University of Southern California Dissertations and Theses

Selection of bridge location over the Merrimack River in southern New Hampshire: a comparison of site suitability assessments

(USC Thesis Other)

Selection of bridge location over the Merrimack River in southern New Hampshire: a comparison of site suitability assessments

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content SELECTION OF BRIDGE LOCATION OVER THE MERRIMACK RIVER IN
SOUTHERN NEW HAMPSHIRE:
A COMPARISON OF SITE SUITABILITY ASSESSMENTS

by

Mark R. Wimmer

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 Mark R. Wimmer

ii

ACKNOWLEDGMENTS
Foremost, I would like extend my sincere gratitude to my thesis advisor, Dr. Karen Kemp, for
without her expertise and guidance, I would have not completed my thesis. I would also like to
thank my committee members Dr. Su Jin Lee and Dr. Katsuhiko (Kirk) Oda for their assistance
and tutelage that gave me the building blocks for my research and study. Last but not least, I
want to thank my wife and children for their support through all the long nights and missed
family time spent on school; without them I could not have made it this far.

i

TABLE OF CONTENTS
ACKNOWLEDGMENTS ii
LIST OF TABLES iv
LIST OF FIGURES v
LIST OF ABBREVIATIONS vii
ABSTRACT viii
CHAPTER 1: INTRODUCTION 1
1.1 Study Area 2
1.2 A Methodology for Site Suitability 4
1.2 Research Objective 5
1.3 Outline of the Methodology 6
1.4 Thesis Organization 7
CHAPTER 2: PRIOR STUDIES - OVERVIEW AND RESULTS 9
2.1 Alternative Bridge Location Site Selection Methods 9
2.2 Alternative Bridge Locations 10
2.2.1 NRPC Alternative #1 10
2.2.2 NRPC Alternative #2 11
2.2.3 NRPC Alternative #3 12
2.3 NRPC Site Suitability Analysis 13
CHAPTER 3: IDENTIFICATION OF CRITERIA AND DATA PREPARATION 16
3.1 Identify Criteria 16
3.1.1 Population Concentrations between the Four Municipalities 17

ii

3.1.2 Distance from the Nearest Major Road 17
3.1.3 Bridge Abutments Cannot Be Located in the 100 Year Floodway 18
3.1.4 Roadway Surface Cannot Exceed 1% Grade between Banks 18
3.1.5 Location Needs To Minimize Environmental Impacts 18
3.2 Assemble Relevant Data 19
3.2.1 Merrimack River Hydrography 20
3.2.2 Digital Elevation Model (DEM) 20
3.2.3 100 Year Flood Zone 21
3.2.4 NH Public Roads 21
3.2.5 Land Cover 22
3.2.6 Population Dataset 22
3.3 Standardization 23
3.4 The 100 Year Floodway Raster 25
3.5 Slope Between Opposite Riverbanks 26
3.6 Land Cover 29
3.7 Distance from the nearest Major Road 32
3.8 Population Concentration 37
CHAPTER 4: WEIGHTED OVERLAY ANALYSIS AND RESULTS 40
4.1 Weighted Overlay with Equal Weights 41
4.2 Process for Selecting Suitable Bridge Locations 43
4.3 Weighted Overlay with Environmental Bias 48
4.4 Weighted Overlay with Engineering Bias 51
4.5 Weighted Overlay with Transportation Bias 53

iii

4.6 Sensitivity Analysis 55
CHAPTER 5: COMPARISON OF RESULTS 59
5.1 Comparison of NRPC Alternative #1 against the Weighted Overlay Results 59
5.2 Comparison of NRPC Alternative #2 against the Weighted Overlay Results 61
5.3 Comparison of NRPC Alternative #3 against the Weighted Overlay Results 61
5.4 Comparison Conclusion 61
CHAPTER 6: DISCUSSION AND CONCLUSIONS 63
6.1 Data Shortcomings and Limitations of Methodology 63
6.2 Future Site Suitability Analysis 65
6.3 Conclusion and Opportunities for Future Work 67
REFERENCES 68
APPENDIX A: IMAGES OF ALTERNATIVE NRPC BRIDGE BUILDING SITES 70

iv

LIST OF TABLES
Table 1: Merrimack River Hydrography Dataset Metadata 20
Table 2: Digital Elevation Model (DEM) Metadata 20
Table 3: 100 Year Flood Zone Dataset Metadata 21
Table 4: NH Public Roads Dataset Metadata 21
Table 5: Land Cover Dataset Metadata 22
Table 6: Population Dataset Metadata 23
Table 7: Suitability Values 24
Table 8: Available Land Covers and Availability to Build 29
Table 9: Selected Land Cover Types and Assigned Scale Values 31
Table 10: Major Road Classes 33
Table 11: Selection Criteria and Associated Weighting Percentage for the Equal Weights
Weighted Overlay 41

Table 12: Selection Criteria and Associated Weighting Percentage for the Environmental Bias
Weighted Overlay 49

Table 13: Selection Criteria and Associated Weighting Percentage for the Engineering Bias
Weighted Overlay 51

Table 14: Selection Criteria and Associated Weighting Percentage for the Transportation Bias
Weighted Overlay 53
Table 15: Original and Tweaked Scale Values of the Land Cover Criterion 56

v

LIST OF FIGURES
Figure 1: Map Study Area 3
Figure 2: Google Earth Image of Merrimack River between Manchester and Nashua, NH 4
Figure 3: Flowchart of GIS Work 6
Figure 4: NRPC Alternative #1 Bridge Building Site 11
Figure 5: NRPC Alternative #2 Bridge Building Site 12
Figure 6: NRPC Alternative #3 Bridge Building Site 13
Figure 7: Alternative #2 Right-of Way Impacts 14
Figure 8: Final 100 Year Floodway Raster Dataset 25
Figure 9: Merrimack River and 100 Year Floodway 26
Figure 10: Points with > 1% Slope between Them 27
Figure 11: Final Slope Raster Dataset 28
Figure 12: Final Land Cover Raster Dataset 32
Figure 13: Final Road Dataset 34
Figure 14: Accumulated Cost Distance through Adjacent Cells 35
Figure 15: Final Road Raster Dataset 36
Figure 16: US Census Block Groups across Study Area 37
Figure 17: Raster Display of Population per Cell 38
Figure 18: Final Population Concentration Raster Dataset 39
Figure 19: Weighted Overlay of Equal Selection Values 42
Figure 20: Northern Favorable Bridge Location 43
Figure 21: Southern Favorable Bridge Location 44
Figure 22: Restricted Areas along the River 45

vi

Figure 23: Less Favorable Areas along the River 46
Figure 24: Highest Suitability Distant from the Merrimack River 47
Figure 25: White Areas Indicating Unsuitable Build Areas 48
Figure 26: Weighted Overlay of Environmental Bias Values 50
Figure 27: Weighted Overlay of Engineering Bias Values 52
Figure 28: Weighted Overlay of Transportation Bias Values 54
Figure 29: Sensitivity Analysis Results 57
Figure 30: Highest Suitability Areas from Weighted Overlays Compared against NRPC
Alternatives 60
Figure 31: 100 Year Floodway and 1% Slope Criteria Check 65
Figure A-1: NRPC Alternative #1 Bridge Building Site 70
Figure A-2: NRPC Alternative #2 Bridge Building Site 71
Figure A-3: NRPC Alternative #3 Bridge Building Site 72

vii

LIST OF ABBREVIATIONS
AHP Analytic Hierarchy Process
CSV Comma-Separated Values
DEM Digital Elevation Map
DOT Department of Transportation
FEE Frederick E. Everett
FEMA Federal Emergency Management Agency
FIPS Federal Information Processing Standards
FIRM Flood Insurance Rate Maps
GIS Geographic Information Systems
GIST Geographic Information Science and Technology
MCDM Multiple Criteria Decision Making
MCE Multi-Criteria Evaluation
NAD North American Datum
NH New Hampshire
NHHD New Hampshire Hydrography Dataset
NLCD National Land Cover Database
NRPC Nashua Regional Planning Commission
SSI Spatial Sciences Institute
US United States
USACE US Army Corps of Engineers
USC University of Southern California

viii

ABSTRACT
The goal of this research was to assess alternative proposed bridge crossing locations over the
Merrimack River between the cities of Nashua, NH and Manchester, NH resulting from two site
suitability analysis studies that employ different criteria. A new bridge will provide an alternate
route for commuters to access the F.E. Everett Turnpike and U.S. Route 3 in southern NH.
Historic traffic count trends show that traffic on bridges and collector roads has increased
substantially due to residential growth. This thesis compared alternatives proposed by a site
suitability study conducted by Nashua Regional Planning Commission (NRPC) in 2003 to new
solutions derived in this thesis through weighted overlay analysis, which took into account
distance between major population concentrations and roads, position with respect to historic
floodways and terrain, and environmental impacts. The comparison shows how the location of
the most suitable bridge locations to span the Merrimack River change when the criteria are
altered and different suitability analysis processes are used. The thesis includes a description of
criteria and data utilized in the research, an explanation of how the standardized input layers
were created, an examination of the methodology for the weighted overlay analysis, and the
comparison results of the two site suitability analysis studies.

1

CHAPTER 1: INTRODUCTION
Since the 1950s, there has been a need for a bridge over the Merrimack River between the cities
of Nashua, NH and Manchester, NH. Urban growth along the east side of the Merrimack River
south of Manchester has produced demand for an alternate route for commuters to access the
F.E. Everett Turnpike and U.S. Route 3 in Nashua. Residents who live on the east side of the
Merrimack have a great interest in finding a solution to this site suitability problem of locating
the optimal location to build another bridge.
Currently there are four bridges over the Merrimack River adjacent to the metro areas of
Nashua and Manchester. Drivers wishing to access U.S. Route 3 from the east side of the river
have the following options (listed here from south to north):
1. The Tyngsboro Bridge, which provides access to US 3 via Massachusetts Route 113.
2. The Sagamore Bridge, connecting the Everett Turnpike (US 3) to New Hampshire Route
3A in Hudson. With urban growth in areas to the south of Nashua, this bridge has seen a
77% increase in vehicular traffic in the last ten years (City of Nashua 2001).
3. The Taylor Falls/Veteran’s Memorial Bridges (otherwise known as the Hudson Bridges),
which guides the traffic NH Route 111 through the city streets of Nashua (Hollis St and
Canal St). With over 40,000 daily vehicle crossings, these bridges lack the capacity to
handle rush hour traffic demand as well as other high volume travel times (City of
Nashua 2001).
4. The Raymond Wieczorek Drive/Manchester Airport Access Road bridge, built in 2011,
which provides a connection between NH Route 3A in the southern tip of Manchester, a
few miles south of Interstate 293, and the Everett Turnpike in the southern corner of
Bedford, with an interchange connecting the road with US 3.

2

Despite having four bridges serving the area in between Nashua and Manchester, there is a
12 mile gap between the Hudson Bridges and the Manchester Airport Access road bridge. A
new bridge would provide an additional river crossing between Nashua and Manchester and
another viable option for commuters to cross the Merrimack. It would also help alleviate normal
and rush hour traffic on the current bridges by providing another viable option across the natural
barrier.
The need for a bridge in the region between Manchester and Nashua is driven by social and
economic factors. Time spent traveling and ease of getting to locations on opposite sides of the
Merrimack River needs to be reduced and improved respectively in order to fulfill the need of
the local residents. Reducing travel costs, diverting heavy traffic flow and more efficiently
transporting people and goods across the river will make a positive economic impact on the
region.
1.1 Study Area
The study area consists of the region between the cities of Manchester and Nashua New
Hampshire. It is dominated by the Merrimack River Valley that extends from the Massachusetts
border north to central NH. The primary landform is eastern New England Upland, which
consists of a hilly landscape with elevation increasing as one moves east or west away from the
Merrimack River valley. The soil is fertile with fruits and hay grown in the region along with a
mosaic of habitats to include large stands of northern hardwoods, coniferous and mixed forests
and wetlands (Netstate 2015). A map of the study area can be seen in Figure 1.

3

Figure 1: Map of Study Area
The Merrimack valley is also home to New Hampshire's most prominent mill and factory
cities and towns. By population, Manchester and Nashua are the first and second largest cities
respectively in NH. The communities of Merrimack and Litchfield are located along the
Merrimack River between the larger cities. Merrimack on the western side of the river is also a
large town with Litchfield located in a more rural area on the eastern side. Figure 2 shows a
satellite view of the study area.

4

Figure 2: Google Earth Image of Merrimack River between Manchester and Nashua, NH
1.2 A Methodology for Site Suitability
Geographic information systems (GIS) can be utilized to assess suitability of bridge
locations by allowing decision makers access to large volumes of spatial data from a multitude of
sources (Ardeshir et al. 2014). GIS can help in the determination of suitable and unsuitable
locations through data analysis and manipulation. Multiple parameters can be analyzed to
determine if they have a higher or lower impact on the results.
Site suitability modeling using raster data is a classic GIS application, utilizing spatial
data to identify sites most suitable for a specific use. A site suitability analysis helps to identify
suitable sites that meet specific criteria. The results of the site suitability analysis produce a
detailed display of the most to least suitable areas for consideration of placement of a certain
enterprise, while filtering out unusable or less favorable sites (Kumar and Kumar 2014).
The number of criteria required in a particular analysis depends on purpose, location and
circumstances surrounding the area under study (Carr and Zwick 2005). Scale values and a

5

weighting system can be applied to the various levels of suitability to assess the overall
suitability for a specific location. The analysis of location suitability is a good method to assist
the decision maker to find a better solution that is based on the latest updated geographical
information data.
The methodology used in this thesis collectively considers many of the criteria,
objectives, and constraints associated with site suitability analysis, while accounting for their
relative importance through weighting and scale values. It is implemented to show how the
process might be applied using Esri ArcGIS version 10.2.1 to create and overlay the criteria
layers and perform the weighted overlay analysis. All GIS work performed in this thesis was
administered using Esri ArcGIS version 10.2.1.
This thesis compares alternatives proposed by a site suitability study conducted by
Nashua Regional Planning Commission (NRPC) in 2003 to new solutions derived in this thesis
through weighted overlay analysis. The methodology of the two analyses is similar, both using
specific criteria and weighting to formulate the most favorable location for a new crossing over
the Merrimack River. A final comparison of the resulting most suitable bridge locations is done
by overlaying all the alternatives from both analyses and assessing how each accomplishes their
own criteria.
1.2 Research Objective
The objective of this research was to assess alternatively proposed bridge crossing
locations over the Merrimack River between the cities of Nashua, NH and Manchester, NH
resulting from two site suitability analysis studies that employ different criteria. A new bridge
will provide a unique route across the river so the population of travelers would be well served to
better understand the different alternative span locations and the benefits and disadvantages of

6

each site. The results of this research can encourage public participation in the urban decision
making process and assist various planners and authorities to formulate a suitable plan for
sustained transportation development of the region.
1.3 Outline of the Methodology
The research was accomplished by first establishing criteria. Criteria help identify and
reduce the number of solutions that need to be considered by establishing requirements that need
to be met. They are derived from needs expressed by customers, experts and users with interest
in the outcome of the analysis. Having the correct criteria will help ensure the end result will
meet the needs and desires of the users. The criteria used in the NRPC site suitability analysis
and the research accomplished in this thesis are found in Sections 2.1 and 3.1 respectively.

Figure 3: Flowchart of GIS Work

7

Preparing input layers is the next step in getting the data in a format that can be used in
the weighted overlay process. Data must be manipulated into rasters that are standardized by not
only projection, but by a consistent scale of suitability. Precise and proper preparation of the
input layers will ensure the results are more accurate at the end of the analysis. A flowchart of
the entire process is shown in Figure 3. Layers are represented by square rectangles, GIS
operations by rounded rectangles, ArcGIS tools by ellipses and order of operations by blue
arrows.
Once the input layers are ready for the weighted overlay process, a weight is applied to
each of the input layers to represent different biases toward each criterion that are deemed more
or less important by the user. The layers are then multiplied by the assigned weighted value and
for each cell, the resulting values are added together. Higher values in the output raster generally
identify those locations as being the best while lower values are less favorable.
The comparison phase is not so much intended to choose one location over another, but
to identify the best suitability for the set of criteria. Rarely is GIS analysis used as a decision
making system that dictates a solution, rather it enables the user with the capability to evaluate
two different results. The final comparison is done between alternatives identified as favorable
in the NRPC site suitability analysis and the locations found most suitable in the weighted
overlay analysis done in this thesis.
1.4 Thesis Organization
Chapter One has introduced the background behind the need for a new bridge across the
Merrimack River and outlined the methodology and objectives of the past and present studies.
Chapter Two offers a background to the site suitability analysis performed by the NRPC and the
alternative bridge locations that resulted. Chapter Three identifies the criteria and data

8

preparation of the weighted overlay analysis as well as outlines the standardization and data used
in the process. Chapter Four discusses the results of the weighted overlay analyses. It also
provides an explanation of the visual process used in determining the most favorable bridge
locations and a sensitivity analysis to show the stability of the model. Chapter Five expounds on
the comparison of the NRPC Alternatives against the weighted overlay results. Finally Chapter
Six discusses the advantages and limitations of the methodology, considers its usability and
suggests potential improvements and future work.

9

CHAPTER 2: PRIOR STUDIES - OVERVIEW AND RESULTS
This chapter provides background regarding bridge location site suitability analysis in general
and the analysis performed by the Nashua Regional Planning Commission (NRPC) specifically.
The three alternative bridge locations formulated by the NRPC are presented. Also, this chapter
provides an overall assessment of the locations of the three alternative bridge locations.
2.1 Alternative Bridge Location Site Selection Methods
Selecting the ideal location for a bridge location is as important as building the bridge
itself correctly. Previous studies have used a representation of static, dynamic, deterministic or
stochastic mathematical programming models to determine the optimal site location. Static and
deterministic location models use fixed criteria, relying on the fact that the demand for a facility
in one location will not change (Plastria 2001). On the other hand, dynamic and stochastic
location models assume that future values such as environmental factors and population shifts
are variable (Owen & Daskin 1998). These formulations attempt to capture the uncertainty of
time-dependent input parameters such as forecast demand or geological factors. Other methods
such as Multi-Criteria Evaluation (MCE) use a multi-objective programming model to allow the
decision maker to take into account present and future goals (Schilling 1980). By using analytic
hierarchy process (AHP) or multiple criteria decision making (MCDM) approaches, site
suitability analysis produces a detailed display of the most-suitable areas for consideration of
placement of a certain facility, while filtering out unusable or less desirable sites (Kumar &
Shaikh 2012).
In respect to the bridge site over the Merrimac River, the NRPC performed a site
suitability analysis in 2003 formulating three possible locations for a new bridge. The future
bridge would be part of a project called the Circumferential Highway, a transportation project

10

first proposed in the late 1950s intended to provide additional crossings of the Merrimack River
and mitigate congestion in NH cities of Hudson and Nashua (NRPC 2003). All three alternative
sites were derived using the same criteria, though sites differ in overall suitability. The
following is a list of criteria used by the NRPC:
1. Must reduce traffic congestion based on NRPC’s regional traffic model
2. Road connector to bridge needs to have a posted speed limit of at least 45 miles per
hour and the capacity to two lanes
3. Cannot place linkage roadway in the Pennichuck Brook watershed
2.2 Alternative Bridge Locations
The following describes the three primary alternative bridge locations determined by the
NRPC. All three use the same NRPC site suitability model to anticipate land development
patterns, needs for traffic to move between the origin and destination places in the region and
how traffic will be represented in 2025 (NRPC 2003).
2.2.1 NRPC Alternative #1
Alternative #1 is a bridge connecting Concord St and NH 3A near the Hudson and
Litchfield border. The NRPC gave the traffic congestion and road capacity criterions higher
weighting in this alternative analysis. Figure 4 shows the location determined by the NRPC. A
larger version of this map can be found in Appendix A.

11

Figure 4: NRPC Alternative #1 Bridge Building Site (Source: NRPC 2003)

2.2.2 NRPC Alternative #2
Alternative #2 prescribes a bridge extending from the east end of Industrial Drive in
Merrimack, in a southeasterly direction across the Merrimack River through Litchfield to an
intersection with NH 111. This connects the new bridge to the Frederick E. Everett (FEE)
Turnpike, a major north-south highway in western NH. The NRPC gave traffic congestion and
impacts on the watershed criterions higher weighting in this alternative analysis. Figure 5 shows
the second location determined by the NRPC. A larger version of this map can be found in
Appendix A.

BRIDGE

12

Figure 5: NRPC Alternative #2 Bridge Building Site (Source: NRPC 2003)
2.2.3 NRPC Alternative #3
Alternative #3 proposes a bridge starting at the Henri Burque Highway, extending east
across the Merrimack River and ending at NH 111 on the east side of Hudson. The NRPC gave
traffic congestion and road capacity criterions higher weighting in this alternative analysis. The
location is similar to Alternative #1, though the road linkage is different in order to access
another suitable crossing site over the Merrimack River by avoiding the Pennichuck Brook
watershed. Figure 6 shows the location of this alternative as determined by the NRPC, and
again, a larger version of this map can be found in Appendix A.

BRIDGE

13

Figure 6: NRPC Alternative #3 Bridge Building Site (Source: NRPC 2003)
2.3 NRPC Site Suitability Analysis
The site suitability analysis accomplished by the NRPC was thorough and resulted in the
three alternatives outlined above. Though as stated through their own admission in the NRPC
Circumferential Highway White Paper, the alternatives proposed from the site suitability analysis
did not always provide the best location. The analysis of the alternatives was weighted heavily
on reducing traffic flow and less on the capacity of the road connectors and placement of the
linkage roadways outside of the watershed. Often times the new bridge location would decrease
traffic rates on major roads but cause detrimental effects on secondary roads feeding the new
bridge (NRPC 2003). Also, by building linkage roads directly through the Pennichuck Brook

BRIDGE

14

watershed on two of the three alternatives, traffic flow was again lowered by using shorter routes
but placed the bridge in an area of low suitability as defined by the third criterion.
Of the three Alternatives submitted, Alternative #1 did produce a future bridge site with
fewer impacts on natural and residential environments than the other two Alternatives by using
the existing road network and not building any addition linkage roads. Alternative #2 will have
substantial right-of-way impacts on agricultural lands and residential development in Litchfield
and Merrimack and will also impact important natural resources along the Merrimack River to
include a mining pit, wetlands and wintering grounds for the bald eagle shown in Figure 7.
Alternative #3 would pass through existing developed areas in both Nashua and Hudson,
resulting in intense right-of-way impacts in existing neighborhoods. The location could also
potentially impact Alvirne High School in Hudson and surrounding residential areas by locating
new supporting roads through school property and rural communities.

Figure 7: NRPC Alternative #2 Right-of-Way Impacts

15

Alternative #2 does reach its traffic flow objectives but the NRPC forecast that it would
never pass the Environmental Impact Statement process due to severe environmental impacts to
wetlands and habitat in the project area (NRPC 2003). Alternative #1 also achieved its traffic
goals by providing the largest reduction in traffic on the Taylor Falls/Veteran’s Memorial
Bridges, though its success was built on increasing traffic volumes on the Daniel Webster
Highway in south Merrimack, Concord Street in Nashua and the Henri Burque Highway to the
highest levels of any other Alternative. Additional problems with NRPC’s suggested sites can be
shown with the extra bridging needed to span an adjacent rail yard in north Nashua in Alternative
#3, which would result in a longer bridge and subsequently increased costs.
The NRPC site suitability analysis does produce three valid alternative river crossings,
though the heavy weighting to reduce traffic congestion results in alternatives that are vulnerable
in other suitability factors. Different weighting biases should be employed showing weights that
favor other criteria instead of primarily the traffic flow reduction. By accomplishing a weighted
overlay analysis that incorporates a more varied and balanced weighting of criteria, it will
provide the population of southern New Hampshire different options to take into consideration.
The next chapter outlines the methodology used here for a site suitability analysis that addresses
these varied factors.

16

CHAPTER 3: IDENTIFICATION OF CRITERIA AND DATA PREPARATION
This chapter outlines the methodology taken to produce the raster layers needed to perform the
weighted overlay analysis to re-assess suitable sites for Merrimack River bridges. The process
begins with identifying the evaluation criteria and then continues by outlining the processes
required to assemble data appropriate for each criteria.
3.1 Identify Criteria
In order to perform the intended analysis, site requirements or criteria are required to
assess the overall site suitability. Each criterion can be expressed in varying degrees of
suitability for the decision under consideration. Criteria can be considered either a factor or
constraint. A factor is a criterion that makes a location favorable for the placement of the facility
in question. If a specific criterion is considered a constraint, the constraining attribute values
make the location an unfavorable site. The selection criteria used in this study to identify
favorable locations for the bridge are:
1. The river crossing should be near population concentrations between the four
municipalities of Manchester, Nashua, Litchfield and Merrimack.
2. The new bridge location should be in the vicinity of major roads, not exceeding five
miles for ease of linkage.
3. Bridge abutments cannot be located in the 100yr floodway
4. Roadway surface cannot exceed 1% grade between banks
5. Location needs to minimize environmental impacts by avoiding wetlands, forested
and residential zones

17

Each of the criteria is described in detail in the next sections while later sections describe the
data sources and preparation of each criterion raster.
3.1.1 Population Concentrations between the Four Municipalities
The river crossing should be located where there are high numbers of nearby residents. It
needs to be near the highest cluster of users, or population concentrations of the four
municipalities of Manchester, Nashua, Litchfield and Merrimack it would serve. By using
population data from the 2010 US Census, population concentrations were found by joining
block groups population count data to the study area block groups boundaries. To assess this
criterion, a surface was created during the weighted overlay process in which each cell in the
study area was given a suitability value based on how many people are within a specified
distance of each cell.
3.1.2 Distance from the Nearest Major Road
The new bridge location should be in the vicinity of major roads, not exceeding five
miles for ease of linkage. When acting as a connector, the bridge needs to provide access to
major roads capable of handling traffic volume that will result from the everyday demand of the
river crossing. A major road is defined as a route that provides largely uninterrupted travel and
is designed for speeds greater or equal to 45 miles per hour. The Hudson River Bridges are at an
optimal location to provide a large population easy access to a river crossing but they only link to
secondary city streets. These secondary streets are unable to handle the flow of users trying to
access the bridges and cause congestion. The new span needs to connect to major roads in order
to facilitate flow of traffic to and from itself. To meet this criterion a surface was created in
which each cell in the study area was given a suitability value based on the proximity to major
roads.

18

3.1.3 Bridge Abutments Cannot Be Located in the 100 Year Floodway
The NH Department of Transportation requires all new bridge construction to have the
roadway surface over the bridge abutments above the floodway of a 100 year event. Bridge
abutments are the substructures at the ends of the bridge deck where the bridge contacts and rests
on the banks of the river. In case of flooding, the bridge roadway should not be under water to
allow continued use of the bridge through the high water event timeframe. Locations in the 100
Year Flood Zone Map indicate areas prone to flooding if a 100 Year flood occurred. Therefore
these would be not suitable locations for the bridge to be located. During the weighted overlay
analysis, locations inside the floodway designated area were given restricted values to ensure the
final location cannot be located in the 100 Year Floodway.
3.1.4 Roadway Surface Cannot Exceed 1% Grade between Banks
The NH Department of Transportation dictates that the bridge roadway surface cannot
exceed a 1% grade. This means that the span between the abutments on the opposing river banks
needs to maintain a less than 1% grade so the construction requirements can be met. The slope
between locations on opposing banks of the Merrimack River was used to determine if this
criterion can be fulfilled at a specific site. During the weighted overlay analysis, the identified
locations on opposite banks that have a slope greater than 1% were given restricted values to
ensure the final site cannot be located where the grade of the bridge roadway would exceed
construction requisites.
3.1.5 Location Needs To Minimize Environmental Impacts
Introduction of new construction always changes the environment around the site, but
city planners and engineers try to avoid disturbing fragile or hard to replace natural environments
which are most likely to occur in wetlands and forested zones. Similar environmental impacts

19

can be seen to occur when uprooting people. Eminent domain can be exercised but engineers try
to avoid positioning a new bridge in the middle of a densely populated residential area.
Involuntary resettlement has been shown to cause destructive repercussions on humans to
include loss of employment, degraded access to equal housing, marginalization and social
disarticulation (Contractor, Madhiwalla, and Gopal 2006). This criterion will be accomplished
by avoiding building the bridge in areas that will cause detrimental impacts on both human and
animal environments. During the weighted overlay process, suitability values were applied to
each different land cover based on impacts that bridge construction could cause on the
environment and residents.
3.2 Assemble Relevant Data
Preparations for site investigations include collecting topographical maps, infrared
photography, remote sensing images and aerial photographs (Groenier & Gubernick 2008). The
process of precisely positioning a river bridge consists of a comprehensive study of preliminary
engineering, hydrology and hydraulics, roadway alignment, along with environmental and
geological surveys (Groenier & Gubernick 2007). The objective of this thesis though is not to
identify an exact final location for a new river crossing, but to show where the suitable locations
are for a potential final plat.
Once the criteria used to measure the degree a particular objective(s) is being met are
determined, data are sought that may be used to represent these criteria. All data was gathered
and stored in a geodatabase that could be manipulated in ArcMap 10.2.1 for suitability value
assessment. The following sections detail the data layers used in the GIS analysis.

20

3.2.1 Merrimack River Hydrography
The New Hampshire Hydrography Dataset (NHHD) contains stream segments that make
up the surface water drainage system for the Merrimack River. It also includes flow direction,
names, stream orders and centerline representations for areal water bodies. The hydrography
dataset was used to reference the location of the Merrimack River and its tributaries. Metadata
for the NHHD is shown below in Table 1.
Table 1: Merrimack River Hydrography Dataset Metadata

Source
NH GRANIT: New Hampshire's Statewide Geographic Information System
(GIS) Clearinghouse, retrieved from http://www.granit.unh.edu/
Originator Complex Systems Research Center, University of New Hampshire
Format Personal Geodatabase Feature Class (.gdb)
Date of compilation Jan 2006

3.2.2 Digital Elevation Model (DEM)
The Digital Elevation Model (DEM) is a terrain elevation data set in a digital raster form
consisting of regularly spaced grid of elevation points. The standard DEM consists of a regular
array of elevations cast on a designated coordinate projection system. Elevation data was used to
calculate slope between the opposing locations along the banks of the Merrimack River to
determine if the roadway surface grade criteria could be met. Metadata for the DEM is shown
below in Table 2.
Table 2: Digital Elevation Model (DEM) Metadata

Source Downloaded through the National Map Viewer (http://nationalmap.gov)
Originator US Geological Survey
Format Digital Elevation Map (DEM) Raster
Resolution 1 Arc Second (approximately 30 meters)
Date of compilation 04 Feb 2014

21

3.2.3 100 Year Flood Zone
A 100 year Flood Zone is derived from Flood Insurance Rate Maps (FIRM) produced by
the Federal Emergency Management Agency (FEMA). The 100-year flood is referred to as the
1% annual exceedance probability flood, since it is a flood that has a 1% chance of being equaled
or exceeded in any single year. The NH Department of Transportation requires all new bridge
construction to have the roadway surface over the bridge abutments above the floodway of a 100
year event. The flood zone data was utilized to determine if the 100 year flood zone criteria
could be met. Metadata for the 100 Year Flood Zone is shown below in Table 3.
Table 3: 100 Year Flood Zone Dataset Metadata

Source Federal Emergency Management Agency (FEMA)
Originator Federal Emergency Management Agency (FEMA)
Format Shapefile Line Feature Class (.shp)
Date of compilation 05 Dec 2011

3.2.4 NH Public Roads
The NH Public Roads dataset is a statewide dataset containing the location of state, local
and selected private roads in NH and their associated attributes, including road names. Rural and
urban arteries, interstates and collectors as well as local and public roads are included in the data.
This dataset was utilized to analyze the location of major roads in respect to proximity of
favorable river crossing locations. Table 4 shows metadata for the dataset.
Table 4: NH Public Roads Dataset Metadata

Source
NH GRANIT: New Hampshire's Statewide Geographic Information System
(GIS) Clearinghouse, retrieved from http://www.granit.unh.edu/
Originator NH DOT Bureau of Planning and Community Assistance
Format Shapefile Line Feature Class (.shp)
Accuracy Horizontal accuracy of approximately 12 meters
Date of compilation 30 Sep 2013

22

3.2.5 Land Cover
National Land Cover Database (NLCD) classification schemes are based primarily on a
decision-tree classification of circa 2011 Landsat satellite data along with ancillary data sources,
such as topography, census and agricultural statistics, soil characteristics, wetlands, and other
land cover maps. NLCD 2011 keeps the same 16-class land cover classification scheme that has
been applied consistently across the United States in previous years. The data used in this thesis
was from the National Land Cover Database (NLCD) map of Superzone 13, which covers all of
New England and Mid-Atlantic US. Table 5 shows metadata for the dataset.
Table 5: Land Cover Dataset Metadata

Source
Downloaded through the National Map Viewer
(http://nationalmap.gov)
Originator US Geological Survey
Format Shapefile Polygon Feature Class (.shp)
Resolution Spatial resolution of 30 meters across the conterminous U.S.
Date of compilation 04 Apr 2014

3.2.6 Population Dataset
Population data presented in Census block groups was used in the calculation of
population density in the thesis research. Census block groups are geographical units used by the
US Bureau of the Census to collect and tabulate decennial census data. They are bounded by
streets, roads, railroads, streams and other bodies of water, other visible physical and cultural
features (US Census Bureau 2010). Population data is released by the Census in a comma-
separated values (CSV) table format and needs to be joined to a block group map in order to use
the population data in spatial analysis. Table 6 shows metadata for the dataset.

23

Table 6: Population Dataset Metadata

Source 2010 US Census
Originator US Census Bureau
Format Joined Shapefile Polygon Feature Class (.shp) and CSV table
Spatial Unit Census Block Groups
Date of Collection 01 Apr 2010
Date of Compilation 19 Aug 2014

3.3 Standardization
When aggregating spatial data in an overlay analysis, the coordinate systems of all layers
need to be uniform. The Projected Coordinate System used for all the layers in this analysis was
the NAD 1983 StatePlane New Hampshire FIPS 2800 (Meters). All datasets were projected to
this coordinate system to ensure proper integration during the various integrated analytical
operations.
All the final rasters used for input to the overlay analysis needed to have a uniform spatial
resolution, or cell size. Determining the appropriate cell size is a balance of resolution and
higher spatial accuracy with faster processing and display as well as smaller file size. Since the
bridge is the primary focus of the spatial analysis, the cell size would need to be large enough to
appropriately cover the lanes of a span while remaining small enough in size to have the spatial
accuracy needed to determine features impacting the construction of a two to four lane bridge.
30 meters was decided to be the scale of analysis as that is approximately the distance across a
two lane road. The 30 meter scale provides resolution small enough to capture roadway
influences, while at the same time being large enough to not cause data or file size problems.
To ensure the cell alignment of all final rasters, a Snap Raster was used during the
processing of the raster. A workable area was clipped out of the available data to include the
area of prime interest. This clipped section makes certain the area of study was included in the

24

processed area while also eliminating the need to crunch data that would never be used in the
spatial analysis due to its distance from the study area. A raster was made from the clipped
region, with cell size the prescribed 30 meters, and designated the Snap Raster. Through the
environment settings in ArcGIS, the output extent can be ensured to be the same as the Snap
Raster. By using the Snap Raster in creating all the final rasters, it ensures the output cell size of
all process rasters is the same as the snap raster cell size, as the cells in the output rasters are all
aligned with the cells of the Snap Raster.
To perform the final weighted raster overlay process, each criterion can be expressed in
varying degrees of suitability for the decision under consideration. This can be done by
reclassifying the data values into a common range of values. The range of 1 to 9 was used with
“1” being Restricted and 9 being the Highest Suitability. For example, distance from the nearest
road would be treated not as a Boolean statement all-or-none buffer zone of suitable locations,
but rather, as a continuous expression of suitability according to a special range of values. A
Boolean constraint can also be used if a criterion has a yes/no tolerance (Eastman 1999), where
the Restricted suitability value is assigned with the no value. Table 7 shows all the Suitability
Values and corresponding definitions.
Table 7: Suitability Values
Suitability Value Definition
1 Restricted
2 Lowest Suitability
3 Lower Suitability
4 Low Suitability
5 Moderately Low Suitability
6 Moderately High Suitability
7 High Suitability
8 Higher Suitability
9 Highest Suitability

25

3.4 The 100 Year Floodway Raster
The 100 year floodway criterion states that the bridge abutments cannot be located in the
100 year floodway. This drives a Boolean dataset with only two possible values, outside the
floodway (suitable) or inside the floodway (not suitable). Creating a raster dataset for this
criterion was relatively straight forward. The polygon for the 100 year floodway was converted
into a raster (ArcGIS Polygon to Raster Tool [Conversion]). The raster was reclassified (ArcGIS
Reclassify Tool [Spatial Analysis]) with values inside the polygon assigned a value of 1 and
values outside a 9. Figure 8 shows the final reclassified raster dataset for the 100 year floodway.

Figure 8: Final 100 Year Floodway Raster Dataset

26

3.5 Slope Between Opposite Riverbanks
In order to determine if the slope of the roadway surface between the opposing banks of
the river meets the criterion of not exceeding a 1% grade, the elevations of the locations along
the river banks was needed. The distance between locations opposite each other on opposing
river banks was also required to calculate the slope between them. Since the bridge abutments
could not be built in the 100 year floodway, the location of the usable river banks would need to
be determined by the area of the 100 year floodway. Figure 9 shows the Merrimack River and
the expanded riverbanks caused by the 100 year floodway.

Figure 9: Merrimack River and 100 Year Floodway

27

To fully capture the extent of the 100 year floodway, positions every 30 meters were
plotted along the outside edge of the 100 year floodway along the river. Then by using the
elevation raster, the value of the cell corresponding with each of these locations along the banks
was recorded to the attribute table of the point feature class (ArcGIS Extract Mutli Values Points
Tool [Spatial Analyst]).

Figure 10: Points with > 1% Slope between Them

Next, the nearest point on the west side of the river was found for each point on the east
side (ArcGIS Near Tool [Analysis]). This process adds both the ID of the nearest west side point

28

and the distance from it to the east side point attribute table. Then, by joining this west side
points by ID to the east side points, the elevation of both sides of the river and the distance
between them can be used to get the slope from any east side point to the nearest west side point.
Slope was calculated between the joined points by using the difference in their elevations divided
by their distance apart. Points with greater than 1% slope between them were identified and can
be seen in Figure 10. The points were buffered (ArcGIS Buffer Tool [Analysis]) out to 30m to
ensure they included any roadway surface that would be built over them. The resulting polygons
from the buffering were converted into a raster dataset (ArcGIS Polygon to Raster Tool
[Conversion]).

Figure 11: Final Slope Raster Dataset

29

The final step was to reclassify (ArcGIS Reclassify Tool [Spatial Analysis]) the values in
the raster. Since the data type for this criterion is Boolean with only two possible values, < 1%
grade (suitable) or > 1% (not suitable), values inside the polygons were assigned a value of 1 and
values outside a 9. Figure 11 shows the final reclassified raster dataset for the slope criterion.
3.6 Land Cover
The physical and biological cover over a parcel of land can be a big determinant of how it
is used for current or future activities. If the parcel has already been disturbed by human action,
it is considered more readily available for continued human use than non-disturbed areas.
Residential zones are the home environment to humans, in the same manner that forests are the
domiciles to deer, raccoons, birds, moose, etc. So when the criteria states that the bridge location
should minimize environmental impacts, this specification means impacts to both natural and
human areas.
In building a raster layer, the first step was to identify what types of land covers could be
used that would not cause a big environmental impact. From the available NLCD, the land
covers were assessed on whether construction would cause detrimental environmental effects to
natural or human habitats. Table 8 shows the available land covers that were included in the
dataset and the rationale for their inclusion / exclusion into the availability to be considered as a
bridge construction site.
Table 8: Available Land Covers and Availability to Build
Land Cover Type Rationale
Availability to
Build Bridge on
Residential /
Commercial
Do not want to uproot
residents or businesses
Cannot
build on
Open Wetlands
Cannot destroy wetland environment
Cannot
build on
Forested Wetlands
Beech / Oak Do not want to disturb Cannot

30

Land Cover Type Rationale
Availability to
Build Bridge on
Paper Birch / Aspen standing timber / forests build on
Other Hardwoods
White / Red Pine
Spruce / Fir
Hemlock
Mixed
Orchards Do not want to remove standing trees
Cannot
build on
Water Need to build bridge over river Can build on
Hay / Pasture
Land disturbed by humans to raise non-
indigenous plants and or livestock
Can build on
Crops
Land disturbed by humans to raise non-
indigenous plants
Can build on
Other Miscellaneous land disturbed by humans Can build on
Transportation Existing roads Can build on
Disturbed
Land disturbed by humans during past event
that left the parcel changed from natural habitat
Can build on

The criterion’s objective is to utilize land covers that would cause only minimal impact
on the environment. The use of wetlands, forested and residential zones were made unavailable
to meet the intention of the criterion. The Pennichuck Brook watershed, on the western side of
the Merrimack River between the towns of Merrimack and Nashua, consists of wetlands and
forested land and therefore it was not set as suitable and consequently avoided. Tearing down
houses and deforestation are not minimal environmental impacts.
Despite the fact the bridge needs to span water, its abutments cannot actually be built in
water. The water land cover cannot be totally excluded because the bridge will cover some
section of water for the end result, but any possible locations cannot be located in the middle of a
water land cover. Setting a scale value to Restricted assigns a value to a cell in the output
weighted overlay result that is the minimum value of the evaluation scale set, minus 1. This will
ensure that no water type land covers will be identified as suitable locations but will not
eliminate the areas as potential span areas. Making the water unsuitable would exclude viable
areas from suitability consideration.

31

For the areas that were deemed buildable, it is next necessary to rank them in terms of
suitability. Transportation and Disturbed areas have the highest suitability as both have already
been impacted by man and are the cheapest and most feasible for construction. Areas designated
as Other were set with High Suitability as these areas have been disturbed by man at some point
in history, but may still have the potential to be returned their natural environment. Hay/Pasture
and Crop areas were deemed buildable areas but only given a Moderately High Suitability as it is
more favorable to raise crops and hay on these parcels than to be the location of supporting
roadwork or the span itself. Table 9 shows the land covers that could be used and their
suitability level of little or no environmental impact.
Table 9: Selected Land Cover Types and Assigned Scale Values
Land Cover Type Scale Value Suitability Level
Water 1 Restricted
Hay / Pasture 6 Moderately High Suitability
Crops 6 Moderately High Suitability
Other 7 High Suitability
Transportation 9 Highest Suitability
Disturbed 9 Highest Suitability

Once the buildable land covers were identified, a layer was built that only included the
land covers listed in Table 8. This layer was converted into a raster (ArcGIS Polygon to Raster
Tool [Conversion]) and reclassified (ArcGIS Reclassify [Spatial Analysis]) with the scale values
shown once again by Table 8. White areas on the map are locations where the land cover was
deemed unsuitable. They were removed from further consideration in this analysis by setting the
associated pixel values to NULL. Figure 12 shows the final reclassified raster dataset for the
land cover criterion.

32

Figure 12: Final Land Cover Raster Dataset
3.7 Distance from the nearest Major Road
In order to properly handle the flow of traffic, the new bridge location needs to be in the
vicinity of major roads, not exceeding five miles for ease of linkage. To best incorporate this
criterion into the spatial analysis, a dataset with suitability values that correspond to the
proximity to a major road was needed. Major roads were those considered to be principal
arteries or collectors of the roadway network in the study area. Table 10 shows the classes of
roads from the NH Public Roads dataset that were included as major roads.

33

Table 10: Major Road Classes
Road Class Definition
17 Urban Collector
16 Urban Minor Arterial
14 Urban Principal Arterial - Other
11 Urban Principal Arterial - Interstate
7 Rural Major Collector
6 Rural Minor Arterial
2 Rural Principal Arterial – Other
1 Rural Principal Arterial - Interstate

There are three major roads, the Everett Turnpike, US Route 3 and NH Route 3A, that
run north and south parallel to the Merrimack River. The Everett Turnpike and US Route 3 are
located on the western side of the river, acting as the major connectors for local and transient
traffic between the large metropolitan areas of Nashua and Manchester. NH Route 3A is a major
thoroughfare on the eastern side of the river, providing the same access to the major urban areas
though more for the local population in the region. Since these three major roads parallel the
Merrimack River and have equal opportunity for connection to a crossing point anywhere along
their path, they were removed from the major road dataset. The connector roads that need to be
considered are those running east and west, as those are the arteries that will bring traffic flow to
a river traverse. Proximity to where the east and west running roads intersect the river is the
primary influence on the placement of a new bridge. A line shapefile was created consisting of
all major roads and can be seen in Figure 13.

34

Figure 13: Final Road Dataset
Once the major roads were identified, the distance of each location to the nearest road
needed to be assessed. Of course, in the real world, roads are not built in straight lines. Rather,
they are constructed around natural barriers such as waterways and terrain. For this analysis,
unsuitable land cover also acted as barriers to straight-line travel since access roads would not be
built across these areas. In order to take into account this non-Euclidean distance, the raster

35

dataset for this criterion was built using a distance with barriers. A path distance tool (ArcGIS
Path Distance [Spatial Analysis]) was used to create a raster in which the assigned cell values are
the accumulative distance of traveling through adjacent cells to the nearest road, avoiding areas
with restricted land cover types (set as NULL). Figure 14 shows the values of the accumulated
cost distance through adjacent, not NULL cells to the nearest major road.

Figure 14: Accumulated Cost Distance through Adjacent Cells

The major roads shapefile and land cover raster were used to produce this raster. The
raster was reclassified (ArcGIS Reclassify [Spatial Analysis]) with scale values from 1 to 9, with
1 being the least suitable and 9 being the most. Suitability was determined by the proximity to a

36

major road, with the highest suitability being the closest (most accessible) and the lowest
suitability assigned to the largest distances with values decreasing by a simple linear decay
function. No cells had distances greater than 5 miles so there was no need to eliminate any as
unsuitable. Figure 15 shows the final reclassified raster dataset for the distance to major road
criteria. The effect of the barriers caused by land cover patches through which travel is not
permitted (white areas) is evident in the isolated red and yellow areas.

Figure 15: Final Road Raster Dataset

37

3.8 Population Concentration
In order to serve the highest number of people, the bridge should be located where there
are high numbers of nearby residents. Areas where people are clustered together are called
population concentrations. Data from the 2010 US Census was utilized to find the population
concentrations by first joining (ArcGIS Spatial Join [Analysis]) block groups population count
data to the study area block groups boundaries, shown in Figure 16.

Figure 16: US Census Block Groups across Study Area
To determine the population concentration for each cell, it was necessary first to generate
a raster in which population counts could be assigned to each cell. The population counts for
each block group were first converted to density per square meter using the field calculator.
Since the size of the eventual raster cell would be 30 meters by 30 meters, the field calculator
was again utilized to multiply that value by 900, resulting in a density value of people per 900
square meters assigned as an attribute to each of the block group polygons. Finally, the polygon

38

shapefile was converted to a raster (ArcGIS Polygon to Raster Tool [Conversion]), allowing the
block group density attribute value to be equal to the count of individuals in each cell. Figure 17
shows the results of the raster conversion and the resulting population per cell created from the
US Census block group population counts.

Figure 17: Raster Display of Population per Cell
The last step was to employ the focal statistics tool (ArcGIS Focal Statistics [Spatial
Analysis]) to produce an output raster where the value for each output cell is a function of the
sum of all values in a specified neighborhood around the input cells. The desired vicinity was
approximately one kilometer so a 30 x 30 cell neighborhood was used to create the final raster.

39

The resulting raster values then indicate how many people are within approximately one
kilometer of each cell. The raster was reclassified (ArcGIS Reclassify [Spatial Analysis]) with
scale values from 1 to 9, with 1 being the least suitable and 9 being the most. Suitability was
determined through a linear function by the amount of population, with the highest suitability
being the greatest numbers and subsequently the lowest suitability with the smallest. Figure 18
shows the final reclassified raster dataset for the population criteria.

Figure 18: Final Population Concentration Raster Dataset
With the set of criteria rasters standardized, proceeding onto the weighted overlay
analysis was now possible. The next chapter discusses the weighted overlay analysis process
and the output derived from the different biases implemented.

40

CHAPTER 4: WEIGHTED OVERLAY ANALYSIS AND RESULTS
This chapter discusses the weighted overlay analysis and its results. While conducting the
overlay analysis, influence percentages were sequentially modified to reflect different biases
resulting in different outcomes regarding favorable bridge locations. The chapter also includes
an explanation of the visual process used in determining the most favorable bridge locations and
a sensitivity analysis to show the stability of the analysis model.
Weighted overlay analysis commonly is used to solve multi-criteria problems such as
optimal site selection or suitability modeling. It is a technique for applying a common scale of
values to diverse and dissimilar inputs to create an integrated analysis that begins with a set of
standardized raster layers that are reclassified with the common evaluation scale. During the
overlay calculation, the cell values in each raster are multiplied by a weighting percentage
representing the importance of that criterion to the final suitability result. Finally, each of these
weighted values is summed for each cell. The result is a raster whose cell values represent the
suitability. Higher values generally indicate that a location is more suitable.
In this weighted overlay analysis, four different biases were assessed by changing the
weighting percentages. The analysis allows the various factors to have different importance by
adjusting the weighting percentage. These biases implemented are:
1. Equal Weights - No bias with all weighting percentages equal
2. Environmental Bias – Bias towards lessoning environmental impact
3. Engineering Bias – Bias with higher weighting on construction criteria
4. Transportation Bias – Bias on roads and population the traffic network would serve

41

Each of these is expanded on in separate sections below. A section to explicate the process for
selecting suitable bridge locations based on the results of the weighted overlay analysis also
follows.
4.1 Weighted Overlay with Equal Weights
For the first weighted overlay (ArcGIS Weighted Overlay Tool [Spatial Analyst]), the
same weighting percentage was applied to all the selection criteria as shown in Table 11. By
using a uniform weight, the resulting map has no bias towards any of the selection criteria since
it uses each equally toward modeling the suitability. Figure 19 shows the resultant overlay.
Table 11: Selection Criteria and Associated Weighting Percentage for the Equal Weights
Weighted Overlay
Selection Criteria Weighting Percentage Used
Central to Populations of Four Municipalities 20%
Distance to Nearest Major Road not Exceed 5 miles 20%
Bridge abutments cannot be located in 100yr floodway 20%
Roadway surface cannot exceed 1% grade between banks 20%
Location needs to minimize environmental impacts 20%

Using a process of visual analysis that is described in detail in the following section, two
suitable areas were found, 1.5 miles north of the Hudson Bridges (Inset #3) and just upstream
from Reed’s Island in Merrimack, NH (Inset #2). The first area is inside the network of major
roads and populace of Nashua, making the land suitable for building a structure across the river
as the area has already been urbanized. The second area also has urbanization and a robust road
network down to the river banks on both sides, though it is farther from a major population.
Both locations are also outside the 100 year floodway and any slope restrictions, resulting in
good, suitable locations. With equal weights for each criterion, there is no distinction between
them.

42

For comparison purposes, Inset #1 in Figure 19 shows an unfavorable area south of big
bend in the river. There is a large area of restricted on both sides of the river. This area is where
the 100 year floodway is very wide, a condition that consequently lowers the suitability.

Figure 19: Weighted Overlay of Equal Selection Values

43

4.2 Process for Selecting Suitable Bridge Locations
After completing the weighted overlay analysis (ArcGIS Weighted Overlay Tool [Spatial
Analyst]), a visual analysis of the map was accomplished to identify the most favorable bridge
locations. Visual analysis consisted of scanning the mapped areas alongside the Merrimack
River’s path from the US 3A Bridge in Manchester south to the NH 111 bridges in Nashua.
Areas that had the highest suitability both on directly opposite sides of the river as well as in a
close proximity to the river were identified as most favorable.

Figure 20: Northern Favorable Bridge Location
The weighted overlay results with equal weights can be used as an example of the area
selection process through visual analysis. By doing a visual assessment of the final map, it can
be seen that areas of the highest suitability ended being located around the Reed’s Island
location. Figure 20 shows how this northern favorable bridge location has areas of high

44

suitability adjacent to the water with matching areas of the same suitability directly across the
river. The thinnest area of the river was selected as keeping the span across the river short will
save on construction costs and requires less of an environment footprint. Figure 21 shows the
southern favorable bridge location also with areas of high suitability adjacent to the river and
matching areas of the same high suitability directly across on the opposite bank.

Figure 21: Southern Favorable Bridge Location
Visual analysis was also used to identify areas along the river that were unfavorable for
the bridge location. By identifying unfavorable areas, sections of the river could be determined
to be unfavorable for a new bridge location and eliminated from the selection process that
determines the most favorable location. Attention could be focused away from these areas and
back to the more favorable ones. Figure 22 shows parts of the river that have a very wide 100
year flood way, which become restricted areas through the weighted analysis process. When the

45

restricted areas become too wide, the areas become less favorable due to the long length of
bridge that would be needed to span them.

Figure 22: Restricted Areas along the River
Figure 23 shows other less favorable areas identified through the weighted overlay
process. A thin area of high suitability runs parallel to the river in the upper left portion of the
figure, though it is located far from the river. The area directly across the river that would match
this area consists of only moderate suitability. So despite it being adjacent to the river, its lower
suitability and the fact it has a poor matching area on the opposite side of the river make this area
less favorable than the ones shown in Figures 20 and 21. Similarly, a matching problem exists in
the bottom part of Figure 23 where only one area of high suitability is close to the river. Though
not as unfavorable as areas shown in Figure 22, the areas in Figure 23 have matching and
suitability deficiencies that make them less favorable than the areas in Figures 20 and 21.

46

Figure 23: Less Favorable Areas along the River
When areas of highest suitability are located distant from the Merrimack River, they are
not considered as favorable build locations. Since the primary objective of the study is to
identify a location for a bridge over the Merrimack River, only highly suitable areas adjacent to
the Merrimack River were considered. Figure 24 shows an area of Highest Suitability on both
sides of the Nashua River. This would be a favorable location for a bridge crossing the Nashua
River. But since the areas are not adjacent to the Merrimack River, they were not considered to
indicate a favorable location for a building site. This exclusion rule was be applied to all highly
suitable areas far from the banks of the Merrimack River.

47

Figure 24: Highest Suitability Distant from the Merrimack River
White areas on maps should not be misconstrued as missing data or unknown suitability.
During the building of the land cover layer, all unsuitable land covers were set as NULL in order
to prevent them from being considered in the process. During visual analysis, any area of white
can be affirmed to be an unsuitable land cover and subsequently should be avoided when
determining favorable locations. Figure 25 shows large areas of white that represent large
expanses of unsuitable land covers and should be not considered as favorable building sites.

48

Figure 25: White Areas Indicating Unsuitable Build Areas
This visual analysis process of identifying adjacency, matching and suitability plus
deficiencies was applied to identify all the favorable bridge locations for each of the weight
overlay bias runs discussed in the sections below. The compilation of all the favorable bridge
locations from the visual analysis performed can be found in Chapter 5.
4.3 Weighted Overlay with Environmental Bias
Environmental concerns play an important role in determining a suitable bridge site. The
second weighted overlay (ArcGIS Weighted Overlay Tool [Spatial Analyst]) was run to put
more weight on the need to minimize environmental impacts. This was done by increasing the
percent weighting of the Land Cover raster. Population and roads were kept somewhat relevant
as areas with roads and population tend to already have an urban footprint, so using the areas

49

lowers the amount of undisturbed land that needs to be developed on. The selection criteria of
slope and the 100 year floodway were set lower since they are more derived from engineering
governances. Weighting percentages used for the environmental bias are shown in Table 12.
Table 12: Selection Criteria and Associated Weighting Percentage for the Environmental
Bias Weighted Overlay
Selection Criteria Weighting Percentage Used
Central to Populations of Four Municipalities 20%
Distance to Nearest Major Road not Exceed 5 miles 20%
Bridge abutments cannot be located in 100yr floodway 10%
Roadway surface cannot exceed 1% grade between banks 10%
Location needs to minimize environmental impacts 40%

Figure 26 shows the Environmental Bias overlay. Two suitable areas were found through
visual analysis, again 1.5 miles north of the Hudson Bridges (Inset #3) and just upstream from
Reed’s Island near Merrimack, NH (Inset #2). The first area was selected under the weighted
overlay with equal weights as well as through this weighted overlay bias as it is located in the
urban zone of northeast Nashua. The area has already been urbanized so there would be little to
no environmental impact if construction would take place at this location to cross the river. The
second area has urbanization down to the river banks on both sides, with light industry on the
west side and a housing development on the east side. Environmentally this site would not
increase the human footprint, though by saving natural areas, the bridge would need to be placed
through homes and businesses. The weighting definitely keeps potential construction away from
environmentally sensitive areas, but will sacrifice urban areas to reach its ecological goals.
Inset #1 shows an unfavorable area south of horseshoe pond. There are large areas of
low suitability on both sides of the river. On the east side moderate suitability is found next to
the river but there is no matching suitability on the western bank. The lower suitability on the
western side forms an area that would have to be traversed with a longer bridge span. Unsuitable
land covers occupy this area, lowering the suitability.

50

Figure 26: Weighted Overlay of Environmental Bias Values

51

4.4 Weighted Overlay with Engineering Bias
For the third Weighted Overlay (ArcGIS Weighted Overlay Tool [Spatial Analyst]), the
weighting percentage values were set to provide an Engineering Bias. This weighted overlay
focuses on finding suitable bridge locations based solely on criteria that are pertinent to the
construction and engineering of the structure. Slope of the bridge roadway between the
abutments, building out of the 100 year floodway and proximity to major roads are all key
engineering criteria. The engineering biased weights used are shown in Table 13.
Table 13: Selection Criteria and Associated Weighting Percentage for the Engineering Bias
Weighted Overlay
Selection Criteria Weighting Percentage Used
Central to Populations of Four Municipalities 05%
Distance to Nearest Major Road not Exceed 5 miles 30%
Bridge abutments cannot be located in 100yr floodway 30%
Roadway surface cannot exceed 1% grade between banks 30%
Location needs to minimize environmental impacts 05%

Figure 27 shows the Engineering Bias weighted overlay. Two suitable areas were found,
the first 2 miles north of the Hudson Bridges (Inset #3) and just upstream from the location
identified by the Equal and Environmentally Biased overlays (Inset #2). The second takes
advantage of electrical power lines crossing the river south of Merrimack, NH. The first area is
shifted a little north from the other dually located area due to more favorable slope along the
river further upstream. The second area also better avoids the slope and 100 year floodway
restrictions. The high bias towards roads cements the second location which is close to two
principal arterial on ramps, or collectors. The second site would be a very ideal location for the
bridge due to its prime location near connecting roads, centrally located geographically between
the existing bridges and acceptable slope and distance outside the 100 year floodway. The
stretch of land right along the river, which is undisturbed forested land, kept this spot from being

52

located until the environmental weight was lowered. The weighting values used before kept
potential construction away from the environmentally sensitive area.

Figure 27: Weighted Overlay of Engineering Bias Values

53

Inset #1 shows an unfavorable area just south of the major bend in the river. There are
large areas of moderate suitability on both sides of the river. On the east side higher suitability is
found next to the river in the southern part of this area but there is no matching high suitability
on the western bank. This area is where the 100 year floodway is very wide and subsequently
lowers the suitability.
4.5 Weighted Overlay with Transportation Bias
For the final Weighted Overlay (ArcGIS Weighted Overlay Tool [Spatial Analyst]), the
weighting percentage values were set to provide a Transportation Bias. The weighted overlay
focuses on finding suitable bridge locations based on criteria that are pertinent to the distance of
major roads with a minor focus on location of population. The primary role of roads is to move
population efficiently and in a timely manner, so the population weighting was left high to reflect
this important relationship. The transportation biased weights used are shown in Table 14.
Table 14: Selection Criteria and Associated Weighting Percentage for the Transportation
Bias Weighted Overlay
Selection Criteria Weighting Percentage Used
Central to Populations of Four Municipalities 25%
Distance to Nearest Major Road not Exceed 5 miles 45%
Bridge abutments cannot be located in 100yr floodway 10%
Roadway surface cannot exceed 1% grade between banks 10%
Location needs to minimize environmental impacts 10%

Figure 28 shows the Transportation Bias overlay and the suitable areas for a bridge
crossing. Two suitable areas were found, the first 1.5 miles north of the Hudson Bridges (Inset
#3), at a very similar location to those identified by the Equal and Environmentally Biased
overlays. The second is very close to the Engineering Bias northern location near the electrical
power lines crossing south of Merrimack, NH (Inset #2). The first area is aligned well with the
urban road structure and population base in northeast Nashua. The second area is close to two

54

principal arterial on ramps, or collectors, on the west side of the river. The fact that this second
site shows as high suitability with the higher population weighting shows that it is geographically
centered location between the larger metropolitan centers of Nashua and Manchester proves to
be a plus for serving the population needs of the region.

Figure 28: Weighted Overlay of Transportation Bias Values

55

Inset #1 shows an unfavorable area just south of the US 3A Bridge. There is an area of
moderate suitability on the east side of the river and nothing on the west side. This is due to no
roads nor population in the vicinity on the west side. Without favorable roads or people in the
area, the location is unfavorable.
Chapter 5 summarizes the locations selected in this set of overlay analyses and compares
them to the NRPC study. Before looking to that final comparison, the next section examines the
stability of these results through a sensitivity analysis.
4.6 Sensitivity Analysis
In order to test the stability of the weighted overlay analysis output, sensitivity analysis
can be performed to determine just how sensitive the results are to small changes in the scale
values of the criterion. Such small variations can be used to reflect the amount of uncertainty in
the subjective assignment of these values. This modified input is then used in the original
weighted overlay analysis so that the results can be compared to the original output to determine
the amount of the difference such changes make. Large differences would show a high
sensitivity and small differences would indicate a low sensitivity to this uncertainty. Lower
sensitivity is desired, making the overall results of the weighted overlay output more stable.
The scale values for the land cover criterion were chosen for the sensitivity analysis since
the assignment of values was somewhat subjective. The slope and 100 year floodway scale
values are Boolean in nature and therefore have no changeable values. The population and roads
criterion scale values are determined by a straight-forward linear decay function. It is logical and
sensible to place the bridge near population centers and existing road networks, and a simple
distance decay from these is relatively objective as a scaling method. However, what type of
land cover is most suitable for a bridge can be argued different ways. A more environmental

56

approach might choose that undisturbed land is not used, while a cost efficient approach might
favor undeveloped land.
The original scale values used during the initial land cover weighted overlay analysis can
be found in Table 15. Water was set to “1” because despite the fact the bridge needs to span
water, its abutments cannot actually be built on water. Transportation and Disturbed areas were
set at “9” as both have already been impacted by man and are prime locations for construction.
Other areas were set to a “7” as these areas have been disturbed by man at some point in history,
but may be on their way to returning to their natural state. Hay/Pasture and Crop areas were
deemed buildable areas but only given a “6” to make them less favorable. However, it is equally
reasonable to set these lower to reflect the environmental value of these lands. Thus in this
sensitivity analysis they were both set to “2”, making them less suitable for site construction.
Table 15 also shows the values used for the sensitivity analysis.
Table 15: Original and Tweaked Scale Values of the Land Cover Criterion

Land Cover
Type
Original
Scale Value
Tweaked
Scale Value
Water 1 1
Hay / Pasture 6 2
Crops 6 2
Other 7 7
Transportation 9 9
Disturbed 9 9

Once the correct scale values for the sensitivity analysis were identified, the raster was
reclassified. The weighted overlay analysis for the environmental bias was rerun as it has the
land cover criterion as its highest weighting percentage, which is shown in Table 12 in Section
4.3. A raster calculator was used to determine the difference between the original and new
weighted analysis output. Figure 29 shows the results of the difference calculation.

57

Figure 29: Sensitivity Analysis Results

58

The results show there was not a major change in suitability levels and thus it is
concluded that the model is stable, despite the uncertainty in some of the rating values. Some
differences can be seen when the changes are made so the results also prove that model inputs do
have an effect on the output. This effect is logical as decreasing the scale value for areas with
Hay / Pasture and Crops resulted in a decrease in suitability for those same areas in the output at
the end of the weighted overlay process.

59

CHAPTER 5: COMPARISON OF RESULTS
It is now possible to compare the results of the weighted overlays against the three alternatives
the NRPC concluded after their spatial analysis. Figure 30 shows the highest suitability areas
from this weighted overlay analysis and it also displays the NRPC Alternatives. Figure 30
allows the results of the weighted overlays to be compared directly against the results of the
NRPC spatial analysis results. Only the NRPC Alternative #3 shares a common result with any
of the weighted overlay results. The following sections break down further how the three NRPC
Alternatives compare directly against the weighted overlay results.
5.1 Comparison of NRPC Alternative #1 against the Weighted Overlay Results
NRPC’s Alternative #1 is the middle option of the three displayed on Figure 30. It is
placed moderately well to be within good distance of major roads. It is also out of the 100 year
floodway. However it is located in a restricted area of slope, meaning the grade between the
opposing river banks have a slope greater than 1% between them, which is not allowed by
construction standards. In order to build at this location, the abutments for a bridge would need
to be pushed back away from the river far enough to relax the future bridge surface grade below
1%. Unfortunately this would further the detrimental environmental impact as both sides of the
river at this location has forested areas that would need to be cut down to make way for the
construction of the bridge. When considered in the context of the criteria examined, Alternative
#1 does not present a favorable location.

60

Figure 30: Highest Suitability Areas from Weighted Overlays Compared against NRPC
Alternatives

61

5.2 Comparison of NRPC Alternative #2 against the Weighted Overlay Results
NRPC’s Alternative #2 is the farthest north option of the three displayed on Figure 30.
Like Alternative #1 it out of the 100 year floodway but has forested areas on both sides of the
river. On the eastern side, farmland would have to be used to build the needed support roads to
make the bridge viable at this location. Unlike Alternative #1, slope is not a restricting criteria
but it is still in a low suitable area, meaning the grade between the opposing river banks is less
than 1% but still close enough to potentially cause problems. Alternative #2 is a better choice
than Alternative #1, though it still does not favorably meet all the criteria.
5.3 Comparison of NRPC Alternative #3 against the Weighted Overlay Results
NRPC’s Alternative #3 is the southernmost option of the three displayed on Figure 30.
Like Alternative #1, it is placed moderately well to be within good distance of major roads and it
is also out of the 100 year floodway. Though unlike Alternatives #1 and #2, there are no slope
issues at this place in the river. In fact, Alternative #3 shares a highly suitable area with the
results of the Engineering biased weighted overlay, meaning slope is highly suitable at this
location. Alternative #3 is the only one of the three NRPC Alternatives to share an area with one
of the weighted overlay results. The western side of the river is forested but the eastern side is
lightly urbanized, making the eastern bank a favorable location for the bridge. This plus
Alternative #3 meeting the other criteria makes it the best of the three NRPC alternatives.
5.4 Comparison Conclusion
The conclusions of both site suitability analyses indicate that there are multiple locations
in the study area that are favorable for bridge construction. The comparison shows how the
location of the most suitable bridge locations to span the Merrimack River changes when the
criteria are altered and different suitability analysis processes are used. The varied alternatives

62

are all valid conclusions for the respective criteria used. Therefore in order to find the most
suitable bridge location, any future plans would need to define the criteria to best meet the
objectives of the users.

63

CHAPTER 6: DISCUSSION AND CONCLUSIONS
This chapter examines the shortcomings of the data and limitations of the methodology and
provides several recommendations for expanding future suitability analysis. Final conclusions
and future work regarding the placement of a bridge over the Merrimack River in southern NH
are also discussed.
6.1 Data Shortcomings and Limitations of Methodology
While there are many advantages of the methodology used in this thesis, there are many
limitations that may reduce the accuracy of the conclusions. The disadvantage of the weighted
analysis is that it provides only areas of suitability and no definitive answer. The classes of
suitability are subjective and can change with the definition of the criterion. How each thing is
rated depends on what is determined important by the user. While an analysis of the results can
conclude that there is a perfectly acceptable final solution, another reasonable conclusion could
be derived through a different perspective with divergent criteria.
Furthermore, the shelf life of the conclusions formulated in this thesis is short. The
findings are tied to land cover assessments, which can change with urban development, natural
disasters or other major shifts in human activity in a specific area. Transportation networks can
change with the building of new roads or expanding current traffic corridors. Rapid changes in
population are also possible in the study area with the ever expanding housing developments and
urbanization in this growing region. Current and accurate data is key to running the site
suitability analysis to produce up-to-date results.
With the ever changing data, comes the inability for this methodology to quickly
incorporate the new assessments. For example if land covers were to drastically change in the
region, the new features would have to be projected, the suitability classes reevaluated and the

64

weighted overlays analysis performed again. A change in any of the base data would result in
the need to repeat many steps. An introduction of new parameters would also drive a repetition
of the overall process.
When assigning suitability classes, many assumptions are made in terms of the driving
forces behind bridge placement. Deciding differently on whether environmental concerns justify
extra construction costs or if proximity to current road networks is a higher priority than being
closer to population centers can lead to considerably different results Expert opinions and
research into what are the most suitable factors will increase the probability that a proposed
location will best fulfill all the required mandates.
One last shortcoming of the weighted overlay analysis is the loss of fidelity when using
low weight percentages in the overlay process. Some criteria have Boolean datasets which
strictly allow or disallow a bridge to be placed in a certain area. If the weighting percentage of a
Boolean criterion is set too low, the other criteria may have enough weight to produce areas of
suitability that may mask a disallowed area into appearing like a favorable location. A simple
check of favorable site locations against any Boolean criteria will ensure this shortcoming does
not occur. Figure 31 shows such a check of the different bias results against the 100 year
floodway and 1% slope do not build areas. No favorable location from any of the biases lies
inside the 100 year floodway nor in any specified areas where slope is greater than 1%.

65

Figure 31: 100 Year Floodway and 1% Slope Criteria Check

6.2 Future Site Suitability Analysis
The site suitability analysis performed in this thesis used specific criteria to formulate
results. Additional criteria could be added and assessed to add a higher resolution to the areas
deemed as suitable. Avoiding sensitive areas such as habitats for endangered species can be
added. Proximity to public facilities such as hospitals, fire houses and police stations seems

66

relevant to helping those who protect and serve get to where they are needed in the most
judicious route possible.
Engineering thresholds and criteria for construction are complex and take into account a
multitude of parameters to ensure the success, safety and longevity of a span across a river.
Capacity constraints, width of the bridge surface and number of lanes feeding the span are
examples of potential criteria that could be used in future suitability analyses. Experts in the
engineering field would need to advise on critical criteria to ensure the needed objectives are
being realized.
Additionally, the methodology used in the thesis only accounts for travel across the river
by automobile. Many commuters or travelers will utilize the bridge on foot or bicycle. Future
rail networks may expand across the Massachusetts border into New Hampshire and need a
bridge to cross the river. All three non-vehicular modes of travel require transportation networks
that differ from motorized vehicles. Incorporating criteria to better accommodate different types
of transportation would increase the scope of overall user suitability.
Increasing the types of land covers that were examined would improve the suitability
assessment. Classification such as depth of bedrock, soil type or high fall level (rapids) in the
river could impact the ability to build the bridge abutments, lowering or improving construction
suitability. Wildlife habitats or nesting sites could be considered as low suitability land cover
while rock out croppings could be considered a natural area of higher suitability.
Future site suitability analysis could incorporate all the additions discussed plus any other
criteria identified by public officials, engineers, users or other individuals or groups involved in
the planning and research phase of the bridge construction. The possibilities are endless as the

67

weighted overlay model is a flexible tool able to accommodate any specific criterion with data to
support it.
6.3 Conclusion and Opportunities for Future Work
This thesis has demonstrated how, by incorporating different criteria into the spatial
analysis process, a varied set of results may be formulated. Alternative approaches taken to
solve the same problem showed that there is no one absolute correct location for a new bridge
across the Merrimack River, but only suitable alternatives based on different criteria.
Determining the criteria has been found to be the most crucial step in the site suitability analysis.
Detailed research and planning needs to be accomplished to refine and amalgamate the criteria to
ensure a future bridge location will indeed serve the all the needs of the intended users.
While the conclusions formulated here may be more helpful in ascertaining a more
suitable bridge location across the Merrimack, it also demonstrates that continued analysis and
up-to-date data can benefit the analysis process and overall result. Since the criteria are so
important to the site suitability process, the data that is used to perform the spatial analysis under
the constraints of each criterion needs to be the latest and most accurate as possible. Errors in
the data can reverberate through the analysis and be detrimental to the end result.
The NRPC will continue their work towards planning and constructing a new span across
the Merrimack River. Funding, permitting, political and public support and environmental
concerns will all be challenges that will need to be addressed and overcome. There is a definite
need for a bridge crossing in the region north of the Taylor Falls Bridge. Site suitability analysis
can point to favorable locations, but it will come down to the will of the people of southern NH
whether the results of this study or any future spatial analysis will ever be utilized to aid in actual
bridge construction.

68

REFERENCES
Ardeshir, A., Mohseni, N., Behzadian, K. and Errington, M. 2014. Selection of a bridge
construction site using fuzzy analytical hierarchy process in geographic information system.
Arabian Journal for Science and Engineering 39 (6) (06/01): 4405-20,
http://dx.doi.org.libproxy.usc.edu/10.1007/s13369-014-1070-2.

Carr, Margaret H. and Zwick, Paul. 2005. Using GIS suitability analysis to identify potential
future land use conflicts in North Central Florida, Journal of Conservation Planning. Vol 1;
58-73.

City of Nashua, NH Planning Department, Community Development Division. 2000. Nashua
Master Plan. Website: www.gonashua.com/AddOns/MasterPlan/welcome.htm.

Contractor, Qudsiya , Madhiwalla, Neha and Gopal, Meena 2006. Uprooted Homes, Uprooted
Lives: A Study of the Impact of Involuntary Resettlement of a Slum Community in Mumbai.
Mumbai, India: Centre for Enquiry into Health and Allied Themes.

Eastman, J. R. 1999. Multi-criteria evaluation and GIS, New Developments in Geographical
Information Systems: Principles, Techniques, Management and Applications. Chapter 35:
493-502.

Federal Emergency Management Agency (FEMA) Map Service Center. 2015. Flood Insurance
Rate Map. FEMA 26 Apr 2015. Website: www.fema.gov.

Groenier, J.S. and Gubernick, R. A. 2007. Choosing the best site for a bridge. Journal of the
Transportation Research Board, 347–354. DOI 10.3141/1989-40.

Groenier, J. S, and Gubernick, R. A. 2008. Locating bridges for sustainability. International
Journal of Transportation Technology Transfer (T2) in Africa, 5, 23–25.

Kumar, Santosh and Kumar, Ritesh. 2014. Site Suitability Analysis for Urban Development of
a Hill Town Using GIS Based Multicriteria Evaluation Technique: A Case Study of Nahan
Town, Himachal Pradesh, India. International Journal of Advanced Remote Sensing and
GIS. vol 3 (1), pp. 516-524.

Kumar, Manish, and Shaikh, Vasim Riyasat. 2012. Site suitability analysis for urban
development using GIS based multi-criteria evaluation technique: A case study of
Mussoorie municipal area, Dehradun district, Uttarakhand, India. Journal of the Indian
Society of Remote Sensing, 41, (2): 417-424.

Nashua Regional Planning Commission (NRPC). 2003. Circumferential Highway White Paper.
NRPC 17 Sep 2003. Website:
www.nashuarpc.org/files/7713/9473/7155/circhwywhitepaper.pdf.

69

New Hampshire Department of Transportation. 2010. Standard Specifications - Road and Bridge
Construction. Aug 2010. Website: www.nh.gov/dot/index.htm.

Netstate. 2015. New Hampshire Geography. NSTATE 29 Mar 2015. Website: www.netstate.com

Owen, Susan Hesse, and Daskin, Mark S. 1998. Strategic facility location: A review. European
Journal of Operational Research, 111, (3): 423-447.

Plastria, Frank. 2001. Static competitive facility location: An overview of optimization
approaches. European Journal of Operational Research, 129, (3): 461-470.

Schilling, David A. 1980. Dynamic location modeling for public-sector facilities: A multicriteria
approach, Decision Sciences, 11 (4): 714-24.

US Army Corps of Engineers (USACE), Hydrologic Engineering Centre. 1998. HEC-1 Flood
Hydrograph Package, User’s Manual, Version 4.1.

US Army Corps of Engineers (USACE), Hydrologic Engineering Centre. 2010. HEC-RAS River
Analysis System, User’s Manual, Version 4.1.

US Census Bureau. 2010. Geographic Areas Reference Manual, Chapter 11. FEMA 04 Mar
2013. Website: www.census.gov.

70

APPENDIX A: IMAGES OF ALTERNATIVE NRPC BRIDGE BUILDING SITES
Figure A-1: NRPC Alternative #1 Bridge Building Site (Source: NRPC 2003)

BRIDGE

71

Figure A-2: NRPC Alternative #2 Bridge Building Site (Source: NRPC 2003)

BRIDGE

72

Figure A-3: NRPC Alternative #3 Bridge Building Site (Source: NRPC 2003)

BRIDGE

Abstract (if available)

Abstract The goal of this research was to assess alternative proposed bridge crossing locations over the Merrimack River between the cities of Nashua, NH and Manchester, NH resulting from two site suitability analysis studies that employ different criteria. A new bridge will provide an alternate route for commuters to access the F.E. Everett Turnpike and U.S. Route 3 in southern NH. Historic traffic count trends show that traffic on bridges and collector roads has increased substantially due to residential growth. This thesis compared alternatives proposed by a site suitability study conducted by Nashua Regional Planning Commission (NRPC) in 2003 to new solutions derived in this thesis through weighted overlay analysis, which took into account distance between major population concentrations and roads, position with respect to historic floodways and terrain, and environmental impacts. The comparison shows how the location of the most suitable bridge locations to span the Merrimack River change when the criteria are altered and different suitability analysis processes are used. The thesis includes a description of criteria and data utilized in the research, an explanation of how the standardized input layers were created, an examination of the methodology for the weighted overlay analysis, and the comparison results of the two site suitability analysis studies.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A site suitability analysis for an inland port to service the ports of Los Angeles and Long Beach

PDF

Site location suitability analysis for a smart grid network

PDF

Installing public electric vehicle charging stations: a site suitability analysis in Los Angeles County, California

PDF

Population disaggregation for trade area delineation in retail real estate site analysis

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

Site suitability analysis for implementing tidal energy technology in southern California

PDF

Community gardens for social capital: a site suitability analysis in Akron, Ohio

PDF

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

PDF

Relocation bay: identifying a suitable site for the Tampa Bay Rays

PDF

Predicting archaeological site locations in northeastern California’s High Desert using the Maxent model

PDF

A comparison of weighted and fuzzy overlays in mapping landslide susceptibility, south-central front range, Colorado

PDF

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

PDF

Providing a new low-cost primary care facility for under-served communities: a site suitability analysis for Service Planning Area 6 in Los Angeles County, California

PDF

Locating the need for financial education

PDF

Integrating spatial visualization to improve public health understanding and communication

PDF

Identifying suitable sites for sheltering outside in Long Beach, California

PDF

Geographic information systems eelgrass (Zostera marina) habitat restoration suitability model, Long Island Sound, USA - a ‘sound-wide’ model

PDF

Preparing for immigration reform: a spatial analysis of unauthorized immigrants

PDF

Site selection of medium density housing in Tacoma, Washington: where to put “missing middle” housing

Asset Metadata

Creator Wimmer, Mark R. (author)

Core Title Selection of bridge location over the Merrimack River in southern New Hampshire: a comparison of site suitability assessments

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Publication Date 06/23/2015

Defense Date 05/19/2015

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

Tag Bridge,GIS,Merrimack,OAI-PMH Harvest,suitability analysis,weighted overlay analysis

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Kemp, Karen K. (committee chair), Lee, Su Jin (committee member), Oda, Katsuhiko (Kirk) (committee member)

Creator Email mwimmer@usc.edu,wxwim@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-576347

Unique identifier UC11299862

Identifier etd-WimmerMark-3508.pdf (filename),usctheses-c3-576347 (legacy record id)

Legacy Identifier etd-WimmerMark-3508.pdf

Dmrecord 576347

Document Type Thesis

Format application/pdf (imt)

Rights Wimmer, Mark R.

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