University of Southern California Dissertations and Theses

Modeling nitrate contamination of groundwater in Mountain Home, Idaho using the DRASTIC method

(USC Thesis Other)

Modeling nitrate contamination of groundwater in Mountain Home, Idaho using the DRASTIC method

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content MODELING NITRATE CONTAMINATION OF GROUNDWATER IN MOUNTAIN HOME,
IDAHO USING THE DRASTIC METHOD

by

Jenni Sue Dorsey-Spitz

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(GEOGRAPHIC INFORMATION SCIENCE AND TECHNOLOGY)

AUGUST 2015

Copyright 2015 Jenni Sue Dorsey-Spitz

ii

DEDICATION
I dedicate this document to my family, who inspired me to work hard and peruse my goals and to
my husband, who supported me through all of my adventures.

iii

ACKNOWLEDGMENTS
I would like to thank my thesis advisor, Dr. John Wilson, for his dedication, mentorship, and
support. To my committee members, Dr. Karen Kemp and Dr. Su Jin Lee, thank you for your
valuable suggestions and advice that greatly contributed to the success of this thesis. I will be
forever grateful to my husband and his unwavering support while I pursed my Master’s degree,
thank you.

iv

TABLE OF CONTENTS
DEDICATION ii

ACKNOWLEDGMENTS iii

LIST OF TABLES vi

LIST OF FIGURES vii

LIST OF ABBREVIATIONS ix

ABSTRACT x

CHAPTER 1: INTRODUCTION 1
1.1 Motivation 1
1.1.1 Mountain Home Air Force Base, Idaho 1
1.1.2 Nitrate Contamination 2
1.1.3 Degraded Groundwater 3
1.2 Water Sustainability 4
1.3 Purpose of this Thesis 5
1.4 Thesis Organization 6
CHAPTER 2: RELATED WORK 7
2.1 Groundwater Quality in Idaho 7
2.1.1 Groundwater Quality in Mountain Home AFB 9
2.2 Nitrate Effects on Human Health 10
2.3 Well Development 11
2.4 GIS-based Analysis Methods 14
2.4.1 DRASTIC Method 16
CHAPTER 3: METHODOLOGY AND DATA SOURCES 19
3.1 Study Area 19
3.1.1 Climate 21
3.1.2 Geology and Soils 21
3.2 DRASTIC Method 21
3.3 Data for DRASTIC Parameters 23
3.3.1 Depth to Groundwater 26
3.3.2 Net Recharge 26
3.3.3 Aquifer Media 27
3.3.4 Soil Media 28
3.3.5 Topography 28
3.3.6 Impact of Vadose Zone 28
3.3.7 Hydraulic Conductivity 29

v

3.4 Aquifer Vulnerability Assessment 29
3.5 Model Validation 31
CHAPTER 4: RESULTS AND DISCUSSION 33
4.1 DRASTIC Parameters 33
4.1.1 Available GIS Data 33
4.1.2 Site-Specific GIS Data 36
4.2 DRASTIC Results 38
4.2.1 Model 1 38
4.2.2 Model 2 40
4.3 Validation 43
CHAPTER 5: CONCLUSIONS 48

REFERENCES 51

APPENDIX A: ADDITIONAL DRASTIC PARAMETER MAPS FOR MODEL 1 -
GENERIC AVAILABLE DATA 59

APPENDIX B: ADDITIONAL DRASTIC PARAMETER MAPS FOR MODEL 2 -
SITE-SPECIFIC DATA 63

vi

LIST OF TABLES

Table 1. Natural and Human Factors Affecting Groundwater Quality 9

Table 2. Well Construction Minimum Separation Distances 13

Table 3. The seven DRASTIC model parameters and their relative weights 22

Table 4. DRASTIC parameters and rating values (adapted from Aller et al. 1987). 23

Table 5. Summary of available groundwater quality data from USGS and Mountain
Home AFB for 16 MWs and nine BPWs 24

Table 6. Descriptive statistics of Model 1, Model 2, and nitrate results from 25
wells. 43

Table 7. Model prediction results using MW and BPW data to validate the two
models. Mean Nitrate color corresponds to USEPA action level (>5.0)
standards, while colors correspond to model vulnerability risk classes.
False = 0, True =1 for correct values. 44

vii

LIST OF FIGURES

Figure 1. The Nitrogen Cycle, as it occurs on land. .............................................................. 3

Figure 2. Altitude of water level, in feet (ft), above NGVD 1929 indicating
groundwater levels declining at a rate of 1.08 ft per year ...................................... 4

Figure 3. Water infiltrating the subsurface flows through the groundwater system
and eventually discharges in streams, lakes, oceans, or is pumped from a
well. The residence time in the subsurface can vary from days to
thousands of years (Winter et al. 1998). .............................................................. 12

Figure 4. Vulnerability Map of the Idaho Snake River Plain (IDEQ 1991) ....................... 17

Figure 5. Probability of groundwater contamination by dissolved nitrite plus nitrate
as nitrogen for the Eastern Snake River Plain, Idaho (USGS 1999) ................... 18

Figure 6. Mountain Home AFB area map. .......................................................................... 20

Figure 7. Location of BPWs and MWs, which have been sampled for nitrates on
Mountain Home AFB, Idaho ............................................................................... 25

Figure 8. Hydraulic conductivity values for selected aquifer media types. ........................ 30

Figure 9. Soil type (symbology) and depth to water in feet (labels) for Mountain
Home AFB. The depth to groundwater is greater than 80 feet across the
entire study area. .................................................................................................. 34

Figure 10. General land cover type, consisting primarily of Urban/Developed Land
(red), Agricultural (including the golf course), and Non-Forested Lands. .......... 35

Figure 11. Site-specific land cover type to obtain Rated Net Recharge, ranging from
Urban/Developed Land (darker red) to barren, rangeland (white). Wells
are depicted in blue. ............................................................................................. 37

Figure 13. Impact of the Rated Vadose Zone, ranging from 9 to 3. Data obtained
from well driller’s logs. Wells are depicted in blue. ........................................... 39

Figure 13. Model 1 – DRASTIC Index of vulnerability for Mountain Home AFB,
using generic, publicly available data and overlaid with average nitrate
sampling results per well. .................................................................................... 41

Figure 14. Model 2 – DRASTIC Index of vulnerability for Mountain Home AFB,
using site-specific data and overlaid with average nitrate sampling results
per well................................................................................................................. 42

viii

Figure A 1. DRASTIC Parameter Aquifer Media, using generic, available data. .................. 59

Figure A 2. DRASTIC Parameter Topography (Slope), using contour elevation data. .......... 60

Figure A 3. DRASTIC Parameter Impact of the Vadose Zone, using generic, available
data. ...................................................................................................................... 61

Figure A 4. DRASTIC Parameter Hydraulic Conductivity, using generic, available
data. ...................................................................................................................... 62

Figure B 1. Rated DRASTIC Parameter Depth to Water, using site-specific data................. 63

Figure B 2. Rated DRASTIC Parameter Aquifer Media, using site-specific data.................. 64

Figure B 3. Rated DRASTIC Parameter Hydraulic Conductivity, using site-specific
data. ...................................................................................................................... 65

ix

LIST OF ABBREVIATIONS
AFB Air Force Base
BPW Base Production Well
CWA Clean Water Act
GIS Geographical Information System
IDEQ Idaho Department of Environmental Quality
IDW Inverse Distance Weighting
MCL Maximum Contaminate Level
MW Monitoring Well
NPA Nitrate Priority Area
NPDWR National Primary Drinking Water Regulations
NO3 Nitrate
NOAA National Oceanic and Atmospheric Administration
NWIS National Weather Information System
SDWA Safe Drinking Water Act
SVOC Semi-Volatile Organic Compounds
TCE Trichloroethylene
USAF U.S. Air Force
USEPA U.S. Environmental Protection Agency
USGS U.S. Geological Survey
VOC Volatile Organic Compound

x

ABSTRACT
Mountain Home Air Force Base (AFB) is located in Elmore County, southwestern Idaho. A
regional aquifer is the primary drinking water source for the base residents. While current
groundwater quality meets regulatory drinking standards, data collected from the U.S.
Geological Survey (USGS) and Mountain Home AFB indicates a significant degradation in
quality, particularly nitrate contamination. The purpose of this study was to implement a
groundwater model to spatially delineate areas by vulnerability to groundwater contamination
risk. The model provides a basis for evaluating the vulnerability to pollution of groundwater
resources based on hydro-geologic parameters, which can help develop management practices to
prevent additional nitrate groundwater contamination in the region. Two Geographical
Information System-based groundwater vulnerability models using the DRASTIC method were
created using generic, available data and site-specific data. The models were compared to each
other, as well as groundwater quality data gathered from 25 wells (16 monitoring wells and 9
base production wells) throughout the study site to validate the model. While the results indicate
that the site-specific model is slightly more reliable (56% prediction accuracy), compared to the
generic data model (48% prediction accuracy), neither set of model predictions seem good
enough to inspire confidence and it is clear that the results produced with the two model runs are
not interchangeable. The greatest cause is relative to the small sampling size (n=25) of the wells.
The small sample size limits the opportunities to conduct statistical analysis to validate the model
outcomes. Additional studies would need to be performed using the same approach, but with
larger sample sizes so that the sample size reported here (n=25) would not negatively affect the
results.

1

CHAPTER 1: INTRODUCTION
Groundwater is a valuable resource that provides a source of drinking water to the human
population. While 70% of the Earth is covered by water, groundwater only makes up a fraction
(0.6%) of all available water on Earth; however, that 0.6% accounts for 98% of the freshwater
available for human consumption (Zaporozec and Miller 2000).
Goundwater is stored in aquifers, which not only provides a water source through
extraction (e.g., pumped wells), but aquifers also contribute to surface waters such as rivers and
lakes (Schwartz and Zhang 2003). Unfortunately, across the globe, groundwater has been
significantly degraded by the human population through unsustainable use and contamination
(Morris et al. 2003). Significant contamination has been contributed by the disposal of human,
animal, and hazardous wastes; byproducts of mining and oil operations; leaks from sanitary
sewer lines and septic tanks; and land use operations such as farming (Schwartz and Zhang
2003).
1.1 Motivation
1.1.1 Mountain Home Air Force Base, Idaho
Groundwater supplies 95% of Idaho’s drinking water supply. However, some communities, such
as the Mountain Home Air Force Base (AFB), rely solely on groundwater due to the dry, arid
climates, and lack of surface water availability (IDEQ 2015).
Mountain Home AFB is located in Elmore County, Idaho and has approximately 6,500
base residents. Mountain Home AFB obtains all of its drinking water from the regional aquifer
through base production wells (BPWs). Since the establishment of the base in 1942, 11 on-base
production wells have been installed to provide drinking water. Of the 11, five wells have been
shut down, permanently or temporarily, due to high nitrate concentrations. Currently, the water

2

from two of the base production wells are mixed in order to reduce nitrate levels (ATSDR 2010).
In 1985, Mountain Home AFB began working with the U.S. Geological Survey (USGS) to
sample and monitor groundwater quality through 16 groundwater monitoring wells (MWs); six
of these wells currently exceed the U.S. Environmental Protection Agency’s (USEPA) maximum
contaminant limit of 10 milligrams per liter for nitrate. Additionally, two of the nine base
production wells that are used to provide drinking water also exceed the limit for nitrates.
Mountain Home AFB has also been listed as a nitrate priority area by the Idaho Department of
Environmental Quality (IDEQ). The purpose of this research was to implement a groundwater
model to spatially delineate areas by vulnerability to groundwater contamination. Data from
groundwater monitoring wells was used to validate the model.
Identifying high risk areas for groundwater concentration will help with long-term
management by identifying the extent of contaminated areas and potential areas susceptible to
contamination.
1.1.2 Nitrate Contamination
Nitrate is one of the most common groundwater contaminates in Idaho (IDEQ 2015). Nitrate
(NO3) is the naturally occurring form of nitrogen and is an integral part of the nitrogen cycle in
our environment (Figure 1). However, due to the capacity of sandy soils to move elements easily
through the soil, nitrate can leach into the groundwater. High concentrations of nitrates, typically
specified as a maximum contaminate level (MCL) [MCL > 10 milligrams per liter (mg/L)], are a
health risk and may cause methemoglobinemia (Blue Baby Syndrome) in infants (USEPA 2013).
Due to the growing number of areas with significant nitrate concentrations in
groundwater, the IDEQ has identified and prioritized areas as nitrate priority areas (NPAs).
Currently, Mountain Home AFB is ranked #14 out of 32 NPAs in Idaho (IDEQ 2008).

3

Furthermore, Mountain Home AFB has taken five drinking water production wells offline due to
the nitrates exceeding the MCL and posing health risks.
IDEQ works with areas identified as NPAs to implement management strategies and
water quality improvement and source protection plans (IDEQ 2014). In order to successfully
implement improvement and protection plans, initial data on the location of current and high risk
groundwater contamination needs to be identified.
1.1.3 Degraded Groundwater
Mountain Home AFB partnered with USGS in the 1980s to conduct groundwater monitoring and
sampling. Sixteen wells around the base are used for monitoring water quality parameters, in
addition to water levels. The data show that: (1) Mountain Home AFB groundwater levels are
declining at an average of 1.08 feet per year (see Figure 2); (2) nitrate concentrations are
increasing in three of the wells, decreasing in three of the wells, and show no consistent trend in
Figure 1. The Nitrogen Cycle, as it occurs on land.
Source: USEPA (2013)

4

the other 11 wells since 1985 to the present-day. Of the 16 MW wells, five exceed the MCL of
10 mg/L for nitrate.

Figure 2. Altitude of water level, in feet (ft), above NGVD 1929 indicating groundwater
levels declining at a rate of 1.08 ft per year
Source: Williams (2014)
The declining water table may alter the direction of groundwater movement (Perlman
2014). This change in flow as the water-level continues to decline might also contribute to some
areas having greater nitrate concentrations.
1.2 Water Sustainability
The combination of water degradation, declining aquifer levels, and Mountain Home AFB’s
dependence on the aquifer for drinking water supply drives the need to find one or more

5

solutions. Possible solutions can be identified by using the data to perform analysis, designing a
plan to improve and properly manage groundwater quality, and ultimately implementing one or
more strategies that lead to the sustainable use of water resources.
1.3 Purpose of this Thesis
Water quality and quantity are long-term and enduring concerns on Mountain Home AFB. While
actions such as installing water meters on houses, using reclaimed water for irrigation, and
educating the residents on the importance of water conservation have improved conditions,
future decisions regarding groundwater protection, well installation, and land management will
be informed by the results of monitoring and groundwater modeling.
The purpose of this thesis study was to assess groundwater vulnerability to pollution in
the regional, confined aquifer that supplies Mountain Home AFB using the DRASTIC GIS-
based groundwater risk assessment model. DRASTIC uses hydro-geological data layers (Depth
of water, net Recharge, Aquifer media, Soil media, Topography, Impact of vadose zone and
hydraulic Connectivity) to assess vulnerability to pollution (Aller et al. 1987). The following
objectives were identified in an effort to create a reliable model and demonstrate applications of
the DRASTIC Model method:
1. Use available GIS data with DRASTIC to generate vulnerability estimates for the
study area and compare predicted results with actual groundwater well
observations.
2. Use site-specific and GIS data with DRASTIC to generate vulnerability estimates
for the study area and compare predicted results with actual groundwater well
observations.

6

3. Compare the two sets of model estimates to determine whether there is a
significant difference between DRASTIC model performance using generic and
site-specific data.
1.4 Thesis Organization
The remainder of this thesis consists of four chapters. Chapter 2 describes past studies conducted
on Mountain Home AFB and related work incorporating GIS-based groundwater modeling.
Chapter 3 summarizes the data and methodology used for this study. Chapter 4 describes the
groundwater contamination risk model results, and Chapter 5 summarizes the conclusions that
can be drawn from the analysis conducted for this thesis project and presents some suggestions
for future work.

7

CHAPTER 2: RELATED WORK
This chapter starts with a description of the groundwater quality in Idaho, followed by a
summary of previous research conducted on Mountain Home AFB. An overview of the human
health risks associated with drinking contaminated groundwater, specifically nitrate
contamination and of groundwater wells as sources of groundwater contamination are then
provided. The chapter closes with a review of GIS-based analyses practices, specifically the
DRASTIC Method to develop risk vulnerability models for groundwater contamination.
2.1 Groundwater Quality in Idaho
Approximately 95% of the population in Idaho depends on groundwater as a source of drinking
water. The other 5% obtains their drinking water from surface water, including streams, rivers,
reservoirs, and springs (IDEQ 2015). Groundwater has many benefits over surface water, such as
minimal infrastructure (e.g., water pipes) and better natural protection from contamination and
drought (MacDonald, Davies, and Dochartaigh 2002). Nevertheless, groundwater can still
become contaminated from sources including human and animal wastes; haphazard disposal of
wastes from industrial practices; leaks from storage tanks, pipelines, or disposal ponds; land use
activities such as farming and fertilizer applications; and through poorly constructed wells
(UNICEF 1999, Schwartz and Zhang 2003).
Federal, state, and local governments have established regulations and guidance to
minimize the contamination of waters. The U.S. Federal Clean Water Act (CWA) regulates
discharges of pollutants into U.S. waters and establishes quality standards for surface waters (33
U.S.C. §1251 et seq. 1972). Additionally, the Safe Drinking Water Act (SDWA) regulates public
water supplies to ensure they are safe; however, the SDWA does not regulate private wells that
serve fewer than 25 individuals (USEPA 2014a). In Idaho, a large fraction of the population uses

8

private wells, which are not regulated under the SWDA (IDEQ 2015). Despite Federal
regulations established by the CWA and SWDA to control pollutants from entering waters,
contamination still occurs through improperly developed and operated private wells and
nonpoint pollution sources such as excess fertilizers, herbicides and insecticides from
agricultural lands and residential areas; oil, grease and toxic chemicals from runoff; and bacteria
and nutrients from livestock, pet wastes, and faulty septic systems (USEPA 2012).
The quality of Idaho’s groundwater is affected by human activities as described above
and by natural processes. These include chemistry and precipitation, dissolution of organic and
mineral substances from vegetation, soil, and rocks as water infiltrates the land surface and
percolates through earth materials, and length of time of contact with soil and rocks. These
natural factors determine the concentrations of dissolved minerals in groundwater (Yee and
Souza 1987). Natural and human factors affecting groundwater quality in Idaho are summarized
in Table 1.
Among the natural and human types of contaminants in Idaho (Table 1), arsenic, coliform
bacteria, and nitrates are the three most common (IDEQ 2015). Nitrate contamination has the
greatest impact on Idaho groundwater quality and is the most widespread. While nitrate, a form
of nitrogen, occurs naturally, urban activities are the primary source of additional nitrates.
Idaho’s groundwater has been significantly degraded and impacted by nitrate contamination,
such that 34 areas across Idaho have been identified as NPAs (IDEQ 2015).
IDEQ works with each of the NPAs to restore the degraded groundwater. For each of the
NPAs, a groundwater quality improvement and drinking water source protection plan is
developed in order to provide information to help prioritize and coordinate water quality related
activities (IDEQ 2014).

9

Table 1. Natural and Human Factors Affecting Groundwater Quality
Natural Factors
Natural Source Types of Contaminant
Precipitation Dissolved gases, dust, and emission particles
Infiltration through vegetation,
swamps, or soil and rocks (above
water table)
Biochemical products, organic materials, color, and
minerals
Aquifer rocks Minerals content (increases with time of contact)
Inter-aquifer mixing of cold water
and thermal water
Minerals and gases
Human Factors
Waste Source Types of Contaminant
Agricultural activities Fertilizers, pesticides, and herbicides
Mining operations (ore-
processing plants
Metallic trace elements and phosphates
Nuclear facilities Radioactive chemicals, heat, and dissolved solids
Urban activities (storm and
sanitary sewers, sewage-disposal
plants, cesspools and septic tanks,
and sanitary landfills)
Organic materials, dissolved solids, suspended solids,
detergents, bacteria, phosphate, nitrate, sodium, chloride,
sulfate, metallic trace elements, and others
Industrial facilities (food
processors)
Biochemical oxygen demand, suspended solids, sodium,
chloride
Geothermal activities Heat, dissolved solids, fluoride, and metallic trace
elements
Hazardous waste- and toxic-
waste disposal sites
Toxic metals, hazardous chemicals, and organic
compounds
Source: Yee and Souza 1987.
2.1.1 Groundwater Quality in Mountain Home AFB
Mountain Home AFB is ranked #14 out of the 34 NPAs in Idaho due to the substantial nitrate
contamination. Groundwater is the primary drinking water source for Mountain Home AFB. Not
only is the groundwater quantity important, but so is the quality.
Nitrate levels at Mountain Home AFB have been steadily increasing since initial
groundwater monitoring efforts began in the 1980s. In 1994, a base production well was taken
out of service due to elevated nitrates above the USEPA’s MCL, and the same scenario occurred
again in 1997 (ATSDR 2010).

10

Due to the concern about elevated levels of nitrates across Idaho, the human health risks
associated with nitrate contamination, and the rapidly decreasing water levels of the aquifer,
Mountain Home AFB has sponsored many studies and reports to address this issue. These
studies provide domain knowledge in geology, hydrology, and chemistry, which influences
nitrate contamination in the area (Norton et al. 1982; IDEQ 2008; Schwarz and Parliman 2010;
IDWR 2013).
In 2010, Schwarz and Parliman took and analyzed groundwater quality data to identify
various constituents in the groundwater. The results revealed high amounts of caffeine, which
suggested the groundwater was contaminated by leaks from sanitary sewer lines and septic
systems (Schwarz and Parliman 2010). Low levels of volatile organic compounds (VOCs), semi-
volatile organic compounds (SVOCs), and metals have also been detected in the groundwater.
During routine drinking water supply sampling, low concentrations of trichloroethylene (TCE)
were also detected. Although VOCs, SVOCs, and TCE were detected, none of them exceeded
USEPA’s MCLs (ATSDR 2010).
2.2 Nitrate Effects on Human Health
Drinking water standards have been implemented through the SDWA to protect public health.
The USEPA has developed the National Primary Drinking Water Regulations (NPDWRs) that
set maximum limits for contaminants or naturally occurring constituents in water (i.e., arsenic) to
fall below a set limit (Schwartz and Zhang 2003; USEPA 2012). Limits are identified as MCLs.
The MCL for nitrate is 10 mg/L or 10 ppm (USEPA 2014b). Ingestion of water in excess
of the MCL for nitrates, in some situations, leads to blue baby syndrome (Methemoglobinemia),
a condition that affects the body’s ability to transport oxygen from the lungs to the remainder of
the body (VanDerslice 2007).

11

The MCL for nitrates can be traced to a study of 139 cases in Minnesota in 1950 and a
survey conducted in 1951 regarding additional cases of Methemoglobinemia. While the study
found that Methemoglobinemia occurred when nitrate levels in infant’s water exceeded 10 mg/L,
only five of the 214 cases occurred when the level of nitrate was less than 20 mg/L. Nonetheless,
the USEPA set the MCL at 10 mg/L since available data was limited and the subpopulation at
risk involves infants, so an additional degree of safety was sought in setting the MCL
(VanDerslice 2007).
From September 2002 through September 2007, the USEPA investigated the dose-
response of nitrate in infants in a nitrate contaminated area in Washington State. The study found
that infants, 1-5 months old, who consumed water with nitrate levels above 5 mg/L had
significantly and substantially increased risks of having physiologically elevated levels of
Methemoglbin. While Methemoglobinemia is multi-factorial, it is clear that water containing
nitrates is a contributing factor and thus, protecting infants from high nitrates will help to protect
them from this potentially fatal disease (VanDerslice 2007).
2.3 Well Development
Groundwater can be contained in a variety of hydrogeological features: confined aquifers,
perched aquifers, aquifuges, aquitards, and aquicludes. The most common are confined and
perched aquifers.
Aquifers play a key role in supplying water to wells due to their transmission and storage
properties. When a pump in a well is turned on, the water level in the well casing is reduced,
causing the groundwater in the aquifer to flow towards and into the well (Figure 3). While most
of this flow comes from the storage characteristics of the aquifer, the transmissivity is also

12

important since it describes how well the water can move through an aquifer (Schwartz and
Zhang 2003).

Figure 3. Water infiltrating the subsurface flows through the groundwater system and
eventually discharges in streams, lakes, oceans, or is pumped from a well. The residence
time in the subsurface can vary from days to thousands of years (Winter et al. 1998).

While wells provide a means of extracting groundwater, they are also susceptible to
contamination at the opening on the surface, the piping from groundwater to surface, and the
groundwater source (Rural Water Supply Network 2010). Contamination can also occur during
the well drilling process since large quantities of drilling water and sometimes chemical
additives are added to the subsurface (Barcelona et al. 1985).
As mentioned above in Section 2.1, groundwater can be contaminated from a variety of
human and natural sources. The IDEQ has established minimum separation distances in order to
protect groundwater contamination and public drinking water systems as described in Table 2.

13

Table 2. Well Construction Minimum Separation Distances
Separation of Wells from:
Minimum
Separation
Distance (feet)
Existing public water supply well, separate ownership 50
Other existing well, separate ownership 25
Septic drain field 100
Septic tank 50
Drainfield of system with more than 2,500 GPD of sewage inflow 300
a
Sewer line – main line or sub-main, pressurized, from multiple sources 100
Sewer line – main line or sub-main, gravity, from multiple sources 50
Sewer line – secondary, pressure tested, from a single residence or
building
25
Effluent pipe 50
Property line 5
Permanent buildings, other than those to house the well or plumbing
apparatus, or both
10
Above ground chemical storage tanks 20
Permanent (more than six months) or intermittent (more than two months)
surface water
50
Canals, irrigation ditches or laterals, & other temporary (less than two
months) surface water
25
Source: Idaho Administrative Code, IDAPA 37.03.09, “Well Construction Standards Rules”
a
This distance may be less if data from a site investigation demonstrates compliance with IDAPA
58.01.03, “Individual/Subsurface Sewage Disposal Rules”, and separation distances
However, contaminates often times emanate beyond the source and create a plume. A
plume of dissolved contaminates can migrate with the flow and create a larger problem. Non-
point sources such as fertilizers and point sources including leaking sewer lines can have similar
contamination effects on an aquifer due to the mobility of the contaminates and formation of
plumes (Schwartz and Zhang 2003).
Groundwater quality data collected from Mountain Home AFB and previous reports
suggest nitrate plumes have formed due to non-point (e.g., golf course fertilization) and point
contamination (e.g., leaking sewer infrastructure). However, the spatial distribution of the nitrate
plumes or areas that are vulnerable to such contamination are unknown (Schwarz and Parliman
2010).

14

2.4 GIS-based Analysis Methods
The application of GIS to assess groundwater vulnerability to contamination has been
successfully practiced since the 1980s (e.g. Merchant 1994, Melloul and Collin 1998, Cameron
and Peloso 2001, Al-Adamat, Foster, and Baban 2003, Vias et al. 2005, Baalousha 2006; Jamrah
et al. 2007, Sener, Sener, and Davras 2009, Massone, Londono, and Martinez 2010). GIS has
been employed to identify and assess groundwater contamination at national, state, and local
scales for decades (e.g. Lake et al. 2003, Ceplecha et al. 2004). Many recent studies use
interpolation methods for groundwater analyses, the most common being Inverse Distance
Weighting (IDW) and kriging. Data collected from monitoring wells is used with one or more of
these interpolation methods to produce interpolated layers to analyze the spatial distribution of
groundwater quality. Tikle, Saboori, and Sankpal (2012), for example, used IDW and the data
contained some data clusters that introduced some errors, suggesting that IDW is sensitive to
outliers.
More commonly, studies compare interpolation methods to determine which produces the
most accurate results (e.g. Sun et al. 2009, Jie et al. 2013, Taghizadeh-Mehrjardi, Zereiyan-
Jahromi, and Asadzadeh 2013). Sun et al. (2009) compared interpolation methods for depth to
groundwater in northwest China. Data was collected from 48 observation wells and used to
compare IDW, the radial basis function and kriging. They found that simple kriging is the best
method since it had the lowest standard deviation between predicted and observed values (Sun et
al. 2009).
A more recent study found IDW produced better results compared to kriging or co-kriging
(Taghizadeh-Mehrjardi, Zereiyan-Jahromi, and Asadzadeh 2013). The researchers selected IDW
and variations of kriging since past research identified kriging to create the best model for

15

groundwater quality parameters, specifically heavy metals. However, the research concluded
IDW was the more suitable method of interpolation to estimate groundwater quality variables in
Urmia, Iran.
Jie et al. (2013) also compared IDW to kriging and used spatial interpolation to identify the
best method. These authors used IDW and kriging and analyzed groundwater depth, salinity, and
nitrate values from 90 monitoring wells throughout the Yinchuan, China area. The semi-variation
function in ArcGIS was used to determine the optimal interpolation method. By comparing
spatial correlation between neighboring observations for each variable through semi-variograms,
they were able to determine that both methods offer advantages. IDW is more suitable in areas
where neighboring locations play a larger part and the spatial correlation is weak, whereas
kriging is more suitable for cases of strong spatial correlation, when the whole trend is being
identified (Jie et al. 2013).
IDW and kriging are two of the most common methods; however, the studies revealed
that although the tools already offered in Esri’s ArcGIS platform offer great convenience, small
sampling sizes introduce errors in the results, which can lead to unreliable models (Sun et al.
2009, Tikle, Saboori, and Sankpal 2012, Jie et al. 2013, Taghizadeh-Mehrjardi, Zereiyan-
Jahromi, and Asadzadeh 2013).
In situations of small sampling data, or no groundwater monitoring data at all, researchers
have taken a different approach where a vulnerability model is first created using site specific
geological data and then field verified using either existing or specially acquired groundwater
data. Some of the first ideas of assessing groundwater vulnerability to contamination can be
traced back to France during the 1960s (e.g., Margat 1968). Since then, several methods for
developing vulnerability maps have surfaced, such as CMLS (Nofziger and Hornsby 1986,

16

1987), DRASTIC (Aller et al. 1987), GOD (Foster 1987), LEACHM (Wagenet and Hutson
1989), AVI (Van Stempvoort, Ewert, and Wassenaar 1993), and SINTACS (Ersoy and Gultekin
2013).
2.4.1 DRASTIC Method
The DRASTIC Method, developed for the USEPA, has become one of the most used methods to
distinguish degrees of vulnerability on a regional scale (Merchant 1994, Melloul and Collin
1998, Cameron and Peloso 2001, Al-Adamat, Foster, and Baban 2003, Vias et al. 2005,
Baalousha 2006; Jamrah et al. 2007, Sener, Sener, and Davras 2009, Massone, Londono, and
Martinez 2010). The DRASTIC Method is named for the seven factors considered in the
method: Depth to water, net Recharge, Aquifer media, Soil media, Topography, Impact of
vadose zone media, and hydraulic Conductivity of the aquifer (Aller et al. 1985, Koterba, Banks,
and Shedlock 1993, Rupert 1994, Barbash and Resek 1996, USGS 1999, Ersoy and Gultekin
2013).
The DRASTIC Method has been used to show areas of greatest potential for groundwater
contamination across the globe. The earliest applications had mixed success, mainly due to their
reliance on the uncalibrated DRASTIC Method (e.g. Rupert et al. 1991). However, throughout
the years, the method has been improved through calibrating the point rating scheme to measure
nitrite plus nitrate as nitrogen concentrations in groundwater and through its integration with
GIS. Statistical correlations suggest a linkage between nitrite plus nitrate as nitrogen and land
use, soils, and depth to water (Ott 1993, Rupert 1994).
Groundwater vulnerability assessments have been conducted for the region surrounding
Mountain Home AFB using the DRASTIC Model (IDEQ 1991, USGS 1999). The Idaho
Groundwater Vulnerability Project (IDEQ 1991) used a modified form of the DRASTIC Method

17

to produce a vulnerability map for the Idaho Snake River Plain (Figure 4). The map was
designed as a tool for prioritization of groundwater management activities in order to allocate
limited resources effectively. The Idaho groundwater vulnerability project also provided
justification for future studies (IDEQ 1991).
Furthermore, the vulnerability map was field verified by overlaying water quality data on
top of the vulnerability map. All of the wells that had anomalous levels of contaminates were
located in areas identified as high or very high risk areas, suggesting a good correlation between
the vulnerability map and field data (IDEQ 1991).
However, the project focused on the Idaho Snake River Plain at a scale of 1:250,000,
which provides generalized information on a regional scale. The vulnerability assessment does
not provide enough information for site-specific locations; therefore, more in-depth studies
would need to be performed for site-specific decisions (IDEQ 1991).

Figure 4. Vulnerability Map of the Idaho Snake River Plain (IDEQ 1991)

18

Similarly, the USGS (1999) produced a groundwater vulnerability map using the
DRASTIC Method for the eastern portion of the Snake River Plain, Idaho (Figure 5). The
DRASTIC point rating scheme was calibrated using groundwater quality data and the results
indicated a significant correlation between elevated nitrate levels and depth to water, land use,
and soil drainage (USGS 1999).
IDEQ (1991) and USGS (1999) have employed the DRASTIC Model to illustrate Idaho’s
groundwater risk; however, the projects were either at small scales (≤ 1:250,000) and/or did not
encompass the Mountain Home AFB area that is the focus of this thesis research project.

Figure 5. Probability of groundwater contamination by dissolved nitrite plus nitrate as
nitrogen for the Eastern Snake River Plain, Idaho (USGS 1999)

19

CHAPTER 3: METHODOLOGY AND DATA SOURCES
The overarching purpose of this study was to produce a GIS-based groundwater vulnerability
model for the Mountain Home AFB using the DRASTIC method. The model provides a basis
for evaluating groundwater vulnerability to pollution based on hydro-geologic parameters, which
can help guide the development of management practices to prevent additional nitrate
groundwater contamination in the region and improve management of water resources. The
model was verified using groundwater quality data to illustrate the efficacy of using the
DRASTIC method in assessing the vulnerability of the Mountain Home AFB to groundwater
contamination.
The remainder of this chapter consists of five sections. The first introduces the Mountain
Home AFB study area. The next two sections offer descriptions of the DRASTIC method and the
data that were used to implement this method. The final two sections describe how the various
factors were combined and how the model predictions were validated using groundwater quality
data.
3.1 Study Area
Mountain Home AFB is located in southwestern Idaho in Elmore County, approximately 50
miles southeast of Boise, Idaho and 8 miles southwest of Mountain Home, Idaho. Mountain
Home is close to both mountains and high desert landscapes, with vast areas of open space. The
6,844 acres of Mountain Home AFB consists of buildings, roads, and runways, which covers 20-
25% of the land (USAF 2012). The remainder of the land includes landscaping, open,
undeveloped fields, and partially disturbed areas (Figure 6).

20

Figure 6. Mountain Home AFB area map.

21

3.1.1 Climate
Mountain Home AFB is situated in the western portion of the Snake River Plain and receives
approximately 12 inches of rain per year. Most of the precipitation falls during late fall to early
spring. The semi-arid climate of Mountain Home AFB consists of hot, dry summers with average
daily temperatures of 90
o
F; however, temperatures may reach as high as 109
o
F during August.
During the winter months, the average temperature is 30-35
o
F (USAF 2012).
3.1.2 Geology and Soils
The Snake River Plain is thought to be an area of crustal rifting that started approximately 16
million years ago and grew southeasterly until about 3 million years ago (USAF 2012). Thick
deposits of rhyolites and basalts dominate most of the geology due to early volcanism.
Additionally, approximately eight million years ago, the area was covered by a lake called “Lake
Idaho”, which has since dried up, leaving thick sedimentary deposits of ash, clays, silts, sands,
and gravels (USAF 2012).
The soils on Mountain Home AFB are typical of semi-arid regions, consisting primarily
of silt and sandy loam. The soils have poor drainage and lack organic matter, with varying
thicknesses, depending on the location of bedrock and hardpans (USAF 2012).
3.2 DRASTIC Method
The DRASTIC method uses seven hydro-geological parameters to assess groundwater
vulnerability: (D) depth to groundwater table, (R) net recharge, (A) aquifer media, (S) soil
media, (T) topography, (I) impact of vadose zone, and (C) hydraulic conductivity (Table 3).
The input information was obtained from online databases and site-specific borehole,
land-use and topography data, and used to develop each DRASTIC parameter. Each of the seven
parameters was weighted and rated due to their relative influence on contamination, which

22

ranged from 1 to 5 and 1 to 10, respectively (Tables 3 and 4). Each parameter was multiplied by
a multiplier to obtain the weighted value. Then, the products were summed up to calculate the
final DRASTIC index (Equation 1), where r = the rated factor and w = the weighted factor. The
DRASTIC index (DI) represents the degree of vulnerability and can be used with GIS to produce
a vulnerability map that represents the hydrogeological setting (Shirazi et al. 2012):
DI = DrDw + RrR w + ArAw + SrSw + TrTw + IrIw + CrC w (1)
where D, R, A, S, T, I, and C are the seven parameters and the r and w subscripts correspond to
the rated and weighted factors, respectively.
Table 3. The seven DRASTIC model parameters and their relative weights
Factors Descriptions
Relative
Weights
Depth to Water
Represents the depth from the ground surface to the water table,
deeper water table levels imply lesser chance for contamination
to occur.
5
Net Recharge
Represents the amount of water which penetrates the ground
surface and reaches the water table, recharge water represents
the vehicle for transporting pollutants.
4
Aquifer Media
Refers to the saturated zone material properties, which controls
the pollutant attenuation processes.
3
Soil Media
Represents the uppermost weathered portion of the unsaturated
zone and controls the amount of recharge that can infiltrate
downward.
2
Topography
Refers to the slope of the land surface, it dictates whether the
runoff will remain on the surface to allow contaminant
percolation to the saturated zone.
1
Impact of
Vadose Zone
Is defined as the unsaturated zone material, it controls the
passage and attenuation of the contaminated material to the
saturated zone.
5
Hydraulic
Conductivity
Indicates the ability of the aquifer to transmit water, hence
determines the rate of flow of contaminant material within the
groundwater system.
3
Source: Babiker et al. (2005)

23

Table 4. DRASTIC parameters and rating values (adapted from Aller et al. 1987).
Rating
Depth to
Water
(ft)
Net
Recharge
(Irrigation)
Aquifer
Media
Soil Media
Topo-
graphy
(%)
Impact of the
Vadose Zone
Hydraulic
Conductivity
(m/day)
10 0 - 5.0 Urban
Karst
Limestone
Thin or
absent,
gravel
0 to 2
Karst
Limestone
1.00E+04
9 5.1 - 15.0 Basalt
Sand Stone
and
Volcanic
2 to 3 Basalt 1000
8
Sand &
Gravel
Peat 3 to 4
Sand &
Gravel
100
7
15.1 -
30.0
Improved
Massive
Sandstone
&
Limestone
Shrinking
and / or
aggregate
clay /
alluvium
4 to 5 Gravel, Sand 10 to 1
6
Bedded
Sandstone,
Limestone
Sandy
Loam,
schist,
sand, karst
volcanic
5 to 6
Limestone,
gravel, sand,
clay
0.1 - .01
5
30.1 -
50.0
Semi-
Improved
Glacial Loam 6 to 10 Sandy Silt .01 - .001
4
Weathered
Metamorp
hic /
Igneous
Silty Loam 10 to 12
Metamorphic
Gravel &
Sand
.001 - .0001
3
50.1 -
75.0

Massive
Shale
Clay Loam 12 to 16
Shale, Silt &
Clay
10E-4 to 10E-5
2
75.1 -
100.0

Muck,
acid,
granitoid
16 to 18 Silty Clay 10E-5 to 10E-6
1 > 100.1
Un-
improved

Non
Shrink and
Non -
aggregated
clay
> 18
Confining
Layer,
Granite
< 10E-6

3.3 Data for DRASTIC Parameters
Several types of data were used to construct two DRASTIC models. The first used generic,
publicly available data and the second used site-specific data.
Model 1 relied on generic data that could be gathered from online GIS databases. The
data for each parameter is described in more detail in the subsections below. Three

24

considerations guided the choice of these data as follows: (1) the data had to be publicly
available; (2) the data had to have been collected or updated within the past five years; and (3)
the source must have contained data within the Mountain Home AFB study area itself.
Model 2 used some of these same data sources plus site-specific data obtained primarily
from well driller’s logs. Mountain Home AFB has partnered with the USGS to collect data on
groundwater quality parameters since 1985. Groundwater quality parameters include nitrates and
static water levels for 16 different MWs located around the base. Additionally, since nitrate is a
NPDWR and regulated by the USEPA, nitrates are sampled at all of the BPWs. Table 5 provides
a list of MWs and BPWs with nitrate sampling data available for validating the model outputs.
The location of each well has been located with GPS coordinates and provides data on the spatial
distribution of nitrates across the base (Figure 7). Furthermore, each well has a well driller’s log
that indicates the depth of the well and water level. This data has been maintained in a Microsoft
Excel spreadsheet, and was used in this thesis research project to populate the attribute table for
the feature file. The WGS 84 / UTM Zone 11 N projection with meter as the unit of measure
was used for both models.
Table 5. Summary of available groundwater quality data from USGS and Mountain Home
AFB for 16 MWs and nine BPWs

Base Production Wells (BPWs)
BPW 1 BPW 2 BPW 4 BPW 6
BPW 8 BPW 9 BPW 11 BPW 12
BPW 13
Monitoring Wells (MWs)
MW 3-2 MW 6-2 MW 7-2 MW 11-2
MW 16-2 MW 17-2 MW 18-2 MW 21
MW 22 MW 23 MW 29 MW 30
MW 34 MW 35 MW 36 MW 40

25

Figure 7. Location of BPWs and MWs, which have been sampled for nitrates on Mountain
Home AFB, Idaho. Green area is a golf course within the study area.

26

3.3.1 Depth to Groundwater
The depth to groundwater for Model 1 was obtained from the USGS National Water Information
System (NWIS). NWIS contains data on active well networks, including statistics about
groundwater level data. These data were interpolated using the IDW method to create a smooth
surface representing the spatial continuity of the groundwater surface for the study area. The
IDW was performed on the point data using the default distance parameter (i.e. distance
squared), cell size of 20 m, and a limiting search radius to capture at least two other points. This
approach ensured that larger weights were assigned to points close to an output pixel and this
same procedure was used for every IDW interpolation in this study. Following the IDW, the map
was classified into ranges defined by the DRASTIC Model (1-10, with 1 representing minimal
impact to vulnerability and 10 representing maximum impact). The deeper the groundwater, the
smaller the rating value. Since the groundwater is deeper than 100 feet for the entire year, a
rating factor of one was assigned to the entire study area.
The well driller’s log data was uploaded into ArcGIS for Model 2. The depth to the
water table was obtained from the data and using IDW to interpolate a smooth surface. Similar
to Model 1, the groundwater level is greater than 100 feet, so a rating of one was applied to the
whole study area for the second set of model runs as well.
3.3.2 Net Recharge
Land use was used as a surrogate for net recharge since Mountain Home AFB does not have any
recharge wells and receives little precipitation, so irrigated areas provide the largest amount of
recharge in southern Idaho (Rupert et al. 1991).
Esri’s ArcGIS Online has a robust database for a variety of data, including the U.S. Land
Cover GAP database (USGS 2011). This data layer contained the land cover classification used

27

by the USGS Gap Analysis Program. These data include detailed vegetation and land use
patterns for the continental U.S. A total of 590 detailed land use classes in the data set are
grouped together into a total of eight general classes, including agricultural, urban/developed,
non-forested lands, water, etc. The USA Land Cover GAP data layer was used for Model 1 and
reclassified using the DRASTIC Model rating scheme based upon the eight general classes.
Mountain Home AFB has a Geobase Office that maintains all local GIS databases. The
land use data are divided into four general categories based upon improvements, specifically
irrigation and maintenance: (1) urban/developed (turf lawns, significant amounts of irrigation);
(2) semi-improved (established irrigation systems); (3) improved (drip-line irrigation); and (4)
unimproved (no irrigation, native landscape). These land-use data were stored as a polygon
feature class, so they were converted into raster and then used to assign the rated net recharge (R)
factor.
3.3.3 Aquifer Media
The U.S. aquifers data layer was also obtained from Esri’s ArcGIS Online database. This US
aquifer layer was produced as part of the Ground Water Atlas of the U.S. (USGS 2009) and
specifies the areal extent of the principal aquifers, including aquifer media for the Snake River
Plain Aquifer on which the Mountain Home AFB is located. The data was imported, clipped to
the installation boundary, and used for Model 1. Since the data was in polygon format, it was
convert to raster using the Polygon to Raster conversion tool Esri’s Spatial Analyst (Version
10.1). The data was then assigned a rating factor in accordance with the DRASTIC Model
parameter fields (Table 4).
The drill logs and sampling provided details of the well construction and the subsurface
materials that were bored through. Since each well is an individual entity, the data provided

28

information for that particular data point. The well driller’s logs were reviewed and the
information on aquifer media was uploaded into a well layer. Rating factors were assigned to
each well point, depending on the aquifer media and interpolated using IDW to create a smooth
surface representative of the aquifer.
3.3.4 Soil Media
Models 1 and 2 used the same data for the soil DRASTIC parameter—the SSURGO database
from the United States Department of Agriculture, Natural Resources Conservation Service
(USDA-NRCS 2015). The SSURGO database contains information about soil as collected by the
National Cooperative Soil Survey over the past century. This information was imported in digital
format into the ArcGIS system and clipped to the installation boundary. The soil media types
were then assigned ratings from 1 to 10 according to their permeability (Table 4).
3.3.5 Topography
The topography layer for both Model 1 and Model 2 was constructed using contour elevation
data at 5 foot intervals. The data was imported from the Mountain Home AFB Geobase. In order
to determine the percent slope, the topography data was converted to a raster (20x20 cell size)
using the Topo to Raster Spatial Analyst tool. Once converted, the Slope tool in Spatial Analyst
was used to calculate the percent slope. Each percent range was reclassified using the DRASTIC
Model rating factors (Table 4).
3.3.6 Impact of Vadose Zone
Similar to the aquifer media, the impact of the vadose zone was obtained from Esri’s ArcGIS
Online U.S. aquifer layer. The same data was imported and clipped to the installation boundary
for use in Model 1. Since the data was in polygon format, it was converted to raster using the

29

Polygon to Raster conversion tool in Spatial Analyst. The data was then assigned a rating factor
in accordance with the DRASTIC Model parameter fields (Table 4).
The procedures used for the aquifer media data obtained from the well driller’s logs were
used to estimate the impact of the vadose zone data for Model 2. Rating factors were assigned to
each well point, depending on the aquifer media and interpolated using IDW to create a surface
representative of the vadose zone values.
3.3.7 Hydraulic Conductivity
Unsaturated zone hydraulic properties can be used to estimate the movement of chemicals into
the aquifer and are typically obtained from pump tests, slug tests, and constant-head tests.
However, these values are sometimes not readily available due to the costs associated with
performing the necessary tests to obtain those values. To balance the need for data and lacking
information, quasi-empirical models, hydraulic data from similar soils, or typical values for most
aquifer material from textbooks or publications were used to predict the unsaturated hydraulic
conductivity (Adams and Jovanovic 2005).
Data from the aquifer media were used with Figure 8 to obtain hydraulic conductivity
values (Heath 1983). These hydraulic conductivity estimates were then added to the aquifer
media attribute table and rated in accordance with the DRASTIC Model (Table 4). The polygon
layer was converted to a raster layer using the hydraulic conductivity rating values. The same
process was performed for both Models 1 and 2, using the associated aquifer media data for
each.
3.4 Aquifer Vulnerability Assessment
Each of the DRASTIC factors was reclassified to incorporate the rated factor in order to
calculate the DRASTIC Index (Equation 1). Once each of the layers were prepared, they were

30

combined using the weighted sum overlay function in ArcGIS to combine the layers using the
weighted factor. The output of the function was a single cell layer signifying the vulnerability
index for the study area.

Figure 8. Hydraulic conductivity values for selected aquifer media types.
Source: Heath 1983
The DRASTIC indices were classified into categories based on the mean value of the
dataset then renamed based on vulnerability risk: low and high. The indices were assigned a

31

color corresponding to the level of risk and these colors were then used to produce a
vulnerability map: green was used to indicate “low” vulnerability and red to indicate “high”
vulnerability.
3.5 Model Validation
Two efforts were made to compare the model predictions and to evaluate the efficacy of the two
sets of model predictions as follows.
For the first test, the predictions generated with both models were compared to one
another to see whether or not the two models tended to produce similar rankings in terms of the
vulnerability risk at the 25 locations with monitoring wells (Figure 7).
For the second test, the two models were individually validated using groundwater
quality data obtained from the 25 monitoring wells scattered across the study site (Figure 7).
Both models were overlaid with the well point layer, displaying the average nitrate
sampling results for each well. The models were analyzed to determine the prediction accuracy
rate.
In order to do so, the vulnerability index values for each model were converted from
raster to point values using the Raster to Point conversion tool and the point values were
compared with the well point feature class containing the nitrate data and collected into a series
of tables: descriptive statistics and model prediction analysis. The data was used to calculate
descriptive statistics, including the minimum, maximum, mean, standard deviation, and
coefficient of variance. The mean for each model was used to identify the classification break
between low and high vulnerability. Values below the mean were classified as low, while
anything equal to or greater than the mean was high risk. Nitrate sampling data was classified
into low or high risk by the USEPA action level of 5.0 mg/L. The prediction accuracy rate was

32

determined by whether the model successfully identified a low vulnerability area for a low
nitrate observation and a high vulnerability area for a high nitrate observation.
Finally, the results from the two models were compared to each other to determine
whether there was a significant difference between the models built using the generic and site-
specific data using cross-tabulated tables.

33

CHAPTER 4: RESULTS AND DISCUSSION
This chapter describes the seven DRASTIC parameter maps and the two vulnerability index
maps that were developed to represent groundwater contamination risk for Mountain Home
AFB, Idaho. The first goal of this project was to create a DRASTIC model using available GIS
data from sources such as USGS, ArcGIS Online, etc. The second goal was to create an
additional DRASTIC model using site-specific data obtained from well drilling logs. The third
goal was to compare the model predictions to the groundwater well observations to determine
whether the use of the site-specific data improved the specificity of the model predictions. The
following sections detail the results and how this thesis project accomplished the aforementioned
goals.
4.1 DRASTIC Parameters
4.1.1 Available GIS Data
The average depth to the water table (D) at Mountain Home AFB is 360 feet, which the rated
metric computes to a value of one for the entire study area (Figure 9).
The arid-dry climate and minimal rainfall at Mountain Home AFB means that the net
recharge to the groundwater aquifer is more dependent on land use and irrigation applications.
The general land cover map reproduced in Figure 10 shows the general land cover types which
control the net recharge to the groundwater aquifer on the Mountain Home AFB. The lowest
recharge rate (rated value of one) was associated with the non-forested unimproved areas with no
maintenance or irrigation systems since the net recharge relies entirely on precipitation (annual >
12in/year). The urban/developed land and other improved areas such as the golf course had
relatively higher recharge rates due to the substantial irrigation applications (rated value of 10).

34

Figure 9. Soil type (symbology) and depth to water in feet (labels) for Mountain Home
AFB. The depth to groundwater is greater than 80 feet across the entire study area.

35

Figure 10. General land cover type, consisting primarily of Urban/Developed Land (red),
Agricultural (including the golf course), and Non-Forested Lands.

36

The aquifer media underlying Mountain Home AFBis primarily metamorphic/igneous
with a small section of sand/gravel in the southwest corner of the study area. These media were
rated four and eight, respectively (Appendix A, Figure A1).
Soils on Mountain Home AFB are mostly silty loam with the exception of a small area of
fine sandy loam (Figure 10). The majority of the study area consists of Bahem silt loam, with a
rating of four. Turning next to slope (S), the overall slope is less than 2% across the entire study
area, so water is more likely to percolate rather than flow to another location (rating score 10)
(Appendix A, Figure A2).
The impact to the unsaturated material in the vadose zone mirrored that of the aquifer
media, consisting primarily of grey basalt (rating score 9) and an area of sand and gravel (rating
score 8) in the southwest corner (Appendix A, Figure A3).
The hydraulic conductivity of fractured basalt is relatively minimal, with a value of 0.1
m/day (rating score 6); however, sand and gravel transmits flow more quickly, with a value of
100 m/day (rating score 8) (Appendix A, Figure A4).
4.1.2 Site-Specific GIS Data
Starting in the same place and moving through the various inputs in the same order, the reader
can see that while more specific measurements of depth to groundwater were obtained from well
drilling logs, the overall rating (rating score 1) did not change from the first model since the
depth ranges from 350 feet to 432 feet (Appendix B, Figure B1).
Net recharge ranged from one to 10, with the highest rating associated with the urban
land use because of irrigation systems (rating score 10) (Figure 11). The remainder of the study
area had lower net recharge rates due to the barren land (rating score 1), rangeland (rating score
3), and agricultural land (rating score 7) as a consequence of the types and amounts of irrigation,

37

Figure 11. Site-specific land cover type to obtain Rated Net Recharge, ranging from
Urban/Developed Land (darker red) to barren, rangeland (white). Wells are depicted in
blue.

38

if any, involved with these land uses. The area with the greatest amount of irrigation use is the
golf course, which uses an average of 1 to 2 million gallons of water per day during the growing
season (rating score 10).
The Mountain Home aquifer geology is primarily basalt (rating score 9) with occasional
intercalated, thin, discontinuous deposits of sand/gravels (rating score 8), mudstone (rating score
7), and shale and clay (rating score 3). The deposits of aquifer media other than basalt occurred
predominately on the western portion of the study area and in the northeast corner (Figure B2).
Both DRASTIC models shared the same soils and topography data, and therefore utilized
the same rating results for soils (rating score 4) and slope (rating score 10) for the reasons noted
earlier.
The impact of the vadose zone varied more than it did for the first model due to the data
obtained from the well driller’s logs. Similar to the aquifer media, basalt (rating score 9) is the
primary media; however, deposits of silt/sand (rating score 5), sand/gravel (rating score 7), and
clay (rating score 3) were located throughout the study area (Figure 12). The Mountain Home
aquifer is characterized by mixed hydraulic conductivity, ranging from 100 m/day (rating score
8) to 0.0001 m/day (rating score 4) (Appendix B, Figure B3).
4.2 DRASTIC Results
4.2.1 Model 1
The DRASTIC vulnerability index generated with generic, publicly available data ranged from
102 to 138. The values were classified into two categories: low (<118) and high (≥118) risk
based on the mean model prediction values (Figure 13).

39

Figure 12. Impact of the Rated Vadose Zone, ranging from 9 to 3. Data obtained from
well driller’s logs. Wells are depicted in blue.

40

More than 50% of the DRASTIC aquifer vulnerability map falls into the middle of the
vulnerability spectrum, with patches of ‘high’ and ‘low’ vulnerability scattered throughout the
Mountain Home AFB (Figure 13). Areas of ‘high’ vulnerability largely correspond to residential
lots, the golf course, and the southwest corner of the study area. This result is due to the
combination of net recharge (irrigation application rates), aquifer media (sand/gravel aquifer
media in the southwest corner) and soil type. Several of the parameters were the same across the
entire study area (e.g., depth to water and slope), which resulted in the other parameters having
greater influence on the patterns evident in the vulnerability map.
4.2.2 Model 2
The site-specific DRASTIC vulnerability index had a greater range (93 to 159) compared to the
first model. The ranges for low and high vulnerability classes were <128 and ≥128, respectively.
The DRASTIC aquifer vulnerability map using site-specific data shows broad areas of
‘low’ to ‘high’ risk (Figure 14). Similar to the previous model, depth to water and slope were
assigned the same rating across the entire study area, so other parameters had greater influence
on the model results. The ‘high’ vulnerability classes in the north and northeastern portion of the
study area are a combination of high irrigation, basalt aquifer media, and fractured basalt vadose
zone, which all have high ratings. The ‘moderate’ vulnerability patches were distinguished from
the ‘high’ risk areas due to the lower rated aquifer media and vadose zones, and areas of the
study area with little to no irrigation and similar hydrogeological properties fell into the ‘low’
vulnerability classification with this particular model.

41

Figure 13. Model 1 – DRASTIC Index of vulnerability for Mountain Home AFB, using
generic, publicly available data and overlaid with average nitrate sampling results per well.

42

Figure 14. Model 2 – DRASTIC Index of vulnerability for Mountain Home AFB, using
site-specific data and overlaid with average nitrate sampling results per well.

43

4.3 Validation
The descriptive statistics for both of the DRASTIC models and nitrate well observations are
depicted in Table 6. In summary, Model 2 had a greater standard deviation (6.9) and coefficient
of variation (0.101) compared to Model 1 (3.65 and 0.042, respectively), indicating a greater
variance amongst the predicted values. Additionally, the range of Model 1 risk scores was only
36, compared to 66 for Model 2.
When compared to the nitrate USEPA contamination limits: below action level (<5.0
mg/L) and above action level (≥5.0 mg/L), above MCL (>10.0 mg/L), the Model 2 vulnerability
classes correctly predicted 56% of the USEPA classes, compared to just 48% for Model 1 (Table
7).
Table 6. Descriptive statistics of Model 1, Model 2, and nitrate results from 25 wells.
Statistics Summary
Model 1 Model 2 Nitrates
Min 102 93 0.4
Max 138 159 27.67
Mean 117 127.14 8.66
Std Dev 3.65 12.28 6.9
CoV 0.042 0.101 0.797

The efforts to validate the models was hampered by the relatively small number of
observation wells for Mountain Home AFB (n=25), irregular distribution, and the relatively
small numbers of unique model scores generated at the locations of these wells (i.e. five unique
risk scores for Model 1 and 13 unique scores for model 2) (Table 7).

44

Table 7. Model prediction results using MW and BPW data to validate the two models.
Mean Nitrate color corresponds to USEPA action level (>5.0) standards, while colors
correspond to model vulnerability risk classes. False = 0, True =1 for correct values.
Well #
Nitrate
Observation
Model 1
Prediction
Accuracy
Model 2
Prediction
Accuracy
MW 16-2 0.4 110 1 139 0
MW 3-2 0.97 110 1 123 1
MW 6-2 1.24 118 0 130 0
BPW 13 1.85 118 0 114 1
MW 17-2 3.16 131 0 123 1
MW 7-2 3.4 118 0 123 1
MW 34 4.57 118 0 131 0
BPW 2 4.87 118 0 122.9 1
BPW 9 5.06 118 0 131 1
BPW 11 5.1 114 0 117.8 0
MW 30 5.47 110 0 139 1
BPW 12 6.1 110 0 139 1
MW 22 6.84 122 1 117 0
MW 23 6.85 122 1 127 0
MW 21 6.99 118 1 114 0
BPW 6 8.62 118 1 158.8 1
BPW 4 8.73 110 0 139 1
MW 18-2 8.8 118 1 131 1
BPW 1 12.97 118 1 147 1
MW 36 13.55 118 1 147 1
MW 40 14.83 118 1 123 0
BPW 8 16.42 110 0 159 1
MW 11-2 18.19 110 0 117.8 0
MW 29 23.77 118 1 123 0
MW 35 27.67 118 1 109 0

Total 12 Total 14

Prediction Rate 48% Prediction Rate 56%

The scores generated with the two models in the cells corresponding to the locations of
the 25 observation wells are cross-tabulated in Table 8. These results simply reaffirm what the

45

discerning reader will have already picked up on in Table 7; namely, that the vulnerability
rankings produced with the DRASTIC method and the two sets of input data were widely
different. For example, the shaded area in the top left-hand corner of Table 8 shows that just two
of the five lowest scores generated with Model 1 coincided with the lowest scores from Model 2
(ignoring the difference in the magnitude of the scores for the time being). The results were even
worse at the top and more important end in which high risk areas are supposedly found – the
shaded cells in the bottom right quadrant shows that none of the three highest risk scores
generated with the two models overlapped with one another. Perhaps the most galling of all, the
location for which the highest risk was predicted with Model 2 was one of seven monitoring well
locations for which Model 1 predicted the least risk.
Table 8. Model predictions in the grid cells corresponding to the locations of the 25
observation wells
Model 2
scores
Model 1 scores
Totals
110 114 118 122 131
109 1 1
114 2 2
117 1 1
117.8 1 1 2
122.5 1 1
123 1 3 1 5
127 1 1
130 1 1
131 3 3
139 4 4
147 2 2
158.8 1 1
159 1 1
Totals 7 1 14 2 1 25

46

These results, taken as a whole, show that the predictions generated with the two models
are not very similar and therefore not interchangeable. The next test was to compare the model
predictions with the nitrate observations at the 25 monitoring wells to see whether or not there
were grounds to say that the predictions generated with one of the two models produced more
robust results than the model results generated with the other input data.
Table 9 takes a similar approach to the previous summary table and reports unique Model
1 prediction values by nitrate level. The shaded cells in the top-left quadrant of this table show
that Model 1 correctly predicted ‘low’ risk at just two of the eight monitoring wells with nitrate
levels ≤ 5.0 mg/L and the shaded cells in the bottom-right quadrant show that this particular
model predicted ‘high’ risk in none of the seven monitoring wells recording nitrate levels > 10
mg/L. In addition, the large scatter of counts in the first and third rows and middle column of
Table 9 shows that Model 1 suggest this model does a poor job of identifying locations with
varying levels of risk and the single value in the last row indicates that this model predicted the
greatest risk in a grid cell that coincided with a monitoring well with nitrate readings of ≤ 5 mg/L
and the cell in the first row and third column indicates that this model predicted ‘low’ risk in two
of the seven locations where the nitrate monitoring results pointed to contamination levels > 10
mg/L. Clearly, this particular model, which used generic, publicly available datasets did not do a
very good job of predicting groundwater contamination potential.
Table 9. Model 1 predictions compared with nitrate levels sampled at 25 monitoring wells
Model 1 scores
Nitrate levels (mg/L)
≤ 5 5-10 > 10 Totals
110 2 3 2 7
114 1 1
118 5 4 5 14
122 2 2
131 1 1
Totals 8 10 7 25

47

Table 10 provides the same comparison using the Model 2 predictions and these results
show that this model did a little better, particularly at the top end where the highest
contamination levels have been reported. These model predictions predicted ‘low’ risk for two of
the eight monitoring wells reporting nitrate levels ≤ 5 mg/L and ‘high’ risk for three of the seven
monitoring wells reporting nitrate levels > 10 mg/L. However, there are still problems with these
model predictions since Model 2 shows a similar large scatter to Model 1 and the lowest relative
risk was predicted for one of the seven monitoring wells reporting nitrate levels > 10 mg/L.
These results, taken as a whole, confirm that Model 2 using site-specific data did
marginally better than Model 1 which relied exclusively on generic, publicly available data as
inputs. However, neither set of model predictions seem good enough to inspire confidence and it
is clear that the results produced with the two model runs are not interchangeable. Some
suggestions for how to overcome some or all of the aforementioned problems are provided in the
final chapter.
Table 10. Model 2 predictions compared with nitrate levels sampled at 25 monitoring wells
Model 2
scores
Nitrate levels (mg/L)
≤ 5 5-10 > 10 Totals
109 1 1
114 1 1 2
117 1 1
117.8 1 1 2
122.9 1 1
123 3 2 5
127 1 1
130 1 1
131 1 2 3
139 1 3 4
147 2 2
158.8 1 1
159 1 1
Totals 8 10 7 25

48

CHAPTER 5: CONCLUSIONS
Mountain Home AFB relies on groundwater to provide drinking water for the residents. Over the
years, the groundwater has degraded, both in quantity and quality, which points to the need to
find a long-term sustainable water supply solution. Information on areas of high contamination
risk would assist in articulating and choosing among those solutions.
This thesis attempted to assess the vulnerability of groundwater to contamination using
the DRASTIC method. The method used seven hydrogeological parameters to create a single
map that classifies areas by the potential risk: low to high. A variety of data is required in order
to create the seven layers that contribute to the final model values. Mountain Home AFB
operates much like a small municipality, which includes a Geodatabase office that stores
substantial amounts of geospatial data. However, every study area may not have site-specific
data available. Since GIS has been used to assess groundwater quality since the 1980s, and on
several scales, generic data can be obtained from online sources (Lake et al. 2003, Ceplecha et al.
2004). In addition to determining groundwater vulnerability, this thesis also aimed to compare
the results of two models, the first created using generic, publicly available data and the second
created with site-specific data, to determine whether or not there was a significant difference
between the two models.
The first model (generic data) suggested that most of the study area was at moderate risk
to groundwater contamination, with some smaller areas of low and high risk. The low risk
corresponded primarily to areas lacking irrigation and silt loam soils. High risk areas were areas
characterized by intensive irrigation, sand/gravel aquifer media, and high impact of the vadose
zone. When compared to nitrate observation data, Model 1 had a 48% prediction accuracy rate

49

when both model predictions and nitrate observations were divided into ‘low’ and ‘high’ risk
classes (Table 7).
The second model (site-specific data) characterized the northeast quadrant of the
Mountain Home AFB to be high risk, the perimeter to be low risk, and predicted several
moderate risk areas located around the airfield and further afar. Similar to Model 1, irrigation
(Net Recharge), the impact of the vadose zone, and aquifer media variables had the greatest
influence on the model predictions. When also validated against observation data, Model 2 had a
slightly greater prediction rate of 56% when both model and nitrate observations were divided
into ‘low’ and ‘high’ risk classes (Table 7).
The results summarized in Tables 8-10 showed that the two models produced different
predictions for the grid cells that were coincident with the 25 nitrate monitoring wells and
confirmed that the second model did slightly better than the first model.
This assessment suggests that the well driller’s logs that were used to develop several of
the DRASTIC parameters, such as depth to water, aquifer media, impact of the vadose zone, and
hydraulic conductivity for Model 2 were helpful. However, Model 2 may also contain additional
errors since IDW was used to develop raster layers for those parameters. IDW calculates cell
values dependent of each other using ≥ 2 nearby observations with the importance of these
observations set by the inverse of the distance between the cell and the observation squared.
However, the values for the different contributing variables on the Mountain Home AFB some
areas may not be directly dependent on each other. For example, the golf course is not native to
the area and had to be formally established by new soil, vegetation, excessive irrigation and
fertilization, which creates a micro-ecosystem significantly different than the remainder of the
study area.

50

Both models were developed using the GIS techniques first described by Aller et al.
(1987), but while the methodology has been standardized, data has significant effects on the
production of the final DRASTIC index as illustrated in this study. The DRASTIC method is
commonly used to analyze groundwater contamination risk; however, the scale and detail of data
used in the model and the accuracy has yet to be analyzed. While this study aimed to expand on
that notion, the results and validation encompass a small sample size that could have
compromised the study. Additional studies would need to be performed using the same
approach, but with larger sample sizes so that the sample size reported here (n=25) would not
negatively affect the results (as may have happened here). One popular option, that offers have
tried, is to calculate the DRASTIC index for a county, rather than a single facility like the
Mountain Home AFB.
Furthermore, the DRASTIC Method evaluates seven hydrogeological parameters;
however, previous studies conducted on Mountain Home AFB indicate leaking sewer lines as a
potential source of nitrates. Human impacts are not a parameter in the DRASTIC method and
therefore, they are not captured in the model. The historical leaking sewers could significantly
increase some of the nitrate contamination and values that were used to validate the models, so
the validation process could be biased. Future studies will also need to pursue other validation
methods outside of human induced contamination or look into including a ‘human impact’
parameter to the DRASTIC method.

51

REFERENCES
Adams, Shafick, and Nebo Jovanovic. 2005. Deliverable: Testing and documenting suitable
vulnerability assessment methods in key study area. Deliverables, Department of Earth
Sciences, Univerity of Western Cape.
Al-Adamat, Rita, Ian L. Foster, and Serwan M. J. Baban. 2003. "Groundwater Vulnerability and
Risk Mapping for the Basaltic Aquifer of the Azraq Basin of Jordan Using GIS, Remote
Sensing and DRASTIC." Applied Geography 303-324.
Aller, Linda, Todd Bennett, Jay H. Lehr, and Richard J. Petty. 1987. DRASTIC--a Standardized
System for Evaluating Ground Water Pollution Potential Using Hydrogeologic Settings.
EPA/600/2-85/018, US Environmental Protection Agency, Robert S. Kerr Environmental
Research Laboratory, Office of Research and Development.
ATSDR (Agency for Toxic Substances & Disease Registry). 2010. Public Health Assessment,
Mountain Home Air Force Base, Mountain Home, Elmore County, Idaho. Public Health
Assessment, Atlanta, GA: Federal Facilities Assessment Branch, Division of Health
Assessment and Consultation, ATSDR.
Baalousha, Husam. 2006. "Vulnerability Assessment for the Gaza Strip, Palestine Using
DRASTIC." Environmental Geology 405-414.
Babiker, Insaf S, Mohamad A.A Mohamed, Tetsuya Hiyama, and Kikuo Kato. 2005. "A GIS-
based DRASTIC model for assessing aquifer vulnerability in Kakamingahara Heights,
Gifu Prefecture, central Japan." Science of the Total Environment 345 127-140.
Barbash, Jack E., and Elizabeth A. Resek. 1996. Pesticides in Ground Water--Distribution,
Trends, and Governing Factors. Chelsea, MI: Ann Arbor Press, Inc.

52

Barcelona, Michael J., James P. Gibb, John A. Helfrich, and Edward E. Garske. 1985. Practical
Guide for Ground-Water Sampling. ISWS Contract Report 374, Champaign, IL: Illinois
State Water Survey.
Cameron, Enrico, and Gian F. Peloso. 2001. "An Application of Fuzzy Logic to the Assessment
of Aquifers' Pollution Potential." Environmental Geology 1305-1315.
Ceplecha, Zachary L., Reagan M. Waskom, Troy A. Bauder, James L. Sharkoff, and Tajiv
Khosla. 2004. "Vulnerability Assessment of Colorado Ground Water to Nitrate
Contamination." Water, Air, and Soil Pollution 373-394.
Ersoy, Arzu F., and Fatma Gultekin. 2013. "DRASTIC-based Methodology for Assessing
Groundwater Vulnerability in the Gumushacikoy and Merzifon Basin (Amasya,
Turkey)." Earth Sciences Research Journal 33-40.
Foster, Stephen. 1987. "Fundamental Concepts in Aquifer Vulnerability, Pollution Risk and
Protection Strategy." The Netherlands Vulnerability of Soil and Groundwater to
Pollutants The Hague. Noordwijk Aan Zee: Netherlands Organization for Applied
Scientific Research. 69-86.
Heath, Ralph C. 1983. Basic Ground-Water Hydrology. Water Supply Paper 2220, U.S.
Geological Survey.
IDEQ (Idaho Department of Environmental Quality). 2014. Elmore County Ground Water
Quality Improvement and Drinking Water Source Protection Plan. Plan, Boise, ID:
IDEQ.
IDEQ (Idaho Department of Environmental Quality). 1991. Ground Water Vulnerability
Assessment Snake River Plain, Southern Idaho. Report, Boise, ID: Idaho Department of
Health and Welfare, Division of Environmental Quality.

53

—. 2008. Idaho Nitrate Priority Areas. Accessed October 11, 2014.
http://www.deq.state.id.us/media/471611-ranking 2008.pdf.
—. 2015. Nitrate in Ground Water. Accessed February 16, 2015.
https://www.deq.idaho.gov/water-quality/ground-water/nitrate.aspx.
IDWR (Idaho Department of Water Resources). 2013. Critical Ground Water Areas, Ground
Water Management Areas & Areas of Drilling Concern. Accessed October 11, 2014.
http://www.idwr.idaho.gov/WaterInformation/GroundWaterManagement/.
Jamrah, Ahmad, Ahmed Al-Futaisi, Natarajan Rajmohan, and Saif Al-Yaroubi. 2007.
"Assessment of Groundwater Vulnerability in the Coastal Region of Oman Using
DRASTIC Index Method in GIS Environment." Environmental Monitoring and
Assessment 125-138.
Jie, Chen, Zhang Hanting, Qian Hui, Wu Jianhua, and Zhang Xuedi. 2013. "Selecting Proper
Method for Groundwater Interpolation Based on Spatial Correlation." Fourth
International Conference on Digital Manufacturing and Automation 1192-1195.
Koterba, Michael T., W.S. Banks, and Robert J. Shedlock. 1993. "Pesticides in Shallow
Groundwater in the Delmarva Peninsula." Journal of Environmental Quality 500-518.
Lake, Iain R., Andrew A. Lovett, Kevin M. Hiscock, Mark Betson, Aidan Foley, Gisela
Sunnenberg, Sarah Evers, and Steve Flectcher. 2003. "Evaluating Factors Influencing
Groundwater Vulnerability to Nitrate Pollution: Developing the Potential of GIS."
Journal of Environmental Management 68(3) 315-328.
MacDonald, Alan., J.R. Davies, and Brighid Dochartaigh. 2002. Simple Methods for Assessing
Groundwater Resources in Low Permeability Areas of Africa. Commissioned Report

54

CR/01/168N, Keyworth, UK: British Geological Survey, Department for International
Development.
Margat, Jean. 1968. Groundwater Vulnerability to Contamination. Doc, Orleans, France: 68
SGC 198 HYD, BRGM.
Massone, Hector, Mauricio Q. Londono, and Daniel Martinez. 2010. "Enhanced Groundwater
Vulnerability Assessment in Geological Homogeneous Areas: A Case Study from the
Argentine Pampas." Hydrogeology Journal (Hydrogeology Journal) 371-379.
Melloul, Abraham J., and Martin Collin. 1998. "A Proposed Index for Aquifer Waterquality
Asssessment: The Case of Israel's Sharon Region." Journal of Environmental
Management 131-142.
Merchant, James W. 1994. "GIS-based Groundwater Pollution Hazard Assessment: A Critical
Review of the DRASTIC Model." Photogrammetric Engineering and Remote Sensing
1117-1127.
Morris, Brian L., Adrian R. Lawrence, John P. Chilton, Brian Adams, Roger Calow, and Ben A.
Klinck. 2003. Groundwater and its Susceptibility to Degradation: A Global Assessment
of the Problems and Options for Management. Early Warning and Assessment Report
Series, RS, 03-3, Nairobi, Kenya: United Nations Environment Programme.
Nofziger, David L., and Arthur G. Hornsby. 1986. "A microcomputer-based management tool
for chemical movement in soil." Applied Agricultural Research 1: 50-56.
Nofziger, David L., and Arthur G. Hornsby. 1987. CMLS: Chemical Movement through Layered
Soils Model Users Maunual. Gainesville, FL: University of Florida.

55

Norton, Marc A., William Ondrechen, and James L. Baggs. 1982. Ground Water Investigation of
the Mountain Home Plateau, Idaho. Publication, Boise: Idaho Department of Water
Resources.
Ott, R. Lyman. 1993. An Introduction to Statistical Methods and Data Analysis. Belmont, CA:
Duxbury Press, Wadsworth Publishing Company.
Perlman, Howard. 2014. Groundwater Flow and Effects of Pumping. March 17.
http://water.usgs.gov/edu/earthgwdecline.html.
Rahman, Atiqur. 2008. "A GIS Based DRASTIC Model for Assessing Groundwater
Vulnerability in Shallow Aquifer in Aligarh, India." Applied Geography 32-53.
Rupert, Michael G. 1994. Analysis of Data on Nutrients and Organic Compounds in Ground
Water in the Upper Snake River Basin, Idaho and Western Wyoming, 1980-91. Water-
Resources Investigations Report 94-4135, Boise, ID: US Geological Survey.
Rupert, Michael G., Tana Dace, M.A. Maupin, and Bruce Wicherski. 1991. Ground-water
Vulnerability Assessment, Snake River Plain, Southern Idaho. Report, Boise, ID: Idaho
Department of Health and Welfare, Division of Environmental Quality.
Rural Water Supply Network. 2010. Siting of Drilled Water Wells: A Guide for Project
Managers. Accessed February 16, 2015.
http://www.sswm.info/sites/default/files/reference_attachments.
Schwartz, Franklin W., and Hobao Zhang. 2003. Fundamentals of Groundwater. New York,
NY: John Wiley & Sons, Inc.
Schwarz, David L., and D.J. Parliman. 2010. Use of Human Pharmaceuticals to Determine a
Sewer Water Nitrate Source to the Regional Aquifer, Mountain Home Air Force Base,
Idaho. Report, Boise: URS Group, Inc.

56

Sener, Erhan, Sehnaz Sener, and Aysen Davras. 2009. "Assessment of Aquifer Vulnerability
Based on GIS and DRASTIC Methods: A Case Study of the Senirkent-Uluborlu Basin
(Isparta, Turkey)." Hydrogeology Journal 2023-2035.
Shirazi, Sharif Moniruzzaman, H M Imran, and Akib Shatirah. 2012. "GIS-based DRASTIC
Method for Groundwater Vulnerability Assessment: A Review." Journal of Risk
Research 991-1011.
Sun, Yue, Shaozhong Kang, Fusheng Li, and Lu Zhang. 2009. "Comparison of Interpolation
Methods for Depth to Groundwater and its Temporal and Spatial Variations in the
Minqin Oasis of Northwest China." Environmental Modeling and Software 24 1163-
1170.
Survey), USGS (US Geological. 2011. Gap Analysis Program (GAP). Version 2, National Land
Cover.
Survey), USGS (US Geological. 2009. Ground Water Atlas of the United States. US Aquifers
Data, USGS Office of Ground Water.
Taghizadeh-Mehrjardi, Ruhollah, Mehdi Zareiyan-Jahromi, and F. Asadzadeh. 2013. "Mapping
the Spatial Variability of Groundwater Quality in Urmia (Iran): Comparison of Different
Interpolation Methods." Journal of International Environmental Application and Science
8(3) 359-368.
Tikle, Suvarna, Mohammad-Jawid Saboori, and Rushikesh Sankpal. 2012. "Spatial Distribution
of Ground Water Quality in Some Selected Parts of Pune City, Maharashtra, India Using
GIS." Current World Environment 7(2) 281.
UNICEF (United Nations Children's Fund). 1999. "Towards Better Programming: A Water
Handbook." United Nations Children's Fund Programme Division. Accessed February

57

16, 2015.
http://www.sswm.info/sites/default/files/reference_attachments/UNICEF%201999%20W
ater%20Handbook.pdf.
USAF (US Air Force). 2012. Integrated Natural Resources Management Plan. Plan, Mountain
Home, ID: Mountain Home Air Force Base.
USDA-NRCS (US Department of Agriculture - National Resources Conservation Service). 2015.
"Web Soil Survey." http://websoilsurvey.nrcs.usda.gov/.
USEPA (US Environmental Protection Agency). 2014. Basic Information About Nitrate in
Drinking Water. February 05. Accessed February 16, 2015.
http://water.epa.gov/drink/contaminants/basicinformation/nitrate.cfm.
—. 2013. Drinking Water Contaminants. Accessed July 5, 2014.
http://water.epa.gov/drink/contaminants/#Inorganic.
—. 2014. Safe Drinking Water Act (SDWA). July 30. Accessed February 16, 2015.
http://water.epa.gov/lawsregs/rulesregs/sdwa/index.cfm.
—. 2012. What is Nonpoint Source Pollution? August 27. Accessed February 16, 2015.
http://water.epa.gov/polwaste/nps/whatis.cfm.
USGS (US Geological Survey). 1999. Improvements to the DRASTIC Ground-Water
Vulnerability Mapping Method. USGS Fact Sheet FS-066-99, Boise, ID: USGS.
Van Stempvoort, Dale, Lee Ewert, and Leonard Wassenaar. 1993. "Aquifer Vulnerability Index
(AVI): A GIS Compatible Method for Groundwater Vulnerability Mapping." Canadian
Water Resources Journal 25-37.

58

VanDerslice, Jim. 2007. Final Report: Dose-Response of Nitrate and Other Methemoglobin
Inducers of Methemoglobin Levels of Infants. Grantee Research Project Results,
Washington: US Environmental Protection Agency.
Vias, J.M., B. Andreo, M.J. Perles, and F. Carrasco. 2005. "A Comparative Study of Four
Schemes for Groundwater Vulnerability Mapping in a Diffuse Flow Carbonate Aquifer
Under Mediterranean Climatic Conditions." Environmental Geology 586-595.
Wagenet, Jeff R., and John L. Hutson. 1989. LEACHM: A Model for Simulating the Leaching
and Chemistry of Solutes in the Plant Root Zone. New York: Water Resources Institute.
Williams, Marshall L. 2014. Groundwater Level and Nitrate Concentration Trends on Mountain
Home Air Force Base, Southwestern Idaho. Open-File Report, Boise, ID: US Geological
Survey.
Winter, Thomas C., Judson W. Harvey, O. Lehn Franke, and William M. Alley. 1998. Ground
Water and Surface Water, a Single Resource. Circular 1139, Denver, CO: US Geological
Survey.
Yee, Johnson, and William Souza. 1987. Quality of Ground Water in Idaho. USGS Water-
Supply Paper 2272, Denver, CO: US Government Printing Office.
Zaporozec, Alexander, and John C. Miller. 2000. Ground-Water Pollution. Report 24, Paris:
UNESCO.

59

APPENDIX A: ADDITIONAL DRASTIC PARAMETER MAPS FOR MODEL 1 -
GENERIC AVAILABLE DATA

Figure A 1. DRASTIC Parameter Aquifer Media, using generic, available data.

60

Figure A 2. DRASTIC Parameter Topography (Slope), using contour elevation data.

61

Figure A 3. DRASTIC Parameter Impact of the Vadose Zone, using generic, available data.

62

Figure A 4. DRASTIC Parameter Hydraulic Conductivity, using generic, available data.

63

APPENDIX B: ADDITIONAL DRASTIC PARAMETER MAPS FOR MODEL 2 - SITE-
SPECIFIC DATA

Figure B 1. Rated DRASTIC Parameter Depth to Water, using site-specific data.

64

Figure B 2. Rated DRASTIC Parameter Aquifer Media, using site-specific data.

65

Figure B 3. Rated DRASTIC Parameter Hydraulic Conductivity, using site-specific data.

Abstract (if available)

Abstract Mountain Home Air Force Base (AFB) is located in Elmore County, southwestern Idaho. A regional aquifer is the primary drinking water source for the base residents. While current groundwater quality meets regulatory drinking standards, data collected from the U.S. Geological Survey (USGS) and Mountain Home AFB indicates a significant degradation in quality, particularly nitrate contamination. The purpose of this study was to implement a groundwater model to spatially delineate areas by vulnerability to groundwater contamination risk. The model provides a basis for evaluating the vulnerability to pollution of groundwater resources based on hydro-geologic parameters, which can help develop management practices to prevent additional nitrate groundwater contamination in the region. Two Geographical Information System-based groundwater vulnerability models using the DRASTIC method were created using generic, available data and site-specific data. The models were compared to each other, as well as groundwater quality data gathered from 25 wells (16 monitoring wells and 9 base production wells) throughout the study site to validate the model. While the results indicate that the site-specific model is slightly more reliable (56% prediction accuracy), compared to the generic data model (48% prediction accuracy), neither set of model predictions seem good enough to inspire confidence and it is clear that the results produced with the two model runs are not interchangeable. The greatest cause is relative to the small sampling size (n=25) of the wells. The small sample size limits the opportunities to conduct statistical analysis to validate the model outcomes. Additional studies would need to be performed using the same approach, but with larger sample sizes so that the sample size reported here (n=25) would not negatively affect the results.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Spatiotemporal patterns of salt and nutrient contamination in Los Angeles County's groundwater basins

PDF

Developing and implementing a GIS-based framework to identify optimal locations for clean water wells in sub-Saharan Africa

PDF

Predicting the presence of historic and prehistoric campsites in Virginia’s Chesapeake Bay counties

PDF

Evaluating surface casing depths of oil & gas operations in an effort to protect local groundwater: a GIS enabled process

PDF

An evaluation of soil sampling methods in support of precision agriculture in Northeastern North Carolina

PDF

Assessing the transferability of a species distribution model for predicting the distribution of invasive cogongrass in Alabama

PDF

A Maxent-based model for identifying local-scale tree species richness patch boundaries in the Lake Tahoe Basin of California and Nevada

PDF

Soil lead contamination from the Exide battery smelter: the role of spatial scale in cleanup efforts

PDF

Effect of spatial patterns on sampling design performance in a vegetation map accuracy assessment

PDF

Modeling prehistoric paths in Bronze Age Northeast England

PDF

Preparing for immigration reform: a spatial analysis of unauthorized immigrants

PDF

Using pattern oriented modeling to design and validate spatial models: a case study in agent-based modeling

PDF

Geological modeling in GIS for petroleum reservoir characterization and engineering: a 3D GIS-assisted geostatistics approach

PDF

Validation of volunteered geographic information quality components for incidents of law enforcement use of force

PDF

Beach morphodynamic change detection using LiDAR during El Niño periods in Southern California

PDF

Spatial and temporal patterns of long-term temperature change in Southern California from 1935 to 2014

PDF

Archaeological least cost path modeling: a behavioral study of Middle Bronze Age merchant travel routes across the Amanus Mountains, Turkey

PDF

Use of remotely sensed imagery to map sudden oak death (Phytophthora ramorum) in the Santa Cruz Mountains

PDF

The role of GIS in asset management: integration at the Otay Water District

PDF

Analyzing earthquake casualty risk at census block level: a case study in the Lexington Central Business District, Kentucky

Asset Metadata

Creator Dorsey-Spitz, Jenni Sue (author)

Core Title Modeling nitrate contamination of groundwater in Mountain Home, Idaho using the DRASTIC method

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Publication Date 06/30/2015

Defense Date 05/28/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag DRASTIC,GIST,groundwater,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Wilson, John P. (committee chair), Kemp, Karen K. (committee member), Lee, Su Jin (committee member)

Creator Email jennihil@usc.edu,jhill1227@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-582629

Unique identifier UC11300235

Identifier etd-DorseySpit-3521.pdf (filename),usctheses-c3-582629 (legacy record id)

Legacy Identifier etd-DorseySpit-3521.pdf

Dmrecord 582629

Document Type Thesis

Format application/pdf (imt)

Rights Dorsey-Spitz, Jenni Sue

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA