport rates are also controlled by the chemical properties of the
contamination plume. Ground-water travel times are calculated conservatively so that no retardation
effects are taken into account For a known chemical plume, the retardation effects can be considered
in determining more accurate travel times.
Production Well Number 5 will be sampled at appropriate frequencies to comply with SDWA
requirements. The mandated SDWA parameters conform to the combined general- and site-specific
water quality parameters listed in Table 3-5. Discharge water from PW-5 represents a composite of
the water quality detected among the ground-water and surface-water monitoring stations.
Surface-Water Stations
To meet state drinking water compliance criteria, total coliform (TC) bacteria tests are
performed monthly at PW-5. Coliform bacteria comprise a specific group of organisms, including
aerobic and faculative anaerobic bacteria, both gram-negative rods from fecal and non-fecal origins.
As an indicator bacteria of sanitary quality, the fecal group includes the organisms present in the gut
and feces of warm-blooded animals. Because the indicator group may become stressed or injured in
natural waters, a substantial portion (from 10% to 90%) are incapable of growth under standard
laboratory conditions and may not be detected (McFeters et al., 1986). Because the presence of
coliform bacteria may infer surface-water intrusion and sanitary contamination, the absence of this
group is only indicative of sanitary quality and cannot, by itself, be used to indicate the absence of
surface-water influence.
The use of heterotrophic bacteria plate counts and microscopic particulate analyses (MPA)
may provide more useful information in assessing the transport mechanism for microbiological
contamination originating from Spectacle Pond. It is recommended that heterotrophic plate counts,
in conjunction with total coliform analyses, be performed monthly, and MPAs be performed
biannually. Heterotrophic plate counts provide an estimate of the total numbers of live heterotrophic
bacteria in the sample. Heterotrophic bacteria constitutes the major bacterial group and include
organisms that derive their nutrition from both living and dead organic (carbon-based) material.
Microscopic particulate analyses are used to identify Giardia cysts, Cryptosporidium oocyst,
diatoms, rotifers, coccidia, and insect parts exceeding 1 micrometer in size (Rapacz and Stephens,
1991). These organisms as a group, or in part, are found in surface-water ecosystems, and their
presence is a good indicator of surface-water influences. In addition, this method can be used to
enumerate green, blue-green, and other chloroplast-containing algae, which are further indicators of
surface-water influence. Recently, EPA accepted the MPA method for determining the natural
filtration capabilities of sand and gravel aquifers that receive recharge from surface water. Results
from two separate MPAs performed on water from Spectacle Pond and PW-5 demonstrate that the
sand and gravel deposits are effectively inhibiting intrusion of surface-water micro-organisms.
3-83
-------
Until more information is available regarding transport rates and dilution effects within
Spectacle Pond, surface-water testing parameters will include general and site-specific water quality
groups. Sampling will be performed biannually for general water quality parameters. Because of the
buffering capacity of Spectacle Pond, the site-specific parameters will be tested on an annual basis.
The detection of any site-specific parameters will initiate more thorough source assessment auditing
and the installation of ground-water monitoring devices, if the source is identified.
QUALITY ASSURANCE AND QUALITY CONTROL
Water quality sampling for wellhead protection will be conducted in strict adherence to the
following guidelines. Ground-water monitoring wells will be purged, at a minimum, three times the
standing volume of water, or until temperature and conductivity have stabilized. Cross contamination
will be avoided or minimized by placing a dedicated, inertial pump, composed of small-diameter
(0.625-inch OD) polyethylene flexible-tubing coupled to a Teflon foot valve within each well. To
minimize quantitative" errors attributed to sample collection and analysis, a QA/QC plan has been
developed. Storage, handling, and analytical procedures will be conducted in accordance with standard
EPA guidelines. Emphasis will be placed on appropriate well development, sample filtration, and
preservation techniques. Chemical analyses will be performed at state-certified laboratories that
adhere to standard analytical protocols, including chain-of-custody documentation, percent recovery
and surrogate spike analysis, and use of trip and laboratory blanks.
MONITORING DATA BASE STORAGE, UPDATE, AND RETRIEVAL SYSTEMS
A data base management system is required to store, update, and retrieve information
associated with source audits and ground-water monitoring. Information collected from these
activities is currently in hard copy in a filing system. A number of computer software programs are
available to provide a more effective and efficient means of data storage, update, and retrieval.
Current options range from standard, off-the-shelf, data base management packages to highly
technical geographic information systems (GISs). The associated costs and complexities of each vary
widely.
The appropriate management system will provide many functional benefits to persons involved
in WHPA management By design, the system should eliminate tedious data entry processes, while
providing ease in data retrieval, manipulation, and evaluation. Communication with analytical
laboratories using computer modems would allow the transfer of results electronically. This mode of
transfer will reduce processing time and data-entry errors. For any computer function, it is important
to consider the specific applications needed to perform management tasks. Sophisticated software and
hardware often require matched technical expertise. Whatever system is selected, the technical abilities
of the WHPP staff and the constraints of the project budget must be considered.
3-84
-------
CONTINGENCY PLANNING
Historically, ground-water contamination threats in Littleton have been addressed as they
occur, with the level of response commensurate to the perceived threat. A catastrophic event such as
the accidental release of a large volume of hazardous material initiates an emergency response
involving many departments and agencies. Responses are coordinated through the Littleton Fire
Chief and Water Department The detection of ground-water contamination at downgradient
monitoring wells at facilities initiates a more methodical, site-specific investigation and remediation.
Catastrophic releases initiate a spill-response plan involving the Police, Fire, and Water
Departments (Figure 3-35). Generally, the notification of a catastrophic release is made to the Fire or
Police Department. If the incident occurs near a water supply well, the spill-response team contacts
the Water Department immediately. Containment and diversion methods are implemented first, and
where appropriate, additional support is sought from the Highway Department, Civil Defense
Authority, or a professional incidence-response contractor. In addition to these measures, the Water
Department shuts down the affected production well and utilizes other wells outside the area of
concern to meet water demands. The well is not put into production again until the threat abates.
Li the event of contamination of PW-5 or another production well, Littleton's contingency
plan contains an element for new ground-water supply development The town's future well site is
located in the southwest corner of the Beaver Brook Watershed (Figure 3-36). The proposed well site
has been approved by the State, and protection Zones I, II and III have been delineated. A pumping
station has not yet been constructed at the proposed well.
The adjacent town of Boxborough shares the recharge area to the proposed well and its
protection zones. Boxborough and Littleton formed an environmental support program in 1986 to
protect their shared resources. The towns will implement effective protection strategies together,
including development screening, source regulation, and ground-water monitoring.
Environmental damage can be avoided or minimized through contingency planning and
action. The primary objective of the spill-response team is to contain or divert released hazardous
substance(s) from the water supply well, under the direction of the Water Department Hay bales, oil
booms, absorptive pads, leak repair kits and other spill containment equipment are stored in
emergency vehicles and commonly utilized by the response teams.
The refined capture zone for PW-5 and the ground-water flowpaths and travel times generated
by modeling provide spill-response coordinators with baseline information to better assess
contamination risks posed to PW-5. The necessary response actions are dictated by the concentration,
chemistry, and location of the identified release with respect to the water supply well.
Priority contaminants include those parameters detected above state or federal primary
drinking water standards that have been identified within Zone II. If contamination is identified, a
pumping management strategy would be designed to reduce the likelihood of drawing the contaminant
toward the production well. Concurrently, resampling, plume mapping, site-specific monitoring, and
site-inspection activities, would be initiated. These activities would be followed by remediation.
3-85
-------
Z ui
t? UJ
go
H S
i£
H °
« P
L«^HP«H~*V
•«•••
7
b
a.
tn
m
j_
£
I —
u.
U
UJ
u
£
IL
UJ
tr
E:
UJ
0)
o
%
Ul
LL
a
O
°,
(0
LHBH^
DC
i
§£
Ofn
°£
|i!§
0 °
M ^
05 nj
3 1—
Q %
a
z
^^^
ui
O)
z
u
D.
in
DC
Q
ec
h-
*
CO
-1
g
3
—
-
ENVIRONMENTAL
CONSULTANT
IARDOUS
VASTE
SPONSE
2 *
Q Q
og
z*
UJ(J
§Q
Q.
O.
0
_,|
O— u.
a
t-
KE
£t
11
0
a
HIGHWAY
DEPARTMENT
I
1
lfe.
J<§
d &•
S 1
« a
II
II
vi t/5
§53
1"
3 ^
S
i
3-86
-------
PROPOSED
WELL
LITTLETON
LEGEND
AQUIFER RECHARGE
PROTECTION AREA
—TOWN BOUNDARY
»• STREAM
1234 5
=551=!= '
SCALE (miles)
Figure 3-36. Proposed well site and protection zones, Bennetts Brook watershed,
Littleton, Massachusetts.
3-87
-------
An alternative pumping regime to redirect the plume migration away from the production well could
be engineered using capture wells to alter the regional hydraulic gradient. This action could greatly
reduce or even eliminate migration of the contamination toward the well.
Nonpriority contaminants are characterized as low-level contamination issues within Zone II,
or elevated concentration levels within Zone IIL In both cases, monitoring well(s) in which
contamination is detected are resampled to ensure the integrity of the original test. If positively
confirmed, a source assessment investigation is initiated. The intent of the investigation is to
determine the source and volume of the contamination. In addition to detailed source audits, ground-
water monitoring frequencies may be increased to allow sufficient early warning for the production
well.
3-88
-------
CONCLUSIONS
In 1980, the Littleton Water Department began researching and implementing a
comprehensive aquifer and watershed protection program for the town of Littleton, Massachusetts,
after the bordering town of Acton lost 70% of their municipal ground-water supply capacity due to
contamination from an industrial facility. This incident initiated development of Littleton's existing
wellhead protection program (WHPP) that now serves as a model for state and local agencies. In a
continuing effort to improve the WHPP, the Water Department delineated a new capture zone for
Production Well Number 5 (PW-5) and designed an optimized monitoring program for its long-term
protection. Ground-water monitoring has played an integral role in the overall success of Littleton's
program.
Several tasks were performed to develop an optimized ground-water and surface-water
monitoring program for the new WHPA for PW-5 in the Bennetts Brook watershed. Existing
hydrogeologic reports were used to develop a conceptual hydrogeologic model accounting for aquifer
properties, boundaries, flow dynamics, and water quality. To validate the conceptual flow model,
short- and long-term aquifer testing, ground-water flow simulations, and water quality analyses were
performed. A thorough source assessment was conducted to augment the water quality
characterization.
Regional geomorphology and aquifer testing demonstrate that the Spectacle Pond aquifer is
generally an unconfined, buried valley system. The regional water-table depth varies from 0 to 25 feet
and ground water flows northeast, parallel to the strike of the fault system within the Bennetts Brook
watershed. Most drawdown and recovery curves generated during the aquifer tests were matched with
Neuman type curves established for unconfined aquifers. Exceptions to the unconfined ground-water
response better approximated the Hantush type curve for leaky aquifers. The locations of the
monitoring stations that exhibit this response coincide with semi-confining peat deposits identified at
the perimeter and along the bottom of Spectacle Pond, immediately adjacent to PW-5.
Locally, the glacial outwash deposits have very high transmissivity values of greater than
4,000 feet per day (ft2/d) and high potential well yields of greater than 300 gallons per minute (gpm).
Zones of lower transmissivity (less than 1,350 ft2/d) are encountered along the perimeter of the buried
valleys, where the outwash materials are poorly sorted and the aquifer pinches-out into till and
bedrock uplands. Results from the short-term aquifer test indicate horizontal hydraulic conductivities
ranging from 127 to 2,220 ft/d, with an average value of 560 ft/d. The ratio of horizontal to vertical
hydraulic conductivities (the anisotropy ratio) was estimated to be 50:1.
Interpretations of the long-term aquifer test revealed a downward hydraulic gradient between
PW-5 and Spectacle Pond. As PW-5 was pumped, the hydraulic gradient increased until it finally
reached stabilization. The presence of the vertical hydraulic gradient, confirmed by a vertical
temperature gradient, demonstrates that the pond is recharging the aquifer when PW-5 is in
3-89
-------
operation. The long-term test also demonstrated that the northwest portion of the aquifer receives
approximately five to six times more water from the pond than from ground-water recharge. This
translates to approximately 20% to 25% of the total recharge to PW-5. Throughout the rest of the
aquifer, responses were similar to those observed during the short-term test. Again, hydraulic
conductivities ranged from 300 to 3,600 ft/d with an average value of 520 ft/d. Regionally, the
prevailing ground-water flow component is horizontal, with an anisotropy ratio of approximately 35:1.
The average specific yield value is approximately 0.11, which is typical for an unconfined aquifer.
The capture zone for PW-5 was delineated by combining results from several ground-water
flow models:
• FLOWPATH, a steady-state, two dimensional (2-D), horizontal plane, ground-water
flow and pathline analysis model
• MODFLOW, a steady-state or transient, three-dimensional (3-D) model which utilizes
a finite-difference numerical method to solve 3-D flow equations in combination with
MODPATH
• MODPATH, a pathline analyses post processor for MODFLOW
• FLOWCAD, an extension of FLOWPATH, used to simulate transient hydraulic head
distributions required by the Massachusetts Department of Environmental Protection
Depending on the modeling objective, each model can be used effectively for delineating WHPAs.
FLOWPATH provided the regional (far-field) approximation of steady-state flow dynamics
for the Spectacle Pond aquifer. MODFLOW was used to address near-field, vertical flow components
and variations in aquifer characteristics (hydraulic conductivity and depth) in proximity to PW-5.
FLOWCAD was employed to generate the transient, Zone II boundary, as required under current
state WHPA delineation regulations. Each simulation utilized a time-related criterion to define the
zone of contribution, also referred to as the time-related capture zone. A second and less
sophisticated criterion, drawdown, generated inappropriate protection results because of uncorrected
regional ground-water flow gradients.
Baseline ground-water monitoring was done to characterize water quality for long-term
program planning. Monitoring goals were differentiated into ambient trend and source assessment
activities. These categories were further sub-divided by analyses of general water quality parameters,
site-specific water quality parameters, and a combination of both parameters.
The water quality of the Spectacle Pond aquifer is slightly acidic and very fresh: a Type I Water.
Sodium and calcium are the most abundant cations in the ground water; bicarbonate is the most
abundant anion. Trilinear Piper diagrams were used to differentiate two hydrochemical fades within
the study area. Above a depth of 25 feet, a sodium-bicarbonate water type predominates; a
calcium-bicarbonate water type predominates below 25 feet.
The land uses identified during the source assessment include underground fuel storage tanks,
metal plating shops, a municipal landfill, agricultural operations, industrial and commercial
developments, and residential subdivisions. Collectively, these sources pose potential ground-water
contamination threats of heavy metals, VOCs, pesticides, and nutrients. Results of source-assessment
monitoring demonstrate that, to date, ground-water quality within the capture zone of PW-5 has been
3-90
-------
virtually unaffected by source operations. Sodium is the only exception. Moderate sodium
concentrations detected within the surface water and shallow aquifer are probably related to highway
de-icing activities. Despite these otherwise favorable analyses, naturally occurring manganese and iron
were detected at elevated levels in proximity to PW-5.
To maintain the drinking water quality of PW-5, the current WHPA will be modified to
include the new Zone I, II, and III criteria determined through the ground-water modeling effort.
Zone I is the 14-acre parcel (400-foot protective radius) adjacent to PW-5, owned by the Littleton
Water Department. Future land use in this area will be restricted to open space, exclusively. Zone II
is established as the union of the capture zones generated using the 3-D steady-state (MODFLOW)
criteria and the 2-D transient (FLOWCAD) criteria required by the state. Ideally, total land-use
restrictions would also be proposed within Zone II. However, because of existing zoning regulations
and associated property values, this option is impractical. Instead, moderate- and high-risk land-use
activities will be prohibited, while other land-use activities such as residential development and existing
commercial operations will be allowed. All activities within Zone II will continue to be closely
monitored through ground-water analysis and periodic site inspections, where appropriate.
Zone III includes the upgradient portion of the Bennetts Brook watershed, which contributes
recharge to Zone II and PW-5. Protection measures within Zone III will be less restrictive. Land-use
regulations, coupled with source assessment monitoring, will provide sufficient protection against
regional ground-water quality degradation. Although effective land-use controls already exist as
bylaws, the newly-delineated protection zones provide a template of ground-water and surface-water
flow dynamics necessary to design an optimized monitoring program.
A long-term, water quality monitoring program is proposed to provide an accurate and
cost-effective means of ensuring protection of PW-5 in the future. Monitoring sites are strategically
located to provide early warning detection within Zone II and compliance monitoring within Zone
III. The use of ground-water flow paths and travel times was critical in designing the long-term
monitoring network for the WHPA of PW-5. To allow for sufficient early-warning response time, a
minimum 300-day TOT criterion has been established for all new monitoring locations. This criterion
will allow biannual sampling events to be conducted, while providing sufficient response time between
initial detection of ground-water contamination and the execution of necessary contingency measures.
Flowpath studies have demonstrated the importance of surface-water monitoring in the overall
WHPA monitoring program. Despite Littleton's commitment to wellhead protection, it can neither
eliminate nor regulate ground-water or surface-water contamination originating from Zone III areas
that fall within the jurisdiction of neighboring towns. Fortunately, regional flowpath analyses indicate
that scheduled monitoring of Spectacle Pond will provide an effective means of monitoring these
politically unaccessible areas. Land-use activities in bordering towns are generally unregulated.
Hydrogeologic boundaries do not often correspond with geopolitical boundaries. Inevitably,
the mode of implementation and technical sophistication of WHPPs from one municipality to the
next will vary considerably. Mutual water quality issues need to be identified and cooperation fostered
toward resolution of these issues, so that a cooperative relationship can be attained. In most
instances, the exchange of hydrogeologic evaluations will greatly enhance the accuracy and
effectiveness of a regional WHPP. To address the present WHPP discrepancies between Littleton and
its neighboring communities, educational outreach programs will be initiated. Specifically, the benefits
3-91
-------
of ground-water protection will be discussed among appropriate town boards and planning
committees. If neighboring communities fail to recognize the benefits of such a program, state or
federal assistance may be sought. Regardless of the approach, the need for the implementation of an
effective, joint WHPP should be emphasized.
3-92
-------
REFERENCES
Achinger, W. C and R. T. Shigehara. 1968. A Guide for Selected Sampling Methods for Different
Source Conditions. Journal of the Air Pollution Control Association 18(9): 605-609.
Bray D., A H. Cathcart, and S. C. Danos. 1990. Nutrient Chemistry of the Spectacle Pond Well,
Littleton, Massachusetts. Open File Report. Littleton Water Department, Littleton, Massachusetts.
Cadwgan, R. M. and J. C. Swallow. 1991. Microwell Installations: Advanced Technology for Ground
Water Monitoring. Paper Presentation. New England Water Works Association Annual Conference,
Mystic, Connecticut.
Cooper, H. H. and C. E. Jacob. 1946. A Generalized Graphical Method for Evaluating Formation
Constants and Summarizing Well Field History. American Geophysical Union Transactions 27: 526-
534.
D. L. Maher Company. 1985. A Report on the Prolonged Pump Test Conducted in May 1985 at the
Spectacle Pond Well Field, Ayer, Massachusetts. D. L. Maher Company, North Reading,
Massachusetts.
Delaney, D. F. and A Maevsky. 1980. Distribution of Aquifers, Liquid-Waste Impoundments, and
Municipal Water-Supply Sources, Massachusetts. U.S. Geological Survey, Water-Resources
Investigation, Open-File Report 80-431. U.S. Government Printing Office, Washington, D.C. de
Marsily, G. 1986. Quantitative Hydrogeology. Academic Press, Inc., San Diego, California, pp. 284-
337.
Duffield, G. M. and J. O. Rumbaugh. 1989. AQTESOLV: Aquifer Test Solver. User's Manual.
Geraghty & Miller Modeling Group, Reston, Virginia.
Franz, T. and N. Guiger. 1992. FLOWCAD, Transient Two-Dimensional Aquifer Analysis Program.
User's Manual. Waterloo Hydrology Software, Waterloo, Ontario, Canada.
Franz, T. and N. Guiguer. 1989. FLOWPATH, Two-dimensional Horizontal Aquifer Simulation
Model. User's Manual. Waterloo Hydrogeologic Software, Waterloo, Ontario, Canada. 72 pp.
Freeze, R. A. and J. A Cherry. 1979. Groundwater. Prentice-Hall, Inc., Englewood Cliffs, New
Jersey. 588 pp.
Frimpter, M. H., J. J. Donahue, and M V. Rapacz. 1988. A Mass-Balance Nitrate Model for
Predicting the Effects of Land Use on Groundwater Quality in Municipal Wellhead Protection Ares.
3-93
-------
Cape Cod Aquifer Management Project Document. Cape Cod Planning and Economic Development
Commission, Barnstable, Massachusetts.
GHR Engineering Associates, Inc. 1989. Patriot Energy Project, Ayer, Massachusetts, Baseline
Hydrogeological Evaluation. GHR Engineering Associates, Inc., Lakeview, Massachusetts.
Goldberg-Zoino & Associates, Inc. 1987. Geohydrologic Study of the Littleton Corporate Center,
Littleton, Massachusetts. File No. V-10122. Goldberg-Zoino & Associates, Inc., Newton Upper Falls,
Massachusetts.
Haley and Ward, Inc. 1981. Prolonged Pump Test at the Vartanian Well Site. Haley and Ward, Inc.,
Waltham, Massachusetts.
Hantush, M. S. 1964. Hydraulics of Wells. In: Advances in Hydroscience. V. T. Chow (Editor).
Academic Press, Inc., New York, New York. pp. 281-432.
Hem, J. D. 1989. Study and Interpretation of the Chemical Characteristics of Natural Water. U.S.
Geological Survey Water Supply Paper 2254, Edition 3. U.S. Government Printing Office,
Washington, D.C 263 pp.
Huibregtse, K R., and J. H. Moser. 1976. Handbook for Sampling and Sample Preservation of
Water and Wastewater. EPA-600/4-76-049. U.S. Environmental Protection Agency, Office of
Research and Development, Cincinnati, Ohio. 257 pp.
Jahns, R. H. 1953. Geologic Map of the Ayer Quadrangle, Massachusetts. U.S. Geological Survey,
Denver, Colorado.
Kick, J. R. 1980. Gravity Investigations, Littleton, Massachusetts. Littleton Water Department,
Littleton, Massachusetts.
Kruseman, G. P. and N. A. de Ridder. 1990. Analysis and Evaluation of Pumping Test Data.
Edition 2. International Institute for Land Reclamation and Improvement, Wageningen, The
Netherlands.
Lapham, W. W. 1989. Use of Temperature Profiles Beneath Streams to Determine Rates of Vertical
Ground-Water Flow and Vertical Hydraulic Conductivity. U.S. Geological Survey Water-Supply
Paper 2337. U.S. Government Printing Office, Washington, D.C. 35 pp.
Massachusetts Department of Environmental Protection. 1991. Guidelines and Policies for Public
Water Systems. 1991 Edition. Massachusetts Department of Environmental Protection, Boston,
Massachusetts. 174 pp.
McDonald, M. G. and A. W. Harbaugh. 1988. A Molecular Three-Dimensional Finite-Difference
Ground-Water Flow Model. U.S. Geological Survey Techniques of Water Resources Investigations,
Book 6, Chapter Al. U.S. Government Printing Office, Washington, D.C. 586 pp.
3-94
-------
McFeters, G. A., J. S. Kippin, and M. W. Le Chevalliers. 1986. Injured Coliforms in Drinking Water.
Applied Environmental Microbiology 51: 1.
Metcalf & Eddy, Inc. 1981. Groundwater Protection Plan, Town of Littleton, Massachusetts. Metcalf
& Eddy, Inc., Boston, Massachusetts.
Moore, B. A., A. H. Cathcart, and S. C. Danos. 1990. Wellhead Protection Area Delineation and
Monitoring Strategy, Littleton, Massachusetts. In: Minimizing Risk to the Hydrologic Environment,
Proceedings of American Institute of Hydrology Conference, Las Vegas, Nevada. A. Zaporozec
(Editor). American Institute of Hydrology, Minneapolis, Minnesota, pp. 182-197.
Neuman, S. P. 1974. Effect of Partial Penetration on Flow in Unconfined Aquifers Considering
Delayed Gravity Response. Water Resources Research 10: 303-312.
Pollock, D. W. 1989. Documentation of Computer Programs to Compute and Display Pathlines
Using Results from the U.S.G.S. Modular Three-Dimensional Finite-Difference Ground-Water Flow
Model. U.S. Geological Survey Open File Report 89-381. U.S. Government Printing Office,
Washington, D.C. 188pp.
Rannie, E. H. and R. L. Nadon. 1988. An Inexpensive, Multi-Use, Dedicated Pump for Groundwater
Monitoring Wells. Groundwater Monitoring Review 8(4): 100-107.
Rapacz, M. V. and H. C. Stephens. 1991. Public Groundwater Supply Filtration: Necessary or Not?
In: Ground Water Management Journal 4: 419-432, Proceedings of the Eastern Ground Water Issues
Conference, Portland, Maine. National Water Well Association, Dublin, Ohio.
Rosenfeld, J. K. 1990. Ground-Water Contamination at Hazardous Waste Disposal Facilities.
Ground Water Management 1: 237-250, Proceedings of the Cluster of Conferences, Kansas City,
Missouri. National Water Well Association, Dublin, Ohio.
Roy, S. P., P. Milner, D. D'Amore, J. Drake, and E. C. White. 1984. Groundwater Monitoring
Handbook. Massachusetts Department of Environmental Protection, Boston, Massachusetts. 143 pp.
Scalf, M. R., J. F. McNabb, W. J. Dunlap, R. L. Cosby, and J. Fryberger. 1981. Manual of Ground-
Water Sampling Procedures. National Water Well Association, Dublin, Ohio. 93 pp.
Silvey, W. D. and H. E. Johnston. 1977. Preliminary Study of Sources and Processes of Enrichment
of Manganese in Water from University of Rhode Island Supply Wells. U.S. Geological Survey Open-
File Report 77-561. U.S. Government Printing Office, Washington, D.C. 33 pp.
Theis, C. V. 1935. The Relation Between the Lowering of the Piezometric Surface and the Rate and
Duration of Discharge of a Well Using Groundwater Storage. American Geophysical Union
Transactions 16: 519-524.
Todd, D. K, R. M. Tinlin, K D. Schmidt, and L. D. Everett. 1976. Monitoring Groundwater Quality:
Monitoring Methodology. EPA-600/4-76-026. U.S. Environmental Protection Agency,
Environmental Monitoring Systems Laboratory, Las Vegas, Nevada. 154 pp.
3-95
-------
Town of Littleton Planning Board. 1988. Aquifer and Water Resources Overlay District Bylaw.
Littleton Water Department, Littleton, Massachusetts. 4 pp.
Town of Littleton Planning Board. 1981. Toxic and Hazardous Materials Bylaw. Littleton Water
Department, Littleton, Massachusetts. 3 pp.
U.S. Environmental Protection Agency. 1990a. A Review of Sources of Ground-Water
Contamination From Light Industry. EPA 440/6-90-005. U.S. Environmental Protection Agency,
Office of Ground-Water Protection, Washington, D.C 48 pp.
US. Environmental Protection Agency. 1987. Guidelines for Delineation of Wellhead Protection
Areas. EPA 440/6-87-010. U.S. Environmental Protection Agency, Office of Ground-Water
Protection, Washington, D.C
U.S. Environmental Protection Agency Region 1. 1989. Protecting Ground Water from the Bottom
Up: Local Responses to Wellhead Protection. U.S. Environmental Protection Agency Region 1,
Boston, Massachusetts. 327pp.
U.S. Geological Survey. 1985. Hydrology and Water Resources of Tributary Basins to the Merrimack
River from Salmon Brook to the Concord River, Massachusetts. Hydrologic Atlas 662. U.S.
Geological Survey, Denver, Colorado.
Walton, W. C 1987. Groundwater Pumping Tests — Design and Analysis. Lewis Publishers, Inc.,
New York, New York. 201pp.
Wetzel, R. G. 1975. Limnology. W. B. Saunders Company, Philadelphia, Pennsylvania, pp. 66-68.
Zen, E-an. 1983. Bedrock Geologic Map of Massachusetts. (Editor). U.S. Geological Survey,
Denver, Colorado.
3-96
-------
ADDITIONAL BIBLIOGRAPHY
Alter L, T. W. Bennett, G. Hackett, R. J. Petty, J. H. Lehr, H. Sedoris, D. M. Nielsen, and J. E.
Denne. 1989. Handbook of Suggested Practices for the Design and Installation of Ground-Water
Monitoring Wells. EPA/600/4-89/034. National Water Well Association, Dublin, Ohio. 398 pp.
Berner, E. K and R. A. Berner. 1987. The Global Water Cycle. Prentice-Hall, Inc., Englewood
Cliffs, New Jersey. 383pp.
Cathcart, A. H. 1990. Hydrologic Investigation of North-Central Massachusetts. Open File Report.
Littleton Water Department, Littleton, Massachusetts.
Canter, L. W., R. C. Knox, and D. M. Fairchild. 1987. Groundwater Quality Protection. Lewis
Publishers, Inc., Chelsea, Michigan. 562 pp.
Davis, S. N. 1988. Where Are the Rest of the Analyses? Ground Water 26(1): 2-5.
Dunne, T. and L. B. Leopold. 1978. Water in Environmental Planning. Freeman and Company,
New York, New York. 818 pp.
Hounslow, A. W. 1988. Contemporary Interpretation of Water Quality Data, A Practical
Geochemical Approach Using Personal Computers. In: Groundwater Modeling Workshop. U.S.
Environmental Protection Agency Region 1, Boston, Massachusetts. 65 pp.
Livingstone, D. A. 1963. Chemical Composition of Rivers and Lakes. Chapter G. In: Data of
Geochemistry. M. Fleischer (Editor). U.S. Geological Survey Professional Paper 440-F, Edition 6.
U.S. Government Printing Office, Washington, D.C. 64 pp. ,
Morrissey, D. J. 1987. Estimation of the Recharge Area Contributing Water to a Pumping Well in a
Glacial-Drift, River-Valley Aquifer. U.S. Geological Survey Open-File Report 86-543. U.S.
Government Printing Office, Washington, D.C. 60 pp.
Pfannkuch, H. 1982. Problems of Monitoring Network Design to Detect Unanticipated
Contamination. Groundwater Monitoring Review 2(1): 67-76.
U.S. Environmental Protection Agency. 1990b. Guide to Ground-Water Supply Contingency
Planning for Local and State Governments. EPA 440/6-90-003. U.S. Environmental Protection
Agency, Office of Ground-Water Protection, Washington, D.C. 83 pp.
U.S. Environmental Protection Agency. 1975. Minimal Requirements for a Water Quality Assurance
Program. EPA 440/9-75-010. U.S. Environmental Protection Agency, Washington, D.C.
3-97
-------
White, D. R, J. D. Hem, and G. A. Waring. 1963. Chemical Composition of Subsurface Waters.
Chapter F. In: Data of Geochemistry. M. Fleischer (Editor). U.S. Geological Survey Professional
Paper 44-F, Edition 6. U.S. Government Printing Office, Washington, D.C. 67 pp.
3-98
-------
EPA-600/R-93/
APRIL 1993
CHAPTER 4
WELLHEAD PROTECTION AND MONITORING OPTIONS FOR
THE SIOUX FALLS AIRPORT WELLFIELD, SOUTH DAKOTA
by
Assad Barari, Derric L. lies, and Tim C. Cowman
South Dakota Geological Survey
South Dakota Department of Environment and Natural Resources
University of South Dakota
Vermillion, South Dakota 57069
Beth A Moore
Lockheed Environmental Systems & Technologies Company
Las Vegas, Nevada 89119
April 1993
Contract Number CR-816204-01-0
Project Officer
Steven P. Gardner
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas, Nevada 89193-3478
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89193-3478
-------
NOTICE
This report is the result of research supported by the U.S. Environmental Protection Agency's
Environmental Monitoring Systems Laboratory in Las Vegas, Nevada, as part of its efforts to provide
technical assistance to state, tribal, and local governments on the implementation of the Wellhead
Protection Program. The specific methods and approaches contained in this document have
undergone peer review but do not constitute official Agency endorsement or policy recommendations.
The Office of Research and Development provides this information to help solve complex technical
problems related to refined delineation and ground-water monitoring of wellhead protection areas in
various hydrogeologic settings. Further assistance is available from the Environmental Monitoring
Systems Laboratory in Las Vegas, from the Office of Ground Water and Drinking Water in
Washington, D.C, and from the ground-water offices in the ten Environmental Protection Agency
regions.
4-ii
-------
ABSTRACT
The city of Sioux Falls is located in the southeast corner of South Dakota. The Big Sioux
aquifer is the primary source of water for about 125,000 persons in the Sioux Falls metropolitan area.
One of the municipal wells in the Big Sioux aquifer, the airport wellfield, is underlain by surficial,
glacial outwash deposits. The Big Sioux River is located directly west of the airport wellfield and flows
south over and through the outwash, draining approximately 4,000 square miles of upstream land.
The city's wells pump most of their water directly from the aquifer and a small quantity from
the Big Sioux River. However, the river is hydraulically connected to the aquifer, and recharge from
the river in the airport wellfield area is significant. In 1988, approximately 79% of the recharge to the
airport wellfield aquifer was induced from the river due to wells pumping. Induced flow from the river
to the aquifer is demonstrated by decreased flow in the river during low recharge periods.
This research was conducted to evaluate (1) the hydraulic connection between the Big Sioux
River and the adjacent aquifer, and (2) the potential impact of the river on aquifer water quality. In
the broader perspective, additional goals included refined delineation of the wellfield protection area
and design of a long-term water quality monitoring program.
Drilling logs indicate that the thickness of the aquifer in the wellfield area ranges from 20 to 50
feet. Aquifer testing results yield an average hydraulic conductivity value of 800 feet per day and a
transmissivity value of approximately 21,000 square feet per day for the aquifer.
Many potential point sources of contamination exist in the study area. These include: industrial
and commercial areas, the South Dakota Air National Guard facility, a petroleum pipeline, the Sioux
Falls Regional Airport, and a decommissioned municipal landfill. The threat of contamination from
these sources is underscored by the recent history of contaminant releases in the area.
To estimate ground-water travel times in the study area, aquifer testing, dye tracing, and ground-
water modeling were employed. During aquifer testing, two dye injections were made. The first dye
was injected in a well approximately 40 feet north of the pumping well. Detectable dye concentrations
first arrived at the pumped well after about 12 hours. The second dye was injected in a well near the
edge of the river, approximately 140 feet north of the pumping well. Detectable dye concentrations
from the second injection site first arrived at the pumped well in 7 to 9 days. Aquifer testing and dye-
tracing results indicate that a contaminant could travel from the river to the wellfield in less than 9
days.
A two-dimensional, steady-state model (FLOWPATH) was used to generate time-related
capture zones for the municipal wells and to simulate contaminant travel times. One-, two-, and five-
year capture zones were calculated for each of the municipal wells in the airport wellfield. Modeling
4-iii
-------
of simulated spill sites from several of the potential point-source contamination areas indicates that
contaminants entering the aquifer at areas to the north and south of the well field could reach the
municipal wells in 1 to 2 years.
The City of Sioux Falls and Minnehaha County have delineated wellhead protection areas by
using the hydrogeologic-mapping method. Wellhead protection ordinances are designed to impose
guidelines and restrictions on new land uses, or proposed changes in existing use, in order to protect
the aquifer water quality.
A wellhead protection monitoring program at the airport wellfield is proposed to document
ambient water quality conditions and to serve as an early-warning detection system. Line-source
monitoring is proposed to monitor the Big Sioux aquifer and the diversion canal for contaminants
that could potentially enter the aquifer. Point-source and nonpoint-source monitoring are proposed
to monitor water quality between the airport wellfield and potential sources. The categories of
parameters for monitoring are general water quality, volatile organic compounds, trace metals,
pesticides, and nutrients. Sampling frequencies for each of the categories were selected as a function
of the type of source to be monitored.
Contingency planning is warranted to establish emergency responses to contaminant releases at
the surface of the aquifer and in the river. Alternative water supply development must also be
continued as part of the contingency planning effort.
4-iv
-------
CONTENTS
Abstract . iii
Figures vii
Tables xi
Abbreviations, Symbols, and Conversion Factors xiii
Acknowledgments ; , xvi
Background for the Sioux Falls Case Study 4-1
Introduction 4_1
Wellhead Protection Program Overview 4-4
Program Inception and Development 4-4
Wellhead Protection Program Objectives 4-4
Implementation of Wellhead Protection 4-5
Hydrogeologic Setting 4.5
Preliminary Wellhead Protection Area 4-8
Source Assessment 4-10
Line Sources 4-10
Point Sources 4-10
Nonpoint Sources 4.13
References for Source Identification 4-13
Research Monitoring Program 4-15
Data Base Limitations 4-15
Monitoring Objectives 4-16
Research Monitoring Tasks 4-16
Data Acquisition 4-17
Monitoring Wells 4-17
Water Quality Sampling 4-25
Water-Level Measurement 4-28
Aquifer Test 4-30
Dye-Trace Test 4-30
Data Interpretation 4.34
Water Levels 4.34
Water Quality and Chemistry 4.44
Aquifer-Test Analysis 4-52
Dye-Trace Analysis 4-58
Ground-Water Flow Model and Flowpath Assessment 4-65
Refined Wellhead Protection Area 4-78
Wellhead Protection Monitoring Program 4-80
Objectives 4-80
4-v
-------
CONTENTS, Continued
Monitoring Types and General Locations
Line-Source Monitoring
Point-Source Monitoring :
Nonpoint-Source Monitoring
Future Studies
Proposed Monitoring Wells
Monitoring Parameters and Frequencies
Quality Assurance and Quality Control Considerations
Monitoring Data Base Storage, Update, and Retrieval
Contingency Planning
Conclusions
Recommendations
References
4-80
4-80
4-81
4-84
4-85
4-85
4-86
4-87
4-88
4-88
4-91
4-95
4-97
4-vi
-------
FIGURES
Number
Page
4-1 Regional location of the city of Sioux Falls within the Big Sioux aquifer and
Minnehaha County, South Dakota 4-2
4-2 Location of the Sioux Falls airport wellfield and study area, South Dakota 4-3
4-3 Hydrogeologic cross section near Sioux Falls municipal well 34A,
South Dakota 4.7
4-4 Water Source Protection Overlay Districts in and near the Sioux Falls
airport wellfield, South Dakota 4.9
4-5 Areas of potential point- and nonpoint-source ground-water contamination
by land-use type, Sioux Falls airport wellfield, South Dakota 4-11
4-6 Locations of monitoring wells and surface-water monitoring points,
Sioux Falls airport wellfield, South Dakota 4-18
4-7 Locations of monitoring wells near municipal well 34A,
Sioux Falls airport wellfield, South Dakota 4-19
4-8 Generalized diagram of vertical monitoring well construction,
Sioux Falls airport wellfield, South Dakota 4-23
4-9 Generalized diagram of horizontal monitoring well construction,
Sioux Falls airport wellfield, South Dakota 4-24
4-10 Water-table elevations for the study area on August 2, 1990,
Sioux Falls airport wellfield, South Dakota ,. 4-35
4-11 Water-table elevations for the study area on August 17, 1990,
Sioux Falls airport wellfield, South Dakota 4-36
4-12 Water-table elevations near municipal well 34A under pumping conditions on
August 2, 1990, Sioux Falls airport wellfield, South Dakota 4-37
4-13 Water-table elevations near municipal well 34A under static conditions on
August 17, 1990, Sioux Falls airport wellfield, South Dakota 4-38
4-vii
-------
FIGURES., Coritinued
Number Page
4-14 Hydrograph of the vertical well in the river, May 16 through 23,1990,
Sioux Falls airport wellfield, South Dakota :..._' . 4-40
4-15 Hydrograph of well R20-89-113, May 10 through June 19, 1990,
Sioux Falls airport wellfield, South Dakota 4-41
4-16 Hydrograph of well R20-89-119, May 15 through June 8,1990,
Sioux Falls airport wellfield, South Dakota 4-42
4-17 Hydrograph of well R20-89-127, May 15 through June 8,1990,
Sioux Falls airport wellfield, South Dakota 4-43
4-18 Trilinear diagram of major cations and anions from January 1990 water samples,
Sioux Falls airport wellfield, South Dakota 4-49
4-19 Trilinear diagram of major cations and anions from April 1990 water samples,
Sioux Falls airport wellfield, South Dakota 4-50
4-20 Trilinear diagram of major cations and anions from July 1990 water samples,
Sioux Falls airport wellfield, South Dakota 4-51
4-21 Water-table elevations near municipal well 34A prior to aquifer testing on
May 15,1990, Sioux Falls airport wellfield, South Dakota 4-54
4-22 Theis analysis of the logarithmic plot of time versus drawdown data from
well R20-89-113, Sioux Falls airport wellfield, South Dakota 4-56
4-23 Cooper-Jacob analysis of drawdown data from well R20-89-113 versus the
logarithmic plot of time, Sioux Falls airport wellfield, South Dakota 4-57
4-24 Dye concentration versus time for samples analyzed onsite from
municipal well 34A, Sioux Falls airport wellfield, South Dakota 4-59
4-25 Dye concentration versus time for samples analyzed by the Ozark Underground
Laboratory from municipal well 34A, Sioux Falls airport wellfield,
South Dakota : 4-60
4-viii
-------
FIGURES, Continued
Number
Page
4-26 Dye concentration versus time from charcoal packs analyzed by the Ozark Underground
Laboratory from municipal well 34A, Sioux Falls airport wellfield,
South Dakota .".. 4-61
4-27 Dye concentration divided by the square root of the sampling interval versus time,
for charcoal packs analyzed by the Ozark Underground Laboratory from municipal
well 34A, Sioux Falls airport wellfield, South Dakota 4-64
4-28 Variable-spaced grid for the ground-water flow model, Sioux Falls
airport wellfield, South Dakota 4-66
4-29 Water-table configuration map generated by the calibrated ground-water flow model,
Sioux Falls airport wellfield, South Dakota 4-68
4-30 One-year TOT capture zones at a porosity of 15% for the Sioux Falls
airport wellfield, South Dakota 4-70
4-31 One-year TOT capture zones at a porosity of 25% for the Sioux Falls
airport wellfield, South Dakota 4-71
4-32 Two-year TOT capture zones at a porosity of 15% for the Sioux Falls
airport wellfield, South Dakota 4-72
4-33 Two-year TOT capture zones at a porosity of 25% for the Sioux Falls
airport wellfield, South Dakota 4-73
4-34 Five-year TOT capture zones at a porosity of 15% for the Sioux Falls
airport wellfield, South Dakota 4.74
4-35 Five-year TOT capture zones at a porosity of 25% for the Sioux Falls
airport wellfield, South Dakota 4.75
4-36 Simulated contaminant flowpaths from discharge sites for a 2-year TOT,
Sioux Falls airport wellfield, South Dakota 4-77
4-37 Recommended monitoring locations for line, point, and nonpoint sources,
Sioux Falls airport wellfield, South Dakota 4-82
4-ix
-------
-------
TABLES
Number Page
4-1 Common Sources of Ground-Water Contamination in Industrial and Commercial
Areas Near the Sioux Falls Airport Wellfield, South Dakota 4-12
4-2 Monitoring Well Locations and Associated Elevation Data,
Sioux Falls Airport Wellfield, South Dakota 4-20
4-3 Inorganic Parameters Analyzed in Water Samples,
Sioux Falls Airport Wellfield, South Dakota 4-27
4-4 Pesticides and Volatile Organic Compounds Analyzed in Water Samples,
Sioux Falls Airport Wellfield, South Dakota , 4-29
4-5 Water Elevations at Ground-Water and Surface-Water Monitoring Points,
August 2 and 17, 1990, Sioux Falls Airport Wellfield, South Dakota 4-39
4-6 Inorganic Analytes for Water Samples, Sioux Falls Airport Wellfield,
South Dakota 4.45
4-7 Pesticides Detected in January and July 1990 Water Samples,
Sioux Falls Airport Wellfield, South Dakota 4-52
4-8 Aquifer Test Summary Results, Sioux Falls Airport Wellfield, South Dakota ...... 4-58
4-9 Categories of Monitoring Parameters and Frequencies,
Sioux Falls Airport Wellfield, South Dakota 4-85
4-xi
-------
-------
ABBREVIATIONS, SYMBOLS, AND CONVERSION FACTORS
ABBREVIATIONS
acre-ft acre-foot
amsl above mean sea level
ASTM American Society for Testing and Materials
d day
DNAPL dense, nonaqueous phase liquid
DOT Department of Transportation
EMSL-LV Environmental Monitoring Systems Laboratory, Las Vegas, Nevada
EPA U.S. Environmental Protection Agency
ft foot
ft/d foot per day
ftVd square foot per day
ft^/min square foot per minute
GIS Geographic Information System
gpm gallon per minute
hr hour
HW horizontal well
in inch
in/yr inch per year
1 liter
mg milligram
mg/1 milligram per liter
Mgd million gallon per day
min minute
ml milliliter
MW monitoring well
|ig/l microgram per liter
liS/cm microSiemen per centimeter
N/A not applicable
ND none detected
PA phenolphthalein alkalinity
psi pound per square inch
PVC polyvinyl chloride
QA/QC quality assurance and quality control
R49W Range 49 West
SC specific conductivity
SCH schedule
SDDENR South Dakota Department of Environment and Natural Resources
4-xiii
-------
SDDOT South Dakota Department of Transportation
SDGS South Dakota Geological Survey
SFE Sioux Falls Elmwood Landfill
T101N Township 101 North
TA total alkalinity
TDS total dissolved solid
TOT time of travel
USGS United States Geological Survey
VOC volatile organic compound
VRW vertical well in the river
WHPA wellhead protection area
WHPP Wellhead Protection Program
ZOC zone of contribution
SYMBOLS
2,4-D
As
Ca
a
C03
Eh
F
Fe
HCO3
K
Mg
Mn
Na
N
NO2
N03
Se
SO<
u
Q
r
S
s
T
t
W
2,4-dichlorophenoxy acetic acid
arsenic
calcium
chloride
carbonate
oxidation-reduction potential
fluoride
iron
bicarbonate
potassium
magnesium
manganese
sodium
nitrogen
nitrite
nitrate
selenium
sulfate
well function
pumping rate
radial distance from pumping well to monitoring well
storativity
drawdown
transmissivity
time
well function
4-xiv
-------
CONVERSION FACTORS
Multiply
acre-foot
cubic foot per second
foot
foot per second .
gallon
gallon
gallon
gallon per day
gallon per day per foot
gallon per day per square foot
inch
inch per year
mile
million gallons. per day
square foot per minute
square foot per second
square mile
9 o
5
By
1230
0.0283
0.3048
0.3048
3.785
0.134
0.00379
0.000003528
0.000207
0.0408
0.0254
25.4
1.609
2.629
0.0929
0.0929
2.59
To Obtain
cubic meter
cubic meter per second
meter
meter per second
liter
cubic foot
cubic meter
cubic foot per second
square meter per day
meter per day
meter
millimeter per year
kilometer
cubic meter per minute
square meter per minute
square meter per second
square kilometer
4-xv
-------
ACKNOWLEDGMENTS
This research was funded through the U.S. Environmental Protection Agency (EPA),
Environmental Monitoring Systems Laboratory at Las Vegas, Nevada (EMSL-LV), under Contract
Number CR-816204-01-0 to the South Dakota Geological Survey (SDGS), Vermillion, South Dakota,
a division of the South Dakota Department of Environment and Natural Resources. We gratefully
acknowledge the support of Steven Gardner (EMSL-LV) who served as the EPA Project Officer.
The assistance of several organizations was instrumental in coordinating and completing this research.
The following are recognized for their contributions: the-city of Sioux Falls, the Ozark Underground
Laboratory, Waterloo Hydrogeologic Software, the U.S. EPA Region VIII, the East Dakota Water
Development District, and Anderson's Truck Stop.
For their careful review of the manuscript and insightful comments, we thank the following
technical reviewers: Dr. Charles Kreitler (University of Arizona), Douglas Heath (U.S. EPA Region
I), and Mike Wireman (U.S. EPA Region VIII). Stan Pence and Patricia Hammond of the SDGS
assisted in hydrologic field work and data reduction. John Nicholson and Carolyn Cameron of
Lockheed Environmental Systems & Technologies Company (LESAT) provided technical writing and
editing support for preparation of the manuscript. Lone Hansen and Dennis Johnson of the SDGS,
as well as Steve Garcia (LESAT) contributed excellent graphics to the report. Shalena Fendzlau
(LESAT) and Colleen Odenbrett (SDGS) are graciously acknowledged for their patience and
expertise on word processing support.
4-xvi
-------
BACKGROUND FOR THE SIOUX FALLS CASE STUDY
INTRODUCTION
In support of the Wellhead Protection Program (WHPP) for the city of Sioux Falls, South
Dakota (Figure 4-1), the South Dakota Geological Survey (SDGS) conducted research in cooperation
with the U.S. Environmental Protection Agency (EPA) Environmental Monitoring Systems
Laboratory in Las Vegas, Nevada (EMSL-LV), to design a long-term monitoring program for the
Sioux Falls municipal (airport) wellfield. The research focuses on characterization of the hydraulic
relationship between the Big Sioux River and the unconfined aquifer providing potable water to the
city of Sioux Falls (Figure 4-2). Potential sources of aquifer contamination in proximity to the
wellfield and the river as a line source were assessed. The study area is located in the upper midwest
region of the United States in the city of Sioux Falls, Minnehaha County, South Dakota (Figure 4-1).
The region is semiarid, and the average annual precipitation is approximately 26 inches. The
evapotranspiration rate is approximately 36 inches per year.
Ground water is the primary source of water used for municipal, domestic, and agricultural
purposes in the region. The predominant aquifer in the area is the Big Sioux aquifer (Figure 4-1).
The use of other aquifers in the area as water supply sources for the city is limited because of inferior
water quality, insufficient water quantity, or distance of the aquifer from the city of Sioux Falls. The
Big Sioux aquifer provides most of the water used by the city. It is the source of drinking water for
about 125,000 persons in the Sioux Falls metropolitan area. In 1988, only about 7% of the water used
by the city was pumped directly from the Big Sioux River; 93% was provided by the aquifer. Recently,
however, the city has increased the amount of water taken from the river after installation of a 44
million-gallon-per-day (Mgd) surface-water intake structure.
The Big Sioux aquifer is composed of surficial outwash deposits that typify the Big Sioux
River valley. The Big Sioux River flows southward over and through the outwash, draining
approximately 4,000 square miles of land upstream from the Sioux Falls municipal wellfield. One of
the Sioux Falls municipal wellfields in the Big Sioux aquifer is the focus of study for this research~the
airport wellfield (Figure 4-2). The airport wellfield consists of 21 of the city's 51 water supply wells.
The land upstream from the Sioux Falls municipal wellfield is used generally for agricultural
purposes and is heavily irrigated in some areas. Several cities upstream from Sioux Falls may also
have an impact on river water quality. From a hydrologic perspective, the Big Sioux River acts as a
central drain for all potential sources of pollution, including surface runoff in the drainage basin and
all point and nonpoint sources of pollution overlying the Big Sioux aquifer.
4-1
-------
MISSOURI RIVER
BIG SIOUX
AQUIFER
^.PIERRE
RAPID CITY
50
100
MILES
BIG SIOUX
SIOUX- FALLS
RIVER
Figure 4-1. Regional location of the city of Sioux Falls within the Big Sioux aquifer and
Minnehaha County, South Dakota.
Discharge from the municipal wells in the airport wellfield induces river water flow into the
adjacent aquifer because of good hydraulic interconnection and because discharge from pumping
exceeds natural recharge rates to the aquifer in the wellfield area. Surface-water inducement is
extensive and is demonstrated by the cessation of river flow in the vicinity of the municipal wellfield
during low-flow periods. Thus, any significant degradation of river water quality will impact the
quality of water pumped from the Sioux Falls municipal wellfield. Therefore, Sioux Falls should be
cognizant of potential pollutants that may enter the river and the aquifer upstream and upgradient
from the city wellfield.
The importance of recharge induced from the river into the aquifer cannot be overstated. In
1988, an estimated 79% of recharge to the airport wellfield was from the river. Although the amount
of recharge from the river varies for different years because of variations in the wellfield pumping rate
and the river flow, this percentage reflects the importance of the river recharge component.
Surface water generally moves at a much faster rate than ground water because of its
unrestricted flow at the surface, as opposed to flow through porous media. Consequently,
contaminants introduced upstream have the potential to reach the wellfield in a relatively short time
4-2
-------
R 50 W R 49 W
2 Miles
27«
0 1000 3000 5000 7000 Feet
Big Sioux aquifer boundary
Area underlain by the Sioux Falls management unit of the Big Sioux aquifer
(Aquifer boundaries from Ellis et al.. 1969)
General area of study for wellhead protection monitoring research
Sioux Falls municipal well and city identification number
Surface water
Intermittent stream
Surface—water intake structure
Map Location
Figure 4-2. Location of the Sioux Falls airport wellfield and study area, South Dakota.
4-3
-------
period. It is very difficult to delineate an effective time-related wellhead protection area (WHPA) for
a specific time period (for example, 1 month) without considering the impact of the river. Therefore,
protection of water quality in the river must'be included in an effective WHPP for the city.
In addition to the threat of contamination from the Big Sioux River, there are several
potential point-source areas of contamination around the wellfield. In particular, contamination has
been documented for the industrial and commercial areas south and east of the wellfield, as recorded
in the South Dakota Department of Environment and Natural Resources (SDDENR) files. Nonpoint
sources are also present in the vicinity of the wellfield. These sources must also be considered in the
development of an effective wellhead protection program.
WELLHEAD PROTECTION PROGRAM OVERVIEW
Program Inception and Development
The 1986 amendments to the Federal Safe Drinking Water Act (SDWA) require states to
develop and implement protection measures for municipal and community ground-water sources. As
a result of this Act and the increasing awareness of the susceptibility of the Big Sioux aquifer to
contamination, the Sioux Falls City Commission adopted multijurisdictional ordinances in April 1990
to establish Water Source Protection Overlay Districts. Minnehaha County adopted a similar
ordinance in September 1990.
City ordinances were developed in a complementary fashion with the county ordinance to
provide the desired areal coverage. Both city and county ordinances were needed to provide
protection because some municipal water supply wells located outside the city boundaries fall under
county jurisdiction. Both city and county ordinances are applicable to parts of the study area.
Wellhead Protection Program Objectives
The overall objective of the WHPP developed by Sioux Falls and Minnehaha County is to
establish a comprehensive water resource protection plan that will prevent and reduce risks to the
public water supply. Resources to be protected include the Big Sioux aquifer system, the Skunk Creek
aquifer system, and associated drainage basins (City Ordinance 15.43.088). The ordinance establishes
procedures and criteria for reviewing and restricting land uses having the potential to pollute water
sources in designated Water Source Protection Overlay Districts.
The intent of the county's efforts is essentially the same as that of the city (County Ordinance
MC16-90). Restrictions apply to land-use activities having the potential to contaminate aquifers and
wellhead sites currently in use and those of potential future use as a public water supply. In addition,
a Water Source Protection Overlay District has been formed. The overlay district is superimposed on
all districts established by the county ordinance. If the Water Source Protection Overlay District
imposes a greater restriction than the underlying zoning district regulations, the more stringent
regulations are to be followed.
4-4
-------
Implementation of Wellhead Protection . -
The WHPP is implemented through use of the Water Source Protection Overlay Districts
(zoning ordinances). The city's ordinances apply within municipal boundaries; the county's ordinance
applies outside municipal boundaries. However, the zoning authority in Minnehaha County allows for
a joint-jurisdictional area 3 miles outside the municipal boundaries, in which both the city and county
ordinances apply. The authority of the city versus that of the county is clear in the joint-jurisdictional
area because ordinances were developed to complement one another.
The city's ordinances apply if new uses or changes in existing uses are proposed for a property
within the Water Source Protection Overlay District. A Screening Committee composed of the City
Planning Director, the City Public Health Director, and the Water and Wastewater Manager must
review the proposed new or changed use prior to its onset. Residential land uses that are connected to
the municipal sanitary sewer system are exempt from the screening process. At a minimum, the
following information is considered:
• Type of new or changed use, development, or operation
• Quality, quantity, and concentration of a contaminant or hazardous material, and its
physical, chemical, or infectious characteristics
• Storage or use of contaminants or hazardous materials
• Leak or spill prevention methods
• Contingency plans
• Adverse effects of contaminants or hazardous materials, including the probability that
any leak or spill would contaminate ground water
• Compliance with other city, state, or federal regulations
The review by the Screening Committee can result in a recommendation of approval or denial
to the City Planning Commission. Or the proposed new or changed use can be referred by the
Screening Committee to another review process by an Environmental Impact Committee.
The Environmental Impact Committee must review all proposed new or changed uses that are
not exempted from further review by the Screening Committee. Members of the Environmental
Impact Committee are appointed by the mayor and city commissioners, and include representatives
from the Health, Planning and Zoning, Engineering, Fire, Attorney, and Utilities Departments.
Factors considered in this review are, at a minimum, the same as those examined by the Screening
Committee. The Environmental Impact Committee will then recommend approval or denial of the
proposed use to the City Planning Commission.
The county's ordinance becomes a factor if new uses or changes in existing uses are proposed
for property within the Water Source Protection Overlay District outside the municipal boundary. A
conditional use permit must be obtained from the County Planning Office. The county review process
considers factors similar to the city review process.
By addressing only new uses or changes in use of property, the city and county are permitting
reasonable and environmentally responsible growth, yet imposing reasonable and adequate safeguards
on land use within the Water Source Protection Overlay Districts. All existing businesses and
activities are allowed to continue in their present form if in conformance with existing federal, state,
4-5
-------
and municipal regulations. Thus, any existing use of property within the Water Source Protection
Overlay District that poses a potential threat to the municipal water supply, in spite of existing
regulations, will continue to be a threat under the enacted ordinances. Any existing threats must be
dealt with by enforcement of existing regulations pertaining to the storage, use, and disposal of
potential contaminants and by monitoring as part of the wellhead protection efforts.
HYDROGEOLOGIC SETTING
The Sioux Falls airport wellfield is underlain by surficial glacial outwash deposits (sand and
gravel) that are part of the Big Sioux aquifer. This aquifer occurs primarily in the Big Sioux River
valley and some of its tributaries. The areal extent of the aquifer is about 770 square miles, including
parts of northeastern South Dakota and extending to the southeastern corner of the state (Figure 4-1).
The Sioux Falls airport wellfield lies within the Sioux Falls management unit of the Big Sioux aquifer.
This management unit occurs in Minnehaha County and encompasses about 34 square miles. Ground
water in this part of the aquifer occurs generally under unconfined conditions.
Outwash deposits associated with the Big Sioux aquifer occur primarily at or near land surface
and are commonly mantled by only a thin layer of soil or fine-grained, alluvial sediments. In the study
area, lithologic logs indicate that the aquifer is overlain by up to 12 feet of alluvial clay (average
thickness, 6.7 feet). The approximate thickness of the outwash in the wellfield area ranges from 20 to
50 feet. The greatest thickness of saturated outwash is reported to be 48 feet near the weir on the
diversion canal (Koch, 1982). Ground-water levels in the study area are about 10 feet below land
surface. In the study area, the Big Sioux River is incised into the Big Sioux aquifer, but does not fully
penetrate the aquifer.
The hydrogeologic relationship between the aquifer and the river in proximity to municipal
well 34A is shown in Figure 4-3. There is direct contact between the Big Sioux River and the aquifer
under generally unconfined conditions, even though the sand and gravel are overlain by a layer of
alluvial clay. The hydraulic relationship between the river and the aquifer depicted in Figure 4-3 is
representative of the entire study area. However, the thickness and extent of silt and clay on the river
bottom and sides are not depicted because their occurrence is uncertain.
Recharge to the Big Sioux aquifer is from the infiltration of precipitation and from the
inducement of river water. In calibrating a digital ground-water model for a large area of the Big
Sioux aquifer, Koch (1982) assumed 6.9 inches per year (in/yr) of recharge from precipitation. Hedges
et al. (1985) estimated the natural recharge rate to be 6.5 in/yr. Using a value of 6.5 for natural
recharge and the actual pumping rate from the Sioux Falls management unit, Barari et al. (1989)
showed that, on average, approximately 59% of the recharge to the entire management unit was from
precipitation and 41% was due to induced recharge from the river. Because a clay layer overlies the
aquifer in the wellfield, it is expected that this section of the aquifer receives less than 6.5 in/yr of
natural recharge. Considering this situation, 6 in/yr of recharge was assumed for this research.
The significance of recharge from the Big Sioux river to the aquifer can be demonstrated. The
area of the aquifer bounded by the diversion canal to the north and east, the Big Sioux River to the
west, and the aquifer boundary to the south (Figure 4-2), is 4.3 square miles. If an annual recharge
rate of 6 inches is assumed, then this part of the aquifer receives approximately 1,400 acre-feet
4-6
-------
(ISOID
NOI1VA313
4-7
-------
(acre-ft) of water from precipitation annually. However, in 1988, the city pumped approximately 6,700
acre-ft (2.2 x 109 gallons) of water from the airport wellfield, which was about 32% of the total water
pumped by the city. No continuous decline in water levels was observed in this wellfield in 1988 or in
1989; therefore, recharge appears to have been equal to the discharge. In order to balance the water
budget, it is assumed that this portion of the aquifer received an additional 5,900 acre-ft of water from
sources other than the direct infiltration of precipitation. While some of this additional water may
have come from portions of the aquifer outside of the 4.3-square-mile area, most probably came from
the Big Sioux River and the diversion canal. Because the source of water in the drainage canal is the
Big Sioux River, it was concluded that induced recharge from the river was 3.8 times the natural
recharge in this portion of the aquifer. Consequently, in 1988, approximately 79% of recharge to the
airport wellfield was from the river. This high recharge percentage demonstrates the significant
contribution of the Big Sioux River to this wellfield. Loss of water from the river to the shallow
ground-water system is also demonstrated by the fact that flow in the river has been observed to cease
in the vicinity of the municipal wellfield during periods of low flow. The most recent occurrence of
this was in July 1988.
Data on stream bed infiltration rates were presented by Jorgensen and Ackroyd (1973). On
the basis of three aquifer tests, they estimated the infiltration rate to range from 0.5 to 1.0 foot per day
(ft/d). The infiltration rate varies depending on the presence of silts and clays, which accumulate
preferentially on the upstream side from the dam and weir across the Big Sioux River and the
diversion canal.
PRELIMINARY WELLHEAD PROTECTION AREA
The WHPAs, or Water Source Protection Overlay Districts defined in the city and county
ordinances, were delineated using the criterion of aquifer boundary and the method of hydrogeologic
mapping (U.S. EPA, 1987). Specifically, the ordinances (City Ordinance 15.43.088; County Ordinance
MC16-90) define an overlay district as "a geographical area overlying a geologic formation, group of
formations, or part of a formation capable of yielding, storing, or transmitting a usable amount of
ground water to wells or springs for domestic or animal use." This definition is appropriate for the
city of Sioux Falls, where the aquifer occurs at, or very near, land surface. The Big Sioux River is
included in the Water Source Protection Overlay District because of its presence within the described
geographic area. One objective of this research is to better define the river-aquifer relationship and to
determine its effect on the preliminary WHPA.
The boundaries of the Water Source Protection Overlay Districts within the study area do not
correspond exactly to the geologic boundaries of the surficial outwash deposits shown in Figure 4-3.
Rather, overlay district boundaries correspond to slightly larger political boundaries, which include the
geologic features of concern. Most of the aquifer within the study area is included within the overlay
district boundaries, with the exceptions of small portions of Sections 17 and 31 (Figure 4-4). For a
3-mile distance beyond the municipal boundaries, the overlay district falls under joint jurisdiction
between the city and the county. The overlay district authorized by the county (not shown on
Figure 4-4), which is outside the 3-mile joint-jurisdictional area follows geologic boundaries rather
than political boundaries.
4-8
-------
R 50 W R 49 W
1
2 Miles
0 1000 3000 5000 7000 Feet
Big Sioux aquifer boundary
City of Sioux Falls overlay district
Minnehaha County and City of Sioux Falls joint jurisdictional overlay district
Surface water
Intermittent stream
*J
Figure 4-4. Water Source Protection Overlay Districts in and near the Sioux Falls
airport wellfield, South Dakota.
4-9
-------
SOURCE ASSESSMENT
Line Sources
From a hydrologic perspective, the Big Sioux River acts as a central drain for all known and
potential sources of pollution in the aquifer, including surface runoff in the drainage basin and
discharges to the river. The relationship of potential line sources of contamination to the airport
wellfield is one important focus of this research. Both the Big Sioux River and the diversion canal act
as potential line sources of contamination.
The Big Sioux River drains agricultural land on which fertilizers and pesticides are commonly
applied; these chemicals are potential contaminants. Accidental chemical spills or illegal dumping of
hazardous and toxic substances into the river are other types of potential pollution. Because of
significant river infiltration to the aquifer, production wells at the airport wellfield are directly
impacted by river water quality. As a result, the city of Sioux Falls needs to be aware of potential
pollution that may enter the river upstream from the city.
Point Sources
Despite the presence of many potential point sources of contamination in proximity to the
airport wellfield, the level of detail for a complete source assessment of each entity is not currently
available. The threat of contamination from these sources is evident from the recent history of
contaminant releases in the area. Several chemical releases involving both underground and above-
ground leaks have occurred and are on record with several agencies including the Sioux Falls Fire
Department, the local Emergency Planning Commission, and the SDDENR. The potential point
sources adjacent to the airport wellfield (Figure 4-5) indicate the varied contaminants present:
Industrial and commercial areas
The South Dakota Air National Guard facility
A petroleum pipeline belonging to the Williams Pipe Line Company
The Sioux Falls Regional Airport
A sand and salt storage area
A decommissioned, municipal landfill
A fire training center
The industrial and commercial areas south of the airport have existed for at least 40 years;
those to the east of the diversion canal have existed for 15 to 20 years. Potential sources of ground-
water contamination within the commercial and industrial areas in the vicinity of the airport wellfield
are listed by facility type in Table 4-1. Many underground and aboveground storage tanks are located
in these areas and contain petroleum-based products. A complete listing of the types of chemicals
used at these facilities is beyond the scope of this research. However, for the types of facilities listed in
Table 4-1, research completed by Rosenfeld (1990) indicates that volatile organic compounds (VOCs)
are likely to be potential, predominant contaminants. In addition, the SDDENR has documented
inorganic contamination of near-surface soils (primarily heavy metals) at one industrial facility.
4-10
-------
R 50 W I R 49 W
1
2 Miles [
*^—•
A
3000 5000 7000 Feet
Big Sioux aquifer boundary
Municipal well
Surface water
Intermittent stream
Williams pipeline
Fire training-center
Location
q Industriot and commercial areas A
\VH Sioux Falls Regional Airport . ^
,:: ,'j South Dakota Air National Guard
-~-J Agricultural land
* + * J Former Sioux Falls Elmwood Landfill
t—±_J
\\\ Elmwood Golf Course
i I I
•!-t.' .1. Residential areas
!. L ' 1 1
@ South Dakota Department of
^ Transportation sand and salt stockpile
Figure 4-5. Areas of potential point- and nonpoint-source ground-water contamination by
land-use type, Sioux Falls airport wellfield, South Dakota.
4-11
-------
TABLE 4-1. COMMON SOURCES OF GROUND-WATER CONTAMINATION IN
INDUSTRIAL AND COMMERCIAL AREAS NEAR THE SIOUX FALLS
AIRPORT WELLFIELD, SOUTH DAKOTA*
Commercial
Industrial
Agricultural equipment dealer
Airport
Auto repair shops
Car washes
Cemeteries
Construction areas
Gas stations
Golf courses (chemical applications)
Laundromats
Lawn care services
Machine and metalworking shops
Paint shops
Photography and printing shops
Road de-icing operations
Road maintenance depots
Toxic and hazardous spills
Wells, operating and abandoned (such as water
supply, injection, monitoring, and exploration)
Chemical manufacture, warehousing and
distribution activities
Electrical and electronic products and
manufacturing
Electroplaters and metal fabricators
Fiberglass and plastic ware manufacturing
Fire retardant and foam products
manufacturing
Jewelry and metal plating
Manufacturing and distribution sites for
cleaning supplies
Petroleum products production, storage, and
distribution centers
Pipelines (such as oil and gas)
Quartzite quarry
Railroad tracks and yards
Storage tanks and pipes (both aboveground
and underground)
Toxic and hazardous spills
* Modified after Eschner et al, 1991.
The South Dakota Air National Guard facility, adjacent to the south edge of the airport
(Figure 4-5), has been in operation since September 1946. Potential contaminants in use at this
facility include aviation fuel (JP4), automobile fuels (diesel and unleaded gasolines), solvents, and
degreasers. Trichloroethylene is a common constituent in the solvents and degreasers, and
perchloroethylene is contained in a commonly-used cleaner.
The Williams Pipe Line Company owns and operates an 8-inch-diameter pipeline used solely
to transport petroleum products. In the past, some agricultural chemicals were transported through
this pipeline; however, that practice was discontinued. The pipeline crosses the aquifer and the Big
Sioux River above the dam (Figure 4-5) and has been in operation since about 1946. It is buried
about 3 to 5 feet below land surface, but is above ground where it crosses the Big Sioux River
upstream from the airport wellfield and the city's surface-water intake structures. About 1 Mgd of
gasoline, diesel fuel, fuel oil, and aviation fuel are currently transported through this line.
In January 1992, a small leak in the Williams pipeline was detected when petroleum product
was found in the glacial till at a site approximately 3 miles northeast of the center of the study area.
To date, about 170,000 gallons of product has been recovered; the total spill is estimated to range
4-12
-------
between 200,000 to 400,000 gallons. This nearby contamination incident occurred in the glacial till
adjacent to the aquifer, rather than in the aquifer. Nonetheless, it underscores the vulnerability of the
aquifer and the need to implement source management practices to protect the airport wellfield.
The Sioux Falls Regional Airport opened in the mid-1930s and was greatly expanded during
World War II. The primary chemicals utilized there are aviation fuels (Jet-A and 100-octane), runway
de-icer (ethylene glycol), and herbicides [Round-up (mono-isopropylamine salt) and 2,4-D (2,4-
dichlorophenoxy acetic acid)]. The herbicides are used to control weeds and to prevent growth of
herbage around the airport fence.
The sand and salt storage area is owned and operated by the'South Dakota Department of
Transportation (SDDOT) and is located near the west edge of the aquifer (Figure 4-5). A sand and
salt mixture used for road de-icing in the winter has been stockpiled in this location for many years.
Historically, the salt and sand mixture was stored uncovered and unlined. Around. 1987, runoff from
the stockpile area endangered nearby trees because of uptake through the root system. Concurrently,
higher levels of total dissolved solids (TDS) were detected in the ground water.
The Sioux Falls Elmwood Landfill was in operation in proximity to the airport wellfield from
1955 through 1958 (Figure 4-5). At that time, there were few restrictions regarding type and quantity
of materials that could be disposed of in landfills. The operators reportedly accepted a variety of
refuse ranging from municipal solid waste to scrap metal.
The Sioux Falls Fire Department maintains and operates a regional training center about 250
feet east of municipal well 34A (Figure 4-5) for departmental personnel. The main chemicals used at
this facility are aviation fuels, to ignite some of the fires, and a fire-fighting foam (FC-203CE Light
Water Brand Aqueous Film Forming Foam), to extinguish some of the fires.
Nonpoint Sources
Three nonpoint sources of potential contamination include residential areas, agricultural land,
and the Elmwood Golf Course (Figure 4-5). Potential ground-water contaminants in the residential
and agricultural areas are nitrogen-based fertilizers and pesticides used for control of weeds and
insects.
The Elmwood Golf Course now covers the old Elmwood Landfill and includes additional land
(Figure 4-5). The golf course and landfill existed concurrently from 1955 through 1958. This golf
course was opened in 1923 and was redesigned to include more area in 1959 after the landfill was
closed. Chemicals or products known to be applied to the golf course area are nitrogen-based
fertilizers, a fungicide, Round-up, and 2,4-D.
References for Source Identification
A complete source assessment has not yet been completed by the city for the airport wellfield.
However, many of the potential categories of sources discussed in this section were identified from the
following information centers:
4-13
-------
Data from the Sioux Falls Fire Department. The Department maintains a
comprehensive, computerized data base of addresses where hazardous materials are
stored. This data base contains information such as type and quantity of hazardous,
onsite materials.
Records of the Sioux Falls Health Department
Records of the Minnehaha County Civil Defense Office
Aerial photographs at the Minnehaha County Office of Planning and Zoning
Reports of spills and leaks on file at the SDDENR and in the files of local and state
law enforcement departments
Records of injection wells on file at the SDDENR
Records of permitted discharges to either ground water or surface water on file at the
SDDENR
4-14
-------
RESEARCH MONITORING PROGRAM
DATA BASE LIMITATIONS
The types of background information available for this study include lithologic logs; recorded
water levels; laboratory analytical data of general water chemistry; pumping records for municipal
wells; aquifer maps; and regional, digital ground-water models that include the study area. The
following technical reports pertaining to the area of study were consulted: Ellis and Adolphson
(1969), Ellis et al. (1969), lies (1992), Jorgensen and Ackroyd (1973), Koch (1982), Koch (1983),
Lindgren and Niehus (1992), Steece (1959a, 1959b), Tomhave (In Preparation), and Vaughn and
Ackroyd (1968).
Within the aquifer boundaries (Figure 4-2), there are approximately 118 test holes or wells for
which lithologic logs are available. In addition, there are 21 municipal wells within the airport
wellfield. Lithologic logs, water-chemistry data, and water-level data are stored in computer data bases
controlled and maintained by the SDGS and the SDDENR. Other related data are found in reports
and files of the city of Sioux Falls. The number and configuration of existing wells in the study area
did not facilitate aquifer testing and sampling in proximity to the river. For this and other reasons,
the existing monitoring network was considered inadequate for the research objectives.
Water-level records are available from chart recorders for each municipal well in the airport
wellfield. These water levels are accurate to the nearest ±0.1 foot and are recorded relative to mean
sea level. However, the accuracy of measuring-point elevations for some critical wells in the research
monitoring network are questionable. Therefore, measuring-point elevations were resurveyed for
select wells.
The quality of water entering the Sioux Falls water-purification facility is representative of the
quality of water in the aquifer and in the Big Sioux River. However, data are not available for
individual production or monitoring wells that demonstrate the spatial or temporal variability in water
quality in proximity to the airport wellfield.
Pumping records in gallons per month are available for each of the municipal wells in the
wellfield. These records are kept on file at the water-purification plant for the city of Sioux Falls.
More detailed records of well discharges, accurate to gallons per minute, are required for ground-water
flow modeling studies.
Surficial maps and cross sections showing the areal extent and thickness of the aquifer have
been prepared by various investigators (Ellis et al., 1969; Jorgensen and Ackroyd, 1973; Steece, 1959a,
1959b), based on surficial geologic mapping, numerous lithologic logs, and water-level measurements.
These maps are regional in scale and are, therefore, deficient in the detail necessary to meet the
objectives of this site-specific research.
4-15
-------
The resolution and predictive capabilities of existing digital ground-water models for the Sioux
Falls management unit of the Big Sioux aquifer are regional in scale. The regional models are
invaluable in demonstrating the overall interaction between the Big Sioux River and the ground-water
system, but utilize too large a grid spacing to simulate detailed flow patterns within the study area.
MONITORING OBJECTIVES
The specific objectives of the research monitoring program are the following:
• To quantify the contribution of stream infiltration to water pumped from the
municipal wells in the airport wellfield
• To calculate and field-verify the time of travel (TOT) between the Big Sioux River and
municipal well 34A
• To estimate the TOT from the Big Sioux River and the diversion canal to selected
municipal wells, under pumping conditions similar to conditions prevailing in 1988
• To estimate the TOT between selected potential point sources of contamination and
some of the municipal wells under pumping conditions similar to those in 1988
• To address monitoring options (ambient trend, source assessment, and early warning
detection) regarding ground-water contamination from the Big Sioux River, as well as
general point source areas in the study area
• To examine the limitations of current wellhead protection guidelines in areas where
there is extensive interaction between surface water and ground water
• To refine the delineation of the existing WHPP, if warranted
RESEARCH MONITORING TASKS
The following tasks were performed to satisfy the specific objectives of the research
monitoring program:
• Installation and development of monitoring wells
• Surveying the elevations of monitoring wells, the Big Sioux River, and municipal well
34A
• Sampling and analysis for the following constituents:
Inorganic parameters from 10 monitoring wells, biannually, and from nine of
these same wells a third time
Pesticides and VOCs from five monitoring wells, biannually
Inorganic parameters from the Big Sioux River, three times per year
Pesticides and VOCs from the Big Sioux River, biannually
• Measurement of water levels
• Aquifer testing using municipal well 34A as the pumping well
• Dye tracing at selected monitoring wells between the river and pumping municipal
well 34A
• Ground-water flow modeling in the vicinity of the airport
• Ground-water monitoring program design
• Research documentation
4-16
-------
DATA ACQUISITION
Monitoring Wells
New Monitoring Wells—
Vertical wells.-Twenty-one vertical monitoring wells were constructed for this study at the
locations shown on Figures 4-6 (five wells marked R-20-90) and 4-7 [16 wells, excluding 34A, the
warS Woi X 3nd ^ VCrtiCal ^ " ^ riV6r (VRW» Sdected welk ™» "^mon
wS m JnLS ?H Tere US6d t0 .T^f b°th W3ter leVdS 3nd P""***"- Ministry. Water levels
w^monitored to determine spatial and temporal trends, as well as to calibrate the ground-water
Sixteen of the 21 monitoring wells were installed near municipal well 34A (Figure 4-7) These
wells were used to accurately establish the configuration of the water-table surface in 5* immediate
vicinity of the municipal well. All 16 wells were monitored during the aquifer and tracer testeTha
rrdS±? ^ P^P^g municipal well 34A Ground-water elevations were carefully determined,
at a detailed scale, between the Big Sioux River and municipal well 34A.
1 000 feeT™ T^S ^ ^T ~119 "* R2°-89-122 ^UIe 4'A are located approximately
1 000 feet northeast and southwest of municipal well 34A, respectively. These wells were used to
^ " area °f the aquifer that was not antidpated to be affected *
River ne ,1A6' '12?' ^ ~128 (FlgUre 4'7)' 10Cated °n the west side of fl» Big Sioux
River near municipal well 34A, were used to document the gaining-versus-losing condition of the river
^ the wTT7 K 10nS' Additiona»y> these wells ™ used to monitor water-level responses
on the west side during the pumping of municipal well 34A
installed o^ide the
nerit f ICVels and the g^d-water chemistry
near the perimeter of the aquifer. Water-elevation data were needed in these areas to determine
ground-water gradients and flow directions, and to support the modeling effort.
The 21 vertical wells were drilled using the hollow-stem auger method. The outside diameter
entr n ^A ?lug was present in the iead auger to prevent native ^^ *™
entering the hollow auger. Cuttings brought to the surface along the outside of the auger were
examined and described. Lithologic logs and well-construction data are stored in the SDGS
computerized data base. The geographical location and elevations of the top of the casing the
screened interval, and the ground surface for each well are listed in Table 4-2.
Prior to drilling, the work area of the drill rig, augers, and other downhole tools and
SZ^h ™Te,,Clean? with high-pressure hot water. This decontamination procedure was repeated
between the drilling of each hole to minimize the possibility of cross-contamination between drill
4-17
-------
See Figure 4-7
for monitoring
wells in this
area.
R 50 W | R 49 W
2 Miles
C^
R20-90-4
0 1000 3000 5000 7000 Feet
Big Sioux aquifer boundary
Letter indicates surface-water monitoring point. Letters A through D are
surface—water elevation measurement points: letter E represents a surface-
water sampling paint.
Monitoring well and well identification number. 'Wells 'with R20—
prefix installed as part of this study.
Surface water
Intermittent stream , ^
Map Location
Figure 4-6. Locations of monitoring wells and surface-water monitoring points,
Sioux Falls airport wellfield, South Dakota.
4-18
-------
119
122
1 16 Monitoring well; number is last 3 digits
" of the well identifier (R20-89- )
A VRW - Vertical well in the river
HW - Horizontal well
• Municipal well 34A
100
200
300 400 feet
SCALE
Figure 4-7. Locations of monitoring wells near municipal well 34A,
Sioux Falls airport wellfield, South Dakota.
4-19
-------
H
O
X
2
CO
I
ig
IS
co O
CO CO
J7 Q"
Z pd
^^ pt*
grf
O W
^^ J^^
Cj *^
o
g
LE
*
.2
W
1
2
O
g %
•S.I
"8 I
C oj
8 W
co
.3 *
o
£
Longitude
I
,3
1
Soo-<±oir-vpcot--lCS
ON ON
00 OO
I I I I t
ON ON ON ON ON
- - - - oo oo op
00
i
aS
1
o
T-5
T-H
•*
3?
-------
H
g
PH
^
^3
i
to
Q
i
Q
g_
^1
pr"] ^
^-4 O
W v.^
Q ^
H O
oQ
^ ffi
0 o
Z co
if
t-J
d
fxj
^?
1^
i
oS
O
o
H
1
•^
a
3
H
la
1
0>
W
T3
|
0
ned Interval
evations*
qj [ij
CO
OD
•S *
S s
•M!
° 5
•o
I
£
42
*yj
^^.
•C
w
S
88859
t*-
~^_
t*i
CO
od
od
si s
•^ V-) CT
vb
^3- r~ fo ?M
«O O O CS
III!
m
o o
CLJ
fe to
oo oo
op -T
CLl S
fc X
oo W
tn
06
«n
TH 50
S
1
T— t <3
I
o.
ex
CO
—j •>->
"~" 2
85
S-B
cd cd
C3 S
ll
.crt Q
^ «H
•2
4-21
-------
A schematic diagram of the vertical well design is shown in Figure 4-8. The wells were
constructed with 2-inch-diameter, schedule 40, flush-threaded (ASTM F480-88A) polyvinyl chloride
(PVC) casing and screen. All PVC parts were cleaned with high-pressure hot water and were
wrapped in plastic prior to transport to the drill site.
These wells were installed by placing the well screen and casing through the center of the
hollow-stem auger after the hole was drilled to the desired depth and the plug was removed from the
auger. The auger was then removed from the drill hole, and the native sediment (sand and gravel)
was allowed to collapse around the screen. A clean filter pack (washed and sorted; predominantly
quartz sand) was added to the annular space. The top of the filter pack ranged from 0.9 to 2 feet
above the top of the well screen. A layer of granular bentonite ranging in thickness from 0.9 to 2 feet
was placed on top of the filter pack. Neat-cement grout placed on top of the granular bentonite
extended up to the land surface. Concurrent with the placement of the cement grout, a locking steel
well protector was installed over the well casing.
New monitoring wells were developed to ensure that all wells would respond physically and
reproducibly to ground-water fluctuations and that they would yield representative water samples.
New vertical monitoring wells were developed at about 10 gallons per minute (gpm) using a jet pump.
The suction pipe for the pump was equipped with a check valve at the bottom to prevent backflow of
water into the well. The suction pipe and check valve consisted of PVC material. Suction pipe and
check valves were cleaned with high-pressure hot water, and were wrapped prior to transport to the
site. A clean check valve and clean suction pipe were used for each well. The suction pipe was
equipped with a valve to allow suction to be broken at the top of the well in order to purge water
between the well and the point of water discharge. This prevented water that had come in contact
with the pump from returning to the well.
A minimum of about 120 well volumes of water was removed from each well during the
development process. Development continued, if necessary, until the temperature and electrical •
conductivity of the water had stabilized in three successive measurements taken at 5-minute intervals.
Horizontal well-A very shallow, horizontal monitoring well (Figure 4-9) was installed
adjacent to the Big Sioux River (designated as HW in Figure 4-7) to monitor the river water level
without actually placing the monitoring point in the river. However, clays and silts deposited on the
bottom and sides of the river greatly reduced the hydraulic interconnection between the river and the
horizontal well screen, thereby making this well a ground-water, rather than a surface-water,
monitoring point. The horizontal well was, therefore, used as a monitoring point for the ground-
water levels adjacent to the river bed.
The excavation for the horizontal well was made using a backhoe to a depth of about 2 feet
below the river, or water-table level. Deeper excavation was prevented because of caving of the sand
and gravel in the trench. After excavation of the trench, the horizontal well was placed on top of the
native sand and gravel. A filter of sand and gravel was placed over the horizontal portion of the well.
The remaining portion of the trench was filled with excavated material. The well was constructed of
2-inch-diameter galvanized steel with a 1-foot-long manufactured well screen at the end of the well
adjacent to the river.
4-22
-------
Q_
LiJ
10
12
14
16
—
—
T
1
/
/
*,• . *.•'
f . . .f :
•• * -'i:
V*
*'> .'**. '"'',
••' •'*•'• .. .- .
.*• • . •*. -i
* ' V '•*•«
.*•' *'••'
"'** . "4 ' -
.. '"* ' ,; •
>,*.'-* i
.*.;"."• 4 • •
==
=
'=^
/
/
/
:
wn i PAP
VV l_l_L_ \sr\\
>M 1 OPKIKIf; 9TFFI
WELL PROTECTOR
\
.4
/'•
-
» «* •
>.'"••"••'
•r • '. • ' *.
• •' ''•' '•'.
'*' '"•': '"f '•':'*•• t'.-'
MPAT /^PKilPMT
INLA 1 v^LMLiM 1
PPAM1 11 AR RFNTONITF
o IM c;pM /in
^ UN, o^n i w,
FLUSH-THREADED; PVC
CASING
MATIVF °FniMFNT /"°AND AND
GRAVEL) OR QUARTZ SAND
FILTER PACK
( 1 1 FT")
v ' ' r ' /
O IM ^P| I 4.O
z. UN, — )L/I 1 T"U,
FLUSH-THREADED, PVC
SCREEN
Figure 4-8. Generalized diagram of vertical monitoring well construction,
Sioux Falls airport wellfield, South Dakota.
4-23
-------
2-IN-DIAMETER STEEL
CASING WITH LOCKING CAP
NATIVE SEDIMENT BACKFILL
FILTER SAND
2-IN-DIAMETER x 1-FT-LONG
STEEL WELL SCREEN
0
l I I I
10
J
APPROXIMATE SCALE IN FEET
Figure 4-9. Generalized diagram of horizontal monitoring well construction,
Sioux Falls airport wellfield, South Dakota.
Water levels in the river and aquifer were higher during installation of the horizontal well
than later, when the aquifer test was conducted. At the beginning of the aquifer test, water levels had
fallen to the point where the horizontal well had no water in it at the bottom of the vertical-well riser.
This necessitated the installation of an alternative surface-water measuring point, which consisted of a
vertical monitoring well in the river.
Vertical well in the river-The vertical well (designated as VRW in Figure 4-7) was installed
near the river's edge when water levels had fallen to the point that the horizontal monitoring well was
dry. The well in the river consisted of 1.25-inch-diameter, galvanized-steel casing and a 2.5-foot-long
manufactured well screen. The screen, attached to the casing, was driven into the river-bottom
sediments with a post driver until it was stable, and some of the well screen was exposed to the surface
water. This well position allowed free entry of surface water into the casing.
Existing Monitoring Wells--
Six existing monitoring wells, installed for previous ground-water investigations, were used to
obtain water-level data outside the immediate area of the airport wellfield. A seventh existing
monitoring well was used to obtain water-level and chemistry data. The locations of these wells are
shown in Figure 4-6 and are indicated by the prefixes DOT, SFE, and MW. Information on these
wells is given in Table 4-2.
4-24
-------
°f thfSC SeVCn weIls"Stone Container MW-10 and -18 and Anderson's Truck Stoo
—.
construction for the collection of representative water qualify samples
Water Qualify Sampling,
Sampling Procedures-
*
manual is an unpublished document intended for in-house use It the SDGS.
4-25
-------
come in contact with the pump from returning to the well. Suction pipe and check valves were
constructed of PVC, and were cleaned with high-pressure hot water and wrapped prior to transport to
the site.
Water samples for inorganic analysis were collected either from the discharge point on the jet
pump or from laboratory-cleaned Teflon bailers. All water samples to be analyzed for pesticides and
VOCs, and some samples for inorganic analysis, were collected using bailers. The bailers were
attached to stainless-steel wire used to raise and lower them. The bailers and stainless-steel wire were
handled with clean gloves during sampling activities to minimize the possibility of introducing
contamination to the water sample. Before connection of the stainless-steel wire to the bailer and
lowering of the bailer down the well, the wire was rinsed with distilled water. The bailer was then
attached to the wire and lowered. While the bailer and wire were being lowered into the well for the
first time, the wire was wiped with a clean towelette wetted with distilled water. The first bailer of well
water was dumped to waste, thereby providing a rinse of the laboratory-cleaned bailer. Subsequent
water withdrawn from the well with the bailer was used for sampling purposes.
Water samples collected for analysis of pesticides and other VOCs were analyzed by the
University of Iowa Hygienic Laboratory, Iowa City, Iowa. Samples were placed in containers supplied
by the laboratory and were shipped in accordance with requirements of the laboratory.
Laboratory Quality Assurance and Quality Control—
The SDGS Basic and Analytical Studies Laboratory analyzed all water samples for inorganic
parameters. The University of Iowa Hygienic Laboratory analyzed water samples for selected
pesticides and VOCs and is certified by EPA for all methods used in the study. The SDGS Basic and
Analytical Studies Laboratory is certified by EPA for the analysis of arsenic, fluoride, nitrate and
nitrite as nitrogen, and selenium. The SDGS Basic and Analytical Studies Laboratory participates in
the U.S. Geological Survey's (USGS) Analytical Evaluation Program twice per year for the analytical
methods used in this study. Both laboratories adhere to strict quality assurance and quality control
(QA/QC) procedures.
Sampling for Inorganic Analysis—
The sampling plan for inorganic constituents was designed (1) to examine spatial and
temporal trends in ground-water and surface-water quality, and (2) to provide baseline data for long-
term monitoring. Ten monitoring wells and a surface-water sample site were chosen for sampling of
inorganic constituents. Five of these wells (R20-89-118, -119, -122, -124, and -127) are in proximity to
municipal well 34A, which is the focal point of the study (Figure 4-7). Five other wells (R20-90-1, -2,
-3, -4, and -5) were located outside of the airport wellfield (Figure 4-6). The surface-water sample site
was located immediately upstream from the divergence of the Big Sioux River and the diversion canal
(Figure 4-6).
The sampling plan included three events during the study period. The sampling events
occurred in 1990 on January 15 through 17, April 9 through 11, and July 9 through 10. The network
was successfully sampled during the first two sampling events. However, after the April sampling
event, well R20-90-124 was struck by construction equipment and sufficiently damaged so that it could
not be used for future sampling. Well R20-90-125 (Figure 4-7) was substituted for the damaged well
4-26
-------
in the July sampling event. No water sample was taken from well R20-90-1 during the July sampling
event. Other than these two differences during the July sampling event, each water-sampling point
was sampled three times for inorganic constituents.
Analyses for inorganic parameters were performed for the parameters listed in Table 4-3.
These parameters can be categorized as major cations and anions, minor constituents, and water
quality indicators such as specific conductivity, oxidation-reduction potential (Eh), laboratory pH, and
so on. The major cations and anions listed in Table 4-3 were chosen for analysis because they
comprise the bulk of the constituents in the water and can, therefore, be used for water typing. The
minor constituents were chosen for analysis because they are important in characterization of ambient
water quality. Selenium and arsenic are of particular concern because they are naturally occurring,
sometimes at significant concentrations, in glacial outwash in eastern South Dakota. The origin of the
selenium and arsenic is believed to be from Cretaceous rocks within the glacial outwash.
TABLE 4-3. INORGANIC PARAMETERS ANALYZED IN WATER SAMPLES,
SIOUX FALLS AIRPORT WELLFIELD, SOUTH DAKOTA
Major Cations and Anions
Minor Constituents
Water Quality Indicators
Bicarbonate
Calcium
Carbonate
Chloride
Magnesium
Potassium
Sodium
Sulfate
Arsenic
Fluoride
Iron
Manganese
Nitrate and nitrite as nitrogen
Selenium
Specific conductivity
Eh
Field pH
Hardness
Laboratory pH
Total alkalinity
Total dissolved solids
The list of parameters given in Table 4-3 represents a baseline water quality characterization
suite. Potential contaminant sources identified in the assessment process did not greatly influence the
choices of parameters for analysis. Key indicators of contamination will be generated for the long-
term monitoring program from the baseline water quality data and from more detailed source
assessment information. These parameters are selectively chosen from long-term water quality data
and tailored to monitor potential, or existing, sources of greatest threat.
Sampling for Pesticides and Volatile Organic Compounds-
The sampling plan for pesticides and VOCs was designed in a manner similar to that for the
inorganic constituents. Five monitoring wells and a river site were chosen for sampling organic
constituents. The five wells selected are a subset of the 10 wells sampled for inorganic constituents.
Wells R20-89-119, -124, and -127 were chosen to provide detailed information near municipal well
34A (Figure 4-7). Wells R20-90-2 and -3 were chosen to provide information from more distant parts
of the aquifer (Figure 4-6). The river sampling site is the same as that used for sampling the
inorganic constituents (Figure 4-6).
4-27
-------
The sampling plan designated two sampling events. Because of budget constraints, five fewer
wells and one less sampling round were scheduled for the pesticides and VOCs than for inorganic
constituents. Sampling occurred on January 15 through 17, 1990, and July 9 and 10, 1990,
concurrently with the first and last sampling events for inorganic constituents. As with the sampling
for inorganic constituents, well R20-89-125 was substituted for damaged well R20-89-124 in the July
sampling event.
Analyses were performed for the pesticides and VOCs listed in Table 4-4. The pesticides and
VOCs were chosen to provide background information on water quality. Pesticide selection was based
primarily on known usage and because some of these pesticides are part of an analytical scan for a
particular chemical class. Volatile organic compound analysis is commonly employed as a screening
technique in association with point-source areas.
Water-Level Measurement
Monitoring-point elevations were established for all stations (land, surface water, and wells)
used in this study, based on existing benchmarks established by the Coast and Geodetic Survey and the
USGS. Elevations were reduced to a common datum to allow direct comparison of lithology and
water levels at various locations. All elevations are relative to mean sea level, thus are directly
comparable with published topographic maps. Reference elevations are accurate to the nearest ±0.1
foot. This accuracy is less than that of the water-level measurements (±0.01 foot) due to an
accumulated error in surveying.
Water levels were measured to define the configuration of the water-table surface relative to
the river elevation during different seasons and under different pumping conditions. These data are
integral to understanding the dynamics of the surface-water and ground-water systems in the wellfield
area.
Water levels were measured with an electric water-level indicator to an accuracy of ±0.01 foot.
Prior to measurement, the probe on the end of the water-level indicator was rinsed with distilled
water. A clean towelette, wetted with distilled water, was then used to wipe the electrical tape as it was
lowered into the monitoring well. The same process was repeated with a clean towelette as the tape
was drawn out of the well.
Ground-water levels were measured in some wells by using a pressure transducer and data
logger. The pressure transducer has a pressure range of either 0 to 15 pounds per square inch (psi) or
0 to 5 psi. Both were utilized in the project. The transducers have a linearity of 0.05%, with a
resolution of better than 0.1%. They are compatible with the data logger, and automatically
compensate for barometric changes on the water column. Each data logger has 32 kilobytes of
memory and is capable of storing approximately 16,000 water-level data points. Pressure transducers
and attached cables were rinsed with distilled water and wiped with a clean towelette before being
placed in a well.
During aquifer testing, data loggers nearest the pumping well (municipal well 34A) were set to
collect data on a logarithmic time scale in the early stages of the test. Loggers more distant from the
pumping well collected data on a linear time scale throughout the test. Data collection at these times
4-28
-------
TABLE 4-4. PESTICIDES AND VOLATILE ORGANIC COMPOUNDS ANALYZED IN
WATER SAMPLES, SIOUX FALLS AIRPORT WELLFIELD, SOUTH DAKOTA
Pesticides
Volatile Organic Compounds
Trade Name
Chemical Name
Atrazine
Bladex
Dual
Lasso
Sencor
Sutan
Treflan
Counter
Dyfonate
Lorsban
Mocap
Thimat
Furadan
2,4-D
Silvex*
Amiben*
Banvel*
Atrazine
Cyanazine
Metolachlor
Alachlor
Metribuzin
Butylate
Trifluralin
Terbufos
Fonfos
Chlorpyrifos
Ethoprop
Phorate
Carbofuran
2,4-D
2,4,5-TP
Chloramben
Dicamba
Chloromethane
Bromomethane
Vinyl chloride
Chloroethane
Methylene chloride
Acetone
Carbon disulfide
1, 1-Dichloroethene
1,1-Dichloroethane
1,2-Dichloroethene (total)
Chloroform
1,2-Dichloroethane
2-Butanone (methyl ethyl ketone)
1,1,1-Trichloroethane
Carbon tetrachloride
Vinyl acetate
Bromodichloromethane
1,2-Dichloropropane
cw-l,3-Dichloropropene
Trichloroethene
Dibromochloromethane
1,1,2-Trichloroethane
Benzene
oiflAw-l,3-Dichloropropene
Bromoform
4-methyl-2-Pentanone (MIBK)
2-Hexanone
Tetrachloroethane
1,1,2,2-Tetrachloroethane
Toluene
Chlorobenzene
Ethylbenzene
Styrene
Xylene (total)
p-Dichlorobenzene
Analyzed in July sampling event only.
4-29
-------
provided for acquisition of the most usable data. Data loggers and pressure transducers proved to be
very beneficial in this study by eliminating much of the staff time required for water-level
measurements.
Aquifer Test
An aquifer test was conducted using municipal well 34A as the pumping well. Aquifer testing
was conducted to provide information regarding (1) the configuration of the water-table surface within
the area of influence of municipal well 34A under pumping conditions, (2) the characteristics of the
aquifer, and (3) the interaction of surface water and ground water. These data were used to calculate
ground-water flow velocities between the river and the pumping well.
The pumping portion of the aquifer test was conducted from noon on May 15,1990 through
2:00 p.m. on June 7,1990 (23 days and 2 hours). The test was scheduled to last 3 days, but was.greatly
extended to accommodate dye-tracing tests. Specifically, it was extended to accommodate a second
injection of dye during a tracer test conducted concurrently with the aquifer test. The recovery phase
of the aquifer test lasted for 72 hours after shutdown of the well on June 7, 1990.
Data obtained for calculation of aquifer transmissivity and storativity values were limited to
the first 23 hours of the test. This is because the initial pumping rate of 396 gpm was lowered to
388 gpm to prevent the water level in the pumping well from reaching the top of the well screen and
because the river level began to rise and impact ground-water levels. Sioux Falls city officials required
lowering of the pumping rate because of their concerns that the screen would become aerated, causing
well maintenance problems.
The depth to water in the monitoring wells and municipal well 34A (pumping well), and the
surface elevation of the Big Sioux River were also measured during the test. Pressure transducers and
data loggers were used on wells R20-89-113 through 118 and for wells R20-89-120, -123, -125, and
-128 (Figure 4-7). Well R20-89-117 was measured manually for a portion of the test because this well
was used for the first dye injection during the tracer test. Wells R20-89-119, -121, -122, -124, -126, and
-127 were measured manually during the aquifer test. Monitoring wells outside of the area shown on
Figure 4-7 were also measured during the test to provide background information on aquifer
conditions during the test.
Dve-Trace Test
A tracer test was conducted in conjunction with the aquifer test to determine ground-water
travel times between the river and municipal well 34A. This information was used to estimate travel
times from the river to other municipal wells.
The first step in preparing the tracer test was to select the appropriate tracer(s) to use in the
hydrogeologic setting of the study area. The main criteria were:
4-30
-------
1.
2.
3.
4.
5.
Toxicity of,the tracer. It was necessary to select a nontoxic, nonradioactive trace
material, even though the tracer would be significantly diluted before reaching the
water-purification plant. The tracer was injected into an area of the aquifer used as a
principal wellfield for the city and exited the aquifer through municipal well 34A,
which discharged to the city's water purification plant.
Adsorptivity of the tracer. It was desirable to choose a tracer that would not readily
adsorb to the aquifer materials. Although the aquifer does not have a relatively high
clay content, the percentage of silt and clay present could enhance adsorption. Thus, a
relatively conservative tracer was desired.
Ability to analyze for the tracer onsite. Field determination of tracer concentrations was
needed so that sampling frequencies could be adjusted to maximize the amount of
useful information. For example, sampling frequency increases at the time a tracer
concentration begins to peak in order to define the shape and duration of the con-
centration curve. Additionally, because the SDGS owns a spectrofluorometer, it was
highly advantageous to select a fluorescent tracer.
Cost of tracer materials and analyses. Some tracer materials are extremely expensive,
while others are relatively inexpensive. Some tracer materials cost more to analyze
than others. Because a relatively large dilution factor was expected between the time
the tracer was injected and the time it arrived at municipal well 34A, a considerable
amount of tracer needed to be injected. Thus, the unit cost of the tracer material
needed to be minimized. Additionally, analyses of tracer concentration would be
performed by a contractual laboratory to check the reliability of the field-determined
concentrations. Because a fairly large number of analyses were anticipated, the unit
cost per analysis for the selected tracer had to be reasonable.
Availability of the tracer. The tracer had to be readily available from a known source in
the event that more was needed in unanticipated circumstances.
These five criteria were best met by the use of fluorescein dye. Fluorescein dye was chosen
because of the following unique characteristics:
• Low toxicity (Caspar, 1987)
• Very low adsorptiviry property (Gaspar, 1987)
• Fluorescence content, which makes it suitable for onsite analysis with the
spectrofluorometric equipment available from the SDGS
• Inexpensive cost per unit of material
• Inexpensive analytical cost
• Availability from the laboratory contracted to perform the analyses for tracer
concentration
In addition, fluorescein dyes can be detected in the nanograms per liter range. This was also a
desirable property because dilution was expected to be significant between the points of dye injection
and detection.
4-31
-------
Determination of the injection point(s) for the dye was the second step in preparing for the
tracer test. Because the objective of the tracer test was to determine ground-water travel times from
the river to the well, the dye had to be injected as close to the river as possible. The most logical
injection points were the monitoring wells between the river and municipal well 34A. Two injection
points were chosen. The first injection occurred in monitoring well R20-89-117, located
approximately 40 feet north of municipal well 34A (Figure 4-7). Injection of dye close to the
pumping well permitted rapid detection and an opportunity to field-test sampling frequencies and
analytical equipment. The second injection point was monitoring well R20-89-123, located
approximately 140 feet north of municipal well 34A. This monitoring well is very close to the river
and provides a representative point by which to determine the travel time from the river to the
pumping well, 34A (Figure 4-7).
For the tracer tests, 1 pound of fluorescein dye was obtained in powder form. The dye was
divided into 0.25-pound allotments, and each was mixed in approximately 0.5 gallon of distilled water
and stored in a polyethylene container in the dark to limit degradation of the dye. After pumping
began at municipal well 34A, water levels were monitored in the two injection wells to ensure that the
cone of influence had reached these wells. Approximately 3 hours after pumping began, one of the
0.5-gallon containers of dye was injected into monitoring well R20-89-117. A length of Tygon tubing
just long enough to reach into the screened area of the monitoring well and remain below the water
table was attached to a plastic funnel. The dye was poured into the funnel and was flushed with
approximately five well volumes of water to help surge the dye from the well casing into the aquifer.
Approximately 72 hours after the first dye injection, field testing with a spectrofluorometer showed
that the majority of the dye pulse had passed through municipal well 34A. At that point, the
remaining three 0.5-gallon containers of dye were injected into monitoring well R20-89-123 in the
same manner as in monitoring well R20-89-117.
Ground-water samples were collected at variable frequencies that coincided with the
anticipated pulse of dye moving through the aquifer. The first arrival time of the dye was estimated
roughly using the pumping rate of the well (the volume of water that would be removed from the
aquifer) and an estimated porosity of the aquifer. A garden hose attached to a spigot on the discharge
line of municipal well 34A provided access to the water for sampling.
Two methods of sampling were used. A Y-adaptor was placed at the end of the hose.
Attached to one end of the adaptor was a PVC coupler in which a charcoal filter pack could be
stored. As ground water passed through the coupler, dye was adsorbed onto the charcoal pack. The
purpose of the charcoal pack is to adsorb and accumulate dye concentrations. Without use of the
charcoal packs, the dye concentrations may be below the detection limit of a spectrofluorometer.
After the charcoal pack was removed from the PVC coupler, it was sealed in a plastic bag, labeled
with the date and time of sampling, and stored on ice in a cooler (in the dark) until it could be
shipped to a laboratory for analysis. Although no specific holding times were specified for either the
water samples or the charcoal packs, it was advisable that analyses be performed within 2 weeks of the
sampling time (Tom Aley, Ozark Underground Laboratory, Protem, Missouri, personal
communication, 1990). It is also very important to keep the samples in a cool, dark environment to
minimize dye degradation.
The other opening on the Y-adaptor was used to fill a 60-miIIiliter (ml) polyethylene bottle
with discharge water from the well. Ground-water samples were obtained by rinsing the 60-ml bottle
4-32
-------
several times with discharge water and then filling the bottle completely. The sample bottle was
labeled with the date and time the sample was collected, and stored on ice in a cooler until it could be
.shipped to a laboratory. The same procedure was performed for a second 60-ml bottle, which was
analyzed onsite by SDGS personnel.
Sampling began immediately after municipal well 34A was turned on, and prior to injection of
the dye, to determine background fluorescence concentrations in the aquifer. Background data are
necessary to document the presence of chemicals that can contribute to fluorescence in ground water
(for example, laundry detergent). Water samples were taken at 30-minute intervals, and charcoal
packs were changed at 60-minute intervals. After the first dye injection, water samples were taken
approximately every 60 minutes, and charcoal packs were changed every 60 minutes. After onsite
analysis of water samples confirmed that the dye pulse had arrived, sampling frequency was increased
to better define the dye-concentration curve. After several hours it became apparent that the second
dye pulse would take a few days to reach municipal well 34A Thus, the sampling frequency was
decreased without compromising the data resolution necessary to accurately define the dye-
concentration curve.
Analysis of the discharge water from municipal well 34A for dye was performed in three
different ways.
1. Onsite analysis of samples by the SDGS. Ground-water samples were analyzed onsite by
SDGS personnel with a fluorescence spectrofluorometer. The fluorometer was set up
inside the well house at municipal well 34A It was calibrated, initially, with
fluorescein standards that bracketed the anticipated concentrations and was
periodically recalibrated to correct for instrument drift. The instrument was also
recalibrated when the concentration of the dye in the water approached the
concentration of the upper standard. The detection limit using this instrument was
about 2 micrograms per liter ((ig/1). The field-analytical setup proved to be invaluable
in adjusting sampling frequencies and determining the duration of the tracer test.
2. Laboratory analysis of samples by the Ozark Underground Laboratory. Some duplicates
samples were sent to the Ozark Underground Laboratory, primarily as a check against
the validity of onsite analyses. This laboratory used a spectrofluorophotometer that
scanned both excitation and emission spectra. The detection limit for the fluorescein
in the water samples was about 0.0005 ug/1.
3. Laboratory analysis of charcoal packs by the Ozark Underground Laboratory. Charcoal
pack samples were also sent to the Ozark Underground Laboratory for analysis.
Upon receipt, the packs were placed in a beaker, and an eluting solution consisting of
5% aqua ammonia and 95% isopropyl alcohol was added. After the charcoal pack
was soaked in the eluting solution for 1 hour, the elutant was poured off into a
separate container. Approximately 3 ml of the elutant was then pipetted into a
cuvette. The cuvette was placed in a spectrofluorophotometer, where it was
synchronously scanned for both the excitation and emission spectra, from 417 to 667
nanometers. The charcoal pack data were very valuable in determining time of first
arrival of the dye because the concentrations were initially too low to be detected in
the water samples.
4-33
-------
DATA INTERPRETATION
Water Levels
The temporal and spatial variability of the water-table surface within the study area are
partially depicted in the following four illustrations. Two of the figures portray the entire wellfield
area (Figures 4-10 and 4-11), and two provide close-up views of the municipal well 34A area
(Figures 4-12 and 4-13). Ground-water and surface-water elevations used in constructing these figures
are presented in-Table 4-5. Water levels depicted in Figures 4-10 and 4-12 were obtained on August 2,
1990; those for Figures 4-11 and 4-13 were obtained on August 17, 1990. Examination of the water-
table contour maps for these dates reveals that ground-water flow directions can change significantly
over a short period of time (say, 2 weeks) in some areas while remaining essentially unchanged in
other areas.
Surface-water recharge from the Big Sioux River and the diversion canal to theJBig Sioux
aquifer in the airport wellfield area remained essentially unchanged during this period (Figures 4-10
and 4-11). Also, the ground-water flow direction is generally to the north in the area just south of the
wellfield and to the west in the area just east of the wellfield (Figures 4-10 and 4-11). Industrial and
commercial facilities exist in these parts of the aquifer (Figure 4-5). Ground-water recharge areas to
the wellfield are coincident with areas of industrial and commercial development. This situation
underscores the potential contamination threat of these areas to the public water supply and,
therefore, the need for adequate regulation and monitoring of sources.
The relationship between the river and the ground water on the northwest side of the airport
differed between August 2 and August 17, 1990 (Figures 4-10 and 4-11). On August 2, 1990, relatively
high flow conditions existed in the river, so that the river level was higher than the ground water on
either side. In contrast, on August 17, 1990, ground water was recharging the river on the northwest
side and the river was recharging the ground water on the southeast side. This relationship is more
easily seen in Figures 4-12 and 4-13, which depict close-up views of the well 34A area.
When water levels were measured concurrently in the vertical well in the river and the
horizontal well near municipal well 34A, differences were discovered between elevations in the two
wells. Instead of measuring the river level, the horizontal well was providing data on the ground water
level immediately adjacent to the river bed. For example, a head difference of 1.32 feet was measured
on August 2, 1990, with the river level being higher than that of the ground water. These
measurements were taken during pumping of municipal well 34A; they illustrate the amount of head
loss that was occurring across an approximately 1-foot-thick layer of alluvial silt and clay on the river
bottom. In spite of the less-than-perfect hydraulic continuity between the river and the ground water,
the ground-water system responded rapidly to changes in river level. The relatively abrupt rises in the
river level on May 18 and 23, 1990 (Figure 4-14), caused nearly equivalent rises in monitoring wells
R20-89-113, -119, and -127 (Figures 4-15 through 4-17), located southeast, northeast, and west,
4-34
-------
R 50 W I R 49 W
0 1000 3000 5000 7000 Feet.
Big Sioux aquifer boundary
Line connecting points of equal water—table elevation.
Number is given in feet above mean sea level.
Contour interval is 2 feet, except around municipal wells 12, 2O. 21.
28, and 34A where contour interval is 6 feet.
Municipal well
Monitoring well
Surface water
Intermittent stream
•jK^mZH Map Location
-^—-i
Figure 4-10. Water-table elevations for the study area on August 2, 1990,
Sioux Falls airport wellfield, South Dakota.
4-35
-------
R 50 W I R 49 W
1
2 Miles
0 1000 3000 5000 7000 Feet
Big Sioux oquifer boundary
Line connecting points of equal water—table elevation.
Number is 'given "in feet above mean sea level.
Contour interval is 2 feet, except around municipal wells 28 and 23
where contour interval is 6 feet.
Municipal well
A Monitoring well
gg;;;;;^TTT> Surface water
•}<^ i Map Location
• — -5
Figure 4-11. Water-table elevations for the study area on August 17,
Sioux Falls airport wellfield, South Dakota.
1990,
4-36
-------
1410.82
1410.18
1410.09
14-12.02 •
1412.98
1411.08 '4/0.0^1410.23
Number is water
elevation in feet
above mean sea
level
Monitoring well
Surface-water
measurement point
Municipal well 34A
1411.ON '-'ne connecting points of equal elevation
' Contour interval is 0.5 feet
0 100 200 300 400 feet
Figure 4-12. Water-table elevations near municipal well 34A under pumping conditions on
August 2, 1990, Sioux Falls airport wellfield, South Dakota.
4-37
-------
•-° x
„ 1410.64
1410.63 •
1410.61
1410.60*- VI4T0.46
Number is water
elevation in feet
above mean sea
level
Municipal well 34A
.... ,, Line connecting points of equal elevation.
141'-u> Contour interval is 0.1 feet
Figure 4-13. Water-table elevations near municipal well 34A under static conditions on August 17,
1990, Sioux Falls airport wellfield, South Dakota.
4-38
-------
TABLE 4-5. WATER ELEVATIONS AT GROUND-WATER AND SURFACE-WATER
MONITORING POINTS, AUGUST 2 AND 17, 1990, SIOUX FALLS
AIRPORT WELLFIELD, SOUTH DAKOTA
Monitoring Point
Monitoring wellt
R20-90-1
R20-90-2
R20-90-3
R20-90-4
R20-90-5
R20-89-113
R20-89-114
R20-89-115
R20-89-116
R20-89-117
R20-89-118
R20-89-119
R20-89-120
R20-89-122
R20-89-123
R20-89-124
R20-89-125
R20-89-126
R20-89-127
R20-89-128
SFE-81-1A
SFE-81-2B
SFE-81-6B
DOT-S6
Vertical river well
Horizontal well
Anderson's Truck Stop
MW-10
Stone Container MW-10
Stone Container MW-18
Water
08/02/90
1409.19
1407.92
1411.59
1413.46
1411.75
1409.76
1410.23
1410.31
1410.09
1410.18
1410.56
1411.63
1411.36
1412.22
1411.14
1410.82
1411.08
1412.02
1411.49
1411.82
1409.19
1409.99
1409.02
1411.31
1413.14
1411.82
1414.70
1408.96
1407.78
Elevation*
08/17/90
1409.07
1407.68
1411.27
1413.20
1411.81
1410.46
1410.40
1410.60
1410.61
1410.63
1410.19
1410.56
1410.59
1409.73
1410.67
1410.64
1410.61
1410.67
1410.73
1410.81
1408.21
1408.13
1408.61
1411.29
1410.63
1410.75
1414.62
1408.62
1407.76
Monitoring Point
Municipal wellt
4
10
11
12
13
14
15
17
18
20
21
23
24
27
28
29
30
Surface Waterf
A
B
C
D
Water
08/02/90
1408.7
1408.3
1409.2
1401.4*
1409.0
1406.2
1407.7
1402.7*
1400.3*
1402.8*
1407.1*
1409.0
1395.1
1408.8
1405.0*
1410.8
1406.4
1407.14
1407.08
1410.11
1413.16
Elevation*
08/17/90
1406.9
1408.0
1407.6
1406.9
1408.8
1405.4
1405.2
1406.7
1404.7
1411.9
1410.7
1409.7
1405.0
1408.8
1401.1*
1403.9t
1408.4
1407.99
1407.94
1407.90
1410.69
* Elevations are in feet above mean sea level.
t See Figures 4-6 and 4-7 for well locations; see Figure 4-6 for locations of surface-water monitoring
points.
* Indicates that the municipal well was pumping at time of measurement
4-39
-------
o
CD
CD
CO
T—
CD
C
=3
CD
CD
C
o
CD
CD
CO
CM
o
5-
CD
CO
3
I
•o
§•
•a
49
,w
to
f
S
1
1
0>
43
a.
§.
(|SUJB y) NOI1VA313 "I3A3"l-y31VM
4-40
-------
11
*3 o-
co to
O OJ
•o E
£ g.
if 5
'- to
O)
'I-
I a
2.-S
"o <»
*l
UJ Q-
o
o>
o>
-------
NOI1VA313 -|3A3-l-y31VM
4-42
-------
I
(isiue $ NOI1VA31313A31-fcJ31VM
4-43
-------
respectively, of the horizontal well (Figure 4-7). Most rises in the ground-water levels on these
hydrographs coincided with a rise in river level. The rapid response of ground-water fluctuations to
changes in river levels confirm good hydraulic interconnection.
Water Quality and Chemistry
Results of analyses for dissolved inorganic parameters for samples collected during January,
April, and July 1990 are presented in Table 4-6. Analyses for arsenic and selenium were performed
only on the April and July samples because the January samples were inadvertently discarded before
analysis.
Trilinear plots of these water quality data, presented in Figures 4-18 through 4-20, indicate
that the ground-water chemistry remains relatively constant throughout the year. The dominant
cation in ground water is calcium, and the dominant anion is bicarbonate. The anion sulfate shows a
significant variation in the ground-water sample analyses, which results in a calculated mean
concentration that is slightly less than the concentration in the surface water (Table 4-6).
Results from analyses of surface water collected in January and April show that the dominant
cation was magnesium, whereas calcium was dominant in July. The dominant anion in surface water
was sulfate in January and April. However, as shown in Figure 4-20, carbonate plus bicarbonate, and
sulfate, were present in equal amounts in July.
Analyses for arsenic show concentrations ranging from less than I ng/1 to 19 ng/1 in ground
water and from 3 to 8 jig/1 in surface water. Only one ground-water sampling location, well
R20-89-122, showed an arsenic concentration greater than 5 jig/l. All concentrations of arsenic are
below the EPA primary drinking water standard of 50 ng/1.
Analyses for selenium show concentrations ranging from less than 1 ng/1 to 30 yg/1 in ground
water, while concentrations were less than or equal to 1 jig/l in surface water (Table 4-6). Only two
(R20-89-127 and R20-90-2) of the 10 ground-water sampling locations showed concentrations greater
than 5 ng/1. Only one location, well R20-90-2, showed a selenium concentration consistently
greater than 5 |ig/l. All concentrations of selenium are below the EPA primary drinking water
standard of 50 jig/l. Based on the results listed in Table 4-6, arsenic and selenium in the ground and
surface waters are not currently present at levels of concern to public health.
None of the VOCs listed in Table 4-4 were detected in analyses except acetone, which was
found in nine of the 12 total water samples. Acetone was used as a cleaning solvent for the Teflon
bailers and probably was introduced to the water during sampling. This conclusion is supported by
the results of analyses of fie\d blanks collected using the Teflon bailers and a shipping blank supplied
by the University of Iowa Hygienic laboratory that accompanied the sample bottles. Acetone was
detected in one of the two field blanks but not in the shipping blank. Acetone is not believed to be
present in the aquifer.
4-44
-------
£
I
I
2
1
6
><
CO
o
I
"8
r
s
it
8
8
•S
o
oo
a
P<
•a
t
tn
O
O c«
si d o d
V V V
3
SSI o| S 8 8 § ^ <3 § 8 IS 3 8 8. 8. 8
O O O O O O O O O O O O O C4 O O* C
vvv vvv vvv vvv
csd esoej ei
o eicje
<=>. <=>. <=>.
ooo o o o oo o ooo oo ooo ooo o
oo ooo ooo oo o ooo oo ooo ooo
r-
«n
es
>n
SS
5!S
I
1
o
?
w «n
4-45
-------
V V
r I
V V
V V
V V
s
s
s
s
ss
ooo ooo ooo oo o ooo
II V V V V V
.
O O O OOO
V V
a. a
od r4
55}
S8S
V V
995 S8
GOOOVO
3 S
ooo
V V V
95
«i
M
to-* o.
csts
oq
'-i
O
S
S
W >—'
s a
PS
4-46
-------
AKOTA
Q
S
If
II
JB
|i,
Is
9 8
SIOUX E/i
ater; Paran
S
|l
BJ §*
S 1
Sg
g§
wj ra
W o
I-
S?
o m
Sgg § gS? ||g g g§8
tS2J *° O-*O Ovrfrl- >O OVTI-TJ-
ssgasss sssssss
3t ^* f^* f^ vi MS ^^ r^ v^ o\ r*4 *o ^^ *o
^J ^3 ^% 'j *O t^ *~* T^ "^ c5 ooodoo QO odooo
s S | s § ^
.53 'Ti «*
1 K 1 1
a 1 * ^
v> ^ i
o "S ^
Q c? SI
I
4-47
-------
a
4
gVD O
M »H
S S 3 S 335
•a.
V V
o vo >n
CS t-l r-<
3 § 2 3
o o o *-«
V V V
s
S S S S S S S
V V
CS OO VO
fi-s.
Pu,
I
1
'I
g-ffl'
if-
-2
4-48
-------
ANIONS
SURFACE-WATER ANALYSIS
GROUND-WATER ANALYSIS
Figure 4-18. Trilinear diagram of major cations and anions from January 1990 water samples,
Sioux Falls airport wellfield, South Dakota.
4-49
-------
100
80
7
•^
60
AAAAf
. AAAAA,.
40 AAAA/yv ..
40A^/V\AAA\\
.„ ,.. - ,A/\AAA-§
. :6^)G(xx/vv^
7 7 / /
^ £ <£ ^ c?
''S,
Ca
CATIONS
•Cl
ANIONS
SURFACE-WATER ANALYSIS
GROUND-WATER ANALYSIS
Figure 4-19. Trilinear diagram of major cations and anions from April 1990 water samples,
Sioux Falls airport wellfield, South Dakota.
4-50
-------
CATIONS
ANIONS
A SURFACE-WATER ANALYSIS
• GROUND-WATER ANALYSIS
Figure 4-20. Trilinear diagram of major cations and anions from July 1990 water samples,
Sioux Falls airport wellfield, South Dakota.
4-51
-------
Of the pesticides listed in Table 4-4, atrazine, cyanazine (Bladex), 2,4-D, and dicamba
(Banvel) were detected in ground water or in the river. The sample collection points and pesticide
concentrations are listed in Table 4-7. Analyses for 2,4,5-TP (Silvex), chloramben (Amiben), and
dicamba (Banvel) were not requested; however, analytical results were provided for these parameters
in the results from the July water samples. Thus, no comparison is available for these pesticides with
respect to the January water samples.
TABLE 4-7 PESTICIDES DETECTED IN JANUARY AND JULY 1990 WATER SAMPLES,
SIOUX FALLS AIRPORT WELLFIELD, SOUTH DAKOTA
—'
Concentration of Pesticide
(ug/1)*
Sample Collection
Point
Well R20-89-119
Well R20-89-125
Big Sioux River
Date of
Sample
Collection
01/16/90
07/10/90
07/10/90
Atrazine
, 0.1
2.2
2.0
Cyanazine
(Bladex)
ND
0.89
0.39
-i=
2,4-D
ND
ND
0.2
—
Dicamba
(Banvel)
—
ND
0.69
„
Abbreviations: ND = none detected; ~ = no analysis performed.
* Dicamba was not analyzed in January water samples. Atrazine, cyanazine, and 2,4-D were
analyzed in both the January and July water samples.
Wells R20-90-119 and R20-90-125 are very close to the Big Sioux River at the airport wellfield
(Figure 4-7). Pesticides in these wells may be attributed to induced movement of contaminants from
the river into the aquifer. The EPA primary drinking water standards for atrazine and 2,4-D are
3 ug/1 and 70 ug/1, respectively. No primary standards currently exist for cyanazine or dicamba. All
pesticide detections are below available EPA standards, although the level of atrazine detected is only
about 1 ug/1 lower than that of the primary drinking water standard.
Aquifer-Test Analysis
Pumping and recovery phases of an aquifer test at municipal well 34A were conducted from
noon on May 15 through 2:00 p.m. on June 10,1990. The test was run under less than ideal
conditions because of rainfell and, therefore, rises in both surface-water and ground-water levels.
However, the test provided adequate data for estimation of aquifer properties and evidence of the
rapid response of water levels in the aquifer owing to changes in river level. It also allowed a
qualitative examination of head loss across the river-bottom sediment.
4-52
-------
To gain an understanding of the relationship between the river and the aquifer near municipal
well 34A, the water-table surface was mapped under pumping and nonpumping conditions.
Figures 4-12 and 4-13 depict times when water-level measurements were available for the entire
airport wellfield area. The water-table surface under nonpumping conditions on August 17,1990
(Figure 4-13), indicates that the river was gaining ground water from the northwest and losing water
on the southeast side. No cone of depression existed around municipal well 34A. For the same area
of the aquifer under pumping conditions on August 2,1990 (Figure 4-12), the river level was higher
than the ground water on either side of the river. The cone of depression around municipal well 34A,
including a slight depression in water levels on the northwest side of the river due to pumping, is
depicted in Figure 4-12. Figures 4-12 and 4-13 complement Figures 4-10 and 4-11, which portray the
water-table configurations for the entire airport wellfield. These figures are significant because they
show that the river is not a complete hydraulic barrier to ground-water movement, under either
pumping or nonpumping conditions.
Values measured with pressure transducers and data loggers indicate that water levels were
rising during the four-day period prior to the aquifer test, as illustrated in the hydrograph of well
R20-89-113 (Figure 4-15). Prior to the test, pressure transducers and data loggers recorded water
levels in wells R20-89-113, -114, -115, -116, -118, -120, -123^ -125, and -128. The water-level rise in
these wells ranged from 0.52 foot in well R20-89-120 to 1.97 feet in R20-89-116, which is the nearest
monitoring well to municipal well 34A (Figure 4-7). Rising, or recovering, water levels resulted from
the shutdown of municipal well 34A four days prior to the test. Further evidence of recovering water
levels are from well R20-89-118; records indicate a 0.02-foot rise during the first half hour of the
aquifer test, before drawdown around municipal well 34A began to influence the well.
The configuration of the water-table surface just prior to the aquifer test is shown in
Figure 4-21. The river was higher than the ground water on either side, and the ground-water
gradients were to the northwest and southeast, away from the river. The difference in the two ground-
water elevations shown on the northwest side of the river in Figure 4-21 may indicate a residual cone
of depression due to pumping of municipal well 34A, similar to the cone of depression shown on
Figure 4-12.
All wells shown in Figure 4-21 responded to pumping of municipal well 34A, except for wells
R20-89-119 and R20-89-122, located approximately 1,000 feet to the northeast and southwest,
respectively, of municipal well 34A (Figure 4-7). The hydrograph record of well R20-89-119
(Figure 4-16) typifies water-level behaviors in areas outside the influence of municipal well 34A. The
hydrograph shows relatively constant water levels during the first 72 hours of the aquifer test, followed
by a small rise on May 18 and 19, 1990; a significant rise beginning on May 23, 1990; and ultimately, a
slight decline and leveling off near the end of the aquifer test.
Hydrographs of wells R20-89-113 and R20-89-127 (Figures 4-15 and 4-17) illustrate the
influence of municipal well 34A on the southeast and northwest sides of the river, respectively. The
hydrograph of well R20-89-113 (Figure 4-15) shows the change from a rising to a declining water level
on the southeast side of the river after the onset of pumping conditions. The hydrograph of well
R20-89-127 (Figure 4-17) illustrates a similar decline in water level on the northwest side of the river
as a result of pumping. Declines in the water level on the northwest side of the river are attributed to
expansion of the cone of depression from municipal well 34A beyond the hydraulic boundary of the
river.
4-53
-------
O 4*1410.4
.1409.34
•1409.27
409.44*
1409.35
Number is water
elevation in feet
above mean sea
level
Surface-water
measurement point
Municipal well 34A
Line connecting points of equal elevation
Contour interval is 0.5 feet
Figure 4-21. Water-table elevations near municipal well 34A prior to aquifer testing on May 15, 1990,
Sioux Falls airport wellfield, South Dakota.
4-54
-------
Data used for calculation of the transmissiviiy and storativity parameters of the aquifer were limited to
the first 23 hours of the test, after which the pumping rate in municipal well 34A was reduced from
395 to 388 gpm. Also, at about the time the pumping rate was lowered, the river level began to rise
and impact water levels in the aquifer. The reduction in pumping was reflected by a small rise in
water levels in nearby wells before water levels continued declining (Figure 4-15). During the first 23
hours of the test, the river level remained essentially unchanged.
The Theis and Cooper-Jacob methods of data analysis were employed in examining aquifer-test
results. Drawdown data were analyzed using AQTESOLV software (Geraghty & Miller Inc 1991)
2T?f ST ^ AQTESOLV software and some example computations are illustrated in Figures 4-
22 and 4-23. Values from manual, pressure transducer, and data-logger measurements were used
Water levels were recovering up to the start of the aquifer test; therefore, the measured drawdowns are
minimal. This situation resulted in higher calculated transmissivity values than would have been
derived had the aquifer been at static conditions prior to the test. This fact is supported by the use of
lower hydraulic conductivity values for calibration of the ground-water flow model than what would
have been computed using average transmissivity values (Table 4-8).
°f aquifer transmissivity (T) and storativity (S), as follows
•Using the Theis method, T = 14.51 ft2/min (20,900 ft2/d) and S = 0.1931
• Using the Cooper-Jacob method, T = 14.59 ft2/min (21,000 ft2/d) and S = 0.1856
The transmissiviiy averages derived from the Theis and Cooper-Jacob methods, 20,900 ft2/d and
A ?^eSP2?tlVdy' 3re ^P"*1 f°r the Big Sioux a1uifer> where transmissivity values normally
exceed 13,400 ft /d. Corresponding storativity values are 0.1931 and 0.1856; these values fall within the
usual range for unconfined aquifers, 0.01 to 0.30 (Freeze and Cherry, 1979). The average saturated
thickness of the aquifer in the area of the airport wellfield is 26 feet. Using an average transmissivity
value ot 20,950 ft /d from these two methods yields an average hydraulic conductivity of 800 ft/d.
The data from wells R20-89-118, -120, and -121 were not used to calculate the averages
presented in Table 4-8 for the following reasons:
• There was less than 1 foot of drawdown in these wells over the duration of the usable
data because of their distance from the pumping well.
There was an overall trend of slightly rising water levels in the aquifer during the test.
The calculated transmissivity values from these wells are significantly higher than
values for other wells.
a
e
A t~, u termmatlon of the Pumping phase of the aquifer test is clearly shown on Figures 4-15 and
4-17 by the abrupt rises in ground-water levels on June 7, 1990. These figures show a general rise in
water level throughout much of the pumping phase of the test. RainfeU in the drainage basin during
portions of the aquifer test caused the river level to rise. The general rise in ground-water levels is the
result of the rising river level and illustrates the sensitivity of the surface-water and ground-water
system.
4-55
-------
(U U| s) NMOQMVaa 031030000
4-56
-------
o
CO
LU
R
CO
DC
UJ
UJ
H-
D_
C UJ
co CD
£2
0%
CM !=$2
•ng^
II ^cn
i^ls
ilES!
ggd
2 ZUJ
l^^r
§««0 m
Q_ Qh- '*
CO
CC
UJ
LU
UJ
CC
DC
UJ
CL
II
O
II
LV- DETERMINED PAR,
URVE MATCHING:
00
^S
H-O
QDC
WELL FUNCTION OF i
li
*~^
3
sT
WELL FUNCTION FACl
li
— j
CHANGE IN DRAWDO)
li
CO
<-•
**—
i
CD
3
INTERCEPT OF STRAI
AT ZERO DRAWDOWN
II
o
C
E
CO
co
II
CO
cc
UJ
UJ
^^
cc
11
p 5
< CO
^co^
qz S
rfQ ^
9!< < °
>§ o: co
:3a H- w
§"J u u
UJ^ |_
i-O
O DC
-
E ti
CM^g O
*%Z
CM CC
•^ O m
«t F «H
i- CO CM
I! II II
CO
O
O
O
o"
•M CM
M in
*- co
CD
II
iiuunilMinniiiniiiiiiilmniinlniiiiiUnniiiininimiiilniiiittiliiiiiiiHlniiinn
o
o
o
o
O
E
C
+-•
UJ
O I-
I
C
oq
CO
II
o
•o
Q.
.a
E
-S
B
-
-2
I
r^
4
li
T3 a
•sl
a,
"
q
c\i
oo
co
CM
oo
•
o
CD
CJ
CM
•
O
0°
(U U| s) NMOQMVaa Q31O3dyOO
4-57
-------
TABLE 4-8. AQUIFER TEST SUMMARY RESULTS,
SIOUX FALLS AIRPORT WELLFIELD, SOUTH DAKOTA
Average
Theis Method
Cooper-Jacob Method
Well
R20-89-113
R20-89-114
R20-89-115
R20-89-116
R20-89-117
R20-89-118*
R20-89-120*
R20-89-121*
R20-89-123
R20-89-124
R20-89-125
Transmissivity
(ft2/min)
14.17
14.89
15.01
14.32
12.84
18.94
18.93
18.90
14.34
14.86
15.66
Storativity
(unitless)
0.1348
0.1685
0.2616
0.2686
0.2767
0.1510
0.1994
0.0878
0.1728
0.1656 '
0.0961
Transmissivity
(ft2/min)
14.12
14.94
15.16
14.23
12.87
—
—
—
14.92
14.87
15.58
Storativity
(unitless)
0.1352
0.1617
0.2501
0.2632
0.2704
—
•
. --
0.1558
0.1570
0.0915
V
14.51
0.1931
14.59
0.1856
Abbreviation: — = no data.
* Transmissivity and Storativity values for these wells were not used in calculating average values. The
Cooper-Jacob method of analysis was not valid for these wells because u (the well function factor) is
not less than 0.05.
Dye-Trace Analysis
Dye-trace tests were conducted to determine ground-water travel times between the river and
municipal well 34A. Dye concentrations were determined (1) from onsite analysis of water by the
SDGS, (2) from laboratory analysis of water by the Ozark Underground Laboratory, and (3) from
laboratory analysis of charcoal packs by the Ozark Underground Laboratory (shown in Figures 4-24
through 4-26, respectively). All dye concentrations were corrected for background fluorescence in the
aquifer. Background fluorescence concentrations ranged from 18 to 33 \igfl in the samples analyzed
onsite.
The first dye injection occurred at well R20-89-117, approximately 40 feet north of pumping
well 34A (Figure 4-7). Detectable dye concentrations first arrived at municipal well 34A after about
12 hours; peaked about 1.3 days after injection, and had moved through the aquifer after 6 days. The
second dye injection occurred at well R20-89-123, approximately 140 feet north of well 34A
(Figure 4-7), 3 days after the first dye injection. Detectable dye concentrations from the second
injection first arrived at municipal well 34A after 7 to 9 days; peaked 16 or 17 days after injection; and
were present, but declining in concentration, after 20 days when the test was terminated. Results of
the three sets of analyses are discussed as follows:
4-58
-------
CO
CM
*.
V •»
O) ™
00 t3
§1
= "c"
<1) —
T3
I
I
z
o
o
LU
ur
*~ Q
to
DC
U_
UJ
O
Z
CO
Q
UJ
CO
CL
HI
o
o
NOI1VU1N30NOO 3AQ
4-59
-------
CO
CM
tr
fi
(D
•a
*
CO
CO
to
CM
O
CM
O
LU
-3
LU
LU
O
z
CO
Q
LU
CO
Q.
LU
1
E
I
^
I
1
P Q
6
^ o
cd cx
* -si
8 J9
"3.33
e fa
"
g
£
a
_g
%-»
I
o
CM
O
CO
O
CO
NOIlVyiN30NOO 3AQ
4-60
-------
CO
CM
£1
_ Q)
?f
tg£«
* L_
5,
Q
o
g
UJ
10
LU
S5
cc
u_
UJ
o
z
03
I
O
UJ
05
a.
ui
in
o
o
oo
§
o
o
NOIlVyiN30NOO 3AQ
4-61
-------
1. Onsite anafysis of samples by the SDGS. Results indicate that the first dye began arriving
at well 34A about 18 hours after it was injected at monitoring well R20-89-117 (Figure
4-24). The dye concentration peaked at 132 ng/1 about 31 hours after injection, and
then steeply declined. About 47 hours after injection, a second dye peak occurred at a
concentration of 120 ng/1. The concentration then began to decline rapidly until it
reached near-background levels approximately 7 days after injection. The reason for
the second peak at 47 hours is unclear. However, it may have been caused by a lens of
lower permeability sediment between the injection well and municipal well 34A. The
subsurface lens could have caused the dye plume to split, with one part of the plume
taking a longer flowpath to the discharge point. The dye injected at well R20-89-123
first arrived at about 7 to 9 days, peaked at about 17 days, and then gradually declined.
Only an estimate can be made for the time of first arrival of the second dye pulse
because of erratic instrument response prior to its arrival at well 34A.
2. Laboratory analysis of samples by the Ozark Underground Laboratory. Results indicate
that detection of the dye at well 34A began about 15 hours after injection at
monitoring well R20-89-117 (Figure 4-25). Earlier detection by the Ozark
Underground Laboratory, as compared with the onsite analyses, can be attributed to a
lower detection, limit of the instrumentation at the laboratory. The dye concentration
peaked at 85 (ig/1 at about 33 hours, and then began a fairly steep decline. The
laboratory analyses did not show a second major peak at 47 hours as the onsite
analyses did, but a small rise in dye concentration did occur at this time. The lack of
correspondence between onsite and laboratory data may be a result of degradation of
the dye between the time the sample was taken and the time it was analyzed. Dye
decay is enhanced by light and warmer temperatures. The dye concentrations shown
in Figure 4-25 reaffirm the hypothesis that at least two different peaks resulted from
the first injection. The second dye injection, which occurred at well R20-89-123, first
appeared about 9 days after injection, peaked at about 17 days, and then began a
gradual decline in concentration.
3. Laboratory anafysis of the charcoal packs by the Ozark Underground Laboratory. Results
indicate that the dye arrived at municipal well 34A about 12 hours after injection at
monitoring well R20-89-117 (Figure-26). Earlier detection by analyses of the charcoal
packs, as compared with water-sample analyses, can be attributed to the accumulator
effect of the charcoal packs. A charcoal pack allows otherwise nondetectable dye
concentrations to accumulate to a detectable level. The dye concentration peaked at
about 33 hours, and then began a fairly steep decline. Similar to the laboratory water
analyses, the charcoal packs did not show a second major peak at 47 hours, as the
onsite water analyses did. However, a small rise in dye concentration did occur at this
time. The dye injected at well R20-89-123 first appeared about 8 days after injection
and peaked at about 16 days, at which time the concentration began to decline.
Dye concentrations from the second injection, determined by analysis of the charcoal
packs, are erratic; the charcoal packs were left in the discharge line for inconsistent
time intervals. Because of the accumulator effect of the charcoal packs, use of
inconsistent sampling intervals had a sporadic effect on the resulting concentration.
Erratic trends do not occur in the concentration curve for the first dye pulse because
4-62
-------
the charcoal packs were left in the discharge line for consistent time intervals. The
amount of dye accumulated in the charcoal pack is not a direct function of the
concentration and sampling interval. Although the exact function is not known, it is
more closely related to concentration and the square root of the sampling interval
(Tom Aley, Ozark Underground Laboratory, personal communication, 1990).
Dye concentrations determined from charcoal-pack analyses, divided by the square
root of the sampling interval, are depicted in Figure 4-27. As expected, the peaks and
concentrations for the first dye pulse in Figures 4-26 and 4-27 are similar, because the
sampling intervals were quite similar during this phase. The major difference in the
dye concentrations for the second dye injection is that the peak in Figure 4-26 occurs
at about 16 days, whereas the peak in Figure 4-27 occurs at about 17 days.
The fluorescein concentrations determined by the laboratory analysis are consistently below
those determined by the onsite analyses. Although the fluorescein concentrations differ, the relative
peaks and troughs essentially correspond. Thus, interpretations of the time of first arrival of the dye
and the time required for the dye pulse to move through municipal well 34A are similar in both cases.
The dye traveled from well R20-89-117 to well 34A in 12 hours, and from well R20-89-123 to well 34A
in about 8 days. In this test, the velocity of the dye in the aquifer was dependent on its location with
respect to the cone of depression created by well 34A. As the dye moved from the outer to the inner
part of the cone of depression, its velocity increased substantially, due to the increase in the water-
table gradient.
The travel time determined from the second dye injection should be close to the travel time
between the river and well 34A because the distances are similar. Specifically, the distance from well
R20-89-123 (second dye injection well) to well 34A is 140 feet, and the shortest distance from the river
to well 34A is 135 feet. Thus, it can be concluded that a contaminant entering the aquifer from the
river would reach municipal well 34A hi approximately 8 days, under similar water-level and pumping
conditions. These calculations assume that contaminant transport is primarily advective. Dispersive
processes of contaminant flow, in addition to advective processes, would reduce the travel time from
the river to the well.
The city of Sioux Falls has many other municipal wells located along the Big Sioux River; this
study focuses on the area in proximity to well 34A and the river. Most of these other wells are high-
capacity Bragstad or Ranney wells. In some cases, the laterals extending from the Ranney wells are
closer to the river than municipal well 34A. Ranney wells typically pump at rates of 1,500 gpm or
more, as opposed to the 400-gpm discharge pumped from municipal well 34A during testing. The
travel time for a contaminant entering the aquifer from the river (through primarily advective
processes) near a high-capacity well is expected to be significantly less than 8 days, and may well be on
the order of hours.
4-63
-------
co
CVJ
(0
•a
CO
I
m
CM
O
HI
—3
LJJ
CO
QC
U_
111
O
CO
o CO
Q
LLJ
CO
D_
LU
in
o
in
CO
o
o
co
o
in
o
o
[w.ji| (|/Br/)]
iooy 3d vnos AS
ONHdlAIVS dO
NOiivyiN30Noo
4-64
-------
Ground-Water Flow Mode! and Flowpath Assessment
A two-dimensional, steady-state flow model of the study area was developed to further evaluate
surface-water and ground-water relationships, to delineate capture zones, and to enable particle
tracking from simulated spill sites. The FLOWPATH modeling code (Waterloo Hydrogeologic
Software, 1989) was used to develop this model. FLOWPATH is a flow and particle-tracking
numerical model designed for an IBM personal computer that allows calculation of steady-state,
hydraulic-head distributions; ground-water velocities; pathlines; travel times; and capture zones.' Some
of the features of the FLOWPATH model are as follows:
•
•
It allows a maximum of 10,000 grid cells.
It handles confined, unconfmed, and leaky confined conditions.
It accommodates heterogeneous aquifer properties.
• It accepts spatial variations in ground-water recharge from infiltration.
• It accounts for surface-water interaction with the aquifer.
• It allows for multiple pumping and injection wells.
• It uses the finite-difference method to calculate ground-water flow, and the particle-
tracking method to calculate pathlines and travel times.
Grid and Boundary Design--
A 64- by 74-npde, rectangular grid with a variable-grid spacing was developed for the model
(Figure 4-28). The grid spacing was more dense in areas where there were pumping wells. A tighter
grid spacing was incorporated in these areas to accommodate exact geographical locations of wells.
The 19 municipal wells used in the wellfield during 1988 were assigned representative locations in the
grid. Average annual pumping rates for 1988 were included as an attribute with each well. Boundary
conditions for the model were defined by nodes coinciding with the east, west, and south aquifer
boundaries. For the initial design of the model, these boundaries were considered impermeable. The
northern boundary of the aquifer is hydraulically connected to the Big Sioux aquifer. To simulate this
condition, a constant-head boundary of 1,423 feet above mean sea level was incorporated in the design
of the model Surface-water nodes were placed along each node in the model where the Big Sioux
River or the diversion canal was present. In all, 119 surface-water nodes were incorporated in the
model. Representative surface-water elevations and river-bottom elevations were included as attributes
of the surface-water nodes.
Aquifer Properties--
An initial hydraulic conductivity of 800 ft/d was assigned uniformly to the aquifer, based on
results of the aquifer test. The average transmissivity calculated from the aquifer test data was
14.6 ft2/min (Table 4-8), or 21,000 ft2/d. The average saturated thickness in the area of the aquifer test
is 26 feet. This yields an average hydraulic conductivity of about 800 ft/d. The actual hydraulic
conductivity may vary considerably, but 800 ft/d but was used as a starting point in calibrating the
model and was modified as necessary during model calibration. A hydraulic conductivity of 0.75 ft/d
was assigned to the river bed. This represents an average of the infiltration rate of 0.5 to 1.0 ft/d, as
reported by Jorgensen and Ackroyd (1973). Results from the aquifer test yielded an average
storativity of 0.19. In an unconfined aquifer, the storativiry is approximately the same as the effective
porosity. Thus, a porosity of 0.2 was used for the modeled area.
4-65
-------
R 50 W I R 49 W
2 Miles
1 1 1
6 1000 ' 3000 5000 7000 Feet
Big Sioux aquifer boundary
Model grid
Surface water
Intermittent stream
»J<^ I Map Location
"^ i
Figure 4-28. Variable-spaced grid for the ground-water flow model,
Sioux Falls airport wellfield, South Dakota.
4-66
-------
2.
Koch (1982) used an average annual recharge value of 6.9 inches per year (in/yr) in calibrating
a digital model for a larger area of the Big Sioux aquifer. Hedges et al. (1985) estimated the natural
recharge rate to the entire Big Sioux aquifer to be 6.5 in/yr. A clay layer overlies the aquifer in the
wellfield. Therefore, an average annual recharge of 6 in/yr was assumed for this study. A recharge
amount of 6 in/yr permitted a good fit of the model to observed water levels. Bottom elevations of the
aquifer were determined from test holes in the area and incorporated into the model. The model
calculated aquifer transmissivities at each node by using the test-hole elevation data.
Model Calibration--
The primary objective of model calibration is to minimize the error of the resultant output by
comparing generated head values with observed head values to improve interpretation of ground-water
flow. The calibration was conducted by modifying the variables of the aquifer properties until the
model generates a set of hydraulic heads similar to observed hydraulic heads. The model was
calibrated to hydraulic conditions observed in 1990 by Waterloo Hydrogeologic Software (designer of
the FLOWPATH software code) according to the following steps:
1. Flux nodes were added to the west, east, and south boundaries of the aquifer to
account for surface-water runoff from topographic highs surrounding the domain.
Runoff becomes ground-water recharge because of infiltration along the domain
boundary. The flux nodes on the west and east were assigned values of 0.07 ft/d; flux
nodes on the south were assigned values of 0.03 ft/d.
Hydraulic conductivity values were modified from the default value of 800 ft/d, which
was initially assigned to all nodes except impermeable boundary nodes. The default
hydraulic conductivity value was changed to 650 ft/d. Additional sets of nodes near the
center of the aquifer were modified to values of 400, 500, and 900 ft/d.
The leakage factor was increased fivefold to a value of 0.3, based on the following
rationale. The leakage factor at surface-water nodes is reflective of how well the
surface water is hydraulically interconnected to the adjacent aquifer. The larger the
number, the better the interconnection. The leakage factor is derived from three
terms: the hydraulic conductivity of the stream bed, the thickness of the stream bed,
and the relative area of the stream to the area of the node that it occupies. A leakage
factor of 0.06 applies to most surface-water nodes in this area based on estimated
stream infiltration rates, stream bed thicknesses, and the area of the river relative to
the surface-water node. During initial calibration of the model, the river did not
significantly influence the adjacent water-table elevations in the aquifer. Based on
observed water levels, changes in river level directly affect ground-water levels. To
simulate this effect in the model, the leakage factor was increased fivefold to a value
of 0.3. This value created the desired effect on the hydraulic heads in the aquifer.
Following calibration of the model, a final simulation was run. The hydraulic heads calculated
by the model for each node were quite similar to observed hydraulic heads, reflecting an acceptable
calibration. Water-table contour lines generated by the model are shown in Figure 4-29. These
contour lines compare favorably with the observed hydraulic conditions depicted by the contour lines
in Figure 4-11. In most cases, the .difference between observed and calculated heads is no more than
0.5 foot, although there are some areas where the difference approaches 1 foot. In these areas, the
3.
4-67.
-------
R 50 W I R 49 W
2 Miles
0 1000 3000 5000 7000 Feet
—1_ - Big Sioux aquifer boundary
• Sioux Falls municipal well
iAin-- Line connecting points of equal water-table elevation.
Contour interval is 1 foot
Surface water-
Intermittent stream
•\^ I Map Location "
•^—J
Figure 4-29. Water-table configuration map generated by the calibrated ground-water flow model,
Sioux Falls airport wellfield, South Dakota.
4-68
-------
calibration is considered acceptable because the degree of accuracy of observed water levels is ±0 5 to
1 foot. Calibrated hydraulic head values show that the Big Sioux River is gaining ground water (a
gaming river) to the north of the wellfield. The river begins losing water to the aquifer (a losing river)
as it flows adjacent to the wellfield. The wellfield induces a significant amount of recharge from the
surface water. After the river flows to the south of the wellfield, it once again becomes a gaining river
The diversion canal reflects a similar hydrologic relationship to the aquifer; it changes from a gaining
mode to a losing mode, and then back to a gaining mode.
Model Output--
Following successful calibration, the model was used to generate time-related capture zones
and perform particle tracking from simulated spill sites. Three separate travel-time criteria were
selected to depict enlargement of capture zones: 1, 2, and 5 years. The Big Sioux aquifer has
characteristically high transmissivity values (greater than 13,400 ft2/d), so 1- and 2-year travel times
were used to depict FLOWPATH results to facilitate monitoring site selection in relation to priority
sources. A 5-year travel-time criteria was selected to assess the validity of the preliminary WHPA.
The current delineation criteria and method are recharge boundary and hydrogeologic mapping
respectively. 6'
Time-related capture zones-One of the major objectives in developing an accurate ground-
water flow model was to calculate capture zones for each well in the wellfield. Each municipal well
was assigned 100 particles to be released from a radius of 150 feet. The model then tracked the
particles upgradient for a specified time period. The velocity of the particle tracking is dependent on
several factors, including hydraulic conductivity, hydraulic gradient, effective porosity, and retardation
factor. The hydraulic conductivity and hydraulic gradient factors were determined during the model
calibration. Two effective porosities, 0.15 and 0.25, were used in the model simulations. These
numbers represent lower and upper values of effective aquifer porosities based, in part, on the average
storatmty value of 0.19 obtained from the aquifer-test analysis. By using different effective porosities
the sensitivity of the model to this parameter can be demonstrated. A retardation factor of 1 was used
in all the model runs. The retardation factor is a variable that can be used to account for the
adsorption of particles to the aquifer materials. Because these simulations are based on the predicted
behavior of a generally conservative contaminant, no attempt was made to determine specific
retardation factors. A retardation factor of 1 indicates that no adsorption occurs. FLOWPATH
assumes that contaminant transport is advective. Therefore, no effects of contaminant dispersion are
reflected in the model results.
One-year TOT capture zones are shown for the municipal wells in the study area in
Figures 4-30 and 4-31 for aquifer porosities of 15% and 25%, respectively. Two-year TOT capture
zones (Figures 4-32 and 4-33) and 5-year TOT capture zones (Figures 4-34 and 4-35) for aquifer
porosities of 15% and 25%, respectively, are also shown. The sensitivity of the model to porosity is
apparent in this set of figures. Significantly larger capture zones result from the 15% porosity than
from the 25% porosity.
The 1-year TOT capture zones of many of the wells intercept either the surface-water bodies
(river or diversion canal) or the industrial and commercial area south of the wellfield. In reality the
capture zones of many of the wells along the river will intercept the Big Sioux River in much less than
a year. As the results of dye-tracing indicate, capture zones from some of the wells will intercept the
river in a matter of hours or days.
4-69
-------
R 50 W I R 4-9 W
2 Miles
0 1000 3000 5000 7000 Feet
Big Sioux aquifer boundary
Sioux Falls municipal well
Municipal well capture zone
ffSSt*n Surface water
Intermittent stream
Map Location
Figure 4-30. One-year TOT capture zones at a porosity of 15% for the
Sioux Falls airport wellfield, South Dakota.
4-70
-------
R 50 W I R 49 W
2 Miles
0 1000 3000 5000 7000 Feet
Big . Sioux aquifer boundary
Sioux Falls municipal well
Municipal well capture zone
Surface water
Intermittent stream
T^C, ' Map Location
^—i
Figure 4-31. One-year TOT capture zones at a porosity of 25% for the
Sioux Falls airport wellfield, South Dakota.
4-71
-------
R 50 W I R 49 W
2 Miles
0 1000 3000 SOOO 7000 Feet
Big Sioux aquifer boundary
Sioux Falls municipal well
Municipal well capture zone
Surface water
Intermittent stream
"k^ I Map Location
"^ i
Figure 4-32. Two-year TOT capture zones at a porosity of 15%
Sioux Falls airport wellfield, South Dakota.
for the
4-72
-------
2 Miles
0 1000
1 - 1 - 1 — I I
3000 5000 7000 Feet
Big Sioux aquifer boundary
Sioux Fails municipal well
Municipal well capture zone
Surface water
Intermittent stream
Mop Location
Figure 4-33. Two-year TOT capture zones at a porosity of 25% for the
Sioux Falls airport wellfield, South Dakota.
4-73
-------
R 50 W I R 4-9 W
2 Miles
0 1000 3000 5000 7000 Feet
Big Sioux aquifer boundary
Sioux Falls municipal well
Municipal well capture zone
Surface water
Intermittent stream
Map Location
Figure 4-34. Five-year TOT capture zones at a porosity of 15%
Sioux Falls airport wellfield, South Dakota.
for the
4-74
-------
R 50 W I R 49 W
2 Miles
0 1000 3000 5000 7000 Feet
Big Sioux aquifer boundary
Sioux Falls municipal well
Municipal well capture zone
Surface water
Intermittent stream
T<^j I Map Location
"^~-*o
Figure 4-35. Five-year TOT capture zones at a porosity of 25% for the
Sioux Falls airport wellfield, South Dakota.
4-75
-------
When the travel time is increased to 2 years, some of the capture zones of wells in the center
of the wellfield intercept the surface water. It can be seen from Figure 4-32 that, at a porosity of 15%,
after 2 years many of the capture zones reach the aquifer boundary.
After 5 years (Figures 4-34 and 4-35), more surface water and a greater length of the aquifer
boundary are intercepted by the capture zones. A capture zone approaching an impermeable
boundary, such as the glacial till, will begin to bend upgradient and away from the boundary. For this
reason, the exact configuration of flow lines near the boundary has a greater error of uncertainty. In
addition, the accuracy and quantity of water-level elevation data near the aquifer boundaries are
insufficient to define water-table contours and, consequently, flow lines. Therefore, capture zones that
approach the boundaries of the aquifer should be considered approximate.
The capture zones presented here are reflective only of ground-water travel times. Because of
the close interaction between the river and aquifer, surface water must also be considered as a mode of
contaminant transport into the wellfield. Surface-water velocities are much greater than ground-water
velocities. If the zone of contribution (ZOC) is viewed in terms of both surface water and ground
water, which it must be, then the zone becomes much larger than that produced by the model results.
The ZOC actually includes all upstream portions of the Big Sioux River and its tributaries. Because
the river and its tributaries act as a central drain for shallow ground water in the basin, all upgradient
portions of the Big Sioux aquifer are also in the ZOC to the airport wellfield. Thus, protection of the
quality of water in the river and upgradient portions of the aquifer is warranted, over as large an area
as is practical.
Simulated spill sites-The model was used to simulate spills with particle tracking for four
scenarios: (1) from the industrial and commercial areas south and east of the wellfield, (2) from the
old Elmwood Landfill southwest of the wellfield, (3) from a sand and storage area west of the landfill,
and (4) from a location at the north end of the wellfield. The threat of a potential point-source spill
is greatest in the industrial and commercial areas (Figure 4-36), therefore, the model was used as an
aid in understanding the impact of an unremediated spill in these areas. Figure 4-36 shows the
particle tracking from these simulated spills into the wellfield after 2 years. A porosity of 15% and a
retardation factor of 1 were assumed to simulate a worst-case scenario. (Larger capture zones result
and, therefore, overprotection occurs.) The travel times calculated by the FLOWPATH model assume
advective transport.
Results from several particle tracking runs indicate that contamination north and south of the
wellfield would readily migrate to the municipal wells in 1 to 2 years. Contamination east of the
wellfield would move west toward the diversion canal and then be diverted south along the canal. At
this point, some of the contamination may be discharged into the diversion canal. This could threaten
the city's surface-water intake structures near the southern end of the diversion canal. It is not likely
that contamination from this area would enter any of the municipal wells after 1 year. However, one
of the 2-year municipal well capture zones and two 5-year capture zones intercept the diversion canal
downstream from the simulated spill site in Section 33 (Figures 4-32, 4-34, and 4-35). Therefore, it is
possible that contaminants could enter the wellfield by way of the diversion canal. Simulated spill sites
in the vicinity of the old Elmwood Landfill and the sand and salt storage area indicate that
contamination from these areas would flow into the Big Sioux River downstream from the wellfield,
and thus would not affect the wellfield.
4-76
-------
R 50 W I R 49 W
1
2 Miles
0 1000
3000 5000 7000 Feet
Big Sioux aquifer boundary
Municipal well
Surface water
Intermittent stream
Simulated contaminant discharge
and flowpaths
Williams pipeline
Fire training center
Map
Location
X>| Industrial and commercial areas JJJ
5>Sj Sioux Falls Regional Airport J£
r^.-.1-';! South Dakota Air National Guard
•_—J Agricultural land
**J Former Sioux Fails Elmwood Landfill
Hi Elmwood Golf Course
1 '
t-i-^j Residential areas
I South Dakota Department of
Transportation sand and salt stockpile
. J
Figure 4-36. Simulated contaminant flowpaths from discharge sites for a 2-year TOT,
Sioux Falls airport wellfield, South Dakota.
4-77
-------
Modeling results aided in understanding the complex surface-water and ground-water
interaction in the wellfield area. These results will also be used as a guide in the development of a
wellfield management scheme and in the formulation of wellhead protection monitoring options.
Modeling results are based on the calibrated model, which, in turn, are based on observed hydraulic
conditions depicted in Figure 4-11 and average annual pumping rates in 1988. The wellfield and
aquifer are a very dynamic system in which pumping rates and hydraulic heads are continuously
changing. Therefore, model results should be used only as a tool in understanding the flow system.
Ground-water flow directions and velocities vary throughout the year in response to pumping regimes.
Refined Wellhead Protection Area
The city and the county chose the hydrogeologic mapping method to delineate the preliminary
WHPAs in Minnehaha County (Figure 4-4); the corresponding delineation criteria is recharge
boundary. This method of delineation resulted in WHPAs that are actually aquifer-protection zones
rather than subunits of the aquifer. Initially, WHPA delineation using the calculated, fixed-radius
method, showed that the majority of the aquifer was included for protection by the areas calculated for
each well. Individual areas calculated by fixed radius either overlapped, were close to one another, or
intersected an aquifer boundary. The decision was made to use the hydrogeologic mapping method
because:
• The areas determined using the fixed-radius method are underprotective because the
method does not account for well interference.
• The actual size of the fixed-radius areas is difficult to calculate accurately because the
calculation is dependent on variable pumping rates and times for each well, as well as
river recharge to the aquifer.
• Implementation and management of wellhead protection are easier for the entire
aquifer than for discrete areas. This approach is also more technically defensible, if
challenged in court.
• Inclusion of the entire aquifer as the WHPA supports effective, long-term protection
of the ground-water supply, considering the complex interconnection between the river
and the aquifer.
The Aquifer Protection Overlay Districts in the study area that were established by the city and the
county prior to this research are illustrated in Figure 4-4.
The results of the modeling exercise also indicate that the capture zones of individual wells
(for 2- and 5-year criteria) cover the majority of the aquifer in the vicinity of the airport wellfield
(as did the fixed-radius method). These results justify the use of the hydrogeologic mapping method
for delineation. There are small portions of the aquifer which occur outside of the current WHPA
that should be included to ensure water supply protection. These excluded areas are the result of using
political boundaries rather than strictly geologic boundaries for delineation. The areas are located in
Section 31 [Township 102 North (T102N), Range 49 West (R49W)] and near the center of the north
edge of Section 17 (T101N, R49W; Figure 4-4). Other than these two inclusions, no modifications to
the WHPA are recommended. Consideration should be given to adding buffer zones along the
aquifer boundaries to regulate activities in areas that contribute surface runoff directly to the aquifer.
4-78
-------
Research results demonstrate that the Big Sioux River is a major source of recharge to the
aquifer. Thus, the river must be considered a potential line source of contamination to the ground-
water supply. Protection of the river from contamination must be an integral part of an effective
WHPP. Agreements with political entities upstream from the wellfield should be negotiated to
maintain or improve the present river water quality (1) because of the river's contribution to the
ground water in the airport wellfield, and (2) because the city of Sioux Falls also uses surface water
directly from the river.
4-79
-------
WELLHEAD PROTECTION MONITORING PROGRAM
OBJECTIVES
The objectives of the Sioux Falls wellhead protection monitoring program at the airport
wellfield are:
• To document ambient water quality conditions during different seasons and
precipitation conditions
• To serve as an early-warning detection system for contamination
Both objectives pertain to ground water in or near the wellfield and to surface water around and
upstream from the wellfield. The degree to which these objectives are achieved is dependent on
economic considerations, the level of protection that is acceptable to the city and enforceable under
current regulations, and the limitations of current technology. Other ideas and examples beyond those
presented in this report which may be of use in formulating a monitoring program, monitoring
network, and data-management system can be found in Todd et al. (1976), Tinlin (1976), and
Hampton (1976).
MONITORING TYPES AND GENERAL LOCATIONS
Three general types of monitoring pertain to the airport wellfield:
• Line-source monitoring
• Point-source monitoring
• Nonpoint-source monitoring
Line-source monitoring should be a combination of surface-water and ground-water monitoring.
Recommended point-source and nonpoint-source monitoring will consist only of ground-water
monitoring.
Line-Source Monitoring
The purpose of line-source monitoring is to examine the water quality of the river, the
diversion canal, and the aquifer. The line sources to be monitored include the Big Sioux River and
the diversion canal, which lie on the west, north, and east sides of the airport wellfield. Contaminants
introduced to the river travel rapidly downstream and may enter the aquifer at the airport wellfield
from three sides in a very short time.
4-80
-------
Currently, the dangers to the water supply exceed any protection afforded by line-source
monitoring strategies and technologies. For example, the time required to collect water samples and
to conduct laboratory analyses exceeds the travel time necessary for water to flow from a line source to
some of the municipal wells. In situ, real-time, analytical capabilities are needed to provide a relatively
continuous and current assessment of water quality for this dynamic flow system. The line-source
monitoring sites recommended here are of limited use for early warning detection of contaminants in
the airport wellfield area. However, frequent line-source monitoring will provide baseline water
quality data. With the development and implementation of newer, in situ technologies, line-source
monitoring sites may become a more important component of an early warning system.
The few water samples taken from the river for the purposes of this research reveal no current
violations of state and federal drinking water standards. Thus, the present river water quality does not
adversely affect the ground-water quality of the airport wellfield. Because there is the possibility of
accidental or illegal discharges of hazardous and toxic substances to the river, and because of the
significant hydraulic interconnection between the river and the aquifer, the river must be considered as
a potential source of contamination to the airport wellfield.
Two surface-water line-source monitoring sites are recommended. One site should be located
upstream from the divergence of the Big Sioux River and the diversion canal along the border of
Sections 29 and 32, as shown in Figure 4-37. A second site should be located at the diversion canal on
the east side of the airport wellfield (Section 4, Figure 4-37); this location is downstream from the
industrial and commercial area, yet upstream from the southern surface-water intake structures. These
two monitoring sites will provide water quality data representative of surface water that is induced to
the aquifer or pumped into the intake structures. Data from these monitoring sites, therefore, may
trigger contingency responses for the wellfield.
Nine line-source sites for ground-water monitoring are recommended at the locations shown
in Figure 4-37 (Sections 32, 5, 6, and 4), between the river or diversion canal and some of the
municipal wells. These locations were selected based on the criterion that the municipal well, or group
of wells, have a capture zone that intercepts either the river or diversion .canal (Figure 4-32).
Monitoring-site selection is based on model predictions for a worst-case scenario, assuming a 2-year
travel time, a porosity of 15%, and a retardation factor of 1 (Figure 4-32). Data from ground-water,
line-source monitoring will provide an early warning detection ranging from hours to more than 1
year. Ground-water, line-source monitoring will function primarily to provide data on the ambient
trend conditions of the surface water that is induced into the ground water. As in the case of surface-
water, line-source monitoring, in situ, real-time devices are required to permit the ground-water
component to function as an effective early warning system.
Point-Source Monitoring
The purpose of the point-source monitoring is to examine the water quality between key
potential point sources and both the municipal wells and the surface-water intake structures. Only
those point sources identified in the source assessment process were considered in siting monitoring
stations. Of the point sources identified, the sand and salt storage area and the former Elmwood
Landfill (Figure 4-37) were not considered a threat to the airport wellfield, based on simulated spill
4-81
-------
R 50 W R 49 W
0 1000
3000 5000
Big Sioux aquifer
boundary
Municipal welt
Surface water
Intermittent
stream
Williams pipeline
Surface—water
intake structure
7000 Feet
Types of monitoring
Line sourc
(~ A surface
ej ^ water
| mm ground
\J™ water
Point source
Nonpoint
source
Q
Map
Location
trial and commercial areas
|^\\^| Sioux Falls Regional Airport A
'..'/ . )| South Dakota Air National Guard Jl
-~-~-J Agricultural land
+ *** + Former Sioux Falls Elmwood Landfill
I I I I Elmwood Golf Course
i i •!—i
1t11111 Residential areas
^ South Dakota Department of
Transportation sand and salt stockpile
A
Fire training center
Figure 4-37. Recommended monitoring locations for line, point, and nonpoint sources,
Sioux Falls airport wellfield, South Dakota.
4-82
-------
results (Figure 4-36). However, monitoring is recommended downgradient of the South Dakota Air
National Guard facility, the Williams pipeline, the industrial and commercial areas, the fire training
center, and the Sioux Falls Regional Airport.
Ground water beneath the South Dakota Air National Guard facility (Figure 4-37) is drawn
into the municipal water distribution system. The Air National Guard is currently remediating known
petroleum ground-water contamination and is investigating other potential sources at their facility
(Kim Kurtenbach, SDDENR, personal communication, 1992). At present, monitoring is
recommended for two contaminant sources on this property. Upon completion of the Air National
Guard's site-specific investigation, the city may want to evaluate the need for a complementary
monitoring program for compliance purposes.
As previously discussed, a soil contamination (petroleum product) incident occurred northeast
of the study area in the glacial till because of a leak in the Williams pipeline in 1992. If petroleum
product from the pipeline (Figure 4-37) reaches the ground water, contaminants could easily enter the
airport wellfield, the Big Sioux River, or the diversion canal. If a leak occurs where the pipeline
crosses the Big Sioux River, the river would transport contaminants to three sides of the wellfield.
Contamination would not only threaten the ground water in the wellfield because of induced
infiltration, but would also require immediate shutdown of pumping at the surface-water intake
structures. Therefore, monitoring is necessary to address potential contamination from the pipeline.
Ground water beneath the industrial and commercial area northeast of the wellfield
(Figure 4-37) may discharge to the diversion canal and be carried downstream to the southern intake
structures. From the diversion canal, this water may recharge ground water on the west side of the
canal due to pumping from the municipal wells. Alternatively, ground water may move under the
diversion canal toward wells in the airport wellfield. For these reasons, monitoring of the industrial
and commercial areas is warranted.
The remaining point sources of concern include the industrial and commercial area to the
south, a fire training center near the west edge of the wellfield, and the airport facility itself
(Figure 4-37). The fire training center and the regional airport present a challenge to early-warning
detection monitoring. These sources are within the wellfield boundaries, rather than some distance
from it, and ground-water travel times are short. The fire training center is only about 250 feet east of
municipal well 34A, and most of the municipal wells in the airport wellfield are inside of the
boundaries of the regional airport.
Twelve locations for point-source monitoring sites are shown in Figure 4-37. Monitoring for
point-source contamination will equally serve to document ambient water quality and to act as early
warning detection for contamination.
Two sources of contamination at the South Dakota Air National Guard facility are
recommended for monitoring: (1) an area near the south edge of Section 5 (Figure 4-37), for
contamination leaked from a heating fuel tank; and (2) a site along the east edge of the Air National
Guard property in Section 8, for contamination leaked from an old underground fuel storage facility.
These two monitoring locations were chosen based on the capture zones shown in Figure 4-30.
4-83
-------
Two locations are recommended for monitoring the Williams pipeline. Both are on the
northwest side of the airport and river in Section 32 (Figure 4-37). These sites were chosen based on
the capture zones shown in Figure 4-32 and serve the dual purpose of point-source and nonpoint-
source monitoring. A third possible location for pipeline monitoring is the surface-water, line-source
monitoring point downstream from where the pipeline crosses the Big Sioux River (Section 29 and 32
border; Figure 4-37).
The industrial and commercial area south of the airport is treated as one point source, rather
than a number of smaller, discrete sources. Three monitoring locations in Section 8 are recommended
along the northern edge of the area (Figure 4-37). These three locations were chosen based on the
capture zones delineated in Figure 4-30.
Two locations are recommended for monitoring the east edge of the airport (Sections 33
and 5; Figure 4-37). These locations are based on the capture zones depicted in Figure 4-30 as they
relate to (1) the airport fuel storage area, and (2) the area near the airport terminal where de-icing of
commercial aircraft occurs. Upon completion of a thorough assessment of the use of de-icing
chemicals at the airport, additional monitoring may be warranted for a larger area.
Another point-source monitoring location is recommended in Section 6, at the west edge of
the airport. The purpose of this location is to examine ground-water quality between the fire training
center and municipal well 34A.
The two point-source monitoring locations east of the airport and the diversion canal
(Sections 33 and 4; Figure 4-37) are intended to examine the commercial and industrial area northeast
of the airport. The rationale for these locations is based on the present understanding of ground-
water flow directions. These sites will permit characterization of the water moving under the industrial
and commercial area (1) before it discharges to the diversion canal and is pumped into the south
surface-water intake structures, or (2) before it enters the wellfield on the west side of the diversion
canal. The surface-water, line-source monitoring point in Section 4, just downstream from these two
point-source monitoring locations, may also be used to monitor the northeastern industrial and
commercial areas.
Nbnpoint-Source Monitoring
The purpose of nonpoint-source monitoring is the same as that of point-source monitoring.
All three types of nonpoint sources, namely, residential areas, agricultural areas, and the Elmwood
Golf Course (Figure 4-37), were considered in choosing monitoring sites.
The residential area southwest of the airport (Sections 7 and 18; Figure 4-37) is not considered
to be a threat to the water quality at the wellfield, based on ground-water modeling results. However,
the residential area east of the airport could possibly impact the ground-water quality. Ground water
beneath the residential area probably discharges to the east side of the diversion canal, and then
partially recharges the aquifer on the west side of the canal. Some ground water may flow into the
wellfield under the canal.
4-84
-------
Agricultural land occupies considerable area on the west, north, and east sides of the wellfield
(Figure 4-37). The contamination threat from agricultural land is considered to be minimal at this
time. However, this short-term perception should not result in ignoring the long-term adverse impacts
of activities on the agricultural land. A major, unanticipated spill of a hazardous or toxic substance
could cause adverse water quality effects in a very short period of time.
The Elmwood Golf Course includes the area of the former Elmwood Landfill (Figure 4-37).
Although the former landfill is not considered a potential source of contamination based on the
modeling of simulated spills (Figure 4-36), the golf course is considered a potential contaminant
source. Two-year TOT capture zones depicted in Figure 4-32 indicate that the golf course area
contributes recharge water to the wellfield.
Seven locations are recommended for nonpoint-source monitoring and are depicted in
Figure 4-37. Few monitoring locations are sited relative to the total area of the nonpoint sources.
This reflects the perception that the nonpoint sources are considered to be a minimal short-term
threat. The nonpoint-source monitoring site located on the east side of the diversion canal in Section
4 will serve a dual purpose: to monitor the residential area in Section 4, and to monitor the industrial
and commercial area in Section 33. The ground-water, line-source point in Section 4 may also be used
to monitor agricultural land east of the airport. Both of the surface-water, line-source points (one at
the border of Sections 29 and 32; one in Section 4) may be used to monitor nonpoint sources.
Nonpoint-source monitoring data may prove more useful in determining ambient water quality trends
than serving as an early warning of imminent and serious contamination problems.
Future Studies
Hydrologic conditions at the wellfield are very dynamic, particularly ground-water flow
directions and gradients. For this reason, it is recommended that transient-state, ground-water
modeling be performed to determine flowpaths and gradients under variable pumping conditions.
A thorough inventory of potential point sources of contamination is also warranted. This additional
information would aid in determining final monitoring locations. However, the monitoring strategies
proposed as a result of this research should be used to implement the program. Refinements in the
program can be incorporated later, based on transient-state model studies and a detailed point-source
assessment.
PROPOSED MONITORING WELLS
Proposed wells for the monitoring system should have a well screen that intersects the water
table and that will accommodate water-level fluctuations resulting from pumping and from seasonal
effects. The screen length should be sufficient to monitor the upper 5 to 10 feet of the saturated sand
and gravel. Monitoring wells with this construction (Figure 4-8) may not be suitable for examining
contaminants of the dense, nonaqueous phase liquid (DNAPL) type, which would likely be at or near
the base of the aquifer. If DNAPL-type contaminants become the source of investigation within the
wellfield area, then the typified well construction diagram (Figure 4-8) will need to be modified. Wells
would need to penetrate the saturated thickness of the aquifer, with screen lengths and depths situated
to intercept the DNAPL at, or toward, the base of the aquifer.
4-85
-------
MONITORING PARAMETERS AND FREQUENCIES
Five general categories of parameters are recommended for monitoring line, point, and
nonpoint sources, as listed in Table 4-9. A tailored list of key water quality parameters and optimum
sampling frequencies cannot be generated from the limited data base of this research. Selection of key
parameters for monitoring and associated frequencies must be based on a quantitative assessment of
potential point and nonpoint sources, as well as a risk analysis.
TABLE 4-9. CATEGORIES OF MONITORING PARAMETERS AND FREQUENCIES,
SIOUX FALLS AIRPORT WELLFIELD, SOUTH DAKOTA
Source Type and Monitoring Frequency
(sampling events per year)
Monitoring Categories
General water quality
VOCs
Trace metals
Pesticides
Nutrients
Line Source
4
2
2*
4*
4*
Point Source
4
2
2
0
0
Nonpoint Source
4
0
2*
4*
4*
Abbreviation: VOCs = volatile organic compounds.
* The timing of sampling events is dependent on the seasonal usage of agricultural chemicals.
The five categories of monitoring parameters listed in Table 4-9~general water quality, VOCs,
trace metals, pesticides, and nutrients—will be used as a starting point for assessment, of water quality
in the wellfield monitoring system. The general water quality parameters include those listed in
Table 4-3, except for arsenic and selenium. The VOCs include the chemicals listed in Table 4-4.
Analysis for trace metals include arsenic, barium, cadmium, chromium, lead, mercury, selenium, and
silver. These trace metals are examined by the South Dakota Department of Health when evaluating
the suitability of water for human consumption. The pesticide compounds include those listed in
Table 4-4. The nutrients include the parameters of nitrate plus nitrite, potassium, sulfate, iron,
phosphorous, and zinc. This list of nutrients is recommended by the South Dakota Department of
Agriculture, Division of Regulatory Services (Bruce Jacobson, personal communication, 1992). Four
of these nutrients—nitrate plus nitrite, potassium, sulfate, and iron—are included in the list of general
water quality parameters for completeness, if the situation arises that one category is analyzed, but the
other is not.
Sampling frequencies for the three source types (line, point, and nonpoint sources) are
proposed in Table 4-9. The minimum recommended sampling frequencies for the five parameter
categories are:
4-86
-------
• Quarterly for general water quality parameters for all three source types
• Biannually for VOC parameters for line and point sources
• Biannually for trace metal parameters for line, point, and nonpoint sources
• Quarterly for pesticide and nutrient parameters for line and nonpoint sources
Monitoring at these frequencies should continue until in situ monitoring technology is commercially
available and affordable, or until other information warrants modification of these frequencies.
Quarterly monitoring is conducted to document water quality trends attributed to recharge
events and to the seasonal use of certain chemicals, such as pesticides. For example, in the Big Sioux
aquifer, quarterly monitoring of pesticides and nutrients is necessary, at a minimum, to understand the
chemical transport rate through the aquifer after field application. Biannual monitoring of metals and
VOCs, which are costly to analyze, is determined primarily as a function of budget limitations. Where
quarterly monitoring is preferred to yield a more complete data base, biannual monitoring represents
a compromise between technical objectives and funding constraints. Analysis for these parameters is
chiefly for early-warning detection monitoring.
QUALITY ASSURANCE AND QUALITY CONTROL CONSIDERATIONS
Quality assurance and quality control considerations are an integral part of a successful
monitoring program and are a function of program objectives. These considerations must be
consistent with the ultimate intent of using the data for legal enforcement of wellhead protection
ordinances. At a minimum, the QA/QC procedures developed for field operations and laboratory
analyses for'the research monitoring program should be continued in the long-term monitoring
program. In addition, QA/QC requirements should be developed for data storage and validation.
Field operations QA/QC must address proper well design (Figure 4-8), construction materials,
installation, and development (discussed in the section Data Acquisition, Vertical Wells) to ensure that
cross-contamination is avoided and that environmental data are representative of the subsurface area
from which they were acquired. After wells are installed, measuring-point elevations must be
determined accurately (preferably to ± 0.01 foot) to determine ground-water flow directions for map
construction and modeling purposes (discussed in the section Data Acquisition, Water-Level
Measurement). Finally, samples must be acquired following strict adherence to established guidelines
(Coker et al., 1988) to avoid inadvertent contamination and to ensure representativeness and
reproducibiliiy. Field sampling QA/QC should include an appropriate number of blanks, spikes,
duplicates, and splits to establish the validity of the hydrochemical data (CDM Federal Programs
Corporation, 1987; Canter et al., 1988).
Analytical laboratories used for determination of water chemistry parameters should be have a
rigorous QC program to guarantee that measurements generate precise, accurate, and reliable data.
These laboratories should be certified by the State or the EPA; certification often aids in lending
credibility to data that may be court-admissible. Laboratories should have a documented, internal
QA/QC program with verifiable effectiveness, often available to the client in report format.
4-87
-------
Data base QA/QC involves methods of data reporting accuracy, storage (computer storage
versus paper files), retrieval, manipulation, and access. Data quality and accuracy should be addressed
by the person who is closest to the level of its acquisition (for example, the driller or field personnel),
before it is entered into the data base. Thereafter, a controller should be responsible for checking the
electronic data base for accuracy, as well as doing backups and inventories. Error check programs
that search for inconsistencies and common errors can be created and run; this is commonly done
following keyboard entry procedures (Rowe and Dulaney, 1991).
Data should be readily available for use in temporal-contaminant studies and be in a form
compatible with anticipated computer software applications. Access to sensitive or compliance
monitoring data should be restricted to select personnel, perhaps as "data retrieval only," to protect the
integrity of the data base.
MONITORING DATA BASE STORAGE, UPDATE, AND RETRIEVAL
The WHPP monitoring program is very data extensive. Large quantities of information are
gathered on a continual basis and need to be kept in an organized format that will allow ready access.
The most applicable format for this task is a Geographic Information System (GIS), together with
complementary relational data bases and electronic spreadsheets.
A GIS allows information to be transferred either electronically or manually. Both modes will
be used for the type of information derived from the monitoring system. Examples of the types of
information that will be generated and incorporated in the GIS are: (1) locations of monitoring wells,
municipal wells, streams, and other surface water; (2) locations and characteristics of potential sources
of contamination; (3) continuous water-level measurements; (4) inorganic and organic analyte data
from water samples; and (5) daily meteorological data.
Much of the tabular data for values such as water levels and water chemistry will be placed in
electronic spreadsheets and relational data bases. This will facilitate an appropriate means for updates
and queries. In addition, this information can be electronically interfaced with GIS software.
Nontabular information, such as location maps and water-table contours, is digitized into the GIS in
the form of a map. Specific attributes such as well depth, water levels, and parameter concentrations
can be related to the entities on the maps.
Developing a GIS for wellhead protection will greatly enhance the capabilities of the
monitoring system. The effectiveness of the system is dependent on the ability to organize and
interpret the data derived from it. Therefore, it is important that appropriate technologies are used to
process the information produced by the wellhead protection monitoring system in a timely and effec-
tive manner.
CONTINGENCY PLANNING
The data presented in this report indicate that there is significant recharge from the Big Sioux
River to the Big Sioux aquifer in proximity to the airport wellfield and that contaminants in the river
can reach the municipal wells in a matter of hours to days. Contaminants in the aquifer, in or near
4-88
-------
the airport wellfield, also can reach the municipal wells in a relatively short time period. Therefore,
contingency planning is warranted in the event that contamination of a well, or wells, occurs.
Contingency plans to respond to a contamination event must address the potential for a release on the
surface of the aquifer and in the river, or the diversion canal.
Contaminants released on the surface of the aquifer will probably be localized in extent and
may affect only one or a few wells. However, depending on where the release occurs and the ground-
water flow direction, the surface-water intake structures may also be threatened by ground-water
discharge to the river north of the wellfield or to the diversion canal.
If contaminants are released at the surface, the wells or surface-water intake structures that
will be impacted may be identified using the research modeling results. However, it is recommended
that a transient-state, ground-water flow and transport model be developed for this wellfield to design
optimum pumping rates and durations of the individual municipal wells in order to minimize the
inducement of contamination into the wellfield. The installation of monitoring wells near the
contaminant release point will assist in defining the extent of contamination and the potential threat
to the municipal water supply. These same monitoring wells can also be used to track the progress of
remediation efforts and to help determine when resumption of normal wellfield operations is
appropriate.
Contaminants released to the Big Sioux River or to the diversion canal will receive some
dilution from the surface water, and then be induced to the aquifer due to pumping of the municipal
wells. The surface-water intake structures are most vulnerable to contamination in the event of a
release to the river north of the wellfield, or to the canal. In addition, a number of wells could also be
contaminated.
The development of a transient-state, ground-water flow and transport model may show that
under certain conditions it is beneficial to pump wells near the river and discharge the water to the
river or another suitable location. In this way, the local hydraulic gradient may be modified, to
eliminate or minimize degradation of the ground-water drinking supply. The pumping of existing, and
possibly additional, wells near the river could be used to create a hydraulic barrier between
contaminated river water and municipal wells in the center of the wellfield. This may be a realistic
contingency plan because data presented in this report (Ground-Water Flow Model and Flowpath
Assessment section) indicate that there is usually a gradient from the river and diversion canal toward
the wellfield. The installation of monitoring wells near the perimeter of the wellfield will assist in
defining the extent of ground-water contamination and the potential threat to the municipal water
supply. These monitoring wells may also be used to track the progress of remediation efforts and to
help determine when the resumption of normal wellfield operation is appropriate.
Another option to deal with contamination of the Big Sioux River or the diversion canal may
be to shut off, or significantly reduce pumping from, select wells in the wellfield. Drinking water
would then need to be obtained from other wellfields in the Big Sioux aquifer, or from other aquifers.
Other aquifers that may serve as supplemental sources of water during remediation efforts are the
Split Rock Creek aquifer (east of Sioux Falls) and the Skunk Creek aquifer (west and northwest of
Sioux Falls).
4-89
-------
-------
CONCLUSIONS
The South Dakota Geological Survey (SDGS) conducted research at the Sioux Falls municipal,
airport wellfield to design a long-term monitoring program for wellhead protection. The research
focused on characterization of the hydraulic relationship between the Big Sioux River and the
surrounding unconfined aquifer which supplies potable ground water to the city. Potential sources of
contamination in the wellfield recharge area were assessed, particularly the river as a line source.
Research monitoring tasks included qualitative source assessment, monitoring well installation and
development, ground-water and surface-water sampling, water-level measurement, aquifer testing, dye
tracing, ground-water flow and simulated spill modeling, and monitoring program design. Major
summary points and conclusions are as follows:
• The qualitative source assessment was used to categorize potential sources within the
recharge area into three types: line, point, and nonpoint sources. The Big Sioux River
and the diversion canal act as potential line sources of contamination, draining
fertilizers and pesticides from agricultural land, as well as hazardous and toxic
substances from spills or discharges to the river.
» Potential point sources include industrial and commercial areas, the South Dakota Air
National Guard facility, the Williams pipeline, the Sioux Falls Regional Airport, a
sand and salt storage area, the former Elmwood Landfill, and a fire training center.
« Potential nonpoint sources include residential areas, agricultural land, and the
Elmwood Golf Course.
• A network of 30 ground-water monitoring points was established to measure water
levels, to collect samples, to conduct aquifer testing, and to perform dye tracing. These
30 points consisted of seven existing and 21 new vertical (single) wells, one new
horizontal well, and one new vertical well located in the river. A surface-water sample
site was located in the Big Sioux River upstream from the diversion canal.
• Inorganic analytes were determined for ground-water samples taken from 10
monitoring wells biannually, and for nine of these wells a third time. Pesticide and
volatile organic compound (VOC) analyses were performed biannually on ground-
water samples taken from five wells. Inorganic analytes were determined for one river
monitoring location which was sampled three sample taken three times during the
study year. Pesticide and VOC analyses were performed for this same river location
twice in the year.
4-91
-------
Water-level measurements indicate that ground-water flow is generally to the north in
the area south of the wellfield and to the west in the area east of the wellfield.
Ground-water recharge areas are coincident with the areas of industrial and
commercial development. This situation underscores the need to regulate and monitor
these source areas.
Rises in ground-water levels often coincide with rises in river level, indicating good
hydraulic interconnection between the river and the aquifer.
Water quality analyses indicate a calcium-bicarbonate ground-water type that is
relatively constant throughout the year. The surface-water quality exhibits more yearly
variation; in January and April, the dominant water type was magnesium-sulfate and
in July the predominant water type was calcium-bicarbonate/sulfate.
Concentration levels for arsenic and selenium in ground-water and surface-water are
below the primary drinking water standard of 50 \igfl and, therefore, are not presently
a public health concern. Volatile organic compounds (VOCs) were not detected in
any water samples, with the exception of acetone. The presence of acetone in the
samples is attributed to the use of this solvent as a cleaning agent for the Teflon
bailers. The pesticides atrazine, cyanazine, dicamba, and 2,4-D were detected in
samples from wells very close to the river, or from the river. Detection levels for
atrazine and 2,4-D are below the EPA primary drinking water standards.
Aquifer-test data were analyzed with the Theis and Cooper-Jacob analytical methods
to yield average values of aquifer transmissivity (T) and storativity (S). These values
for the Theis method are T = 20,900 ft2/d and S = 0.1931; for the Cooper-Jacob
method they are T = 21,000 ft2/d and S = 0.1856. The aquifer has an average
saturated thickness of 26 feet in the area of the airport wellfield, yielding an average
hydraulic conductivity of 800ft/d.
Dye tracing was conducted between the river and municipal well 34A to determine
ground-water travel times. Dye from well R20-89-117 (40 feet north) reached the 34A
in 12 hours, based on the first time of arrival; dye from well R20-89-123 (140 feet
north) reached 34A in approximately 8 days. The travel time from well R20-89-123 is
considered to be most representative of the travel time between the river and
municipal well 34A because the separation distances are similar. For a conservative
contaminant entering the aquifer from the river, an estimate of several hours to 8 days
is given as the travel time range to municipal well 34A.
A two-dimensional, steady-state flow model of the airport wellfield was developed
using FLOWPATH to evaluate surface-water and ground-water relationships, to
delineate capture zones, and to simulate contaminant spills from potential sources.
4-92
-------
One-year and 2-year time-of-travel (TOT) capture zones were generated for aquifer
porosities of 15% and 25%. A worst-case scenario of a 2-year TOT capture zone,
using a 15% porosity, was prepared to assess the relative threat of existing and
potential source of contamination to the capture zones of wells within the airport
wellfield.
Simulated spill modeling indicated that the threat of contamination from point sources
is greatest in the industrial and commercial areas east and south of the airport Spills
from the salt storage area and old Elmwood Landfill would flow into the Big Sioux
River downstream from the wellfield and, therefore, are considered a minor threat.
Contamination north and south of the wellfield could easily migrate to the municipal
wells in 1 to 2 years.
A 5-year TOT capture zone map was generated to compare to the existing wellhead
protection area (WHPA—the area within the aquifer boundaries) so that the present
delineation criteria (recharge boundary) and method (hydrogeologic mapping) could
be verified or improved. The 5-year TOT capture zones for individual wells cover the
majority of the aquifer in proximity to the wellfield. These results verify the use of the
recharge boundary criteria and hydrogeologic mapping method to delineate the
WHPA.
A long-term monitoring program is proposed (1) to document ambient water quality
conditions during different seasons and recharge events, and (2) to serve as an early-
warning detection system for contamination. Line-source monitoring consists of both
surface- and ground-water monitoring; point- and nonpoint-source monitoring will
consist primarily of ground-water monitoring with wells.
Two surface-water line-source monitoring sites are recommended: one upstream from
the divergence of the Big Sioux River and the diversion canal, and one in the diversion
canal on the east side of the airport wellfield. Nine ground-water line-source
monitoring sites are recommended between the river, or diversion canal, and some of
the municipal wells.
Twelve monitoring sites are recommended downgradient from five point sources of
concern; that is, between five point sources and downgradient municipal wells. The
five point sources targeted for monitoring include the South Dakota Air National
Guard facility, the Williams pipeline, a fire training center, the industrial and
commercial areas (south and east), and the Sioux Falls Regional Airport. The sand
and salt storage area and the former Elmwood Landfill are considered minimal threats
to the wellfield based on simulated spill modeling.
Nonpoint-source monitoring focuses on three areas: the residential area east of the
airport; agricultural land on the west, north, and east sides of the airport; and the
Elmwood Golf Course. Seven sites are recommended for well installation.
4-93
-------
Proposed wells for ground-water monitoring should be designed primarily to sample
water-soluble or partially-soluble contaminants in the upper 5 to 10 feet of the aquifer.
Well screens should be long enough to accommodate water-level fluctuations due to
recharge events or due to pumping. Wells of this construction are not suitable for
monitoring dense, nonaqueous-phase liquid contaminants, which are considered of
least importance for monitoring at the present time.
Five categories of parameters are recommended for monitoring line, point, and
nonpoint sources: general water quality, VOCs, trace metals, pesticides, and nutrients.
The minimum recommended sampling frequencies for the five parameter categories
are:
Quarterly for general water quality parameters for all three source types
Biannually for VOC parameters for line and point sources
Biannually for trace metal parameters for line, point, and nonpoint sources
Quarterly for pesticide and nutrient parameters for line and nonpoint sources
Quality assurance and quality control (QA/QC) objectives and procedures were
implemented for the research monitoring program for field operations, laboratory
analyses, data interpretation, and report review. At a minimum, QA/QC procedures
for field operations and laboratory analyses should be maintained for the long-term
monitoring program. Additional QA/QC should be developed to address data
accuracy, precision, and reproducibility if compliance monitoring emerges as an
integral part of the long-term program.
Monitoring data should be stored in a Geographic Information System (GIS), to
accommodate anticipated volume and diversity and a QA/QC program should be
established to account for data accuracy and completeness. A GIS will allow
electronic and manual information storage and transfer, data analysis through interface
with spreadsheets and other software, and enhanced cartographic capabilities.
Contingency planning is warranted to establish emergency responses to contaminant
releases at the surface of the aquifer and in the river, or the diversion canal. It is
recommended that a transient-state, ground-water flow and transport model be
developed for the airport wellfield to predict optimum pumping rates and durations,
and to minimize the contamination threat to individual wells.
New water supply development and wellhead protection should be initiated for the
Split Rock Creek and Skunk Creek aquifers as part of the contingency planning effort.
4-94
-------
RECOMMENDATIONS
The byproducts of the research monitoring program are (1) validation of the present WHPA
delineation criteria (recharge boundary) and method (hydrogeologic mapping), and (2) a proposed
long-term, surface-water and ground-water monitoring program. In a continuing effort to improve
the Sioux Falls Wellhead Protection Program, the following recommendations are offered:
• Protection of the Big Sioux River from upstream contamination should be
implemented to the extent that is technically practical and financially feasible. This
would include such actions as:
Establishing a buffer zone along both sides of the river to reduce direct runoff
of contaminants from adjacent lands, particularly pesticides and nutrients from
agricultural areas
Posting wellhead protection warning signs at transportation route and river
junctions to stress that the Big Sioux River contributes significantly to the
Sioux.Falls drinking water supply
• A detailed assessment of potential point sources in proximity to the wellfield should be
conducted by the city. This assessment should include a tabulation of the type of
industry or activity at each location, the type and quantity of chemicals used or stored
at each location, and the precise location and manner or storage of each chemical.
Some of this information can be derived from the computerized data base maintained
by the Sioux Falls Fire Department.
• At a minimum, early warning and ambient trend monitoring should be implemented
as described in the wellhead protection monitoring program. Items associated with
this effort are:
Educating the public about the importance and benefits of wellhead protection
so that the South Dakota Department of Environment and Natural Resources
and the city of Sioux Falls are informed at the earliest possible time about
accidental chemical spills near or in the river. In this way, surface remediation
may be expedited to reduce or minimize aquifer contamination.
Establishing permanent and protected surface-water and ground-water sample-
collection sites. A network of monitoring wells should be established along
the river and diversion canal, as well as between the airport wellfield and
potential sources of contamination to the south. Frequent, manual sampling
should be conducted until in situ, real-time monitoring devices are
commercially available and affordable.
4-95
-------
Contingency plans should be adopted in the event that surface-water intake structures
or the municipal wells become contaminated, or if it is determined that contaminants
are moving toward the wells, requiring them to be shut down. At present, the Big
Sioux aquifer and the river are the primary sources of drinking water for the city. The
SDGS and the U.S. Geological Survey are conducting new water supply investigations
of the Split Rock Creek aquifer. Another potential water development area for the
city is the Skunk Creek aquifer. If it is determined that these aquifers can sustain
further development, it is recommended that they be used as contingency water
suppliejs and that wellhead protection be evaluated for these aquifers, also.
It is recommended that a transient-state, ground-water flow and transport model be
developed for the airport wellfield for the following reasons:
To determine flowpaths and gradients under variable pumping conditions so
that the location of monitoring sites can be refined or verified
To determine pumping times, rates, and durations of municipal wells that may
provide water in a contamination event
4-96
-------
REFERENCES
Barari, A., D. L. lies and T. C. Cowman. 1989. Assessment of Water Resources and Conceptual
Evaluation of a Regional Water Supply for Southeastern South Dakota. South Dakota Geological
Survey Open-File Report 60-UR. Vermillion, South Dakota. 18 p.
CDM Federal Programs Corporation. 1987. Data Quality Objectives for Remedial Response
Activities, Development Process. EPA 540/G-87/003. U.S Environmental Protection Agency, Office
of Emergency and Remedial Response and Office of Waste Programs Enforcement, Washington,
D.C.
Canter, L. W., R. C Knox, and D. M. Fairchild. 1988. Ground Water Quality Protection. Lewis
Publishers, Inc., Chelsea, Michigan, pp. 438-466.
Coker, M. K, K Wilkie and M. Coker. 1988. South Dakota Geological Survey Water Sampling
Manual. Internal Document. South Dakota Geological Survey, Vermillion, South Dakota.
Ellis, M. M. and D. G. Adolphson. 1969. Basic Hydrologic Data for a Part of the Big Sioux
Drainage Basin, Eastern South Dakota. Water Resources Report 5. South Dakota Geological Survey
and South Dakota Water Resources Commission, Vermillion, South Dakota. 124 pp.
Ellis, M. M., D. G. Adolphson and R. E. West. 1969. Hydrology of a Part of the Big Sioux Drainage
Basin, Eastern South Dakota. U.S. Geological Survey Hydrologic Investigations Atlas HA-311. U.S.
Geological Survey, Denver, Colorado. Map + 5 pp.
Eschner, E. 1991. Monitoring Technologies for Wellhead Protection. (Editor and Co-author).
Internal Report. U.S. Environmental Protection Agency, Environmental Monitoring Systems
Laboratory, Las Vegas, Nevada.
Freeze, R. A. and J. A. Cherry. 1979. Ground Water. Prentice-Hall, Inc., Englewood Cliffc, New
Jersey, p. 61.
Gaspar, E. 1987. Modern Trends in Tracer Hydrology, Volume I. CRC Press, Boca Raton, Florida.
•CRC Press. 145 pp.
Geraghty & Miller, Inc. 1991. AQTESOLV Aquifer Test Design and Analysis Computer Software,
Version 1.1. Geraghty & Miller Modeling Group, Reston, Virginia.
Hampton, N. F. 1976. Monitoring Groundwater Quality: Data Management. EPA 600/4-76/019.
U.S. Environmental Protection Agency, Office of Research and Development, Environmental
Monitoring Systems Laboratory, Las Vegas, Nevada. 62 pp.
4-97
-------
Hedges, L. S., J. Allen and D. Holly. 1985. Evaluation of Ground-Water Resources, Eastern South
Dakota and Upper Big Sioux River, South Dakota and Iowa; Task 7, Ground Water Recharge.
Internal Report prepared for U.S. Army Corps of Engineers, Contract DACW 45-80-C-0185. South
Dakota Geological Survey, Vermillion, South Dakota. 36 pp.
lies, D. L. In preparation. Ground-Water Study for the Sioux Falls-Brandon Area. South Dakota
Geological Survey Open-File Report 34-UR. Vermillion, South Dakota.
Jorgensen, D. G. and E. A. Ackroyd. 1973. Water Resources of the Big Sioux River Valley Near
Sioux Falls, South Dakota. Water-Supply Paper 2024. U.S. Geological Survey, Denver, Colorado.
50pp.
Koch, N. C 1982. Digital Computer Model of the Big Sioux Aquifer in Minnehaha County, South
Dakota. Water-Resources Investigations 83-4064. U.S. Geological Survey, Denver Colorado. 49 pp.
Koch, N. C 1983. Evaluation of the Response of the Big Sioux Aquifer to Extreme Drought
Conditions in Minnehaha County, South Dakota. Water-Resources Investigations Report 83-4234.
U.S. Geological Survey, Denver Colorado. 6 pp.
Lindgren, R. J. and C A. Niehus. 1992. Water Resources of Minnehaha County, South Dakota.
Water-Resources Investigations Report 91-4101. U.S. Geological Survey, Denver, Colorado. 80 pp.
Rosenfeld, J. K 1990. Ground-Water Contamination at Hazardous Waste Disposal Facilities. In:
Ground Water Management 1: 237-250, Proceedings of the 1990 Cluster of Conferences, Kansas City,
Missouri. National Water Well Association, Dublin, Ohio.
Rowe, G. W. and S. J. Dulaney. 1991. Building and Using a Ground Water Data Base. Lewis
Publishers, Inc., Chelsea, Michigan. 218pp.
Steece, F. V. 1959a. Geology of the Sioux Falls Quadrangle, South Dakota (15 minutes). South
Dakota Geological Survey, Vermillion, South Dakota.
Steece, F. V. 1959b. Geology of the Hartford Quadrangle, South Dakota (15 minutes). South
Dakota Geological Survey, Vermillion, South Dakota.
Tinlin, R. M. 1976. Monitoring Groundwater Quality: Illustrative Examples. (Editor). EPA 600/4-
76/036. U.S. Environmental Protection Agency, Office of Research and Development, Environmental
Monitoring Systems Laboratory, Las Vegas, Nevada. 81 pp.
Todd, D. K, R. M. Tinlin, K D. Schmidt and L. G. Everett. 1976. Monitoring Groundwater Quality:
Monitoring Methodology. EPA 600/4-76/0260. U.S. Environmental Protection Agency, Office of
Research and Development, Environmental Monitoring Systems Laboratory, Las Vegas, Nevada.
154 pp.
Tomhave, D. W. In preparation. Geology of Minnehaha County, South Dakota. South Dakota
Geological Survey Bulletin 37. Vermillion, South Dakota.
4-98
-------
U.S. Environmental Protection Agency. 1987. Guidelines for Delineation of Wellhead Protection
Areas. EPA 440/6-87/010. U.S. Environmental Protection Agency, Office of Ground-Water
Protection, Washington D.C
Vaughn, K. D. and E. A. Ackroyd. 1968. A Preliminary Report on a Recently. Discovered Aquifer at
Sioux Falls, South Dakota. South Dakota Academy of Science Proceedings 47: 68-74.
Waterloo Hydrogeologic Software. 1989. FLOWPATH User's Guide. Waterloo, Ontario, Canada.
72 pp.
4-99
-------
-------
EPA-600/R-93/
APRIL 1993
CHAPTER 5
WELLHEAD PROTECTION AREA DELINEATION AND MONITORING STRATEGIES
FOR A FRACTURED BEDROCK AQUIFER
DOVER, NEW HAMPSHIRE
by
W. James Griswold* and James H. Vernonf
BCI Geonetics, Inc.
Laconia, New Hampshire 03247
Beth A Moore
Lockheed Environmental Systems & Technologies Company
Las Vegas, Nevada 89119
*Present address: GEI Consultants, Inc.
Concord, New Hampshire 03301-8500
tPresent address: HydroSource Associates, Inc.
Ashland, New Hampshire 03217
April 1993
Contract Number CR-816207-01
Project Officer
Steven P. Gardner
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas, Nevada 89193-3478
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89193-3478
-------
NOTICE
This report is the result of research supported by the U.S. Environmental Protection Agency's
Environmental Monitoring Systems Laboratory in Las Vegas, Nevada, as part of its efforts to provide
technical assistance to state, tribal, and local governments on the implementation of the Wellhead
Protection Program. The specific methods and approaches contained in this document have
undergone peer review but do not constitute official Agency endorsement or policy recommendations.
The Office of Research and Development provides this information to help solve complex technical
problems related to refined delineation and ground-water monitoring of wellhead protection areas in
various hydrogeologic settings. Further assistance is available from the Environmental Monitoring
Systems Laboratory in Las Vegas, from the Office of Ground Water and Drinking Water in
Washington, D.C, and from the ground-water offices in the ten U.S. Environmental Protection
Agency regions.
5-ii
-------
ABSTRACT
A wellhead protection area and ground-water monitoring strategy were established for a
fractured bedrock aquifer in Dover, New Hampshire. The delineation and monitoring studies
proceeded in tandem with an exploration project to expand the municipal ground-water supply. This
study describes how the conceptual hydrogeologic model for the site was developed and refined.
Techniques used to conceptualize the site hydrogeology include a survey of existing geologic and
hydrogeologic information, lineament analysis, surficial and bedrock geologic mapping, surface
geophysical surveys, test well drilling, overburden and bedrock monitoring well installation, borehole
geophysics and other surveys, chemical sampling of surface and ground water, aquifer testing, dye
tracing, and data analysis.
The bedrock aquifer consists primarily of quartz monzonite and metasedimentary rocks that
interfinger along a fractured, faulted contact zone trending north 60 degrees east (N60°E). A N5-
10°W trending lineament and fracture zone intersects the N60°E zone at the site. The bedrock aquifer
is directly overlain by Pleistocene-age sands and gravels. These sediments are overlain by low-
permeability marine clay and lodgement till. It is estimated that 20% of .the water produced from the
bedrock aquifer is derived from overburden sediments in the watershed area.
A test well was installed to a depth of 400 feet in the bedrock aquifer as part of the ground-
water exploration program. Four overburden and bedrock well pairs constitute the present
monitoring network for the test well. Two well pairs lie along the N60°E faulted contact zone, and
two well pairs lie along the perpendicular N30°W trend. The test well and four of five bedrock wells
airlift in excess of 150 gallons per minute. Few contaminant threats exist near the site. Baseline
sampling indicates that minor, elevated levels of iron, manganese, and radon pose the only water
quality problems at present.
Test drilling and borehole surveys (caliper, video camera, acoustic televiewer, thermal-pulse
flowmeter, and hydrophysical logging) indicate that fracturing and ground-water flow are highly
discrete. Flow occurs at isolated, definable depths rather than uniformly along the length of the
borehole. Hydrophysical logging indicates that the borehole water is distinctly layered with respect to
the fluid electrical conductivity parameter. Most borehole water is produced by moderately- to steeply-
dipping fractures and fracture zones that intersect the wells.
Aquifer testing and dye-trace results indicate that the N30°W and N60°E directions have
higher aquifer transmissivities relative to the surrounding bedrock matrix. Drawdown contours are
elongate about the N30°W well alignment, suggesting preferred flow in this direction. Dye-trace
results indicate more rapid travel of injected dye along the N30°W direction than the N60°E direction.
Dye traveled 152 feet in 130 minutes (the time of first arrival of the dye) from injection in a bedrock
monitoring well along the N30°W trend to the test well, which was pumped at 200 gallons per minute.
5-iii
-------
This represents a velocity of 1,680 feet per day. Dye injected in a bedrock monitoring well located 596
feet from the test well arrived there in 148 hours, indicating a velocity of 96 feet per day along the
N60°E direction.
Flowmeter and acoustic televiewer surveys indicate that a moderately west-dipping fracture
zone provides interconnection between the test well and bedrock well R2 along the N30°W trend.
Lacking discrete flow information beyond the test well and well R2, statistical fracture descriptions
become good approximations of flowpaths at increasing distances from the site. Therefore, prominent
fracture peaks along the N5-10°W and N60°E trends represent the most probable flow directions
within the bedrock fracture system at Blackwater Brook. The N60°E trend is substantiated by the
existence of the faulted, fractured contact zone along this strike. Evidence to suggest preferred flow
along the N5-10°W direction is structural and hydrogeologic. Structural control is inferred by strong
expression of the lineament on several platforms of photography and in outcrop fracture trends.
Enhanced transmissivity along the N30°W direction is attributed to the proximity and similar
orientation of the N5-10°W fracture zone.
A quadratic equation is derived from accepted hydrogeologic relationships (Darcy's Law and
the Thiem equation). In this equation, ground-water travel time (determined using the time of first
arrival of dye at the test well) is directly proportional to the square root of distance from the test well.
Constants of proportionality for the quadratic relationship are calculated for the N30°W and the
N60°E directions based on dye-trace velocities. Distances for the 200-day and 1,000-day time-of-travel
thresholds were then calculated for the two fracture zone directions: N5-10°W and N60°E.
Three wellhead protection zones are delineated within the recharge area for the test well using
a variety of criteria and methods. Zone I is the state-mandated 400-foot sanitary radius. Zone IIA
consists of two 1,000-foot-wide "arms" along the N5-10°W and N60°E directions, extending to the
200-day time-of-travel distances. Zone IIB is the area within a smooth curve connecting the outer
boundaries of Zone IIA, producing an oval shape. Zone III is the upgradient area contributing to the
1,000-day time-of-travel distance modified by hydrogeologic features. Recommended regulation of the
wellhead protection zones varies from complete control and restriction of activities in Zone I to public
education in Zone HI.
A major component of wellhead protection program management is long-term ground-water
monitoring. Under present conditions, monitoring of the test well and existing monitoring wells will
focus on a moderate effort to assess ambient water quality and physical parameters. After the
production well is developed, the monitoring frequency and list of monitoring parameters increases.
Proposed frequencies, parameters, and new sites for monitoring derive from technical and
management goals. Action levels are proposed to trigger contingency responses.
5-iv
-------
CONTENTS
Abstract ui
Figures vii
Tables ix
Abbreviations, Symbols, and Conversion Factors xi
Acknowledgments xiv
Background for the Dover Case Study 5-1
Introduction 5-1
Wellhead Protection Program Overview 5-1
Hydrogeologic Setting 5-3
Preliminary Wellhead Protection Area Delineation 5-5
Contaminant Source Assessment > 5-9
Source Inventory List 5-9
Source Characterization and Prioritization 5-12
Research Monitoring Program 5-13
Data Base Limitations 5-13
Monitoring Objectives .... 5-13
Research Monitoring Tasks 5-14
Data Acquisition and Interpretation 5-14
Lineament Analysis ...- 5-14
Geologic Mapping 5-15
Surface Geophysics 5-16
Test Drilling 5-17
Well and Piezometer Installation , 5-19
Borehole Geophysics 5-23
Aquifer Testing and Characterization 5-29
Dye Tracing 5-38
Water Quality Analysis 5-41
Refined Conceptual Hydrogeologic Model 5-43
Refined Wellhead Protection Area Delineation ._ 5-46
Existing Wellhead Protection Programs 5-46
Delineation of the Wellhead Protection Area for the Blackwater Brook Site 5-47
Wellhead Protection Monitoring Program 5-52
Monitoring Objectives 5-52
Monitoring Sites 5-54
Existing sites 5-54
Recommended Sites 5-54
Monitoring Parameters and Frequencies 5-56
5-v
-------
CONTENTS, Continued
Quality Assurance/Quality Control Considerations
Monitoring Data Base
Contingency Planning
Monitoring Action Levels
Contaminant Incident Response
Conclusions
Recommendations
References
5-59
5-60
5-60
5-60
5-61
5-63
5-67
5-69
Appendix 5-A 5-73
5-vi
-------
FIGURES
Number
5-1 Location of the Blackwater Brook ground-water development site,
Dover, New Hampshire 5-2
5-2 Conceptual hydrogeologic block diagram, Blackwater Brook site,
Dover, New Hampshire 5-4
5-3 Lineament map, Blackwater Brook site, Dover, New Hampshire 5-6
5-4 Surficial geology map, Blackwater Brook site, Dover, New Hampshire 5-7
5-5 Preliminary wellhead protection area, Blackwater Brook site, , ,
Dover, New Hampshire 5-8
5-6 Contaminant source assessment map, Blackwater Brook site,
Dover, New Hampshire 5-10
5-7 Rose diagrams of (A) lineament mapping, and (B) fracture fabric of bedrock,
Blackwater Brook site, Dover, New Hampshire 5-15
5-8 Plot of magnetic geophysical survey data with fitted model and mafic dike
hypothesis, Blackwater Brook site, Dover, New Hampshire 5-17
5-9 Diagram of caliper, lithologic, and borehole video camera logs of the test well,
Blackwater Brook site, Dover, New Hampshire 5-18
5-10 Locations of the test well and monitoring wells, Blackwater Brook site,
Dover, New Hampshire 5-20
5-11 Schematic diagram of overburden and bedrock monitoring well designs,
Blackwater Brook site, Dover, New Hampshire 5-22
5-12 Geologic fence diagram, Blackwater Brook site, Dover, New Hampshire 5-24
5-13 Plot of water-level data from domestic well Dl, Blackwater Brook site,••
Dover, New Hampshire 5-29
5-vii
-------
FIGURES, Continued
Number Page
5-14 Plot of ambient water levels in overburden and bedrock
monitoring wells, Blackwater Brook site, Dover, New Hampshire 5-30
5-15 Ambient hydraulic gradient and water-level data for the bedrock aquifer,
July 1990, Blackwater Brook site, Dover, New Hampshire 5-32
5-16 Ambient hydraulic gradient and water-level data for the overburden aquifer,
July 1990, Blackwater Brook site, Dover, New Hampshire 5-33
5-17 Semi-log plot of drawdown versus the log of time for observation well R4 when
the test well is pumping, Blackwater Brook site, Dover, New Hampshire 5-34
5-18 Contour map of the expansion of the 100-foot water-level contour in time during
constant-rate aquifer testing, Blackwater Brook site, Dover, New Hampshire 5-36
5-19 Logarithmic plot of time versus drawdown and Gringarten-Witherspoon type curves
for observation well R4 when the test well is pumping, Blackwater Brook site,
Dover, New Hampshire 5-37
5-20 Plot of drawdown versus the square root of time and Jenkins-Prentice linear-
flow model calculations for monitoring well R3, Blackwater Brook site,
Dover, New Hampshire 5-39
5-21 Dye-trace breakthrough curves of rhodamine-WT injection at well R2
to the test well, and fluorescein injection at well R3 to the test well,
Blackwater Brook site, Dover, New Hampshire 5-40
5-22 Trilinear diagram of ground-water types from overburden and bedrock
monitoring wells, Blackwater Brook site, Dover, New Hampshire 5-44
5-23 Wellhead protection area and zones, Blackwater Brook site,
Dover, New Hampshire 5-50
5-24 Potential contaminant sources in proximity to wellhead protection zones,
Blackwater Brook site, Dover, New Hampshire 5-53
5-25 Existing and proposed monitoring sites, Blackwater Brook site,
Dover, New Hampshire 5-55
5-viii
-------
TABLES
Number
Page
5-1 Potential Sources of Ground- Water Contamination and Degradation in Proximity to
the Blackwater Brook Site, Dover, New Hampshire . . . ..................... 5-11
5-2
5-3
5-4
5-5
5-6
Information Regarding Bedrock and Overburden Monitoring Well Construction,
Lithology, and Testing, Blackwater Brook Site, Dover, New Hampshire ......... 5-21
Flow Rates and Fluid Electrical Conductivities at Discrete Zones Within
Boreholes from Hydrophysical Logging, Blackwater Brook Site,
Dover, New Hampshire ..................... . ............. .
5-27
Statistical Analysis of Rhodamine-WT Dye Trace from Well R2 to the Test Well,
Blackwater Brook Site, Dover, New Hampshire . . . ............... .......... 5-41
Wellhead Protection Zones at the Blackwater Brook Site, Dover, New Hampshire . . 5-48
Time-of-Travel Distances for Dye-Trace Directions, Blackwater Brook Site,
Dover, New Hampshire ........................................ ; ...... 5-49
5-7 Sampling Parameters and Frequencies for the Test Well and Associated
Monitoring Wells, Blackwater Brook Site, Dover, New Hampshire
5-57
5-8 Sampling Parameters and Frequencies for the Proposed Production Well and
Associated Monitoring Wells, Blackwater Brook Site, ~'
Dover, New Hampshire ................. . ........... . ................ 5-58
5-ix
-------
-------
ABBREVIATIONS, SYMBOLS, AND CONVERSION FACTORS
ABBREVIATIONS
amsl above mean sea level
cfe cubic foot per second
cps count per second
d day
Dl domestic well number 1
EPA U.S. Environmental Protection Agency
FEC fluid electrical conductivity
ft foot
ft/d foot per day
ft/d172 foot per square root of day
f^/s square foot per second
gpd/ft gallon per day per foot
gpm gallon per minute
gpm/ft gallon per minute per foot
gr gram
hr hour
1 liter
m meter
m2/d square meter per day
MCLG Maximum Contaminant Level Goals
fiS/cm microSiemen per centimeter
MftVd million square feet per day
mg milligram
mg/1 milligram per liter
mi mile
min minute
N/A not applicable
N5-10°W north 5 to 10 degrees west
N60°E north 60 degrees east
NHDES New Hampshire Department of Environmental Services
NHSGS New Hampshire State Geological Survey
PCB polychlorinated biphenyl
pCi/1 picocurie per liter
ppb part per billion
Ol overburden well number 1
QA quality assurance
QC quality control
Rl bedrock well number 1
5-xi
-------
SDWA Safe Drinking Water Act
TDS total dissolved solids
TOT time of travel
TW test well
USGS U.S. Geological Survey
UST underground storage tank
VOC volatile organic compound
WHPA wellhead protection area
WHPP Wellhead Protection Program
ZOC zone of contribution
SYMBOLS
Ca
a
C03
Fe
HCO3
K
Mn
Na
SO4
calcium
chloride
carbonate
iron
bicarbonate
potassium
manganese
sodium
sulfate
C
D
dh/dt
Eh
F
Y
h
K
q
Q
r
s
S
t
to
T
V
x
constant term or coefficient
hydraulic diffusivity
hydraulic gradient
oxidation-reduction potential
well function
gamma
hydraulic head
hydraulic conductivity
distance between the injection well and the test well
porosity
Darcy velocity
discharge
radius or radial distance
drawdown
storage coefficient
time of first arrival of dye
initial time
transmissivity
average velocity
distance from a given well to the linear extended well
5-xii
-------
CONVERSION FACTORS
Multiply
acre-foot
cubic foot per second
foot
foot per second
gallon
gallon
gallon
gallon per day
gallon per day per foot
gallon per day per square foot
inch
inch per year
mile
million gallons per day
square foot per minute
square foot per second
square mile
By
1230
0.0283
0.3048
0.3048
3.785
0.134
0.00379
0.000003528
0.000207
0.0408
0.0254
25.4
1.609
2.629
0.0929
0.0929
2.59
To Obtain
cubic meter
cubic meter per second
meter
meter per second
liter
cubic foot
cubic meter
cubic foot per second
square meter per day
meter per day
meter
millimeter per year
kilometer
cubic meter per minute
square meter per minute
square meter per second
square kilometer
°F = —(°C) +32
5-xiii
-------
ACKNOWLEDGMENTS
This study was funded under Contract Number CR-816207-01 to the City of Dover, New
Hampshire, by the U.S. Environmental Protection Agency (EPA), Environmental Monitoring
Systems Laboratory in Las Vegas, Nevada (EMSL-LV). The authors express their gratitude to the
following individuals for their help during this project. Steven Gardner (EMSL-LV) served as the
EPA Project Officer and provided technical, logistical, and moral support throughout the research.
Assistance and support given by Pierre Bouchard (former Director of Public Works for the City of
Dover) and Pierre LaVoie (Superintendent of Water and Sewer for the City of Dover) were
appreciated. We thank Dr. Frederick Paillet (U.S. Geological Survey-Denver), William Pedler (GZA
GeoEnvironmental, Inc., Special Wellbore Services), Douglas Heath (U.S. EPA Region I),
Dr. Eugene Boudette (New Hampshire State Geologist), and Thomas Aley (Ozark Underground
Laboratory) for technical assistance and information. The authors appreciate technical reviews
completed by Dr. Charles Kreitler (University of Arizona), Douglas Heath (U.S. EPA Region I),
Dr. Philip Berger (U.S. EPA, Office of Water), Dr. Jeffrey Rosenfeld and John Rotert of Lockheed
Environmental Systems & Technologies Company (LESAT), and William Murray (ABB
Environmental). John Nicholson and Carolyn Cameron (LESAT) provided technical writing and
editing support for preparation of the manuscript. Robin Roth (BCI Geonetics, Inc.) and Steve
Garcia (LESAT) are both responsible for excellent graphical contributions and report 'production.
Shalena Fendzlau (LESAT) is graciously acknowledged for her patience and expertise in the word
processing and formatting of this manuscript. The collective exploration and research experience of
BCI Geonetics in fractured bedrock aquifers provided important philosophical, technical, and
informational bases for the current project.
5-xiv
-------
BACKGROUND FOR THE DOVER CASE STUDY
INTRODUCTION
As part of an effort to provide assistance to the states for wellhead protection, and recognizing
the different problems posed by varying geologic and hydrologic conditions that exist throughout the
United States, the U.S. Environmental Protection Agency (EPA) provided several grants to
municipalities to examine the issue of developing monitoring strategies in a number of hydrogeologic
environments. This document reports the results of the EPA grant to the city of Dover, New
Hampshire, to develop wellhead protection monitoring strategies in a fractured bedrock aquifer at
Blackwater Brook in the northern part of the city (Figure 5-1).
The report first examines some of the steps Dover has taken to protect its ground-water
resources, including the development of temporary protection zones for areas yet to be explored and
developed for ground-water supplies. A brief description of the regional hydrogeologic setting of the
city is presented. The report focuses on the Blackwater Brook site, examining the process of
delineating an appropriate wellhead protection area (WHPA) and of developing monitoring strategies
for the fractured bedrock aquifer. Contaminant sources are examined, characterized, and prioritized,
both for existing threats and for potential future threats.
Details of the steps taken to characterize the hydrogeology of the Blackwater Brook site are
presented, including results from test drilling, surface and borehole geophysics, aquifer testing, and dye
tracing. Based on this information, a refined WHPA and protection zones are described for the
fractured bedrock site. Monitoring strategies proposed for the WHPA include additional monitoring
sites, monitoring parameters, sampling frequencies, and contingency planning. The report concludes
with recommendations both for the delineation process and for the development of further monitoring
strategies.
WELLHEAD PROTECTION PROGRAM OVERVIEW
Dover, New Hampshire, a city of 26,000 people located in the seacoast region of New
Hampshire (Figure 5-1), depends entirely on ground water for its municipal water supply. A
2.5-miIlion-gallon-per-day withdrawal comes from seven gravel-packed wells located in kame and esker
deposits. Dover has grown substantially during the past 5 years, and additional water supplies are
needed. Also, the seacoast area of New Hampshire is undergoing substantial changes in its economic
character. The closing of Pease Air Force Base in 1991 in nearby Newington has affected the entire
region profoundly. Many experts predict that following the current economic recession, the regional
5-1
-------
Figure 5-1. Location of the Blackwater Brook ground-water development
site, Dover, New Hampshire.
5-2
-------
economy will expand as the land and base facilities redevelop in the private sector. To meet the
anticipated growth, the city of Dover has embarked on a number of options to augment its water
supplies, including artificial recharge of several of its existing well sites and water exploration in
fractured bedrock aquifers.
The city passed a ground-water protection ordinance in 1988 that stipulates permitted and
nonpermitted activities within primary and secondary protection areas around the Dover wells. The
ordinance offers a means of minimizing contaminant threats to sensitive hydrogeologic areas that
might adversely affect the quality of water in the Dover municipal well system. A corollary objective is
to monitor ground-water quantity and quality to provide a solid management basis for this resource.
The State of New Hampshire had not yet implemented a Wellhead Protection Program
(WHPP) when the city of Dover passed its ground-water protection ordinance. Primary and
secondary wellhead protection zones were therefore delineated for the city wells using as sophisticated
a method as existing hydrogeologic data would support. The wellhead protection zones were viewed
as a preliminary effort, subject to additional hydrogeologic information that might lead to
modification or improvement of the boundaries.
The primary protection zone is defined as the area encompassed by a 400-foot radius around a
well. Municipal ownership or control of this land is mandated by New Hampshire state law, which
prohibits all activities other than recreational activities and maintenance of the well facilities.
Secondary protection zones for the existing wells were established on the criterion of the effective
radius from a 180-day, no-recharge pumping scenario or the distance to the 1,000-day time-of-travel
contour, depending on the level of information available. The zones were drawn and modified on the
basis of analytical modeling of available aquifer test data, hydrogeologic mapping, and a mapping
effort for contaminant threats. The ordinance restricts land-use activities, specifically focusing on
secondary protection zones. For example, solid and toxic waste disposal is prohibited, as is the
establishment of salvage yards, motor vehicle repair shops, salt storage, animal feedlots, and
commercial storage of fertilizer, manure, and pesticides.
The development and implementation of Dover's Protection Ordinance and the WHPP
involves several key agencies. The city's Department of Public Works is responsible for the
construction, maintenance, and production of water from Dover's well system. The Department has
primary responsibility for both the physical and hydrochemical monitoring of the wells. The
Department of Planning developed the protection ordinance language; this department reviews and
governs challenges to the ordinance. The Planning Department also controls zoning regulations for
the city. The city's Fire and Police Departments have responsibility for the public safety elements of
wellhead protection contingency plans.
HYDROGEOLOGIC SETTING
The 28 square miles that constitute the city of Dover (Figure 5-1) lie in a regional
hydrogeologic environment of Pleistocene-age glacial sediments ranging from high-permeability
outwash sands and gravels (eskers and kames) to low-permeability marine clays and lodgement tills.
The surficiai glacial and marine sediments overlie fractured Precambrian and Paleozoic
metasedimentary rocks intruded by younger Paleozoic granites and diorites (Figure 5-2).
5-3
-------
LL
s.
PQ
CQ
.2,
8
•a
a
•o
a
5-4
-------
Younger intrusions of gabbro and possibly other mafic rocks occur in the area. The bedrock geology
was mapped and described by Novotny (1968), with updates of the mapping by Lyons et al. (1986).
Bradley (1964) mapped the surficial geology and provided details of the hydrogeology of the region.
The surficial mapping is currently, being updated by the New Hampshire State Geological Survey
(NHSGS) and the U.S. Geological Survey (USGS).
The Blackwater Brook site was selected for water supply development in bedrock based on a
suite of criteria including photolineaments, mapped geologic features, bedrock outcrop fractures,
surface geophysical surveys, and a contaminant threats assessment. Strong photolineaments striking
N5-10°W as well as northeast-trending photolineaments, occur in the area (Figure 5-3). These
photolineaments are present on several platforms of imagery and suggest structural bedrock control.
Surface geophysical surveys at the Blackwater Brook site reveal no geophysical signature for the
N5-10°W lineaments, but do show the trace of a strong magnetic anomaly coincident with a
northeast-trending lineament (Figure 5-3). Together, the lineaments and magnetic anomaly correlate
with an inferred geologic contact at the site. This northeast-trending feature is expressed as a geologic
contact (Figure 5-4), photolineament, and a magnetic anomaly. The new municipal water supply well
(the test well shown in Figures 5-1 and 5-2) was sited along this northeast feature.
The Blackwater Brook site has a 5-square-mile watershed, containing primarily surficial
marine geologic deposits ranging from clays to fine- and medium-grained beach sands and outwash
gravels (Figure 5-4). Drilling data support a depositional model of Pleistocene marine transgression.
Marine clays, up to 30-feet-thick at the site of the new supply well, are underlain by 4 to 10 feet of
permeable sand and gravel. Highland areas are either rock-cored drumlins covered with till or
bedrock exposures where till has been eroded. Ground-water flow in both the overburden and
bedrock at the site moves generally from the northeast to the southwest.
Ground-water recharge to the Blackwater Brook site occurs as precipitation, which averages
42 inches per year. Recharge to the bedrock aquifer occurs where open bedrock fractures are in
contact with overlying weathered bedrock and permeable sand and gravel. Recharge of the
overburden occurs in areas where the low-permeability clays at the site pinch-out and where
permeable sand, gravel, or bedrock outcrop to receive direct precipitation. Based on estimates of
recharge rates for similar geologic and climatic environments in central Connecticut (Randall et al.,
1966), the total potential recharge to the watershed ranges from 380 to 520 million gallons per year.
According to initial conceptual models, the bedrock well initially receives water from overlying
sediments during pumping. However, storage within the surficial deposits in the immediate area
probably cannot sustain long-term municipal-level production requirements. Sustainable recharge to
the well comes from leakage from overburden deposits, which travels along water-bearing fractures in
a zone of contribution (ZOC), reflecting directional transmissivities.
PRELIMINARY WELLHEAD PROTECTION AREA DELINEATION
To provide temporary protection of the ground-water development site, a WHPA (Figure 5-5)
was delineated at Blackwater Brook. At the time of the preliminary delineation, the test well was sited
along the N5-10°W lineament (Figure 5-3) that exhibited strong surface expression. Early in the
study, this lineament was identified as a potential, high transmissivity zone within the bedrock aquifer.
5-5
-------
I
8
.59
JS
I
3
*
O
ti
CO
s
ex
c
1
5-6
-------
s
s
1
•«
1
1
•8
o
a
s
i3
I
1
Q
1
S3
^
ffl
S
^
m
^
a
§3
ao
CO
1s
>o
§
Ml
ffi
5-7
-------
I
Q
•35
I
W •->
qq
-
« .s
.
II
a o^
o.
1
E
1
PH
5-8
-------
To represent the directional properties of flow in the aquifer, the Cooper-Jacob equation
(Cooper and Jacob, 1946) was applied along the potentially transmissive N5-10°W lineament and the
perpendicular direction. Representative transmissivities derived for similar hydrogeologic regimes
were assigned for the N5-10°W fracture zone and the perpendicular direction of 10,000 gallons per day
per foot (gpd/ft) and 1,000 gpd/ft, respectively. It was assumed that the N5-10°W fracture zone may
have as much as a 10-fold greater transmissivity than the perpendicular direction. The Cooper-Jacob
solution used a discharge of 200 gallons per minute (gpm), a storage coefficient of 0.01 (a combined
value for the fractured bedrock and overburden storage capacities), and a drawdown criterion of
0.10 foot. The resulting protection zone formed an oval extending 2,200 feet on either side of the test
well along the N5-10°W fracture zone and 700 feet on either side of the test well perpendicular to the
zone.
At the time of the preliminary delineation, the exact location of the test well was
undetermined. Therefore, a 300-foot-wide zone corresponding to the potential width of the lineament
(the fractured, faulted contact zone in Figure 5-2) was chosen as the center of the WHPA with the test
well at its center. To either side of this zone, 700 feet of protection was added (calculated using the
Cooper-Jacob solution), resulting in a 1,700-foot-wide protection zone (Figure 5-5). The 2,200-foot-
long protection zone on either side of the test well along the N5-10°W fracture zone was modified for
regulatory reasons. The north-northwest side of the WHPA was shortened from 2,200 feet to
1,800 feet because it extended approximately 400 feet beyond the town boundary. Since the city of
Dover has no regulation authority in the adjacent town, the preliminary WHPA boundary coincided
with the northern town boundary. The preliminary WHPA consisted of a rectangle with dimensions
of 4,000 feet (north-northwest) by 1,700 feet.
CONTAMINANT SOURCE ASSESSMENT
Source Inventory List
A number of organizations including the EPA Region I office in Boston, the New Hampshire
Department of Environmental Services (NHDES), and the New Hampshire State Department of
Agriculture provided information relating to possible ground-water contamination sources in the
region. Additionally, other sources of information pertaining to possible ground-water pollution were
identified from contacts with Dover authorities (for example, fire department and public works
officials), local landowners, and construction site personnel to obtain information on potentially
contaminated sites. At the NHDES, the Waste Management Division, the Underground Storage
Tank (UST) Program, and the Ground Water Protection Bureau provided sources of information.
After a detailed review of available information had been completed, known sites of potential or actual
contamination were checked by means of field surveys. This study identified 80 sites throughout the
city (excluding the downtown, developed areas) as potentially hazardous. These include landfills, salt
storage areas, gasoline stations, petrochemical storage facilities, agricultural concerns, and industrial
parks.
Few contaminant sources pose an immediate threat to a future production well at the
Blackwater Brook site (Figure 5-6). This is one of the primary reasons for the site selection. A listing
of the existing and potential sources of ground-water contamination in proximity to the site
(Table 5-1) includes existing road de-icing operations, spills, and illegal disposal associated with major
5-9
-------
* SEWAGE
TREATMENT
PLANT
-•
GAS
_ STATION
-" '
TILE
SUPPLY*
r
COUNTY HOME
(USTs)
_ AUTOMOTIVE
<• SERVICE
FUEL j
STORAGE
GAS
STORAGE 'STATION
Figure 5-6. Contaminant source assessment map, Blackwater Brook site, Dover, New Hampshire.
(USTs = underground storage tanks)
5-10
-------
TABLE 5-1. POTENTIAL SOURCES OF GROUND-WATER CONTAMINATION AND
DEGRADATION IN PROXIMITY TO THE BLACKWATER BROOK SITE, DOVER, NEW
HAMPSHIRE*
Source Type
Existing Sources
Potential Sources
Commercial and Industrial
Residential and Waste
Management
Naturally Occurring
Airport
Automotive services
Landfill
Quarry
Road de-icing operations
Unanticipated toxic and
hazardous spills
Household hazardous products
Septic systems
Storage tanks
Wells (operating and
abandoned)
Iron and manganese (Berwick
Formation source)
Radon (quartz monzonite
source)
Sodium and chloride (trapped
sea water or a marine clay
source)
Uranium (quartz monzonite
source)
Auto repair, paint, and
machine shops
Car washes
Dry cleaners
Gas stations
Golf courses
Household lawns (chemical
applications)
Sewer lines
Storm water drains and
retention basins
Swimming pools (chlorine)
Transfer stations
Wastewater treatment facilities
*Modified after U.S. EPA, 1990, p. 330.
transportation routes. A landfill exists about 3 miles northwest of the site, and a rock quarry lies
about 2 miles to the southwest. A few residences currently exist near the site. The primary residential
threats include storage and disposal of household products (for example, heating oil), septic system
leachate, and abandoned or operating wells. Naturally occurring iron, manganese, radon, sodium, and
chloride are concerns for municipal water treatment.
It is anticipated that a significant amount of land near the well site will be zoned for
residential use. Lawn chemicals, chlorine from swimming pools, and leakage from sewer lines, storm
water drains, retention basins, and municipal waste transfer stations (Table 5-1) may add to existing
residential threats. Light-commercial and light-industrial facilities that support residential
developments such as auto repair shops, laundromats, car washes, golf courses, medical institutions,
5-11
-------
dry cleaners, and warehouses represent other possible future threats to the well site. However, the city
of Dover's ground-water protection ordinance ensures that controls against hazardous and toxic
sources will be implemented to protect the new water supply well at the Blackwater Brook site. The
advantage of implementing wellhead protection prior to urbanization is that the city can specify
growth locations and development procedures that will minimize later problems with ground-water
quality (U.S. EPA, 1990a).
Source Characterization and Prioritization
The degree of the perceived threat and the need for controls determined the prioritization of
existing contaminant sources at the Blackwater Brook site. The highest priority threat is a spill of
hazardous substances along the Spaulding Turnpike, a heavily traveled, north-south thoroughfare
(Figure 5-6). Another threat is salt contamination from highway de-icing activities. Because the area
is rural and undeveloped, illegal dumping in open fields along the numerous access roads to remote
areas (for example, old logging trails) and possibly along the Turnpike is also a priority concern.
A quartz monzonite formation mapped in the area (Novotny, 1968) is known to contain
pockets of water with high radon levels from elsewhere in the region. The other major bedrock unit is
the Berwick Formation, a "rusty" schist characteristically high in iron and manganese. Sodium and
chloride levels are unusually high in the bedrock aquifer and are believed to be from trapped sea
water. It is anticipated that these levels will decrease in time as water is discharged through
production. These naturally occurring constituents will be the targets of ambient" trend monitoring to
determine if water treatment is necessary.
The landfill located approximately 3 miles to the northwest (Figure 5-6) is not considered to
be a major threat. It does not occur along the identified hydraulically conductive fracture zones in the
bedrock aquifer. Contaminants from a 3-mile distance are unlikely to impact a well at Blackwater
Brook unless their source occurs along a preferentially conductive flowpath.
At present, the nearest septic leachate field, situated 1,000 feet from the test well, would
probably contribute negligible amounts of nitrate to the future well. However, as development
encroaches on the WHPA, residential threats will increasingly take on more importance as a result of
density impacts. Permeable sand and gravel deposits approximately 1,000 feet to the northeast and
east of the test well, in areas where residential development could occur, could provide a direct avenue
for contamination to the bedrock aquifer.
5-12
-------
RESEARCH MONITORING PROGRAM
DATA BASE LIMITATIONS
Prior to exploration work at the Blackwater Brook site, comparatively little data existed
regarding aquifer parameters, flowpath directions, and flow rates in the area. However, the region was
reasonably well described in terms of its geologic and geophysical characteristics. Information is
available concerning the surficial geology (mapped in detail at the 1:24,000 scale) and the bedrock and
structural geology, including an extensive data base of fracture fabric analysis, photolineament analysis,
and structural interpretations developed as part of the ground-water exploration process. Published
well logs on file with the USGS, the NHSGS, and the NHDES were compiled, along with files of well
logs from drillers. From this data base, the objectives and tasks for the research monitoring effort
took form.
MONITORING OBJECTIVES
The research monitoring program leading to long-term monitoring of the well site area had
two basic objectives. The first objective was to characterize the hydrogeology of the area, including a
detailed examination of physical and chemical parameters. This information-gathering phase resulted
in the delineation of a wellhead protection zone based on knowledge about the hydrogeology of the
region and site.
The second objective of the research monitoring program was the development of a
technically defensible and reliable long-term monitoring plan to protect the public water supply. Both
physical and hydrochemical monitoring were used to achieve this objective. Physical monitoring
examines the behavior of the aquifer over time to acquire a better understanding of the impact of
water withdrawal as well as ground-water flow dynamics. Information collected from physical
monitoring provides valuable data to support well and aquifer performance and management.
Hydrochemical monitoring focuses on the chemical character of the ground water. Ambient
or background hydrochemical monitoring examines the system on a regional basis, recording changes
in ground-water chemistry over comparatively large intervals of time and space. Source-assessment
hydrochemical monitoring concentrates on impacts to ground-water quality from an identified
potential pollution source near the wellhead. An early-warning hydrochemical monitoring system
provides a mechanism for initiating contingency plans to protect the public from dangerous exposure
to pollutants in the well water if a contamination event should occur (Canter et al., 1987).
Some of the work undertaken in the Dover project is not commonly identified as "monitoring"
per se. In particular, the hydrogeologic characterization (information-gathering) phase is frequently
not considered part of an overall monitoring effort. However, to determine effective strategies for
5-13
-------
field monitoring, the rate of propagation of both anticipated and unanticipated contaminants must be
understood and the physical properties of the site known. An effective research monitoring program
requires both hydrogeologic characterization and the development of long-term monitoring strategies
(Pfannkuch, 1982).
RESEARCH MONITORING TASKS
The process of ground-water exploration is one of developing and refining a conceptual
hydrogeologic model of a designated site and the surrounding area. The exploration for a test well site
at Blackwater Brook employed photolineament analysis, fracture fabric mapping, and surficial .
geologic and geomorphic mapping. Geophysical surveys helped further isolate the structural
anomalies that form the major conduits of flow. Wellhead protection area delineation involves
essentially the same process of building and revising a conceptual hydrogeologic model. Accordingly,
the scope of work for this research program involved similar tasks, including:
Lineament analysis of high- and low-altitude photography
Geologic mapping, including fracture fabric analysis of bedrock and geomorphologic
analysis of watersheds
Surface geophysics to isolate suspected structural anomalies
Test drilling
Observation well and piezometer installation
Borehole geophysics
Chemical sampling of surface and ground water
Aquifer testing
Dye tracing
Data compilation and interpretation
DATA ACQUISITION AND INTERPRETATION
Lineament Analysis
Extensive use of aerial photography and images was employed at the Blackwater Brook site,
with a number of scales and image types for lineament analysis ranging from high-altitude, 1:126,000
color infrared images to 1:7,200 black-and-white photographs. These were examined with a mirror
stereoscope, and lineaments were drawn on mylar overlays attached to the photographs. The image
analyst discriminated among strong, moderate, and weak lineaments.
Two lineament orientations predominate in the area (Figures 5-3 and 5-7A). One trends
approximately N5-10°W and was the focus of the original exploration efforts. Part of its intensity and
visibility may be due to glacial enhancement of the original fracture pattern. The other predominant
trend, N60°E, is much more subdued but appears to be significantly larger in extent. Additional
photolineaments trend N45°E (parallel to the regional foliation), N30°W (perpendicular to the N60°E
trend), N70°E, and N80°E (Figures 5-3 and 5-7A).
5-14
-------
n = 31 READINGS
LINEAMENTS
n = 149 READINGS
FRACTURES
(A)
(B)
Figure 5-7. Rose diagrams of (A) lineament mapping, and (B) fracture fabric of bedrock,
Blackwater Brook site, Dover, New Hampshire. (N = north; n = total number of observations.)
Geologic Mapping
Existing geologic maps and publications (Lyons et al, 1986; Lyons et al., 1982; Novotny, 1968;
Bradley, 1964; and Billings, 1956) provide a general description of bedrock geology and structural
features. To better define the relationship between lineaments seen on the photographs and mappable
structural conditions on the ground, rock fractures were measured at several outcrops near the
proposed wellfield. Fractures observed at outcrops were measured with a pocket transit, and the data
were reduced and compared with lineament results (Figures 5-3, 5-7A, and 5-7B).
The fracture fabric rose diagram (Figure 5-7B), a statistical display of fracture strikes, shows
that the N5-10°W trend is well represented. Strong northeasterly trends are also present. With the
knowledge of a fault outcrop in a quarry 2 miles southwest of the site with a similar strike
(Figure 5-6), attention focused on the N60°E lineament as a potential water-bearing zone. The quarry
fault has a N54°E strike and cuts a quartz monzonite formation intruded by a variety of pegmatite and
quartz veins. The N60°E photolineament coincides approximately with the interpreted fault projected
northeastward from the quarry. Extensive alteration of the rocks, particularly along fractures observed
at the quarry outcrop and in drill cuttings at the Blackwater Brook site, indicates weathering and
possibly hydrothermal alteration of the rocks. Quartz monzonite at the quarry contains quartz veins
with pyrite and extensive iron oxide staining along fractures.
5-15
-------
Rather than one discrete fault connecting the quarry with the Blackwater Brook site, it is
likely that both the quarry and the well site lie within a fault zone. The interpreted fault underlying
the Blackwater Brook site may be one of a system or network of northeast-trending faults linking the
Norumbega Fault System in Maine with the Clinton-Newbury Fault System in Massachusetts (Lyons
et al., 1982; Gotten, 1988). As such, the interpreted fault and contact zone could be a previously
unmapped link in a major fault system trending approximately northeast in eastern New England. If
this fault is part of a regional fault system, the zone probably contains a number of individual faults
and a high fracture density that would promote ground-water flow.
The lineament and interpreted fault zone also coincide with a geologic contact between the
Devonian quartz monzonite and the Precambrian or Cambrian Berwick Formation (gray phyllite,
biotite schist, gneiss, quartz-mica schist, and calcareous schist). The contact, as mapped by Novotny
(1968), trends about N46°E. However, the contact is uncertain and probably exists as a zone in which
the two rock types interfinger and are cut into a melange of fault and fracture-bounded slivers
(Figure 5-2).
In addition to the lineament and fracture fabric data collection and analysis, the area was
analyzed for local drainage basin size, soil type, and recharge potential. Surficial geologic mapping
done by the NHSGS and the USGS aided in the geologic characterization (Mack and Lawlor, 1991).
A variety of surficial deposits occur at the site and include glacial till, swamp deposits, and marine silt,
sand, and clay (Figure 5-4). Drilling logs from the test well indicate that marine clay is underlain by
permeable sands and gravels (marine and glacial outwash deposits). These permeable deposits, which
outcrop along the fringes of the Blackwater Brook watershed (Figure 5-4), are capable of storing
water and recharging the bedrock aquifer. With the potential for drilling into a major structure
(a fault) that cuts across a number of small surface-water drainage basins, work focused on assessing
the surficial deposits that could potentially recharge the fault zone.
Surface Geophysics
Geophysical surveys initially targeted the N5-10°W and N60°E lineament features (Figure 5-3)
using instruments that measure microgravity, magnetics, and electromagnetics. The location of the
geophysical survey lines is shown in Figure 5-6. Attention centered on the N60°E-trending lineament
when it was recognized that this lineament coincides with a pronounced magnetic high (Figures 5-3
and 5-8). From a lower base level representative of the less magnetic quartz monzonite, the magnetic
signature increases approximately 600-gammas (Y) before stabilizing at a higher base level
characteristic of the iron-rich Berwick Formation (Figure 5-8). While the broad north-to-south
increase in magnetic field can be explained by the change in rock formations, the explanation for the
600-y peak is less certain. Possible explanations include a steeply-dipping mafic dike (86° to the
southeast) along the contact zone (Figure 5-8); a horizontal or subhorizontal mafic dike that
terminates at the contact zone; or, in part, alteration of chlorite to magnetite along the faulted contact
zone. The N5-10°W fracture zone showed no geophysical signature on the surveys performed. The
results of the surface geophysics, coupled with the evidence from lineament and fracture fabric
analysis, warranted test drilling to determine whether the site could produce sufficient quantities of
water for municipal needs.
5-16
-------
N
2,900
— 2,800
a 2,700
£ 2,600
o
JZ 2,500
LU
O 2,400
s 2,300
2,200
Q-
LU
OBSERVED
- FITTED
BASE LEVEL
HORIZONTAL DISTANCE (ft)
BERWICK
FORMATION
MAFIC DIKE
(MODELED)
Figure 5-8. Plot of magnetic geophysical survey data with fitted model and mafic dike hypothesis,
Blackwater Brook site, Dover, New Hampshire.
Test Drilling
A test well (TW) was drilled near the N60°E lineament, near the peak of the magnetic
anomaly (Figure 5-3). The well penetrated 32 feet of clay and 6 feet of permeable sand and gravel
overburden (Figure 5-9), which were cased off. The initial bedrock encountered at 40 feet was the
quartz monzonite; next, the Berwick Formation was logged at a depth of 107 feet. The upper portion
of the Berwick Formation encountered is differentiated as the phyllitic Gonic Member of the Berwick
Formation. Significant water-bearing fractures were encountered in the Berwick Formation at 220 feet
and at 342 feet. Evidence of these producing fractures can be seen as greater-diameter openings in the
caliper log and as dipping fractures recorded in the video camera log (Figure 5-9). The well was .
developed with airlift techniques for approximately 12 hours, removing cuttings that had fallen into
the fracture zones. Final airlift yields exceeded 300 gpm from the 6-inch borehole.
5-17
-------
CALJPER
LITHOLOGIC
LEGEND
LITHOLOGIC LOG
><><:-
!>,»•<> *»J
MARINE CLAY,
GRAY
COARSE SAND
AND GRAVEL
QUARTZ
MONZONITE
GONIC MEMBER,
BERWICK
FORMATION
BERWICK
FORMATION
-^,. ..g^ FRACTURE/
-~ "^ FRACTURE ZONE
VIDEO CAMERA LOG
OPEN,>45°
DIPPING FRACTURE
OPEN. <45 °
DIPPING FRACTURE
0 CAVITY
QUARTZ VEIN OR
CLOSED, <45°
DIPPING FRACTURE
1 1 VERTICAL FRACTURE
' 0
20
40
60
80
100
120
140
240 •
260
280 •
300 •
320 •
340 '
360 '
380 '
400
420
?
6 6.5 7 7.5
DIAMETER
(in)
VIDEO
CAMERA
AIRLIFT
YIELDS
BOTTOM F
'" CA NG ""
•1-2 gpm
-7gpm
• 33 gpm
-40gpm
• 44 gpm
•150 gpm
• 300 gpm
N E S W N
AZIMUTH
Figure 5-9. Diagram of caliper, lithologic, and borehole video camera logs of the test well,
Blackwater Brook site, Dover, New Hampshire.
5-18
-------
No mafic bedrock was encountered during drilling. It is possible that a steeply dipping mafic
dike, wide enough to cause the magnetic anomaly but narrow enough to be missed by drilling, exists at
the site. Recent blasting at the quarry 2 miles southwest of the site (Figure 5-6) has exposed a large
body of gabbro (mafic rock). It may be that the sharp magnetic anomaly reflects, in part, shallow-
depth alteration of chlorite to magnetite along the faulted contact. Examination of drill cuttings from
the wells show ample evidence of iron-bearing mineralization (pyrite), as well as the magnetic
minerals pyrrhotite and magnetite in the sections of borehole corresponding to the Berwick
Formation. No magnetic minerals were found in the cuttings from the quartz monzonite.
Alternatively, the step-like anomaly may be a deeper-seated feature. The geologic significance of the
magnetic increase remains speculative.
Well and Piezometer Installation
Bedrock Observation Wells-
With the success of the test drilling, the city elected to pursue the Blackwater Brook site as a
future municipal water source. Four observation wells were drilled into bedrock: two, R3 and R4,
along the N60°E trend and two, Rl and R2, in an orthogonal orientation along N30°W (Figure 5-10).
All of these wells were drilled to approximately 400 feet (Table 5-2), and well construction was
performed according to standard operating procedures outlined in the American Water Works
Association publication "Standard for Water Wells" (1988) (Figure 5-11). The domestic well, Dl,
located about 2,000 feet north of TW (Figure 5-10), was used for water-level monitoring during
aquifer testing. It penetrates about 10 feet of overburden and then enters bedrock. The well was
drilled to about 350 feet and is estimated by drillers to yield 10 gpm.
Lithologic logs, obtained during drilling and by subsequent description of drill cuttings at a
5-foot interval, reveal that all the bedrock wells penetrate both the Berwick Formation and the quartz
monzonite. The mapped contact between the two formations (Lyons et al., 1986; Novotny, 1968;
Billings, 1956) coincides approximately with the N60°E lineament and is evident in all of the bedrock
wells (Figure 5-10). Wells Rl and R2 along the N30°W direction, penetrate fracture zones that bear
water. Wells Rl and R2 airlifted 170 and 300 gpm, respectively. Thus, the contact between the
Berwick Formation and the quartz monzonite is actually a zone at least 300 feet wide (the distance
between wells Rl and R2, transverse to the N60°E lineament). This zone contains an interfingering,
or a melange, of the two rock types illustrated in the lithologic log of TW (Figure 5-9). In some wells
(TW, Rl, and R4) the Gonic Member of the Berwick Formation is distinguishable, and 20- to
40-foot-thick pegmatites are present in wells R3 and R4. In all the bedrock wells, quartz veins and
monzonite intrusions from a few inches to a few feet thick are present.
Similar to TW, producing fracture zones in the bedrock observation wells are highly discrete.
During drilling, the presence of larger cuttings and drill stem drops indicated fractures or fracture
zones at one or more depths in all the bedrock wells. In all wells except R4, significant water
production during drilling and airlifting indicated flow from one or more discrete fracture zones.
Caliper logs, exemplified by Figure 5-9, showed increases in wellbore diameter at major fracture zones.
These fracture zones correlate with fracture zones and water-producing zones indicated on lithologic
logs.
5-19
-------
QUARTZ MONZONITE
BERWICK FORMATION
TW = TEST WELL
D1 = DOMESTIC WELL AND NUMBER
—140
LEGEND
O1 = OVERBURDEN WELL AND NUMBER
R1 = BEDROCK WELL AND NUMBER
TOPOGRAPHIC CONTOUR
(INTERVAL IS 20 FEET)
Figure 5-10. Locations of the test well and monitoring wells, Blackwater Brook site,
Dover, New Hampshire.
5-20
-------
3
{2
a §5
S |ji
o K
|1
o [>
S o
250
w *•
^^ ui4
nr C
33^
o< O
w o
>ff!
o m
Q e5
25 W
W ^
w o
8§
o2"
*— ( K
Q fe
|S
§4 S«>
U.D
TAB
CONSTR
4-»
1 | S g1
S O COG, "^
•a *§ o P ^i-
J 2 « CM
^ Q
jll
2 S
vj
s
.i'S
QH
o —
'5b -S
,2 S
o S
T3
li^s
li^
o,
Q S^
3 ^
f2
1
V
^ o\ a\ f*$ ^_ *«. . _*. Q. 53
t~- >n«oco«nt'S co T— i T-I ,O
"^35
v^ cj c? ^? ^o r^ *n ^^ ^D t"^
CO-S®lOt~;C~;'/^ t-J Oq to *T
^^ co C1^ ^P *n to oo cs ON ON
-•.
4- 1 4- « - « 1
^o QJ *n vo JM feb
1 1 1 1 1 fl f ^ f]| ll
S «««.pq^|^|^g|*|g
o cs o o «n o
"^ ^coS g? 2^- ,2^ 2^.^
o opo>no p p p p
^! IpcJj^i^!^ ""* ^ °4 ^
{^ OiC^OiOiO O O O Q
d
J3
.§
"E.
'" Q,
1
2^
"tf
j§
1
g<
f
1
o
«
II
*i
if ^
Is
«\s
S3 g
?2
J3
a °
g-'w
jviations: j
ic water le^
5B
Jl 00
•< *
1
8
f— ^
B
"o.
S
.S
Q
-------
OVERBURDEN
WELL
BEDROCK
WELL
LOCKING CAP
CEMENT
4-IN STEEL CAS ING
BENTONITE
NATURAL FILL -=^c
2-IN SCH 40 g="^-
PVC RISER ?
BENTONITE
FILTER PACK SAND -f
SCH 40
PVC WELL SCREEN
(0.01-IN SLOT)
END CAP
BENTONITE -
LOCKING CAP
CEMENT
>'£ 6.25-IN STEEL CASING
OPEN BOREHOLE
BEDROCK
Figure 5-11. Schematic diagram of overburden and bedrock monitoring well designs, Blackwater
Brook site, Dover, New Hampshire. (ANSI/AWWA A100-84;
American Water Works Association, 1988.)
5-22
-------
The importance of surface geophysical results in this environment was particularly apparent
during the drilling of the observation wells along the N60°E zone. One of the two bedrock wells
located along the N60°E zone, R3, was sited approximately 600 feet N60°E of TW, based on results of
the magnetic survey. Well R3 airlifted more than 150 gpm and demonstrated interconnection to TW
when the water level in TW dropped approximately 15 feet from its normal flowing-artesian condition
during development of well R3. The second well, R4, was sited approximately 365 feet west of TW
without rigorous use of the geophysical results. It is 425 feet deep, with an airlift yield of only
1.5 gpm. A fence diagram of the site geology is presented in Figure 5-12.
Overburden Observation Wells--
Overburden wells (Ol, O2, O3, and O4) were placed next to bedrock observation wells
(Figure 5-10). The surficial geologic mapping conducted by the USGS, along with field checks during
the exploration process, resulted in a high degree of confidence in the conceptual model of surficial
material deposition and distribution. Overburden wells were installed primarily to provide water-level
measurements in the bedrock and overlying aquifers at the same locations. A secondary goal was to
refine the model for surficial material deposition and distribution. All well construction (Figure 5-11)
was performed according to standard operating procedures outlined by the American Water Works
Association (1988).
The overburden material was sampled using a split-spoon device. Overburden thicknesses
range between 19 and 45 feet (Table 5-2) and include a diversity of material: coarse sand and gravel,
marine clay, weathered bedrock, fine sand, topsoil, rocks, and boulders. The marine clays are 30 or
more feet thick in wells TW, Ol, and O2, and thin to the east and the west (15 feet in well O3; 5 feet
in well O4) (Figure 5-12).
Borehole Geophysics
Fractures provide the most important pathways for ground-water flow in bedrock aquifers. •
An understanding of the character of bedrock fractures (for example, depth, orientation, and
morphology) is vital to conceptualizing the nature of flow in such aquifers. The primary means of
obtaining information on fractures is through borehole logging, particularly with geophysical tools.
Borehole geophysics can then be correlated with other hydrogeologic investigations, such as aquifer
testing and dye tracing, to better understand the flow system. The suite of borehole logs and
geophysical surveys acquired in this study include the following:
Video camera logs
Flowmeter surveys
Hydrophysical logs
Borehole acoustic televiewer logs
Gamma logs
5-23
-------
QSUIB y) NOU.VA313
5-24
-------
Video Camera Logging--
All bedrock wells were examined with a 4.75-inch-diameter borehole video camera. The
camera has two lenses: one capable of looking directly downhole, and one that rotates to examine the
sides of the borehole. Low-rate pumping (5 to 10 gpm) of the well during logging ensured a clear
picture by minimizing turbid water. Another advantage of pumping is the ability to see particles of
rock (for example, drill cuttings) moving into the borehole from large fracture zones. The camera is
oriented by suspending a rope down the magnetic north side of the wellbore before the tool begins to
descend. Visual inspection of the videotapes suggests that the camera generally did not rotate with
respect to the wellbore. If the camera did spin, it did so slowly. Such spinning would invalidate
interpretations of fracture orientation, but not of fracture depth or morphology.
The video camera revealed discrete water-bearing fractures and complex zones of intense,
intersecting vertical and dipping fractures (Figure 5-9). Major increases in water production,
determined during drilling with airlift tests, occurred where the borehole intersected high-angle
water-bearing fractures (Figure 5-9, at 220 and 342 feet). Low-angle fractures do not appear to be
major flow conduits in TW.
Flowmeter Surveys--
Two heat-pulse flowmeter surveys were conducted in the bedrock wells at the Blackwater
Brook sites: one using a commercially available tool and, later, one supplied and operated by the
USGS, Denver office. Analysis of the data attest to some of the complexities of dealing with flow
through fractured bedrock. Water enters and exits the borehole at very discrete intervals rather than
- uniformly along its length. Portions of a borehole may show downward flow, while other sections
exhibit upward flow. Correlation of ground-water movement between boreholes is complicated
because of the large distance involved and convoluted flowpaths. Flow directions within the fracture
system, as determined from the combination of the two flowmeter surveys, are graphically depicted in
Figure 5-12.
Results were obtained with commercially available vertical and horizontal flowmeter
equipment for well R2 during the pumping of TW at about 30 gpm. From the surface to about
270 feet, flow in well R2 is downward. Downward flow generally increases from less than 0.10 gpm at
the surface to more than 0.75 gpm at 265 feet, as fractures at several discrete depths contribute water
to downward flow in the wellbore. From 280 to 300 feet, upward flow occurs at about 0.5 gpm. Data
were not obtained below 300 feet. A net outflow of over 1.3 gpm must occur between 270 and
280 feet. Drilling and caliper logs indicate a major water-bearing fracture zone at this depth, and the
horizontal flowmeter indicates flow in a northwesterly direction.
The USGS flowmeter, equipped with inflatable packers, was used to survey wells TW, R2, R3,
and R4. This tool can only be used to perform vertical-flow surveys. Preliminary results indicate that
the TW has upward flow from the main producing fracture at 342 feet. Nearly stagnant conditions
prevail below 342 feet. Under static conditions (no pumping, with the natural, 5-gpm artesian flow in
TW), well R2 receives water from several fractures between 100 and 270 feet. Downward flow occurs
in this well to 275 feet. Pumping TW at 16.5 gpm intensifies the same effects. Under static
conditions, flow in well R4 is undetectable. When TW was pumped at 16 gpm, a barely detectable
downward flow of 0.01 gpm was measured between 0 and 200 feet in well R4. When well R4 was
5-25
-------
pumped at 1.5 gpm, the water level dropped rapidly, but upward flow was detected with inflow at 90 to
100 feet. Inflow to well R3 occurs between 350 and 380 feet; water exits the well between 260 and 300
feet, with weak upward flow between these two zones (Vernon et al, 1993)
The major direction of flow appears to occur along the well R2 and TW alignment (N30°W),
with lesser amounts along the N60°E trend. Well R2 probably contributes water directly to TW from
the outflow in R2 at 275 feet (Figure 5-12). Flow moves down the borehole from the surface and up
the borehole from the bottom to the 275-foot level hi well R2. From this information, coupled with
analysis of the acoustic televiewer logs, a direct connection is seen to exist between wells R2 and TW,
with water entering TW at the major fracture at 342 feet (Figure 5-12). Wells R3 and R4 also show
interconnection with TW, but precisely which fractures or system of fractures intersect is difficult to
ascertain.
Hydrophysical Logging—
Hydrophysical logging involves replacing the standing column of formation water with
uniformly deionized water, and then profiling the physical and chemical changes of the replacement
water. These changes occur when contrasting formation water is drawn back into the wellbore by
continuous, low flow rate pumping (Pedler, 1991). Downhole wireline water quality instrumentation,
which simultaneously measures fluid electrical conductivity (FEC), temperature, pH, and
oxidation-reduction potential (Eh), is used to log the physical and chemical changes of the
replacement water. The FEC logs obtained from these wellbores are used to evaluate the location,
rate of production (hydraulic conductivity), and FEC of the formation water for each of the identified
hydraulically conductive intervals (Pedler, 1991).
The wellbores tested (TW, R2, R3, and R4) display numerous, discrete producing intervals
(Table 5-3). Vertical flow within two of the wellbores, TW and R2, was observed under ambient
conditions. The TW displayed very strong upward flow (up to 6 gpm), with the major inflow at
345 feet Well R2 displayed moderate downward vertical flow (as much as 0.5 gpm), with fluids
exiting the wellbore at the 274-foot interval (Pedler, 1991).
Three types of formation water were observed at the Blackwater Brook site based on the
results of the temperature-corrected FEC logs. Type I water is of low FEC, with values of 100 to
150 microSiemens per centimeter (nS/cm); Type II water is of moderate FEC, with values of 450 to
600 uS/cm; and Type III water is of high FEC, with values of greater than 800 nS/cm. In general, a
trend of increasing FEC with depth was observed in each wellbore (Table 5-3) with abrupt, not
continuous increases.
The abrupt changes in FEC with depth may be attributed to a combination of local,
intermediate, and regional flow systems interacting with the discrete character of flow within the
bedrock fracture system (Toth, 1963). The water quality of the local system more closely reflects that
of rain water with its low conductivities. For example, the fracture at 40 feet in well R3 lies just below
the end of the well casing. Well R3 lies near permeable overburden that receives direct precipitation,
and its low FEC reading probably reflects recent recharge. The local system extends to as deep as
200 feet in TW (Table 5-3), but does not appear in well R2 or in well R4, indicating that flow within
the fracture system is extremely discrete. The intermediate flow system, interpreted as Type II water,
also demonstrates isolated flow within the bedrock aquifer. Type II water is seen at 345 feet in TW
5-26
-------
TABLE 5-3. FLOW RATES AND FLUID ELECTRICAL CONDUCTIVITIES AT
DISCRETE ZONES WITHIN BOREHOLES FROM HYDROPHYSICAL LOGGING,
BLACKWATER BROOK SITE, DOVER, NEW HAMPSHIRE
Well Depth
(ft)
TW 194
199
218
345
362
R2 106
128
156
172
230
265
274
R3 40
358
R4 100
198
245
365
Flow Rate
(gpm)
0.17
0.17
0.17
6.65
0.84
0.21
0.26
0.05
0.16
0.22
0.33
0.52
0.07
3.43
0.10
0.10
0.19
0.03
FEC*
OS/cm)
130
130
130
580
780
500
500
500
500
500
soot
500t
150
460
1,000
600
1,250
900
Abbreviations: FEC = fluid electrical conductivity; gpm = gallon per minute; (iS/cm = microSiemens
per centimeter.
* Flow rate and FEC values are depth-specific.
t FEC increases from 5 to 500 [iS/cm during pumping, then stabilizes at 500
and at 358 feet in well R3. It comprises the entire borehole in well R2, but is completely absent in
well R4. The regional flow system is reflected by the Type III water, and may represent older, very
slow moving water recharged at greater distances than the local or intermediate flow systems. The
higher FEC readings may also be influenced by the presence of trapped sea water, vestiges of periods
of marine inundation during the Pleistocene era. High readings of FEC throughout the length of well
R4 reflect the lack of interconnection of fractures that could bring in younger, more recently
recharged water of lower conductivity.
Borehole Acoustic Televiewer Logging--
The borehole acoustic televiewer is a wireline tool that emits an acoustic signal and receives
the signal's reflection from the wellbore. The presence, attitude, and, sometimes, morphology
(fracture width and degree of filling) of fractures can be determined using the televiewer (Paillet et al.,
1987; Plumb et aL, 1985; Zemanek et al, 1970).
5-27
-------
The televiewer logs show fewer fractures than those shown on the video camera logs.
However, the fracture orientations can generally be determined with more confidence using the
televiewer, because the instrument is oriented and controlled, whereas slow spinning of the video
camera cannot be ruled out. In addition, the televiewer tends to select the more prominent fractures.
The televiewer showed a number of distinct fractures and fracture sets in each well surveyed (TW, R2,
R3, and R4). Analysis of televiewer logs indicates good correspondence with the occurrence of
fractures as interpreted by other means, such as drilling, caliper logging, video camera logging, and
flowmeter surveying.
The televiewer logs reveal major fractures in TW and well R2 at 342 and 275 feet, respectively.
In both cases, the fractures strike approximately N20°E and dip about 45°W. According to flowmeter
results, 342 feet is the main depth of inflow to TW, and 275 feet is the main depth of outflow from
well R2. Thus, it is likely that one fracture or set of fractures directly connects well R2 and TW,
intersecting well R2 at 275 feet and TW at 342 feet. Because the fractures appear to be dipping to the
west on the televiewer logs and TW lies west of well R2 (Figures 5-10 and 5-12), the fracture may
intersect TW at greater depth. If the estimated N20°E strike and 45°W dip for this fracture or
fracture set is correct, the fracture (or fracture set) would intersect TW at about 350 feet—the
approximate depth of the main producing fracture (Figure 5-12). The discrepancy in the exact depth
is within a ±20° precision error of fracture orientation determined from the televiewer logs (Vernon et
al, 1993)
The televiewer log for well R4 shows a prominent fracture at 356 feet, but no water is
produced from this fracture. While this fracture appears to have a large aperture, it does not show
evidence of alteration or reflective infilling minerals, as do the producing fractures in TW and well R2.
Also, the fracture at 356 feet in well R4 occurs as an isolated fracture rather than a fracture set. If the
water-bearing fracture that connects TW and well R2 has a strike of N20°W and a dip of 40°W, its
projection would intersect well R4 below its completion depth.
The televiewer log for well R3 suggests the existence of a weathered fracture zone between 350
and 380 feet in depth, which is the major water producing zone for the well (Figure 5-12). This zone
also appears to contain fractures that dip to the west (Vernon et al., 1993)
Gamma Logging--
Natural gamma logs depict gamma-ray emissions from different depths in the wellbore.
Different rock types produce characteristic gamma-ray counts. In the Blackwater Brook wells, the
Berwick Formation generally produces about 300 counts per second (cps), and the quartz monzonite
generally produces about 600 cps. Sharp changes in gamma-ray counts generally accompany
formation changes.
The gamma-ray logs tend to confirm the depths of bedrock contacts in the lithologic logs. A
gamma-ray count of several thousand per second at 225 feet in well R4 corresponds to the presence of
a low-grade uranium ore in pegmatite logged at that depth. This uranium ore is not considered to be
a water quality threat to a future production well that would be sited next to TW. Nevertheless,
periodic monitoring of radioactive constituents at the production well may be required under the Safe
Drinking Water Act (SDWA).
5-28
-------
Aquifer Testing and Characterization
Ambient Water-Level Monitoring—
Ambient water-level data for the bedrock and overburden wells were gathered before aquifer
testing. The purpose of this task was to document water-level changes under nonstress conditions.
Water levels vary for several reasons, including recharge events, barometric pressure differences, and
diurnal variation. Aquifer-test data should be corrected for natural fluctuations in water level prior to
analysis. Water-level trends were recorded before and during aquifer testing in well Dl,
topographically upslope from TW, and in wells Rl, R2, R3, Ol, O2, and O3 (Figure 5-10).
Transducers were not available to place in wells R4 and O3 to monitor ambient water levels. Water-
level monitoring at 3-hour intervals began 16 days before constant-rate pumping, changing to hourly
intervals 2 days before testing. Water levels in well Dl were monitored hourly throughout pumping
and recovery.
Water levels in well Dl fluctuated substantially, as might be expected for a low-yield (10 gpm)
domestic bedrock well (Figure 5-13). The specific capacity of a low-yield bedrock well can be
exceeded during periods of frequent, maximum usage resulting in large drawdowns and fluctuations
such as those typified in Figure 5-13. A slight drop in the overall trend of the water levels from the
beginning of monitoring to the start of aquifer testing can be explained by the fact that the property
changed ownership. The single owner sold the property to a couple, and their higher level of water
use may be reflected in the amplitude of the curve. However, neither the onset of aquifer testing at
Blackwater Brook nor its cessation produced meaningful changes in the overall trend of water levels in
well Dl. Water-level data in Figure 5-13 indicate that the 15-day pumping and recovery test at TW
showed no response in well Dl.
185 -
G-
2 180 -
ca
UJ
s -
1 4~9C
HI
BSOLUTE V\l
3
165
......
...5
./.
......
..„.
- -
.
•
--A
-..*>
•*;••
•••»•'
!.....
.m....
"
&
LL
O
5
i •?
: a
1 CL
• ^
".-..»
.
...
...
f f
•
jf
-
.."ji..
r
sJ
*
K
'
f
i.?,.
%
«
?..
*\
.'-..
> f
I'T1'
'
-20.000 -10,000 0 10,000 20,000 30.000 40,000
TIME RELATIVE TO START OF CONSTANT-RATE PUMPING TEST (min)
Figure 5-13. Plot of water-level data from domestic well Dl, Blackwater Brook site,
Dover, New Hampshire.
50,000
5-29
-------
A plot of water-level trends in the overburden and bedrock wells monitored 2 weeks prior to
pumping is given in Figure 5-14. Most of the water levels cluster around the 136-foot level. The
higher water levels in O3 and R3 reflect the fact that these wells are upgradient and about 600 feet
away from the others. Diurnal variations are seen in the water-level record of well R3 (denoted as a
solid diamond in Figure 5-14), particularly during the time just prior to the beginning of pumping
when measurements were taken every hour. The slight depression in water levels at approximately
11,000 and 8,500 minutes before pumping was caused by some minor pumping required to perform
flowmeter testing at the site. However, the data show no discernible trend prior to aquifer testing
(pumping phase). As a result, no correction for ambient trends was made to the pumping data
gathered during the test.
145
135
-20,000
-12,000 -8,000 -4,000
TIME RELATIVE TO START OF CONSTANT-RATE PUMPING TEST (min)
LEGEND
03 = OVERBURDEN WELL AND NUMBER
R3 = BEDROCK WELL AND NUMBER
Figure 5-14. Plot of ambient water levels in overburden and bedrock monitoring wells,
Blackwater Brook site, Dover, New Hampshire.
5-30
-------
Ambient ground-water flow in the bedrock and overburden aquifers at the Blackwater Brook
site is from the northeast to the southwest (Figures 5-15 and 5-16, respectively; Table 5-2), generally
along the trend of the N60°E fracture zone. Blackwater Brook is perched above the overburden
marine clay layer (Figure 5-2) and is not believed to be hydraulically connected to either the
overburden or the bedrock aquifers. Northeast of TW, the hydraulic gradient in the bedrock aquifer is
approximately 0.008, while that in the overburden aquifer (the sands and gravels beneath the clay
layer) is 0.005. Southwest of TW, however, the gradient drops to 0.0005 in the bedrock aquifer and to
0.001 in the overburden aquifer. The decrease in ground-water gradient coincides approximately with
the intersection of the N5-10°W and N60°E fracture zones (Figure 5-3). The ambient water-level
contouring results suggest that the N5-10°W fracture zone is a partial barrier to ground-water flow.
Assuming that the flux of water (the Darcy velocity, q) across that plane is constant, then the
difference in hydraulic gradient is explained by a 16-fold increase in hydraulic conductivity in the
bedrock aquifer and a fivefold increase in the overburden aquifer west of the N5-10°W fracture zone.
The cross-cutting N5-10°W fracture zone may act as a partial barrier boundary restricting flow along
the N60°E trend.
Constant-Rate Aquifer Testing—
The test well was pumped at a constant rate of 200 gpm for 15 days. Water was discharged to
a marshy area downgradient from the monitoring wells that drains to Blackwater Brook. With 20 to
30 feet of marine clay overlying the gravel and bedrock in this area, it is unlikely that the discharge
water recharged the aquifer during testing. Pumping was suspended briefly four times between 6,500
and 7,200 minutes (4.5 to 5 days into pumping) because of generator malfunction (a total of about
2 hours shutdown). Drawdown levels returned to expected levels after resumption of continuous
pumping (Figure 5-17). Discharge rates were monitored throughout the test using both an inline
flowmeter with totalizer and an orifice weir. A total of 4,285,000 gallons of water was discharged from
TW over the course of the test, averaging 199.5 gpm (correcting for 2 hours of shutdown).
Water levels in six of the Blackwater Brook Site wells (TW, Rl, R2, R3, O2, and O3) were
monitored with automatic data loggers. The remaining three wells, R4, Ol, and O4, were monitored
manually with dedicated electronic water-level probes. All readings were collected to tolerances of
±0.01 foot. Readings were taken on a logarithmic schedule appropriate for aquifer-test analysis. This
schedule was followed for 15 days of pumping, and repeated for 14 days of recovery.
Diagnostic plots (logarithmic and semi-log) of time-drawdown and time-recovery were created
for all observation points. An example of the time-drawdown semi-log plots for well R4 is shown in
Figure 5-17. The rise in water level at approximately 7,500 minutes is due to the 2-hour pump
shutdown resulting from generator problems. Barrier-boundary conditions could not be detected on
these plots, possibly because the pumping well is so close to the boundary represented by the
N5-10°W fracture zone and its impact was felt immediately at the pumping well.
The pumping well (TW) stabilized at 74 feet of drawdown (Table 5-2) approximately
10,000 minutes (7 days) after pumping began, for a specific capacity of 2.7 gallons per minute per foot
(gpm/ft). The two bedrock wells nearest TW, Rl and R2 (located 150 feet away), responded
immediately to pumping. Wells R4 and R3 (at 365 and 596 feet from TW, respectively) began to
5-31
-------
\
*k
fo^
LEGEND
TEST WELL O1 = OVERBURDEN WELL AND NUMBER
DOMESTIC WELL AND NUMBER R1 = BEDROCK WELL AND NUMBER
138 WATER-LEVEL CONTOUR (INTERVAL 1 FOOT; DASHED WHERE INFERRED)
TOPOGRAPHIC CONTOUR (INTERVAL 20 FEET)
(135.77) WATER LEVEL IN BEDROCK WELL (FEET AMSL)
D1
Figure 5-15. Ambient hydraulic gradient and water-level data for the bedrock aquifer,
July 1990, Blackwater Brook site, Dover, New Hampshire.
5-32
-------
LEGEND
TW = TEST WELL O1 = OVERBURDEN WELL AND NUMBER
D1 = DOMESTIC WELL AND NUMBER R1 = BEDROCK WELL AND NUMBER
138 WATER-LEVEL CONTOUR (INTERVAL 1 FOOT; DASHED WHERE INFERRED)
TOPOGRAPHIC CONTOUR (INTERVAL 20 FEET)
(135.37) WATER LEVEL IN OVERBURDEN WELL (FEET AMSL)
Figure 5-16. Ambient hydraulic gradient and water-level data for the overburden aquifer,
July 1990, Blackwater Brook site, Dover, New Hampshire.
5-33
-------
o -
-10 -
-20 -
-30 -
•40 -
•SO -
............
, , I , . . . . . . 1 . t . i t i i 1 . i t p i t i r t 1 i i r .r_i_i_
••
•60 -| 1 —
1
WE
^
r =
I
E
I
i
ELL
i
596
"'
"H
II
• •
1!
it
i!
R4
200 gpm
>ft
"
ill
tit
HI
: : :
: : :
ill
ill
in "
-•-.
•>
j
j
i
! .
I
j
V
".
I
'
i-
I
i I '
g
.
"•»
i
PUMP SHUTDOWN
FROM GENERATOR
MALFUNCTION
s
•
X
i
'""•*.
>.
t,
I
,
.., .
\
10 100 1,000 10,000 100,(
TIME SINCE PUMPING BEGAN (min)
WHERE: BEDROCK WELL =
TEST WELL DISCHARGE =
DISTANCE FROM R4 TO TEST WELL =
R4
QTW
r(ft)
Figure 5-17. Serai-log plot of drawdown versus the log of time for observation well R4 when the test
well is pumping, Blackwater Brook site, Dover, New Hampshire.
draw down after 8 and 10 minutes of pumping, respectively. All bedrock wells reached steady-state or
pseudo steady-state conditions (a constant change in head with time; McWhorter and Sunada, 1977)
by about 10,000 minutes (7 days) of pumping. Recovery was rapid; 90% of drawdown in TW
recovered after 10,000 minutes (7 days) following pumping shutdown.
Overburden wells also responded quickly to the pumping of TW and exhibited substantial
drawdown (Table 5-2). These wells also stabilized approximately 10,000 minutes (7 days) into the test.
One overburden well, O4, dewatered completely after approximately 2,450 minutes (1.7 days) of
pumping. The lithologic log from well O4 indicated the thinnest section of permeable sands and
gravels under the marine clays, which is interpreted as a pinch-out of a beach facies.
5-34
-------
Water-level contours noted under pumping conditions at TW suggest the presence of a partial
barrier boundary between TW and well R4 (Figure 5-18). The TW could not draw the water it needed
from the northeast; hence, the cone of depression expanded in an oval shape. The resulting oval
potentiometric surface typifies the 15-day constant-rate test. Contouring of overburden well water
levels generally mirrored the patterns seen in the bedrock aquifer. Water-level data from well Dl,
(approximately 2,000 feet from TW; Figure 5-10) showed no response during the test. The ambient-
trend monitoring data indicate that Dl was in use during the test, but there was no significant trend of
dropping or rising water levels associated with the stress due to the pumping of TW (Figure 5-13).
An estimate of the percentage of discharge from TW contributed by the sands and gravels
above the bedrock during constant-rate aquifer testing was calculated volumetrically. The elevations of
the water-level surface during pumping are subtracted from the elevations of the ambient water-level
surface. The resulting volume between pre- and post-pumping conditions in the overburden was
multiplied by a calculated storage coefficient (S) of 5 x 10's for those areas that remained confined,
and by an apparent specific yield of 0.15 (Morris and Johnson, 1967) where water levels drew down
below the clay. From this analysis, approximately 20% of the discharge came from the permeable
sands and gravels beneath the confining clay in the immediate vicinity of TW.
Aquifer Coefficient Calculations--
Analytical models provide powerful tools for determining aquifer properties and for predicting
aquifer behavior under differing hydrogeologic conditions. The principal challenge in using analytical
models correctly lies in matching the proper analytical model with the conceptual model of the system.
That is, it is important to find an analytical model whose attributes and assumptions most closely
match conditions in the field.
The results of the hydrogeologic investigation prior to aquifer testing demonstrate the complex
character of the system. Rather than an equivalent porous medium, the aquifer at Blackwater Brook
is a regime of separated flow systems within the fractured bedrock. Discrete flow is reflected in the
isolated fractures noted in well logs as well as in video and acoustic televiewer analyses; it is also
corroborated by distinct hydrochemical characteristics seen in the hydrophysical logging. Application
of radial flow models such as the Theis (1935) or the Cooper-Jacob (1946) solutions would not
appropriately match the conceptual model for the site. Accordingly, several other analytical
approaches were used to try to quantify the aquifer coefficients and to serve as predictive tools for
WHPA delineation and ground-water monitoring design.
Initially, the time-drawdown curve for the pumping well (TW) was analyzed by using an
analytical model conceived by Gringarten, Ramey, and Ranghaven for a single-plane, vertical fracture
in an otherwise homogeneous, isotropic, confined aquifer (Kruseman and de Ridder, 1989). The bulk
trahsmissivity calculated for the aquifer is 2,200 gpd/ft without taking into consideration directional
variables associated with fracturing and faulting.
5-35
-------
LEGEND
TW = TEST WELL O1 = OVERBURDEN WELL AND NUMBER
D1 = DOMESTIC WELL AND NUMBER R1 = BEDROCK WELL AND NUMBER
TOPOGRAPHIC CONTOUR (INTERVAL 20 FEET)
EXPANSION OF 100-FOOT WATER-LEVEL CONTOUR IN TIME
(DASHED WHERE INFERRED)
Figure 5-18. Contour map of the expansion of the 100-foot water-level contour in time during
constant-rate aquifer testing, Blackwater Brook site, Dover, New Hampshire.
5-36
-------
An analytical model by Gringarten and Witherspoon also assumes a single-plane, vertical
fracture, but uses observation data to calculate a storage coefficient and directional transmissivities
(Kruseman and de Ridder, 1989). Different type curves are used to calculate directional hydraulic
conductivities for the more permeable and less permeable directions. An example calculation is
provided in Figure 5-19 for well R4. Applying these curves resulted in transmissivity values of
4,700 gpd/ft for well Rl and 4,100 gpd/ft for well R2 along the N30°W well alignment. The
transmissivity is 1,800 gpd/ft in the less permeable direction along the N60°E trend (wells R3 and R4).
Storage values calculated using data from wells R3 and R4 are 0.05 and 0.015, respectively. Water
levels in both wells were drawn down below the confining layer of clay into bedrock, so the high
storage values (greater than 0.01 for semi-confined to confined aquifers) probably reflect a
combination of confined and unconfined conditions.
Storage values calculated for wells Rl and R2 using the Gringarten-Witherspoon method are
0.56 and 0.77, respectively. A storage coefficient in the traditional sense cannot be determined for this
setting. Wells Rl and R2 are probably completed in the same fracture system as TW, in the manner
of an extended well, suggesting the applicability of the linear flow model by Jenkins and Prentice
(1982). This method provides values of hydraulic diffusivity (D), where D = T/S, for the aquifer, given
good knowledge of the structural conditions that produce the linear flow patterns in the aquifer.
10
MATCH
. POINT
CL.,=200gpm
GRINGARTEN-WITHERSPOON
TYPE CURVES
TIME(min)
i I i I iiiii i—i i i iii'ii i—j—H ""
I Mlffl.il mfflll . Ml.Tlll I Ml?!!?. I M.ffln . M.ffil Mill.
10
10
WHERE:
BEDROCK WELL
TEST WELL DISCHARGE
DISTANCE FROM R4 TO TEST WELL
WELL FUNCTION AND FACTORS
TIME
TRANSMISSIVITY
STORAGE COEFFICIENT
= R4
= o (gpm)
= r (ft)
= F (u^, r')
=t(min)
= T (gpd/ft)
=S
Figure 5-19. Logarithmic plot of time versus drawdown and Gringarten-Witherspoon type curves
(Kruseman and de Ridder, 1989) for observation well R4 when the test well is pumping,
Blackwater Brook site, Dover, New Hampshire.
5-37
-------
The Jenkins-Prentice method could not be applied to data from wells Rl or R2. Automatic
data loggers indicated drawdown impact at both wells at 0.001 minute following the beginning of
pumping. While plots of drawdown versus the square root of time for wells Rl and R2 show linear
flow patterns at an early time (a straight-line plot), the extrapolated time of zero drawdown necessary
to perform the calculations is impossible to determine.
The linear flow model was applied to the data for wells R3 and R4 using the radial distance
from the pumping well as the distance from the fracture (the extended well). Values for D were
approximately 3.4 million feet squared per day (Mft2/d) for well R4 and 8.5 Mft2/d for well R3
(Figure 5-20) using early- to middle-tune data. The method does not allow independent calculation of
a storage coefficient. However, because the linear flow portion of the curve occurred before
unconfined conditions were obtained, a low storage coefficient can be assumed. Assuming a calculated
value of 5 x 10'5 for S, the resulting transmissiviry values are 1,300 gpd/ft for well R4 and 3,200 gpd/ft
for well R3.
The results of analytical modeling are ambiguous, and the exercise is subject to various
interpretations. None of the analytical approaches adequately accounts for the complexities of the
system. Perhaps for an aquifer of this type, purely analytical solutions are inadequate for the task of
describing the system. Other analytical models (for example, Papadopolus, 1965) should be tried as
other possible approaches and may be applicable in less complicated systems.
The results of the analytical modeling indicate that it is not practical to use such models as a .
means of quantifying the aquifer properties and then employ the model to predict a particular
criterion (such as drawdown or time of travel) for a wellhead protection zone. To develop a predictive
tool for WHPA delineation and monitoring at the Blackwater Brook site, a more direct approach was
taken. Dye tracing was used to determine seepage velocities in the aquifer empirically from selected
observation wells to TW.
Dye Tracing
Two dye traces were conducted, with dye injected as a slug into the major water-bearing
fracture in two bedrock observation wells known to have hydraulic connection to TW (wells R2 and
R3, which are 152 and 596 feet from TW, respectively; Figure 5-12). The dyes chosen, rhodamine-WT
and fluorescein, are the most conservative of the available fluorescent dyes for tracing (Mull et al.,
1988) . They are readily adsorbed onto charcoal samples and have high detectibiliry. Analysis for
these two dyes is reliable and inexpensive.
Two gallons of rhodamine-WT solution (20% solution) were injected into well R2 through a
280-foot garden hose. The 280-foot-length hose was chosen to emplace the dye as a slug at a depth of
275 feet, corresponding to the major water-bearing fracture identified in well R2 (Figure 5-12). The
volume of the 280-foot, 5/8-inch-diameter hose is about 4.5 gallons. Thus, the 2 gallons of dye
solution were followed by 2 gallons of water rinsed from the dye-solution jugs and 3 gallons of
deionized water to completely replace the hose volume. Rhodamine-WT dye emplacement began
4.5 hours after initiation of test pumping, to avoid disturbing water levels during the early period of
the aquifer test. The dye injection in well R2 was completed in approximately 12 minutes.
5-38
-------
3 -20
WELLR3
x = 596ft
•^ = 6.92171(15
t,, = (6.92 min)2 = 47.9 min
S «„
K(S96ftla
4(47.9 min/1,440 min/d)
D = 8,500.000 tfld
l\\ = 6.92 min
so 75 too
SQUARE ROOT OF TIME (-t\. in min)
WHERE: INITIAL TIME
DISTANCE FROM R3 TO LINEAR EXTENDED WELL
HYDRAULIC DIFFUSIVITY
TRANSMISSIVITY
STORAGE COEFFICIENT
= t0(tnin)
= x(ft)
= D (ftVd)
= T(gpdffi)
= S
Figure 5-20. Plot of drawdown versus the square root of time and Jenkins-Prentice (1982) linear-flow
model calculations for monitoring well R3, Blackwater Brook site, Dover, New Hampshire.
Fluorescein dye solution (75% solution) was injected into well R3 in a similar manner, using a
sufficient length of clean hose to emplace the slug of dye near the main water-bearing fracture at a
depth of 357 feet (Figure 5-12). Because this well required a longer length of hose, the hose was
flushed with 6 gallons (the approximate hose volume) of water (including the jug rinse) following the
2 gallons of dye solution. Fluorescein dye injection into well R3 began 6 hours after test pumping
began and was completed in approximately 13 minutes.
Monitoring for the presence and the level of dye at TW under pumping conditions was
accomplished in two ways. Individual water samples were taken at prescribed intervals during
pumping. The sampling interval was frequent in the early phase of the test (every 15 to 30 minutes)
to detect fast arrivals from the near injection well, R2, and was less frequent later (every 4 hours).
Additionally, cumulative samples of the dyes were taken in charcoal samplers for quality assurance
purposes. A small amount of discharge water was continuously diverted into a clean, plastic bucket
which housed the charcoal samplers. These samplers were changed every 4 hours during pumping.
Analysis of the dye samples was performed by the Ozark Underground Laboratory (Protem,
Missouri) using a spectrofluorophotometer. A complete description of the sampling and analysis
methodology has been provided by Aley (1990).
Rhodamine-WT arrived at TW 130 minutes (2.2 hours) after injection in well R2 and was
detected at a concentration of 0.64 ppb (Figure 5-21, top). This indicates a seepage velocity of
approximately 1,680 ft/d over the 152-foot distance with a gradient between R2 and TW of 0.11 (along
the N30°W well alignment) under pumping conditions of 200 gpm. The rhodamine-WT breakthrough
curve resembles a typical dye-trace curve, with a steep rising limb, a relatively sharp peak, and a
shallow falling limb (Figure 5-21, top). The jagged character of the curve may reflect the irregular
5-39
-------
1,200
Q.
a.
O
O
z
O
O
111
Q
O
DISTANCE FROM WELL R2 TO TEST WELL
152ft
FIRST ARRIVAL:
130min(2.2hr)
AFTER INJECTION
SEEPAGE VELOCITY OF FIRST ARRIVAL:
1,680 ft/d
TIME TO MASS ;f"
•'" CENTROID If
(85.3 hr) \
1.000- •
800 - -—
600
400
200 • ••
5.000
10.000 15.000
TIME SINCE PUMPING BEGAN (min)
20,000
15
25,000
CC
01
O
O
u
O
V)
111
cc
O
..4 * 4.
DISTANCE FROM WELL R3 TO TEST WELL:
596ft
SEEPAGE VELOCITY OF FIRST ARRIVAL:
96 ft/d
I
...1-
FIRST ARRIVAL:
8,880 min (148 hr)
AFTER INJECTION
5,000 10.000 15,000
TIME SINCE PUMPING BEGAN (min)
20.000
25.000.
Figure 5-21. Dye-trace breakthrough curves of rhodamine-WT injection (top) at well R2
to the test well, and fluorescein injection (bottom) at well R3 to the test well,
Blackwater Brook site, Dover, New Hampshire.
5-40
-------
character of the bedrock aquifer; fractures widen and narrow over the length of the flowpath.
However, mass balance analysis of the rhodamine-WT curve (Table 5-4) indicates comparatively little
dispersion of the dye [0.02 foot squared per second (fi^/s)] and very direct connection between the
boreholes. The statistical analysis of the breakthrough curve given in Table 5-4 was performed using
the BASIC code provided by Mull et al. (1988).
TABLE 5-4. STATISTICAL ANALYSIS OF RHODAMINE-WT DYE TRACE FROM WELL
R2 TO THE TEST WELL, BLACKWATER BROOK SITE,
DOVER, NEW HAMPSHIRE
Distance from well R2 to TW, ft
Total dye injected, gr
Total dye recovered, gr
Mean discharge, cfs
Time of first arrival at TW, hr
Velocity (from first arrival), ft/d
Time to centroid of mass, hr
Time to peak, hr
Standard deviation of time to peak, hr
Dispersion coefficient, cfs
Skewness coefficient
152
1,800
506
0.446
2.2
1,680
85.3
42.9
64.3
0.02
1.41
Abbreviations: TW = test well; ft = foot; ft/d = foot per day; gr = grams; cfs = cubic foot per
second; hr = hour.
The fluorescein dye placed in well R3 first arrived at TW approximately 148 hours (6.2 days)
after injection, yielding a seepage velocity of 96 ft/d over a distance of 596 feet under pumping
conditions of 200 gpm (Figure 5-21, bottom). The gradient in water level at that time between the
two wells was 0.07. The fluorescein dye trace shows only the rising limb of the breakthrough curve,
which bears little resemblance to the rhodamine-WT curve. Where the rise in rhodamine-WT
concentrations was sharp, that of the fluorescein was slow, moving from the detection limit to slightly
less than 1.2 ppb over a period of approximately 7 days. No peak was detected, and as a result,
statistical analysis of the fluorescein curve was not possible. It appears that the dispersion factor in
the N60°E direction is much greater than that of the N30°W direction. This is probably due to the
longer flowpath and the lower transmissivity of the aquifer in the N60°E direction. In addition, there
may not be direct connection between well R3 and TW. Fluctuations in the fluorescein curve seen in
Figure 5-21 (bottom) may indicate arrival of dye from different flowpaths between well R3 and TW.
Water Quality Analysis
Hydrochemical monitoring occurred during several phases during the project. Several months
before aquifer testing, water from TW was sampled for general chemistry parameters [cations, anions,
coliform bacteria, alkalinity, hardness, total dissolved solids (TDS), sediment, turbidity, color, odor,
pH, and specific conductivity] along with pesticides, radon, volatile organic compounds (VOCs), and
heavy metals. Hydrochemical analyses served as a screening tool early in the exploration process to
5-41
-------
determine acceptable water quality. Later, all wells, several nearby springs, and Blackwater Brook
were sampled for basic cations and anions. These analyses were done as part of the ambient-trend
monitoring effort to characterize water types.
During aquifer testing, water samples were collected from the discharge line of TW three times
during the course of pumping in order to monitor variations in water quality parameters at the
beginning, middle, and end of the test. The final sample also served as the state-required sampling for
new source approval of a municipal supply well. The first sample was taken 20 minutes after the test
began and was analyzed for general chemistry parameters, nitrates, arsenic, and lead. The second
sample was collected at approximately the midpoint of pumping (at 9,981 minutes or 6.9 days), and
was analyzed for general chemistry parameters and nitrates. For the second sample, arsenic and lead
were not analyzed because these constituents were detected at nonsignificant levels in the first sample
round. The final sample was collected just prior to shutdown of the pump (at 21,550 minutes or
15 days), and was analyzed for all of the previously mentioned constituents plus VOCs, insecticides,
polychlorinated biphenyls (PCBs), and radon.
Grab samples were collected periodically during pumping and field-analyzed for temperature,
specific conductivity, pH, chloride, and sodium concentrations to monitor variations in water quality
over time. The grab samples were obtained at a higher frequency than the laboratory samples. More
frequent sampling was conducted as part of the ambient-trend monitoring effort to establish the
chemical character of the water. The instruments used for analysis were a specific conductivity and
temperature meter, a chloride titration kit, and sodium and pH pens. These instruments do not
guarantee a high degree of accuracy or resolution; however, they do permit frequent testing and rapid
turnaround of results. During the first 2 days of testing, these analyses were conducted at 1.5-hour
intervals. During the next 13 days, the sampling interval was decreased to approximately every
4 hours to reduce labor and budget requirements.
Analytical results show that the bedrock ground water differs in character from the surface
water and ground water in the overburden material. Total dissolved solid concentrations average
268 milligrams per liter (mg/1) in the bedrock water and 113 mg/1 in the overburden water. Average
sodium (Na) and chloride (Cl) concentrations are much higher in the bedrock water, 86.8 mg/1 Na and
107 mg/1 Q, compared with an average of 22.4 mg/1 Na and 9.67 mg/1 Cl in the overburden water.
Iron (Fe) and manganese (Mn) levels are higher than EPA secondary standards, averaging
1.40 mg/1 Fe and 0.06 mg/1 Mn in bedrock water and 3.91 mg/1 Fe and 0.11 mg/1 Mn in the
overburden and surface water. All wells contain water that is slightly to strongly alkaline (pH = 7.9 to
9.2). Two samples taken showed slightly acidic water, one from Blackwater Brook (pH = 6.5) and
one from Dl (pH = 6.7). Specific conductivity is generally lower in overburden water (average of
198 uS/cm) than in bedrock water (average of 445 |iS/cm).
No VOCs, pesticides, or heavy metal compounds were detected in any surface- or ground-
water samples. No detections for these compounds were anticipated, as few contaminant sources exist
to threaten the site. Radon levels in TW were 3,200 picocuries per liter (pCi/1) after the 15-day
aquifer test, compared with 2,080 pCi/1 following 2 days of pumping several months earlier. The rise
in radon levels indicates that significant amounts of water may have been drawn from fractures in the
quartz monzonite during the 15-day aquifer test. The EPA has not yet established a maximum limit
for radon, but is considering options in the range of 200 to 2,000 pCi/1 (U.S. EPA, 1990a). In view of
this possible requirement, water treatment for radon may be required at the Blackwater Brook site.
5-42
-------
A Piper trilinear diagram (Piper, 1944) displaying the variation of the major inorganic
ground-water constituents allows characterization of and distinction between the shallow ground water
in the overburden wells and the older, deeper ground water within the bedrock wells (Figure 5-22).
Envelopes around the bedrock (dashed line) and overburden (solid line) inorganic constituent data
indicate water types of sodium chloride and sodium bicarbonate, respectively. Levels of sodium,
chloride, and specific conductivity tend to be higher immediately following drilling of bedrock wells
and to decrease within a few days of well completion. In general, levels of these parameters increase
from north to south in the bedrock aquifer. The variation in water quality among samples taken from
TW when it was first drilled (September 22, 1989), in the middle (July 11,1990), and at the end (July
26, 1990) of the pumping test shows a trend of declining sodium and chloride and improving water
quality. During 15 days of pumping, sodium levels in TW dropped from 116 mg/1 to 81.7 mg/L This
evidence indicates that the bedrock system is connected to active recharge from overburden sands and
gravels.
Overburden marine clays may contribute sodium to the ground water; sodium would be
desorbed by acid precipitation as it slowly infiltrates through the soil. The overburden sand and
gravels are quickly flushed of residual salts as water recharges the system. The recharge area of these
gravels, above the 190-foot elevation of maximum marine transgression, has fairly thick forest soils
that.probably contain abundant carbon dioxide-producing bacteria. Consequently, these waters tend
to have high concentrations of bicarbonate alkalinity. An exception to the overburden water type is
well O3, where the dominant cation is calcium rather than sodium. This well is positioned closest to
the deposit of sand and gravel to the north and east. The lower levels of sodium probably reflect
more active recharge from water percolating through the sands and gravels and a less significant
component of water passing through the sodium-rich marine clay.
REFINED CONCEPTUAL HYDROGEOLOGIC MODEL
In the earliest conceptual model for the Blackwater Brook site, the structural influence of the
N5-10°W fracture zone was seen as the most likely conduit for ground-water flow. Based solely on
lineament analysis, the N5-10°W feature was protected under the city of Dover's ground-water
protection ordinance (Figure 5-5) until further work could be done to refine the model. Later the
focus shifted. According to the revised model, the primary flowpath for ground water in the bedrock
aquifer coincided with a photolineament, contact zone, and magnetic anomaly, all trending roughly
N60°E (Figures 5-3 and 5-4). Analysis of data gathered from imagery, fracture mapping, and surface
geophysics, coupled with drilling of a high-yield test well, supported the interpretation that the
northeast-trending feature is the significant water-bearing zone.
Subsequent work revealed that the structural and hydrogeologic pictures are much more
complex than originally suggested. As opposed to the earlier model of a discrete contact in which two
bedrock formations are juxtaposed along a single plane, the N60°E contact is conceptualized as a
somewhat irregular, or possibly curvilinear, zone. Along this contact, there is extensive interfingering
of the quartz monzonite and the Berwick Formation metasedimentary rocks. Instead of one preferred
direction of fracturing, there are two: one trending N60°E and one trending N5-10°Wl The
Blackwater Brook site lies at the intersection of these two fracture zones. It follows that ground water
flows preferentially along both fracture zones with primary flow along the N5-10°W trend.
5-43
-------
CLASS OF WATER (mg/l) /x /
Reid I -Sodium Chloride /C?A
Reid II -Stagnant
Reid III - Recent Recharge
Reid IV-Dynamic
LEGEND
WELL DESIGNATION
TW. TEST WELL
R1» BEDROCK WELL
AND NUMBER
Ol» OVERBURDEN WELL
AND NUMBER
WATER SOURCE
+ BEDROCK
O OVERBURDEN
^ STREAM
,~.—.BEDROCK
«• • WaL ENVELOPE
OVERBURDEN
WELL ENVELOPE
—Ca
CATIONS
ANIONS
Figure 5-22. Trilinear diagram of ground-water types from overburden and bedrock monitoring wells,
Blackwater Brook site, Dover, New Hampshire.
5-44
-------
The flow field is anisotropic, and ground-water flow is highly discrete within the bedrock
fracture system. Water enters and exits boreholes at distinct intervals rather than uniformly along the
length of the borehole. Portions of a borehole may show downward flow, while other sections of the
same borehole may have upward flow. While cross correlations of water entry and exit points between
boreholes are complicated by the large distances involved, the major direction of flow is N30°W with
lesser flow along the N60°E trend. Wells TW, Rl, and R2 are contained within the intersection of the
N5-10°W and N60°E fracture zones. Wells TW, R3, and R4 lie approximately along the N60°E
fracture zone.
Analysis of aquifer-test data yielded values of low to moderate hydraulic conductivity, with the
N30°W trend exhibiting higher values than the N60°E trend. The zone of highest transmissivity is in
the N30°W direction with moderate transmissivities in the range of 4,000 to 5,000 gpd/ft, while the
N60°E trend has transmissiviry values of 1,000 to 3,000 gpd/ft. Storage is difficult to quantify for two
reasons: (1) the change from confined to unconfined conditions during pumping, and (2) the
quantification of storage within a fractured aquifer rather than a porous one. From dye-trace results,
the movement of water due to pumping is rapid, with flow velocities greater than 1,600 ft/d along the
zone of highest transmissivity (N30°W) and greater than 90 ft/d along the N60°E trend at'prevailing
hydraulic gradients within the well's zone of influence.
Extrapolation of ground-water flow characterization beyond TW and well R2 is difficult.
Lacking discrete flow information beyond TW and well R2, a valid interpretive approach is to assume
that statistical fracture descriptions become increasingly good approximations of flowpaths at
increasing distances from the site. Therefore, prominent fracture peaks along the N60°E and
N5-10°W trends (Figure 5-7) represent the most probable flow directions within the bedrock fracture
system at Blackwater Brook (Vernon et al, 1993). The N60°E trend is substantiated by the existence
of the faulted, fractured contact zone along this strike. Evidence to suggest preferred flow along the
N5-10°W direction is both structural and hydrogeologic. Structural control is inferred by strong
expression of the N5-10°W lineament on several platforms of photography and in outcrop fracture
trends. Greater aquifer transmissivity in the N30°W direction is demonstrated by analytical
calculations, faster dye arrival times, and elongation of drawdown contours in this general direction.
Enhanced transmissivity along the N30°W direction is attributed to the proximity and similar
orientation of the N5-10°W fracture zone.
The original conceptual model suggested that overburden materials in the vicinity of the well
would not be able to provide sufficient recharge to sustain large volume production. Analysis of the
drawdown data in the overburden wells, however, indicates that a large percentage, 20%, of the water
discharged from TW during pumping is derived from these sediments. Moderate drawdowns in the
overburden wells and establishment of steady-state or pseudo steady-state conditions during aquifer
testing indicate that appreciable recharge comes directly from overlying sediments. The future
production well (near TW) will produce intersecting zones of contribution extending preferentially
along the two trends of highest transmissivity. The ZOC for the production well may not be as
extensive as previously believed, due to recharge from the overburden sediments.
The primary water-bearing fractures at the Blackwater Brook site are steeply dipping. In
general, the deeper the flow system, the longer the time required for water to move through the system
in a manner analogous to Toth's (1963) local, intermediate, and deep flow systems. Vertical profiles
of FEC and pH measured during the hydrophysical logging procedure indicate that water in the
5-45
-------
boreholes is distinctly layered. Generally, FEC increases with depth, but not in a continuous manner.
Rather, sharp gradients in FEC and pH occur at discrete depths in the bedrock wells, supporting the
conceptual model of isolated, discrete flow zones. The deeper portions of fracture systems impacted
by the marine transgression may have higher concentrations of sodium and chloride than the more
shallow fractures.
Ground-water quality at the Blackwater Brook site is dependent on the mineralogy of the
bedrock and overburden, as well as the history of marine transgression, human activities, and
biological processes. Chemical dissolution of minerals within the bedrock by infiltrating water
contributes potassium, calcium, magnesium, and sodium to the chemical makeup of the ground water.
The large concentrations of sodium and chloride found in several of the bedrock wells may be
attributed to connate water within deeper bedrock fractures as a result of Pleistocene marine
transgression.
Water quality sampling results (Figure 5-22) showed distinct drops in the levels of sodium and
chloride at TW during the 15-day aquifer test. This and other evidence of active recharge from the
overburden to the bedrock system suggest that the water quality within the bedrock system will evolve
over time in the following manner:
• The dissolved iron and manganese concentrations will increase in the bedrock well
water.
• The sodium, chloride, and TDS concentrations will decrease as a result of recharge of
the bedrock system by infiltrating overburden water.
• Hardness should decrease because of recharge of the bedrock system by infiltrating
overburden water.
• Radon will probably increase slightly, but should stabilize or decrease with well usage.
If radon remains at its present level, water treatment may be necessary.
With the exception of iron, manganese, and possibly radon, the bedrock water quality should
improve over time as long as the WHPA is enforced. This protection should eliminate, minimize, or
regulate potential sources of contamination related to human activity, ensuring maintenance or
improvement of water quality.
REFINED WELLHEAD PROTECTION AREA DELINEATION
Existing Wellhead Protection Programs
The New Hampshire State WHPP was approved by the EPA in September 1990. This
program defines a WHPA as a wellfield management area that includes all of the contributing area to
a well under expected conditions of recharge and well pumping. The criteria for establishing WHPAs
are (1) a calculated, fixed radius based on the size of the well for the primary protection zone
(Zone I) up to a 400-foot radius; and (2) ground-water flow boundaries for the secondary protection
zone (Zone II). The NHDES is the lead agency in wellhead protection for the state and is
responsible for implementation of the program.
5-46
-------
The NHDES has established an interim protection polity, establishing specified radial
distances around wells until final WHPAs can be delineated and the program fully established.
Specified radial distances for the interim policy period are 2,500 feet for major community wells,
1,500 feet for small community wells, and 1,000 feet for noncomrnunity wells. The state program will
identify potential contaminant sources, inform owners and operators of potential sources and their
location within a WHPA, and use periodic inspections to ensure compliance with best management
practice rules. These regulations cover existing supplies of water and treat bedrock and sand-and-
gravel wells in a similar manner. Regulations are still under consideration for new source approval. It
is likely that overburden and bedrock supply wells will not be approved under precisely the same
criteria (P. Currier, NHDES, personal communication, April 1991).
The adjoining New England states also have WHPPs in place. For instance, Massachusetts
requires a 400-foot radius for Zone I; a 180-day, no-recharge ZOC for Zone II; and the upgradient
contributing area to Zone II as Zone III. The State of Maine proposes to use a 200-day time of travel
(TOT) for Zone I, and a 2,500-day TOT boundary for Zone II for unconsolidated aquifers. Zone III
in the Maine system is also the contributing area to Zone II. For fractured bedrock aquifers in
Maine, Zone I is a 300-foot protective radius, Zone II a 1,000-foot protective radius, and Zone III the
remainder of the upgradient watershed.
Delineation of the Wellhead Protection Area for the Blackwater Brook Site
Criteria and Methods for Wellhead Protection Area Delineation—
An approach often used to determine the ZOC in a porous medium involves a numerical flow
model calibrated to steady-state and to transient conditions with data from an aquifer test. The
calibrated model is then extrapolated to identify the ZOC for an extended time period of pumping.
For a fractured bedrock aquifer, the ZOC for a particular time frame is extremely difficult to establish.
Specifically, the theoretical difficulties of modeling flow through fractured media using numerical
models designed for porous media make such efforts inappropriate in many situations.
The approach used to delineate the WHPA at the Blackwater Brook site employs TOT rather
than the ZOC as the primary criterion for delineation. The criteria and threshold levels chosen for
the Blackwater Brook WHPA and protection zones (Table 5-5) reflect the efforts of the EPA, the
NHDES, and the Maine Department of Environmental Services to establish statewide criteria for
bedrock wells, coupled with information gained in this research monitoring project.
The TOT criterion and analytical solution for the delineation utilize the seepage velocity data
gathered empirically from dye tracing. The time and distance the dye traveled from the injection wells
(R2 and R3) to TW are used to determine the constant in the analytical solution, which varies with
direction. An analytical equation was derived (Appendix 5-A) and was used to calculate the distances
for the 200-day and 1,000-day TOT thresholds. Corresponding distances for these thresholds were
calculated for the two dye trace directions: N30°W (R2 to TW) and N60°E (R3 to TW). The
calculated distances were then used to establish protective distances along the two preferred flow
directions (two fracture zones): N5-10°W and N60°E. The resultant WHPA and protection zone
boundaries were finalized using the hydrogeologic mapping method drawing upon the conceptual flow
model and characterization information from aquifer testing, geologic mapping, photolineament and
fabric fracture analysis, and surface and borehole geophysical surveys (Vernon et al., 1993).
5-47
-------
TABLE 5-5. WELLHEAD PROTECTION ZONES AT THE BLACKWATER BROOK SITE,
DOVER, NEW HAMPSHIRE
Zone
Description
Criterion
Purpose
Area encompassed by a 400-
foot radius around the well
HA Area covered by the 200-day
TOT limit along directions of
highest bedrock transmissivity
ID3 Area between the directions of
highest transmissivity extending
to the 200-day TOT limit
III Area contributing to the 1,000-
day TOT contour
Distance
TOT and physical features
for protection against
pathogens in ground water
Same as IIA
TOT and physical features
to ensure adequate tune for
remediation
State-mandated
sanitary protection
area
Protection against
pathogens in ground
water
Same as IIA
Ensure adequate time
for remediation
Abbreviation: TOT = time of travel.
The theoretical approach to determine the TOT distances for the protection zones utilizes a
quadratic relationship in which the time of first arrival of the dye, t, is proportional to the square of
the distance between the injection well, R2 or R3, and the test well, i (Appendix 5-A):
i = Cxt2
This relationship is derived from Darcy's Law (Equation 1, Appendix 5-A) and the Thiem Equation
(Equation 2, Appendix 5-A). It represents a simple analytical expression to interpret the dye-trace
and aquifer test information. The constant, C, varies with direction.
The value for C in the N30°W direction was determined from the empirical results of the dye
tracing at well R2, where t=130 min (0.09 d) and 1=152 ft (Figure 5-21). Solving for C, a value of
506 feet per day"2 (ft/d172) results. The equation for the TOT distance in the N60°E direction is
determined in a similar manner, where t=148 hr (6.2 d) and 1=596 ft (R3 to TW), resulting in a
value for C of 240 ft/d172. Calculated distances corresponding to the 200-day and 1,000-day TOT
thresholds for the N30°W and N60°W directions are given in Table 5-6.
5-48
-------
TABLE 5-6. TIME-OF-TRAVEL DISTANCES ALONG DYE-TRACE DIRECTIONS,
BLACKWATER BROOK SITE, DOVER, NEW HAMPSHIRE
Calculated Distances (ft)
TOT Criteria
(d)
N30°W
(R2 to TW)
N60°E
(R3toTW)
200
1,000
7,200
16,000
3,400
7,600
Abbreviations: TOT = time of travel; TW=test well.
Delineation of Zones I, II, and Ill-
Protection zones within the WHPA (Zones I, II, and III) were established by incorporating
the State of New Hampshire delineation criteria, as well as factors inherent to the conceptual flow
model (Figure 5-23). The TOT distances for the 200-day and 1,000-day thresholds along the two dye-
trace directions, N30°W and N60°E (Table 5-6), were applied to the two principal flow directions,
N5-10°W and N60°E, respectively. (A complete discussion of the selection of the two principal flow
directions is given in the Refined Hydrogeologic Conceptual Model section of Chapter 5.) The TOT
distances form the basis for the Zone II and III delineations. The final Zone III boundary was
determined using the hydrogeologic mapping method.
Zone I—The method of delineation for Zone I is a simple matter of drawing a circle on the
map. The State of New Hampshire proposes a calculated fixed radius for the sanitary protection area
based on well production rate. However, 400 feet is the largest radius used.
Zones IIA and IIB-Zone IIA is the area within the 200-day TOT limit along the directions
of highest bedrock transmissivity: N5-10°W and N60°E (Figure 5-23). The two "arms" of Zone IIA
are 1,000 feet wide and radiate from TW, or the future production well. The choice of a 1,000-foot
width was based on facts unique to this site. The N5-10°W lineament passes between two bedrock
ridges approximately 3,000 feet northwest of TW. The distance between the bedrock ridges is
approximately 1,000 feet.
The domestic well, Dl, located east of the ridges (Figure 5-10), shows no response to pumping
during testing (Figure 5-13), so it may be assumed that the ridges represent relatively unfractured
bedrock on either side of the N5-10°W fracture zone. From lithologic and borehole geophysical
logging results, the faulted contact along the N60°E trend is actually a zone of interfingering rock
types and numerous fracture sets at least 300 feet wide (the distance between wells Rl and R2;
Figure 5-10). Considering the 1,000-foot distance between ridge tops along the N5-100 W trend, the
larger of the two widths (1,000 versus 300 feet) was selected for the width of the 200-day TOT
protection zone in both directions. Zone IIA is subject to the zoning requirements of the Dover
Municipal Aquifer Protection Ordinance. This ordinance takes a somewhat more restrictive approach
to aquifer protection than do the proposed New Hampshire State regulations for a secondary
protection zone.
5-49
-------
Figure 5-23. Wellhead protection area and zones, Blackwater Brook site, Dover, New Hampshire.
(Contour interval is 60 feet.)
5-50
-------
The calculated TOT distance values determined from the derived equation and the dye-trace
results used in determining the secondary protection zone should not be construed as absolute.
Bedrock aquifers are extremely complex. The addition of a surrounding zone of protection reflects a
healthy respect for the intricacies of the system and attempts to account for some of the limitations of
this approach. Consequently, Zone IIB was introduced (Figure 5-23) as a smooth curve connecting
the "arms" of the Zone IIA. By incorporating the oval as another zone of protection, consideration is
given for other possible directions of flow within the bedrock system which may not yet be resolved, as
well as for inhornogeneities along the directions of higher transmissivity already identified, the
N5-10°W and N60°E trends. This site zone would be regulated by the secondary protection
regulations proposed by the State of New Hampshire.
Implementation of the 200-day TOT criterion represents a conservative approach to wellhead
protection. Both TOT calculations and dye tracing were employed to predict contaminant transport
directly through the bedrock aquifer. As an example of this scenario, at a rock outcrop, a spill or
illegally dumped substance may penetrate quickly into the bedrock aquifer. In other situations,
however, some time would elapse before a contaminant released at the surface passed through the
overburden into the bedrock aquifer.
Zone Ill-Zone III is defined as the area contributing to the 1,000-day TOT distances,
calculated to be approximately 16,000 and 7,600 feet (Table 5-6) in the two principal directions of
flow, N5-10°W and N60°E. The contributing area was mapped using surface drainage divides
corresponding closely to the TOT distances (Figure 5-23). The downgradient limit of Zone III was
placed at the southern end of Zone IIB rather than extending the protection zone further. This step
was taken to account for the ground-water gradient and the stagnation point downgradient from TW.
Extending the protection zone an additional 8,400 feet downgradient is not realistic hydrogeologicaliy,
and would overprotect the resource. Protection strategies within Zone III should focus on public
education and awareness for the production well at Blackwater Brook.
5-51
-------
WELLHEAD PROTECTION MONITORING PROGRAM
MONITORING OBJECTIVES
Once the assessment of contaminant sources has been made, flowpaths have been adequately
defined, and WHPAs have been delineated (Figure 5-24), monitoring in the traditional sense of the
term takes place. The monitoring program can be divided into two primary facets: physical
monitoring and hydrochemical monitoring.
Monitoring of physical parameters such as water levels, flow volume, and temperature play a
continuing role in the overall monitoring strategy in several ways. A prime objective of the Dover
ground-water monitoring program is to discern the longrterm impact of water production on water
levels within the fractured bedrock and the overburden aquifers. At the Blackwater Brook site, the
surficial materials have moderate storage and transmissivity capabilities because of the presence of
marine clays and silts. The ability of the N60°E faulted contact zone and the intersecting N5-10°W
fracture zone to draw water from outside the immediate drainage basin to TW remains unknown, but
can be assessed by physical monitoring. Once a production well is brought on line, steady-state
ground-water flow conditions may no longer prevail because of the varying pumping rates. Variation
in pumping rates may expand the ZOC to include previously excluded contaminant threats. A
physical monitoring program could provide the data needed to predict expansion of the ZOC and
possibly indicate a need for increased hydrochemical monitoring near a potential contaminant source.
Physical monitoring results may justify hydrochemical monitoring objectives and parameters.
For example, techniques such as flowmeter and hydrophysical logging, used to develop a conceptual
model for delineation, indicate that ground-water flow at the Blackwater Brook site is three-
dimensional. Water flows up or down in boreholes, following convoluted pathways as it moves
through bedrock fractures. Observation of vertical flow, as well as FEC and pH stratification,
indicates that vertical averaging of physical and chemical parameters in a monitoring well will not
adequately describe either the ambient trends or the characteristics of a possible contaminant plume.
If a contaminant plume is small, or if a plume enters the aquifer with a vertical flow component, the
monitoring network should be optimally designed to detect the contaminant concentration and its
associated flowpath. These objectives may be compromised if vertical averaging in the monitoring
well occurs.
Hydrochemical monitoring serves to determine ambient water quality trends, identify sources
of contamination, and provide early warning detection of pollutants. Hydrochemical monitoring
focuses management, remediation, and prevention efforts in municipal and adjacent water districts.
Location strategies for hydrochemical monitoring that apply to the Blackwater Brook site include
monitoring at the wellhead, at WHPA or zone boundaries, and at point sources.
5-52
-------
* SEWAGE
TREATMENT
PLANT
i L
3RAVEL^
3UARRY
ROCK
QUARRY
•f
COUNTY HOME
(USTs) /
CAR
[ DEALER
WASH < GAS
AUTOMOTIVE
SERVICE
STORAGE
GAS
STATION
.SALT
STORAGE
Figure 5-24. Potential contaminant sources in proximity to wellhead protection zones,
Blackwater Brook site, Dover, New Hampshire.
5-53
-------
At Blackwater Brook, the objective of early warning monitoring is difficult to achieve because
of the high average flow velocities within the fracture system. Despite strong evidence that the
primary ground-water flow directions have been identified and can be monitored, early warning
detection of unanticipated contamination with a monitoring well system is extremely difficult, given
the rapid and highly discrete character of water movement in the bedrock aquifer. The focus of
hydrochemical monitoring in this hydrogeologic environment is on source assessment and ambient
conditions along with contingency planning to achieve the overall objective of ground-water
protection. To address these concerns, the following strategy for monitoring has been designed,
dealing specifically with monitoring sites, parameters, frequencies, quality assurance and quality
control (QA/QC) considerations, and the data base.
MONITORING SITES
Existing sites
All bedrock and overburden wells installed in the research monitoring program will serve in a
physical and a hydrochemical monitoring program at the site (Figure 5-25). Sampling parameters and
frequencies for these wells are discussed in the subsection titled "Monitoring Parameters and
Frequencies."
Recommended Sites
Additional wells for physical monitoring should, first, be placed at a greater distance from TW
than the existing wells. Optimally, new wells should be completed in bedrock along the N60°E
lineament and along the N5-10°W trend as much as a mile from TW to better characterize the
fracture system and to provide distant monitoring points for water levels. At a minimum, two
recommended bedrock well sites are R5 and R6 (Figure 5-25). Of secondary, but significant,
importance is locating some new monitoring wells in areas where data gaps exist: that is, in other
directions from TW than along the N5-10°W and N60°E trends. Additional wells placed in other
directions could be used to verify water levels, drawdown, and the TOT contours. Domestic bedrock
wells such as Dl (Figure 5-25), or others drilled in the future, should be used for physical monitoring.
New overburden wells should be placed in areas known or suspected to be sources of direct
recharge to the bedrock system. Marine clay in the overburden thins from 32 feet thick at TW to only
6 feet in well O3. The clay deposit thins eastward and pinches out. Marine sand deposits northeast,
east, and uphill of the site constitute a major recharge area for TW. The estimated extent of marine
clay in the vicinity of the site is shown in Figure 5-25. Consequently, the hillslope and hilltop
northeast and east of the site are appropriate locations for one or more overburden wells for physical
monitoring (for example, well O8 as shown on Figure 5-25).
The site is reasonably well monitored for hydrochemical parameters by the existing four
bedrock and four overburden wells (Figure 5-25). However, the location of another overburden well,
well O5, nested with well R5 just east of the Spaulding Turnpike would improve source assessment
monitoring (Figure 5-25). Wells O5 and R5 would serve to monitor unanticipated spills along the
Spaulding Turnpike.
5-54
-------
LANDFILL
f, INDUSTRY
§ FUEL STORAGE
X QUARRY
„. BEDROCK
+R5 MONITORING WELL
TESTWELL
DOMESTIC WELL
"\ TOWN LINES
—,v CLAY-SAND CONTACT
IP SURFACE WATER
Figure 5-25. Existing and proposed monitoring sites, Blackwater Brook site, Dover, New Hampshire.
5-55
-------
To monitor the threat from septic systems and domestic storage tanks, single monitoring wells
should be installed in the overburden directly downgradient (approximately 300 feet) from the homes
and upgradient of the future production well. One or more overburden wells on the hillslope or
hilltop northeast and east of the site (for example, well O8 shown in Figure 5-25) would be very
useful. Permeable marine sand deposits in these areas provide significant recharge to the system.
Such overburden wells would be necessary to monitor septic systems if the area is residentially
developed.
A landfill 3 miles northwest of the site lies within Zone III (Figures 5-24 and 5-25). It may be
prudent to monitor this landfill as a potential contamination source as a secondary objective of the
program. To monitor a potential point source, one monitoring well should be placed upgradient of
the source and two or more monitoring wells downgradient from the source. The upgradient well
should be located and sampled to obtain background levels for potential contaminants from the source
(U.S EPA, 1986).
In many circumstances, WHPAs cross political boundaries. Optimally, monitoring and
compliance with protection regulations can be coordinated and implemented by adjoining towns. In
some cases it may be necessary to monitor the political boundaries. To ensure wellhead protection,
monitoring wells should be placed where Zone II, and possibly Zone III, are truncated by town
boundaries. Zone H is truncated by the town of Somersworth and the city of Rochester boundaries
along the N5-10°W trend and by the town of Somersworth boundary along the N60°E trend
(Figure 5-25). If Zone II is truncated politically, at least one overburden and one bedrock monitoring
well should be located adjacent to the city limits at about the center of each arm of Zone IIA. Wells
O6 and R6 should be positioned at the center of the N5-10°W trend, and wells O7 and R7 at the
center of the N60°E trend (Figure 5-25). New wells along the N5-10°W trend should be located as
close as possible to the intersection of a line drawn from TW to the city limits. Along the N60°E
trend, it is recommended that surface geophysical surveys be used to locate new monitoring wells in
the same manner that TW was sited.
MONITORING PARAMETERS AND FREQUENCIES
A production well has not yet been established at the Blackwater Brook site. Only the test
well and the monitoring wells installed as part of the research monitoring program are currently in
place. Accordingly, two monitoring programs, consisting of detailed parameters and. frequencies, have
been developed to reflect the current scenario and to reflect conditions of an on-line production well.
Table 5-7 presents a summary of monitoring parameters and frequencies recommended for existing
conditions. Recommendations for monitoring a water-supply production system at the site are given
in Table 5-8.
Recommended physical monitoring parameters (parameters that bear physical units of
measure) include water level, temperature, and specific conductivity. A chemical parameter often
categorized with these physical parameters is pH. Together, these four parameters constitute a field-
screening procedure for water quality changes. Automated instrumentation and data recorders should
be considered for acquiring physical and chemical parameter data, where possible. The initial high
cost of field-dedicated equipment is usually offset by the increase in frequency and quality of the data,
and the rapid turnaround of results.
5-56
-------
TABLE 5-7. SAMPLING PARAMETERS AND FREQUENCIES FOR THE TEST WELL
AND ASSOCIATED MONITORING WELLS, BLACKWATER BROOK SITE,
DOVER, NEW HAMPSHIRE
Well(s)
Parameters
Frequency
Test well
Overburden wells
Bedrock wells
Water level
Temperature
Specific conductivity
PH
Cation and anion suite
Bacteria
VOCscan
Heavy metals
Water level
Temperature
Specific conductivity
PH
Cation and anion suite
Bacteria
VOC scan
Heavy metals
Water level
Temperature
Specific conductivity
PH
Cation and anion suite
Bacteria
VOCscan
Heavy metals
Radon
Uranium
Monthly
Biannually
Annually
Monthly
Biannually
Annually
Monthly
Biannually
Annually
Abbreviation: VOC = volatile organic compound.
5-57
-------
TABLE 5-8. SAMPLING PARAMETERS AND FREQUENCIES FOR THE PROPOSED
PRODUCTION WELL AND ASSOCIATED MONITORING WELLS,
BLACKWATER BROOK SHE, DOVER, NEW HAMPSHIRE
Well
Parameters
Frequency
Production well
Overburden wells
Bedrock wells
Water level
Applicable state, local, and
SDWA requirements
Iron
Manganese
Water level
Temperature
Specific conductivity
pH
Sodium
Chloride
Cation and anion suite
Bacteria
VOC scan
Heavy metals
Water level
Temperature
pH
Specific conductivity
Sodium
Chloride
Iron
Manganese
Radon
Uranium
Cation and anion suite
Bacteria
VOC scan
Heavy metals
Daily*
Weekly
Monthly
Biannually
Annually
Monthly
Biannually
Annually
Abbreviations: SDWA = Safe Drinking Water Act; VOC = volatile organic compound.
* The city of Dover conducts daily water-level monitoring of all production wells.
5-58
-------
Monitoring of the test well should include monthly water-level measurements to establish
ambient ground-water conditions (Table 5-7). Water levels for all on-line production wells in the city
system are monitored every day by the Dover Department of Public Works. The production well at
Blackwater Brook should also be measured daily (Table 5-8).
Temperature, specific conductivity, and pH should be monitored monthly at the test well, at
bedrock and overburden wells used in the research phase, and at any new wells installed for long-term
monitoring (Table 5-7). After the production well is in service, the bedrock and overburden
monitoring wells should continue to be monitored monthly for water level, pH, temperature, and
specific conductivity.
Under present conditions, existing monitoring locations and the test well should be examined
biannually for general chemical character, including cations, anions, and bacteria, to establish the
natural hydrochemistry of the area. Overburden and bedrock wells should undergo annual scanning
for total VOCs and for heavy metals. Additionally, bedrock wells and the test well should be sampled
annually for radon and uranium.
Once the production well is on line, hydrochemical monitoring of the production well will be
dictated by applicable state and local stipulations, as well as the SDWA requirements (Table 5-8).
Because of the documented high levels of iron and manganese at the test well, it may be prudent to
sample the production well weekly for these elements. Overburden monitoring wells should be
sampled biannually for sodium and chloride, while bedrock wells should be sampled for sodium,
chloride, iron, manganese, radon, and uranium (Table 5-8). Monitoring for sodium and chloride is
important for two reasons: (1) to determine water quality degradation from road salting in both
aquifers, and (2) to track the movement of the deeper bedrock flow regime, which has high levels of
chlorine from recharge water induced by pumping. Results from the gamma borehole geophysical
logging indicate that uranium is present in well R4. While the threat of water quality degradation
from uranium is considered to be minimal, cautious monitoring is recommended. Radon levels are
elevated, and treatment may be considered when the production well is developed. Spatial and
temporal trends in radon levels may provide important information in determining future water
treatment requirements. All monitoring wells should be sampled biannually for cations, anions, and
bacteria and should be scanned annually for VOCs and heavy metals (Table 5-8).
QUALITY ASSURANCE/QUALITY CONTROL CONSIDERATIONS
Samples taken during monitoring are subject to acceptable QA/QC protocol. City personnel
will receive specific training in sampling and monitoring methodologies. Some analysis can be done in
the field. Field kits can be used to measure chemical parameters such as chloride, sodium, iron, and
manganese in the monitoring wells. When the production well is brought on line, sampling and
subsequent laboratory analysis for SDWA parameters will occur. To complement this regulatory
monitoring program, laboratory analysis of select chemical parameters from the monitoring wells is
recommended. In this way, field analysis of chemical parameters can be verified and improved.
Particular attention must be paid to proper ground-water sampling methodology, which
includes purging of monitoring wells, filtering of samples, proper calibration of instruments, and
collection of duplicate samples (Garrett, 1988). The duplicate sample should be refrigerated and kept
5-59
-------
as backup until successful analysis of the first sample is completed. At least one full water quality
sample from the production well should be analyzed by an EPA-certified laboratory each year.
Laboratory results should be reviewed by performing a mass balance calculation to compare total
milliequivalents per liter of cations and anions. Agreement should be within 10%. Chain-of-custody
for sample preparation and tracking will be the responsibility of the city of Dover Department of
Public Works.
MONITORING DATA BASE
Collection of data is not an end in itself. To be of use, the data should be compiled and stored
in an efficient and meaningful manner for later analysis. A great deal of information has been and
will continue to be generated by the research and long-term monitoring programs, including an
extensive data base on well installation and completion, aquifer testing, physical parameters, and
hydrochemical parameters. To store, manipulate, retrieve, and output such information easily requires
the use of a microcomputer with spreadsheet and interactive graphics or interpretive software. The
field data from the long-term monitoring conducted by the Department of Public Works personnel
should be entered into a spreadsheet data base on the day the data are collected. Hydrochemical data
received from the laboratory should also be entered into the spreadsheet to facilitate analysis, with the
original documentation stored in files. Review of these data should be undertaken periodically by
Public Works personnel to identify spatial and temporal trends, to pinpoint possible water quantity or
quality problems that can be discerned from the information, and to perform necessary QA/QC
checks. This may be accomplished in association with consultants.
CONTINGENCY PLANNING
Because ground-water flow travel times are so rapid and the flowpaths are so highly discrete
within the fractured bedrock aquifer, installment of an early-warning detection network of monitoring
wells is probably inappropriate. Based on the dye-trace results, ground water is predicted to travel
roughly 16,000 feet and 7,600 feet in 1,000 days along the N5-10°W and N60°E directions, respectively.
A WHPA of this size extends into adjoining towns (Figure 5-25). Therefore, it may be politically and
legally quite difficult to protect a bedrock water supply well with Dover's present zoning and
regulatory structure. Therefore, beyond identifying and monitoring known sources of potential
contamination, the optimal approach to protecting the production well from unanticipated
contamination is through public education and contingency planning. Contaminants may threaten the
wellhead from a major accident along the Spaulding Turnpike, from illegal dumping of toxic or
hazardous materials along a roadside into permeable sands and gravels that recharge the well, or from
an unknown "preregulation" source.
Monitoring Action Levels
Action levels are needed to determine when to put a contingency plan into effect. The
contingency plan for the Blackwater Brook site contains a three-tiered system of action levels for
water quality parameters based on the Maximum Contaminant Level Goals (MGLGs) set by EPA
(U.S. EPA, 1990b and 1990c), along with the SDWA standards. The focus of the action levels is
sampling results from the monitoring wells, because SDWA requirements regulate the production well.
5-60
-------
• Level A: Samples show secondary standard MCLGs exceeded. If a secondary
standard MCLG is exceeded at a monitoring well anywhere in the WHPA, sampling
frequencies increase and field or in-house analyses are supplemented by mandatory
analyses at certified laboratories that can provide court-admissible data.
• Level B: Samples show primary standard MCLGs exceeded. If a primary standard
MCLG is exceeded in a monitoring well anywhere in the WHPA, a contamination
event is suspected and a site investigation and monitoring strategy may be initiated,
possibly with drilling of new boreholes for monitoring.
* Level C: Samples show SDWA standards exceeded. If SDWA standards are exceeded
in a monitoring well anywhere in the WHPA or if the Level B action level is reached
at the pumping well, the production well is immediately taken off line.
Thus, a Level A situation occurs when a water sample from a monitoring well shows values
greater than the EPA MCLG for a secondary drinking water standard. Two actions are then initiated:
monitoring frequency increases appropriately (for example, from yearly to quarterly), and analysis by a
certified laboratory is initiated in conjunction with field or in-house sampling. If an MCLG is
exceeded for a primary drinking water standard in a monitoring well (Level B), a contamination event
such as a spill or UST leak is suspected. A site investigation and a new, intensified monitoring
strategy are initiated (Pfannkuch, 1982). If Level C action occurs at a monitoring well, or if Level B
action occurs at the production well, the introduction of contaminants near the production well or
within the WHPA is assumed. Level C action, in which the SDWA standards at any monitoring well
are exceeded, will trigger immediate shutdown of the production well until the problem is understood
and remedied. Any level of violation may necessitate review and selection of additional or different
monitoring parameters tailored to the character of the contamination incident.
Contaminant Incident Response
Contingency planning begins with assignment of responsibility for specified actions in the
event of an emergency. If a spill or major accident with the chance of pollution takes place, police
will notify the Director of Public Works regarding the time, place, and severity of the accident, giving
a preliminary assessment of the character of the spilled material. The Director will, in turn, order
immediate shutdown of the production well. Public Works personnel will begin water quality
sampling at monitoring wells that are between the spill and the production well. These samples will
then be analyzed for VOCs, both in-house (if a gas chromatograph is available) and through a
certified laboratory. The rationale to screen for aquifer contamination with VOC analysis is based on
the unique source assessment and prioritization that is conducted in proximity to the test well.
Monitoring should proceed, with frequent samples taken in the early phases, perhaps every hour or
2 hours for the first 24 hours following the incident. The frequency of continued sampling should be
based on the presence or absence of VOCs in the wells.
5-61
-------
Unanticipated contamination events may require rapid emplacement of a source-assessment
monitoring program to determine the contaminant(s), extent of contamination, and effect on the
ground water. Such a monitoring program would include the following steps (Pfannkuch, 1982):
• Rapid preliminary survey
• Phased drilling of observation boreholes
• Phased monitoring of boreholes.
For the rapid preliminary survey, topographic maps would be used and a site visit conducted for
qualitative assessment of local surface drainage, topography (probable water-table configuration),
water levels in nearby wells, soil characteristics, and background levels of the contaminant.
Representative water samples for these assessments would be taken from upgradient or unaffected
wells.
Borehole testing and monitoring well installation would then begin. The first phase of
borehole drilling involves placing a minimum of three boreholes in a triangular pattern around the
spill site or contaminant source. Initial monitoring of these boreholes should provide data on ambient
levels of the contaminant (from the upgradient borehole) and the transport direction of the
contaminant. Monitoring results would aid in determining locations for drilling the next boreholes:
drilling downstream on the flowline, with triangulation determined from information obtained from
the first wells. Transport velocities in the downgradient direction can be obtained at this phase of the
assessment (Pfannkuch, 1982). Thus, initial monitoring and drilling proceed in an iterative manner,
with monitoring results determining the decisions for drilling locations and whether or not to
construct a permanent monitoring well at any specific drill hole. Also, the drilling and initial
monitoring allow refinement of the conceptual hydrogeologic model, which in turn allows (1) design
of the main monitoring program to track the evolution of the contaminant plume, (2) possible
numerical simulation of contaminant flow and transport, and (3) establishment of a clean-up plan.
Long-term and final monitoring for certain wells at selected and increasing time intervals
should track the rehabilitation and recovery (or lack thereof) of the aquifer (Pfannkuch, 1982). The
contaminated well should remain out of use until the surface spill is cleaned up and information from
exploratory wells or test pits is analyzed to determine the location and movement of contaminants.
Water supplies from other city wells will be used until the site is pronounced safe by city officials.
5-62
-------
CONCLUSIONS
Fractured bedrock aquifers pose unique challenges to the tasks of delineating and monitoring
wellhead protection areas (WHPAs). Some aquifers that possess sufficient secondary porosity and
permeability warrant conceptualization as equivalent porous media, and a number of analytical and
numerical models are applicable to simulate ground-water flow in these systems. Frequently, igneous
and metamorphic bedrock aquifers do not exhibit hydraulic behavior consistent with that of a porous
medium. For these types of aquifers, other approaches must be sought to characterize the physical
flow system.
This document offers an approach to delineating wellhead protection zones and establishing
monitoring strategies for a fractured bedrock aquifer in Dover, New Hampshire. A combination of
different geologic and hydrogeologic tools was used in the process of developing, testing, and refining
the conceptual model for the Blackwater Brook well site in Dover. The characterization approach
taken at the Blackwater Brook site may be used elsewhere, with .modification, to reflect different
geologic environments in other fractured bedrock aquifers.
The aquifer at the Blackwater Brook site is confined by up to 30 feet of marine clay in the
overburden and consists of both unconsolidated overburden and fractured crystalline bedrock.
Permeable sands and gravels, above the bedrock but beneath the clay, provide recharge to the
fractured bedrock aquifer. Two directions of relatively high bedrock transmissivity exist at the
Blackwater Brook site: one trending north 60 degrees east (N60°E) and one trending north 5 to
10 degrees west (N5-10°W). The N60°E trend is a fractured, faulted contact zone between the quartz
monzonite and the Berwick Formation metasedimentary rocks. This zone is expressed as a
photolineament and as a magnetic anomaly. The intersecting N5-10°W fracture zone is expressed
with strong signature on several platforms of aerial photography and is present in an outcrop area
north of the site.
Aquifer testing indicates that the direction of highest transmissivity within the bedrock is
north 30 degrees west (N30°W) along wells aligned perpendicular to the N60°E trend. A seepage
velocity of 1,680 feet per day (ft/d) was demonstrated with dye-trace results in this direction.
Drawdown contours are elongate in this direction during pumping of the test well, indicating preferred
flow along N30°W. Enhanced flow along this direction is attributed to the proximity and similar
orientation of the N5-10°W fracture zone. The N5-10°W fracture zone is believed to act as a conduit
for flow along the trend, but may be a partial barrier boundary to flow moving east across the trend.
Dye-trace results indicate a seepage velocity of 96 ft/d along the N60°E trend.
The hydrochemistry at the Blackwater Brook site is characterized by stratification of water
types; pH and fluid electrical conductivity increase at abrupt intervals with depth. These
characteristics reflect the presence of discrete flow zones at depth within the bedrock aquifer. The
ground water at the site is free of contamination from bacteria, volatile organic compounds (VOCs),
5-63
-------
pesticides, and heavy metals. The majority of water quality parameter levels are within state and
federal standards, with only a few exceptions. Iron and manganese levels exceed secondary standards
in both the bedrock and overburden aquifers due to naturally occurring sources in the subsurface.
Additionally, elevated levels of sodium and chloride exist in the bedrock aquifer, probably from
trapped sea water. These levels are expected to drop as the production well is developed and used.
The only serious contaminant threat is naturally occurring radon, which may require treatment. The
potential for contamination at the site is low, but future threats include spills along the Spaulding
Turnpike, highway de-icing, illegal dumping, and domestic septic system leachate.
Time-of-travel (TOT) contours for the delineation process were obtained, in part, by analysis
of aquifer-test results, including test pumping and dye tracing. A quadratic equation was derived
relating the time of first arrival of the dye at the pumping well, t, to the distance from the injection
well to the pumping well, t:
i = Ct2
the constant, C, in the equation was determined from dye-trace results along the two directions of
higher transmissivity in the bedrock aquifer, N5-10°W and N60°E. The constant, C, varies with
direction.
The 200-day and 1,000-day TOT thresholds applied at the Blackwater Brook site represent a
conservative approach to WHPA delineation in this fractured bedrock aquifer. Time-of-travel
calculations assume that a contaminant is already in the bedrock aquifer system, as was the injected
dye. At the bedrock outcrop, a spill or illegally-dispersed substance could penetrate very quickly into
the bedrock aquifer. In another scenario, however, a contaminant released through the overburden
could take significant time to reach the bedrock aquifer.
The criterion for delineating the Zone I protection area is the state-mandated 400-foot
protective radius. The area encompassed by the 200-day TOT contour along the direction of highest
bedrock transmissivity is defined as Zone IIA. This zone will be protected under the City of Dover
Ground Water Protection Ordinance. Zone IIB was delineated to compensate for the technical
limitations and difficulties of precise definition of Zone IIA. Zone IIB is the area defined by a
smooth curve connecting the outer boundaries of Zone IIA. This zone will be protected under the
State of New Hampshire Wellhead Protection Program regulations, which are somewhat less
restrictive than the City of Dover Ordinance. Zone III is the upgradient contributing area to Zones
IIA and IIB and the 1,000-day TOT distance measured along the zones of highest transmissivity,
using surface drainage contribution areas to delineate the zone. The downgradient limit of Zone III
was placed at the southern end of Zone IIB, rather than extending the zone to the nearest surface-
water divide. This step was taken to account for the ground-water gradient (toward the southwest)
and the stagnation point downgradient from the test well
Recommendations for long-term monitoring of the WHPA include locations for new
monitoring wells, physical and hydrochemical parameters, and sampling frequencies. All existing
overburden and bedrock wells installed as part of the research monitoring program will serve as long-
term monitoring stations. Three pairs of overburden and bedrock monitoring wells should be
5-64
-------
installed at a greater distance from the test well, each coinciding with an identified subsurface fracture
zone. These well pairs will serve (1) to better characterize the fracture system, (2) to provide more
distant monitoring points for water levels and select hydrochemical parameters, and (3) to monitor
upgradient activities in bordering towns. One new overburden well is recommended for installation
downgradient from a residential area to monitor septic systems and domestic storage tanks.
Two programs, with detailed sampling parameters and frequencies, are recommended: (1) to
monitor the current test well and associated observation wells, and (2) to monitor the future
production well and associated observation wells. Monitoring of the test well and existing observation
wells focuses on gathering ambient-water quality and physical parameter data. Hydrochemical
parameters of interest include the cation and anion suite, pH, bacteria, VOCs and heavy metal scans,
radon, and uranium. Key physical monitoring parameters include water level, temperature, and
specific conductivity. Sampling frequencies vary for categories of parameters from monthly to
annually.
After the production well is operational, hydrochemical monitoring will be dictated by
applicable state and local stipulations, including the Safe Drinking Water Act (SDWA) requirements.
Levels of iron and manganese should be monitored weekly at the production well for decisions
regarding water treatment. Levels of sodium and chloride should be monitored in the overburden and
bedrock wells biannually to assess the impact of road-salting and to connate water movement. Levels
of iron, manganese, radon, and uranium should be monitored biannually in the bedrock wells to
determine potential water quality degradation caused by natural bedrock sources. All overburden and
bedrock wells should be sampled biannually for cations, anions, and bacteria, as well as undergo
annual VOC and heavy metal scans.
Early-warning detection monitoring is inappropriate in the fractured bedrock aquifer because
ground-water velocities are rapid and flowpaths are discrete. The optimum approach to protecting the
production well from unanticipated contamination is through public education and contingency
planning. A three-tiered system of action levels is proposed for water quality parameters to trigger
select contingency responses. Action levels are based on the Maximum Contaminant Level Goals set
by the EPA, in conjunction with SDWA requirements.
• Level A violations trigger an increase in sampling frequency for target parameters, or
additional parameters if warranted. All analyses are procured through certified
laboratories.
« Level B violations signify a potential contamination event and justify a site-specific
remedial investigation.
• Level C violations indicate that SDWA standards are exceeded, and the production
well is shut down.
It is important to emphasize that the unique TOT distance values determined for Zones IIA
and IIB from the quadratic equation and the dye-trace results should not be construed as absolute.
Bedrock aquifers are immensely complex. The calculated TOT distances for both the 200-day and
1,000-day thresholds are large, and the hydrogeologic conceptual flow model of the test well area
becomes much less reliable at greater distances. Fractures recharging the test well may, or may not, be
5-65
-------
regionally continuous. Water flow through the fractured bedrock aquifer is likely to be contributed
from additional, unmapped structures. However, employing the TOT criterion in conjunction with the
dye-trace method, then modifying the protection zones based on a sound conceptual model of the
aquifer system, provides a defensible approach to wellhead protection in this fractured bedrock regime.
5-66
-------
RECOMMENDATIONS
As the city of Dover develops new water supply sources, the test well at the Blackwater Brook
site will be replaced with a production well. The city purchased 50 acres of property surrounding the
test well and plans to promulgate secondary and tertiary protection zones for this parcel. Other steps
that may be considered as part of the city's ongoing efforts to protect their water supply include:
• Expansion of the recommended physical and hydrochemical monitoring program
• Joint implementation of wellhead protection management and regulation with adjacent
municipalities
• More complete source assessment, possibly including door-to-door surveys, screening,
and risk ranking as the area is further developed
The recommendations made in this report for long-term monitoring should be implemented.
Of particular concern is the sand unit mapped to the north and northeast of the well, a deposit that
may supply considerable recharge to the bedrock aquifer. The area should be monitored, in part, to
evaluate the role of the sand unit in the recharge cycle.
The best scientific attempt to characterize the site hydrogeology and potential contaminant
pathways can be verified or improved only through monitoring of the hydrologic cycle in an iterative
procedure. The long-term monitoring program and its resultant data base will verify or refute the
technical reliability of the WHPA delineation.
Zones II and III of the WHPA extend into the jurisdiction of adjoining municipalities. For
this reason, negotiations between municipalities should occur to ensure that governing bodies enforce
protection of shared water supplies. The optimum approach is for all municipalities to implement
wellhead protection in sensitive areas. If critical differences in implementation strategy cannot be
resolved through negotiation, then the city of Dover should install and monitor wells at municipal
boundaries, as previously described.
As the area to the north of Dover continues to develop, improvements in the preliminary
source assessment will be required. This area is currently unsewered, and developments will utilize
onsite septic systems until sewer lines are extended into the rural areas. As residential areas develop,
particularly on land identified as important recharge to the well, additional source assessment
monitoring for nitrates may be needed. Another important target for improved source assessment is
the landfill to the northwest. The present hydrogeologic conceptual model indicates that flow from
the landfill toward the test well is unlikely — the landfill is outside of Zone II. Nevertheless, source
monitoring is appropriate for baseline characterization until a future time when monitoring data verify
the correctness of the Zone boundaries.
5-67
-------
To improve the conceptual model of the site, numerical simulation of ground-water flow may
be considered, taking advantage of the extensive data base generated in the research monitoring
program. The Blackwater Brook flow system should not be modeled with conventional numerical
codes such as MODFLOW or AQUIFEM because they employ a mathematical foundation for
porous, rather than fractured, media. However, recently developed model codes that simulate dual
porous media hold promise for understanding ground-water flow in complex, fractured bedrock
aquifers.
5-6S
-------
REFERENCES
Aley, T. 1990. Procedure for Analysis of Fluorescein and Rhodamine-WT Dyes in Water or
Charcoal Samplers. Ozark Underground Laboratory, Protem, Missouri. 6 pp.
American Water Works Association. 1988. Standard for Water Wells. ANSFAWWA A100-84.
Denver, Colorado.
Billings, M. P. 1956. The Geology of New Hampshire, Part II: Bedrock Geology. New Hampshire
Department of Resources and Economic Developments, Concord, New Hampshire. 207 pp.
Bradley, E. 1964. Geology and Ground-Water Resources of Southeastern New Hampshire. Water
Supply Paper 1695. U.S. Geological Survey, Denver, Colorado. 80 pp.
Canter, L. W, R. C Knox, and D. M. Fairchild. 1987. Ground Water Quality Protection Lewis
Publishers, Inc., Chelsea, Michigan. 562pp.
Cooper, H. H., Jr. and C. E. Jacob. 1946. A Generalized Graphical Method for Evaluating
Formation Constants and Summarizing Well Field History. American Geophysical Union
Transactions 27(4): 526-534.
Gotten, J. E. 1988. Ground-Water Resources of the Lamprey River Basin, Southeastern New
Hampshire. Water Resources Investigations Report 84-4252. U.S. Geological Survey Denver
Colorado. 25 pp. '
Garrett, P. 1988. How to Sample Ground Water and Soils. National Water Well Association
Dublin, Ohio. 37pp.
Jenkins, D. N. and J. K. Prentice. 1982. Theory for Aquifer Test Analysis in Fractured Rock Under
Linear (Nonradial) Row Conditions. Ground Water 20(1): 12-21.
Kruseman, G. T. and N. A. de Ridder. 1989. Single Vertical Fractures. In Analysis and Evaluation
of Pumping Test Data (Chapter 18). International Institute for Land Reclamation and Improvement
Wageningen, The Netherlands, pp. 263-287.
Lyons, J. B., W. A. Bothner, R. H. Moench, and J. B. Thompson. 1986. Interim Geologic Map of
New Hampshire. Open File Map 86-1. Office of New Hampshire State Geologist, Concord, New
Hampshire.
Lyons, J. B., E. L. Boudette, and J. N. Aleinikoft. 1982. The Avalon and Gander Zones in Central
Eastern New England in Major Structural Zones and Faults of the Northern Appalachians. Special
5-69
-------
Paper 24. In: Major Structural Zones and Faults of the Northern Appalachians. St. Mien, P. and J.
Beland, editors. Geological Society of Canada, Waterloo, Ontario, Canada, .pp. 43-66.
Mack T. J. and S. Lawlor. 1991. Geohydrology and Water Quality of Stratified Drift Aquifers in the
Bellamy, Cocheco, and Salmon Falls River Basin, Southeastern New Hampshire. Water Resources
Investigation Report 91-4161. U.S. Geological Survey, Denver, Colorado. 91 pp.
McWorter, D. B. and D. K Sunada. 1977. Ground-Water Hydrology and Hydraulics. Water
Resources Publications, Fort Collins, Colorado. 290 pp.
Morris, D. A. and A. I. Johnson. 1967. Summary of Hydrologic and Physical Properties of Rock: and
Soil Materials as Analyzed by the Hydrologic Laboratory of the United States Geologic Survey, 1948-
1960. Water Supply Paper 1839-D. U.S. Geological Survey, Denver, Colorado.
Mull D. S., T. D. Liebermann, J. L. Smoot, and L. H. Woosley, Jr. 1988. Application of Dye-Tracing
Techniques for Determining Solute-Transport Characteristics of Ground Water in Karst Terrains.
EPA 904/6-88-001. U.S. Environmental Protection Agency, Region 4, Atlanta, Georgia. 103 pp.
Novotny R F 1968. Geologic Map of the Seacoast Region, New Hampshire, Bedrock Geology.
New Hampshire Department of Resources and Economic Development, Concord, New Hampshire.
Paillet F. L., A. E. Hess, C. H. Cheng, and E. Hardin. 1987. Characterization of Fracture
Permeability With High-Resolution Vertical Flow Measurements During Borehole Pumping. Ground
Water 25(1): 28-40.
Papadopolus, L S. 1965. Nonsteady Flow to a Well in an Infinite Anisotropic Aquifer. In
Proceedings Dubrovonick, Symposium on Hydrology of Fractured Rocks. International Association
of Scientific Hydrology, Dubrovonick, Yugoslavia, pp. 21-31.
Pedler, W. 1991. Hydrophysical Logging of Wellbores BW-1, BW-3, BW-6, and BW-7, Blackwater
Brook Site, Dover, New Hampshire. GZA GeoEnvironmental, Inc., Newton Upper Falls,
Massachusetts. 38 pp.
Pfannkuch, H. O. 1982. Problems of Monitoring Network Design to Detect Unanticipated
Contamination. Ground Water Monitoring Review 2(1): 67-76.
Piper, A. M. 1944. A Graphic Procedure in the Geochemical Interpretation of Water Analyses.
American Geophysical Union Transactions 25: 914-923.
Plumb, R. A., A. Brie, and K Hsu. 1985. In Situ Stress, Natural Fracture Distribution, and Borehole
Elongation in the Auburn Geothermal Well, Auburn, New York. Journal of Geophysical Research
90: 5497-5512.
Randall, A. D., M. P. Thomas, C E. Thomas, Jr., and J. A. Baker. 1966. Water Resources Inventory
of Connecticut, Part 1: Quinebaug River Basin. Connecticut Water Resources Bulletin 8. 102 pp.
5-70
-------
Theis, G V. 1935. The Relation Between the Lowering of the Piezometric Surface and the Rate and
Duration of Discharge of a Well Using Groundwater Storage. American Geophysical Union
Transactions 16: 519-524.
Toth,J. 1963. A Theoretical Analysis of Groundwater Row in Small Drainage Basins. Journal of
Geophysical Research 68: 4795-4812.
U.S. Environmental Protection Agency. 1990a. Guide to Ground-Water Supply Contingency
Planning for Local and State Governments. EPA 4GO/6-90-003. U.S. Environmental Protection
Agency, Office of Ground-Water Protection, Washington, D.C 83 pp.
U.S. Environmental Protection Agency. 1990b. Fact Sheet for Drinking Water Regulations Under
the Safe Drinking Water Act. U.S. Environmental Protection Agency, Office of Drinking Water
Criteria and Standards Division, Washington, D.C 43 pp.
U.S. Environmental Protection Agency. 1990c. Drinking Water Regulations and Health Advisories.
US. Environmental Protection Agency, Office of Drinking Water, Washington, D.C 10 pp.
U.S. Environmental Protection Agency. 1986. RCRA Ground Water Monitoring Technical
Enforcement Guidance Document. OSWER-9950.1. U.S. Environmental Protection Agency, Office
of Solid Waste and Emergency Response, Washington, D.C 208 pp.
Vernon, J. H., F. L. Paillet, W. H. Pedler, and W. J. Griswold. 1993. Application of Borehole
Geophysics in Defining the Wellhead Protection Area for a Fractured Crystalline Bedrock Aquifer.
The Log Analyst 34(1): 41-57.
Zemanek, J., E. Glenn, Jr., L. J. Norton, and R. L. Caldwell. 1970. Formation Evaluation by
Inspection With the Borehole Televiewer. Geophysics 35: 254-269.
5-71
-------
-------
APPENDIX 5-A
DERIVATION OF THE TIME-OF-TRAVEL EQUATION,
BLACKWATER BROOK SITE,
DOVER NEW HAMPSHIRE
5-73
-------
-------
DERIVATION OF THE TIME-OF-TRAVEL EQUATION
When using dye-trace breakthrough curve results for volume measurements of a hydraulic
system, the time to the peak or the time to the mass of the centroid of the trace generally provides the
best estimate of the mean travel time (Mull et al, 1988). However, assuming that a contaminant
travels principally by way of advection in a fractured bedrock aquifer, the most useful aspect of the
dye-trace results for purposes of protecting the well is the time of first arrival of the dye. Additionally
utilizing the time of first arrival aids in data analysis and interpretation and reduces costs.
The first arrival times of rhodamine-WT and fluorescein (2.2 and 148 hours, respectively see
Figure 5-21) reflect variations in the direction and distance of travel from the injection wells to the test
well (TW, the detection well) where samples were taken. In order to contour the time-of-travel
(TOT) from the injection wells to TW on a map, the effects of direction and distance must be
separated. The theoretical approach to determining TOT distances presented here justifies a quadratic
relationship in which the time of first arrival (or the TOT) is proportional to the square of the
distance between the injection and detection wells. The validity of the quadratic relationship depends
upon the validity of the following assumptions:
• Darcy's Law holds
• The Thiem equation holds
• Steady-state conditions prevail
• Discharge, transmissivity, and porosity are constant in space and time
Any mathematical description or model of a natural system must be based on simplifying
assumptions. The applicability of the model depends on the appropriateness of the assumptions and
on the recognition or prediction of circumstances under which the assumptions will be valid Darcy's
Law applies to ground-water flow except in areas of rock with large openings or in areas of steep
hydraulic gradients, such as the vicinity of the pumped test well. Flow in fractured bedrock is often
different from that in a porous medium. Fracture openings may be large with turbulent water flow
Darcy's Law assumes laminar flow. Darcy's Law may be violated close to the pumping well at the '
Blackwater Brook site where hydraulic gradients are steeper. Flow in the fracture directly connecting
TW to well R2, for example, is probably non-Darcian. However, over the larger distances to be
considered in the wellhead protection area (WHPA) delineation process, the bulk rock properties and
flow may be nearly approximated by Darcy's Law. At large distances, radial flow patterns may
predominate even when pumping analysis indicates a linear flow system near the well (Jenkins and
Prentice, 1982).
The derived equation is taken from standard, accepted hydrogeologic relationships (Darcy's
Law and the Thiem equation). It represents the most simple analytical relationship to interpret the
hydrogeologic data. Graphical plots of distance versus drawdown from pumping TW at 200 gpm
indicate that the well stabilized about 7 days (10,000 minutes) after pumping began. Therefore the
steady-state assumption required to use the Thiem equation is satisfied. Discharge was constant at
200 gallons per minute throughout the 15 days of pumping. Hydraulic conductivity and porosity
should not vary appreciably in time, but may vary in space as a result of local heterogeneities in the
bedrock. However, the use of empirically determined seepage velocities from the dye-trace results
eliminates the need to estimate the parameters of hydraulic conductivity (K) and porosity (0) required
5-75
-------
to calculate seepage velocities independently of the average velocity, V, where V = (K dh/di)/<|>, and
dh/dt is the hydraulic gradient.
The results represent an intermediate approach between TOT distances calculated using pipe-
flow assumptions and those determined using an equivalent porous medium approach. The Thiem
equation assumes radial flow, which is not appropriate in the case of the Blackwater Brook bedrock
aquifer. The aquifer is best characterized as a linear flow system along the N5-10°W fracture zone,
which justifies an intermediate approach in the analytical velocity calculation. However, this problem
is solved to some degree by presenting separate solutions and using different constants in the TOT
equation for the two dye-trace directions, N30°W and N60°E. Additionally, uncertainties of the
exercise are mitigated by using the results of other research technologies, such as borehole geophysics
and geologic mapping, to modify the calculated TOT distances to create wellhead protection zones for
the well.
Assuming that Darcy's Law holds,
_ -
-K dh
d\
(1)
where V is average velocity, K is hydraulic conductivity; is porosity; and dh/di is the hydraulic
gradient, the change in head (h) with respect to distance (t) in a radial direction. We also know that
Q =
la
(Thiem equation)
(2)
where s_ and smw are changes in head in pumping and monitoring wells, respectively; Q is discharge;
T is transmissivity, and rw is the radius of the pumping well. Rearranging yields
and
(3)
Because head difference, s, equals change in head (h0 -
Under steady-state conditions, all of these parameters (except t and hmw) are constant with
respect to distance. Substituting the (h0 - hj) for s values and differentiating with respect to distance
(t) yields
d_
di
- V -
5-76
-------
which yields
dh
mwl
d\
- pn
[£ On
rl, .
[— ]» s
\
- (In
is constant.
Letting
= -
C,, we obtain
dh _ -Ci
di i '
(4)
-c,
Substituting the 1 term from Equation 4 as — in Darcy's Law (Equation 1) yields
v di
V =
(5)
Since the average velocity, V, equals —, where dt is the change in time, then
at
A
dt
Separating variables,
idi = —- dt
/fK
idi = I— C^dt yields
(6)
5-77
-------
The dye travels no distance at a time equal to zero, therefore, substituting 1=0 and t=0 into
Equation 6 indicates that C2=0, and
.
—
~^
2
v-
— Ct
^^ l/i*
4> >
(7)
Rearranging yields
*
2JLC,
Setting = C3 yields the following quadratic equation relating the time of first arrival of the
dye (t) to the distance from the injection well to TW (i). The constant, C3, varies with direction.
i = C3f2
(8)
For simplicity, Equation 8 can be expressed as
(9)
Values for C in Equation 9 were determined from the empirical results of the dye tracing for
the N30°W and N60°E directions at Blackwater. For the N30°W direction,
t = 130 min (0.09 d) and
t = 152 ft (well R2 to TW).
Given that i = Ct2 (Equation 9), and replacing,
1 1
152 y* = C (0.09)2 (d)2
c -
C = 5Q6ft/d2.
5-78
-------
The equation to determine TOT distances in the N30°W direction is therefore
l = 506 t2.
The equation for the TOT distance in the N60°E direction is determined in a similar manner,
using tm = 148 hr (6.2 d) and l = 596 ft, resulting in a value for C of 240 ft/d172. The equation in
the N60°E direction is
i = 240 t*.
Calculated distances corresponding to the 200-day and 1,000-day TOT for each direction are
presented in Table A5-1.
TABLE A5-1. TIME-OF-TRAVEL DISTANCES ALONG DYE-TRACE DIRECTIONS,
BLACKWATER BROOK SITE, DOVER, NEW HAMPSHIRE
Calculated Distances (ft)
TOT Criteria
(d)
N30°W
(R2 to TW)
N60°E
(R3 to TW)
200
1,000
7,200
16,000
3,400
7,600
Abbreviations: TOT = time of travel; TW = test well.
*~ '" •A'U.S. GOVERNMENT PRINTING OFFICE: 1994 - 523-865/81380
-------
-------
-------
-------
-------
-------
|