Dever Joseph MRP ( .Pdf )

EMBRACING CAPITAL INVESTMENT: AN ANALYSIS OF DEVELOPMENT ORIENTED TOWARDS MIAMI’S METRORAIL

JOSEPH DEVER

A RESEARCH PROJECT PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER IN URBAN AND REGIONAL PLANNING

UNIVERSITY OF FLORIDA

2021

To public transit users

ACKNOWLEDGMENTS

Thank you Professor Abhinav Alakshendra, Professor Ruth Steiner, and Aaron

DeMayo for your assistance and guidance in completing this research project. More

importantly, thank you for your patience in what became a freewheeling document filled

with several changes amid the pursuit to connect so many of public transit’s challenges.

Even prior to this research project, in my classes with each of you, thank you for helping to shape my view of planning, economics, transportation, and urban design.

To Professor Thomas Hawkins and Kyle Dost, thank you for helping the Online program be more accessible and improve its delivery capacity. The importance of broad-based planning knowledge is essential for so many people and this program offers a quality delivery mechanism that other programs lack. I look forward to seeing this program develop and am hopeful for new successes in a profession that has

created so many unintended consequences in the past several decades.

Within the MURP program, thank you to the faculty and students for emparting

your focus, wisdom, and passion for planning. There is truly much work to be done and I

am optimistic in our impact on the success of our communities.

To my closest friends and colleagues, thank you for listening to my greatest

urban planning gripes, including perpetually pointing out good and bad urbanism at the

least appropriate and generally irrelevant times. Additionally, thank you for your support

in my endeavors and for being there for me through the highs and lows.

Thank you to my parents, Bernadette Houghton and Joe Dever, for showing and

helping me to be a better person. Without your advice, patience, and support, I certainly

would not be where I am today.

To Donna, thank you for your support through this program and your enduring patience with my perpetually evolving plans. Thank you for willingly taking public transit whenever we go on vacation—just to see how it works. I promise it is worth the journey.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 8

LIST OF FIGURES ...... 10

LIST OF ABBREVIATIONS ...... 11

CHAPTER

1. INTRODUCTION ...... 14

2. LITERATURE REVIEW ...... 17

Ridership ...... 18 Transit-Oriented & Transit Adjacent Development ...... 21 Measuring the Built Environment ...... 23 Summary ...... 25

3. METHODOLOGY ...... 26

Selecting a City ...... 26 Qualitative Analysis: Content ...... 28 Quantitative Approach: The Built Environment ...... 31 Quantitative Analysis – Demographic and Parcel-Level Data ...... 32 Statistical & GIS Analyses ...... 33 Summary ...... 35

4. DATA ...... 39

Content Analysis ...... 39 Demographic and Parcel Data ...... 42 Built Environment Data ...... 46

5. RESULTS & DISCUSSION ...... 55

Station Identities & Context ...... 55 Content Analysis ...... 63 Quantitative Analysis ...... 66 Future Research & Summary ...... 72 Qualitative Assessment, History & Discussion ...... 73

6. CONCLUSION ...... 82

APPENDIX

A. INDEPENDENT VARIABLE DESCRIPTIONS ...... 91

B. VARIABLE DESCRIPTIVE STATISTICS ...... 94

C. SUPPLEMENTAL DATA ANALYSIS FOR 2012 PARCELS ...... 95

D. CENSUS DATA MD COUNTY ...... 98

E. SHAPEFILE & RIDERSHIP DATA SOURCES ...... 100

LIST OF REFERENCES ...... 107

LIST OF DATA SOURCES ...... 110

LIST OF NEWSPAPER REFERENCES ...... 110

BIOGRAPHICAL SKETCH ...... 137

LIST OF TABLES

Table page

Table 3-2. This table shows Built Environment variables to be considered in this research...... 37

Table 3-3. This table shows data sources, file formats, and relevant information contained in the dataset...... 37

Table 3-4. This table shows demographic and parcel variables to be considered in this research...... 38

Table 4-2. Summary Categories showing the number of sub-categories classified as Negative, Neutral, or Positive...... 48

Table 4-3. Results from Ordinary Least Squares Model at the ¼ mile radius with Land Value and Land Improvement Values removed...... 50

Table 4-4. Results from Ordinary Least Squares model using a ½ mile radius and with Just Value and Effective Year Built variables removed...... 51

Table 5-1. Table showing station name, designation by Renne and Ewing (2013) as TOD/TAD/Hybrid, and the 2019 ridership ranking from the Miami-Dade Technical Ridership Reports (2019)...... 79

Table 5-2. Table showing Correlated variables categorized by impact...... 79

Table 5-3. Table showing percentage of parcels developed after 1983 (excluding vacant and unbuilt lots) and the boarding rank of the 6 stations studied...... 81

Table A-1. Independent Variable Descriptions...... 91

Table B-1. This table shows the summary statistics for the analysis of FDOR parcel data within ¼ mile of the six selected stations...... 94

Table B-2. This table shows the summary statistics for the analysis of FDOR parcel data within ½ mile of the six selected station...... 94

Table B-3. This table shows the summary statistics for the analysis of 2010 Census data within ¼ mile of the six selected stations...... 94

Table B-4. This table shows the summary statistics for the analysis of 2010 Census data within ½ mile of the six selected stations...... 94

Table C-1. This table shows the ArcGIS Pro statistical output...... 96

Table C-2. This table shows the ArcGIS Pro statistical output...... 97

Table D-1. This table shows the summary statistics for the analysis of 1990 Census Data within ¼ mile of the six selected stations...... 98

Table D-2. This table shows the summary statistics for the analysis of 1990 Census Data within ½ mile of the six selected stations...... 98

Table D-3. This table shows the summary statistics for the analysis of 1990 Census Data within ½ mile of the six selected stations...... 98

Table E-1. GIS Datafiles & Shapefiles ...... 100

Table E-2. Miami-Dade County Ridership Technical Reports ...... 101

LIST OF FIGURES

Figure page

Figure 4-1. Percentage of summary categories identified in the 209 articles reviewed for the content analysis...... 48

Figure 4-2. This chart shows the distribution of positive, negative, and neutral sub- categories mentioned by year...... 49

Figure 4-3. Figure shows each ¼ mile station catchment area with parcel data showing standard residuals from the OLS regression model...... 51

Figure 4-4. Figure shows each ½ mile station catchment area with parcel data showing standard residuals from the OLS regression model. This figure shows the adjusted model identified in Table 4-4...... 52

Figure 4-5. Figure shows each ¼ mile station catchment area with built environment survey data showing standard residuals from the OLS regression model. This figure shows the adjusted model identified in Table 4- 5...... 54

Figure 5-1. Overview of the geographic locations for the six selected stations...... 78

Figure 5-2. An image from Google Streetview captured by the researcher showing obstructed sidewalk connectivity due to a grocery store driveway. This obstruction occurs 372’ from a station entrance...... 80

Figure 5-3. An image from Google Streetview captured by the researcher showing obstructed sidewalk connectivity due to a private development...... 80

Figure 5-4. On the left, 2 street trees in Brownsville are shown. On the right, 3 street trees near Government Center are shown. Note the size difference and landscape difference among two intersections that scored similarly...... 81

Figure 5-5. Five policy recommendations for improving the built environment located near the selected stations...... 81

LIST OF ABBREVIATIONS

CBD Central Business District

DOT Department of Transportation

EIS Environmental Impact Statement

FDOR Florida Department of Revenue

FGDL Florida Geographic Data Library

FLM First-Last Mile

GIS Geographic Information System

GWR Geographically Weighted Regression

NTD National Transit Database

OLS Ordinary Least Squares

TAD Transit-Adjacent Development

TOD Transit-Oriented Development

UMTA Urban Mass Transit Administration

US United States

VIF Variance Inflation Factor

Abstract of Research Project Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master in Urban and Regional Planning

EMBRACING CAPITAL INVESTMENT: AN ANALYSIS OF DEVELOPMENT ORIENTED TOWARDS MIAMI’S METRORAIL

Joseph Dever

April 2021

Chair: Abhinav Alakshendra Cochair: Ruth Steiner Cochair: Aaron DeMayo Major: Urban and Regional Planning

Public transit ridership determines the public perception of the success of transit systems. When proposing capital improvements to create or expand these systems, the consideration of increasing ridership is frequently cited as a reason for funding these construction projects. Because system expansions can be costly, the importance of system design, construction and integration within the built environment can yield lessons for other systems. While key variables are typically considered for correlation among the built environment and ridership, this research will focus on a case study in

Miami, Florida to determine how Metrorail was integrated into the built environment.

This case study considers development within station catchment areas at quarter and half-mile intervals. This research considers how local communities have leveraged this transportation asset to encourage ridership or not. Finally, this case study examines parcel development and demographic factors within the station catchment areas. The research uses a mixed-method approach to provide insight into both qualitative and quantitative characteristics of the study area. First, a content analysis evaluating

newspaper articles will identify perceptions and expectations. Second, a survey examining some built environment elements will designate key neighborhood characteristics that determine how integrated stations are within their neighborhoods.

Third, an Ordinary Least Squares (OLS) regression will evaluate data from the US

Census, Miami-Dade Department of Transportation, and Florida Department of

Revenue to consider local characteristics that may impact ridership. The intent of these

methods is to fully evaluate how development decisions have impacted how transit

facilities serve neighborhoods and encourage ridership.

CHAPTER 1 1. INTRODUCTION

Rail transportation in the United States faces criticism for high construction costs, low ridership, low farebox recovery, and poor connectivity (Pickrell, 1989). These issues impact how rail transit is designed, planned, and operated, with consequences for communities long after construction is completed. Rail transit, more so than other transit services like shuttles or buses, requires significant capital expenditures prior to operations. Often, this capital expenditure is funded through Federal budget allocations, local sales taxes, or local property taxes with the promises of easing car congestion, improving commute times, and creating economic development activity along the rail corridor (Pickrell, 1989).

Due to political, design, and budget changes during the project delivery period, accomplishment of some of these objectives may be sacrificed for other criteria. As an example, the Los Angeles Metro system’s proposal to extend along Wilshire Avenue shows that alignments along dense population corridors are sometimes prevented for purely political reasons (Taylor et al, 2009). In this case, the proposed segment was re- aligned to a less-dense area because of political pushback. Through evaluating a specific system, political decisions like this example can be compared with other

decisions that have impacted how stations developed over time. These political

decisions can determine how well stations support walkability, accessibility, mobility,

and safety near stations (Taylor & Morris, 2015). This comparison also allows a review

of development patterns over time that may make the rail asset more appealing for

riders.

Complex relationships impact the outcomes of transit ridership, and ultimately the perceptions of both riders and non-riders for whether the transit service is effective or

worth the operational cost (Taylor & Morris, 2015). The regional political decision to

build a rail line, the municipal land use decisions to allow or encourage development in

proximity to this rail line, and the local commitment to develop the built environment that

comprises the station catchment area profoundly impact the experiences of riders and

the likelihood of a successful rail system. These successes are demonstrated by the

correlation between higher ridership in dense, walkable neighborhoods with interesting

destinations (Cervero, 1994 & 2000; Ryan & Frank, 2009).

Even though rail transit offers a transportation solution that can move many

riders quickly, this mode requires origins and destinations that encourage travel

between them and allow riders to access those places. In recent years, this accessibility

has been described as the First-Last Mile (FLM) problem, and it has implications for the

connectivity of all transportation modes (Lesh, 2013). Many FLM constraints are driven

by the built environment creating barriers for pedestrians or cyclists. Some of these

barriers may not just be physical constraints, such as the lack of sidewalks, but might

also be driven by the perception of safety or feasibility (Ewing and Clemente, 2013).

Despite these positive correlations associated with higher transit ridership and the more

recent discussions of the FLM problem, there is less public scrutiny in how local

communities support higher density land uses and improved built environment

landscapes.

This case study focuses on built environment and parcel development changes

near transit stations in Miami, FL to determine the relationships among these complex

variables. Because Miami’s system is more than thirty years old, an evaluation of outcomes can be performed by looking at the built environment, system ridership, and changes in the development pattern over time. This case study will seek to explore how the Miami Metrorail was constructed, how the built environment around Metrorail impacts accessibility, and whether density has been impacted along the rail corridor due to parcel development. These inquiries will explore relationships between the built environment, land use, and rail ridership, specifically considering barriers to mobility in a

FLM context and challenges for accessing activity centers in the Miami region.

CHAPTER 2 2. LITERATURE REVIEW

The literature related to variables that impact transit ridership has a long history with broad findings. Frequently, the outcomes do not yield clear answers transferable to transit systems in different cities. Local geography, development patterns, history, climate, political decisions, and local regulations have the potential to impact transit ridership in unique ways. Because of these unique characteristics, cities do not always succeed in attracting ridership for transit systems in the same ways. In larger cities, rail transit is frequently identified as an option for improving transit options as space becomes more limited or land becomes more valuable. Management of road congestion, parking, and, according to Vuchic (2007), allocation of “proper roles to different transportation modes” become essential for managing urban streets.

According to Vuchic (2007), the sizes of cities determine the types of public transportation modes that can be supported by the population. These designations approximate demand for each type of mode based on distance, congestion levels, network complexity, and general cost of providing the service. Based on these approximate thresholds, separated guideway services become appropriate once cities exceed 500,000 people, though this threshold may differ depending on the spatial structure of the city (Vuchic, 2007).

While the separated guideway system is identified as appropriate, the specific form, or mode, is not stipulated. Typically, the greatest difference between modes occurs due to the projected capacity of the system, with rail technologies typically carrying more passengers per hour during peak periods (Pickrell, 1989; Stutsman,

2002; Vuchic, 2007). Beyond capacity, other objectives for choosing certain modes

could include criteria related to reduced traffic congestion, improved public transportation, better accessibility to the city center, better accessibility to other activity center, pollution reduction, and economic development (Mackett & Edwards, 1997).

Ridership

City size, spatial qualities, and mode selection are important because it can best align system service to ridership needs. Despite this ideal, Don Pickrell (1989 and 1992)

identified the political goals of constructing a rail line as being more important than

accurate forecasts of ridership or cost. His conclusions held true for all rail projects

opened in the 1970’s and 1980’s, including Atlanta, Baltimore, Buffalo, Miami,

Pittsburgh, Portland, Sacramento, and Washington DC. For each city, final costs

exceeded budgeted costs. Actual ridership lagged forecasted ridership (Pickrell, 1992).

The input variables of the models were not considered explanatory for the variances in

projected and actual ridership. This lack of explanatory variables suggests that the

entire modeling structure or the interpretation of that data caused the variance (Pickrell,

1989).

The transit goals and the political decisions that influence how projects are delivered determine the layout of rail transit infrastructure. The system routing for rail projects creates a fixed-guideway system with considerable up-front infrastructure costs. Intuitively, the investment in specific station locations has significant impacts on how extensively the system is used. However, in practice, transit investments do not have a clear or predictable relationship with land use (Giuliano & Hanson, 2017).

The complexity of variables that impact the interaction of density, walkability and the built environment is not fully understood and the amount of data is difficult to compare across cities. Ryan and Frank (2009) found that walkability was a statistically

significant variable impacting transit ridership in San Diego. This research focused on

bus transit centers for ridership data. While this study ignored the origin and destination

pairings, other authors have found that these pairings have also been statistically

significant. Specifically, Cervero (1994) found that workers are much more likely to use

rail transit if their origin and destination are both in close proximity to rail.

More recently, studies have tried to model travel decisions using various

mathematical methods, such as Path Analysis, Direct Ridership Models (DRM), and a

Gradient Boosting Regression Tree (GBRT) algorithm (Gan et al., 2020, Kuby et al.,

2004; Kain & Liu, 1999). The GBRT algorithm analysis provides a compelling look at the

Chinese city of Nanjing because the metro system provided station pairing data to the

researchers (Gan et al, 2020). The researchers were able to analyze data by time of

day and subsequently compare characteristics of origins and destinations. The research

supported previous findings by Kuby et al (2004) and others (Parsons Brinckerhoff

Quade & Douglas, Inc., 1996) showing density as a primary contributor to transit usage

(Gan, et al., 2020).

Gan (2020) identified the importance of bus feeder systems at station origins in

the morning and station origins in the evening. This finding supports the “pearls on a

necklace” concept of development provided by Cervero (1994, 2000, 2007). This

concept aligns the idea of nodes with mixed-uses and higher density that can be used

as a conduit and collector for a linear transportation system in either a radial or webbed

form. The best practice according to this theory is for rapid transit to operate along a

linear path, with each node collecting passengers and providing unique qualities for the

destination, such as jobs, parks, or special use districts (Cervero, 2000). Even though

Filion (2001) disagrees with a nodal approach, the researcher does argue for linear

corridors of medium- and high-density development in a format like Cervero’s

recommendations across multiple articles.

Beyond supportive bus service levels, the GBRT pairing data showed transfer

times and route distance to be significant contributors to ridership (Gan, et al, 2020).

These variables are included because both can be considered proxies for rider utility.

This concept is supported by the idea that shorter transfer times between other rail lines

and bus routes will reduce overall travel times. This reduction in time cost is a de facto

increase in utility for the rider (Ben-Akiva & Morikawa 2002). A similar relationship exists

for route distance because higher automotive congestion, car ownership costs and

parking costs reduce the utility of driving to center cities (Shoup, 2005; Cervero &

Morikawa, 2010; Pickrell, 1992; Ryan & Frank, 2009).

Various research yields conflicting ideas about the relationship between land use

and transit ridership. Kim et al (2007) find that commercial and industrial properties are

associated with higher transit usage, while Gan, et al (2020) find that these properties

are not significant contributors. Though these compare different cities in different

countries, the literature conflicts even on the Central Business District (CBD) being a

major contributor, with Kuby et al (2004) finding that the CBD was not statistically

significant with number of boardings and Cervero & Murakami (2010) finding that CBD’s

are major contributors to boardings. Gan, et al. (2020) found that Nanjing’s CBD was a

major contributor, as could be seen from their charts showing rush hour data and the

destination land use qualities. Gan, et al (2020) found that land use did not contribute

significantly to transit ridership, but that a lack of land use mixes did contribute to a reduction in transit ridership for specific stations.

Transit-Oriented & Transit Adjacent Development

Land use can have an impact on station ridership due to numerous variables.

Efforts to categorize rail transit stations based on combinations of density, land uses,

and walkability characteristics have established typologies of station environments.

These typologies include Transit-Oriented Development (TOD), Transit-Adjacent

Development (TAD), and Hybrid Development (Renne & Ewing, 2013). This research

created a comprehensive definition and parametric analysis of all stations in the United

States using multiple criteria. While other researchers had identified some criteria for

stations, there was little consensus about what qualified as TOD (Cervero et al, 2004).

Peter Calthorpe (1993) is credited for establishing the term as a concept explaining how

a walkable, mixed-use environment could encourage non-motorized or transit trips.

TOD frequently has numerous connotations and meanings that are not

consistently applied across places (Renne, 2019). The work by Ewing and Renne

(2013) seeks to break down the concept of TOD into meaningful components that signal

the strength of the built environment and land use patterns in proximity to transit

stations. The concept of TOD, TAD and Hybrid station areas is based on the principles

of Density, Distance, and Diversity and key measurement characteristics for each of

those concepts (Renne & Ewing, 2013). Based on these performance indicators, TOD provides the highest measurement of locations that might support transit, with higher densities, greater land use mixes, and proximity of many activities. Hybrid areas provide moderate densities with some land use mix and general proximity of activities. TAD areas exhibit lower densities, lower land use mixes, and less proximity of activities.

The designation for TOD, TAD, and Hybrid station areas is important because it can direct development and built environment improvements to specific areas that might best benefit walking, cycling, and transit modes. Renne (2019) considers that the idea of TOD was important from a branding perspective when Cervero et al. (2004) and

Calthorpe (1993) were creating a vision of improving the quality and focus around transit stations. While Cervero and Calthorpe both built on the ideas of creating

walkable environments that were in proximity to transit, these density and connectivity

goals were evident in Environmental Impact Statements (EIS) for Urban Mass

Transportation Administration (UMTA) projects in the 1970’s and 1980’s (UMTA, 1980).

Despite the lofty goals identified in the development of many newer transit systems in

places like Atlanta, Dallas, Miami, and Phoenix, there are many transit stations across

the United States that still lack adequate development that would justify the high cost of

transit infrastructure (Renne et al., 2016).

Beyond the difficulty in developing areas supportive of walking and transit use

near rail stations, measuring the success of those TOD efforts is also difficult (Hale,

2014). Specifically, Hale (2014) identifies the lack of metrics for quantifying successful

TOD’s and the difficulty with incorporating TAD benchmarks inappropriately. Hale

(2014) suggests that a 50% rule be utilized for determining success of a TOD, a metric

that would indicate at least half of all trips would be made through a sustainable mode.

This focus on mode-share aligns with Calthorpe’s vision (1993) of creating sustained

pockets of activity where mobility could be concentrated. As Renne and Ewing (2013)

note, diverse nodes with multiple activity centers are important because only 22% of all

travel trips involve commuting for work. Many remaining trips therefore involve

shopping, leisure, or other personal needs that may or may not be obtained in proximity to one’s house. The mode split of these other choices can have significant impacts on traffic, congestion, and personal automobile use.

Measuring the Built Environment

Improving accessibility for public transit riders requires understanding the complexity of land use decisions and the integration of those land use decisions into the urban fabric. These needs are demonstrated by the differences between TOD, TAD, and Hybrid station catchment areas, including the built environment characteristics that comprise their differences. As highlighted, the idea of connectivity can also be considered an FLM problem, which highlights the need for efficient and safe

transportation choices for people to travel to and from public transit (Boarnet et al,

2017). This idea has significant ramifications for equity planning and ridership because of the opportunities that exist to expand the catchment areas of transit stations for both potential riders and destinations of employment or special interest. While much of the literature focuses on bike or scooter sharing that would increase range for a user, many of these services still require infrastructure and built environment qualities that make the services accessible and safe for a variety of users (Boarnet et al, 2017; Venter, 2020;

Zuo et al, 2020).

These concepts for enhancing mode accessibility and quality of FLM connectivity, while gaining popularity in recent years, encouraged the development of the Maryland Inventory of Urban Design Qualities (MIUDQ) by Otto Clemente and Reid

Ewing (2013). The MIUDQ developed built environment surveys that analyzed criteria that are related to the walkability of a place. These five criteria are imageability, enclosure, human scale, transparency and complexity (Ewing & Clemente, 2013). The

MIUDQ uses a physical survey of the streetscape to assess qualities such as sightlines, sidewalk widths, number of pedestrians, transparency of storefronts, variability in structure age and type, as well as other qualities that impact a user’s perception of the space (Ewing & Clemente, 2013).

Ewing and Clemente developed the survey by identifying features through a panel of urbanists to determine relevant qualities that impacted their ranking and rating of a standardized set of urban images. These qualities have been tested in two large surveys to further validate the reliability of variables. Because physical surveys can be considered intrusive by some residents (Caughy et al, 2001), Ewing and Clemente sought to validate digital data collection against the Purciel (2009) survey. Digital information was correlated with imageability, transparency and complexity, though less effective for measuring enclosure and human scale. Several of these criteria, such as street furniture, building colors, and long-sight lines, could not be effectively measured at that time due to technological constraints (Ewing & Clemente, 2013). Improvements to Google Earth’s Streetview may improve the ability to evaluate some of these criteria at this time.

The MIUDQ criteria are important because they align closely with efforts to solve

the FLM problem, particularly according to Venter (2020), who identifies personal

security, personal safety, comfort of waiting areas, sidewalk comfort, and ease of

access as broad categories for improving accessibility. While the naming conventions

differ from those in the MIUDQ, the functional outcome is that these broad categories

are aligned to the MIUDQ built environment characteristics because they focus on

comfort, safety, and quality of traveling to or from transit services.

Summary

Rail transit systems constructed in the 1970’s and 1980’s faced numerous criticisms for cost overruns and low ridership. Despite the criticisms, the constructed systems were completed as originally intended to serve their communities. A single evaluation of ridership does not explain why a system has experienced lower ridership than forecasted and it also does not explain what the highest possible ridership might be. Those explanations require better understanding of local geography, land use, and the travel patterns of residents. By evaluating certain elements, neighborhoods should be developed in ways that support walkability and FLM goals. Frequently, these goals can be achieved through designing neighborhoods with newer development typologies like Transit Oriented Development. In cases where Hybrid Development or Transit

Adjacent Development exists, other design improvements may be required to enhance transit usage. While some literature has evaluated station pairings to clearly show travel patterns among places, this data is more difficult to obtain in the United States for privacy reasons. Until pairing data can more clearly show desired start and finish points, improving connectivity to and between transit stations is essential for developing an effective network of transit solutions.

CHAPTER 3 3. METHODOLOGY

This research performs a mixed-method approach to understand the impact of station locations, certain built environment characteristics and parcel development on rail transit ridership. Performing this research requires both a qualitative and

quantitative analysis of available documents and data. To ensure an accurate

understanding of these documents and data within the context of the chosen city, this

research will focus on performing a content analysis on the history of a rail transit

system, a quantitative analysis on built environment characteristics, and a quantitative

analysis on key demographic and parcel-level data. This research is significant because

rail transit ridership has been stagnating nationally and greater understanding of

cultivating successful transit systems is needed. Considering rail transit as part of a

larger, connected transportation system could also better leverage existing

infrastructure to improve accessibility for people using newer FLM mobility solutions,

like scooters and bikeshare. To allow for this improved connectivity, however, land use

decisions and built environment improvements need to be evaluated based on quality

research that identifies variables that have the greatest impact.

Selecting a City

The City of Miami, Florida was selected for the study area for four distinct

reasons. First, the original Metrorail line is the only one that has been constructed.

Second, the Miami Metrorail was constructed more than thirty years ago. Third, Miami

does not have a long history of large public transit infrastructure as seen in older cities

like New York or Philadelphia. Fourth, Miami’s population has continued to grow over

the past several decades. These four criteria make Miami a compelling case study

because of its relevance to many other American cities.

Studying a single rail service line simplifies ridership analysis because all

variations in ridership must be isolated along that line. Miami expanded its system by

adding 3 stations in 2012, but ridership reports can isolate and control for usage by

station in years after this expansion. Tri-Rail commuter rail operates service in

Metropolitan Miami, but this service does not provide a direct connection to downtown,

requiring a transfer at Tri-Rail/Metrorail Station. The singular nature of the Miami

Metrorail system thus remains strong because its ridership is not diffused.

The age of the system is an important consideration because it takes time for

development to occur in proximity of the train stations. A period of development of at

least 30 years is desirable. This span of time allows consideration for how development,

redevelopment, and potential building abandonment occurs. Developments originally

spawned by the construction of rail facilities will have had a generation to assess the

viability of the development, perhaps encouraging similar, new developments, or

perhaps in discouraging new investments. Miami’s system is appropriate because it is

37 years old, entering operations in 1984. Miami’s system can be evaluated within

context of the Pickrell report data (1989), which provided an early analysis of ridership

trends.

Since so many Southern and Western cities have grown substantially during the

age of the automobile, these cities define more recent experiences of most other United

States cities. Miami does not have a similar transit history to cities in the Northeast like

Boston or New York. The development of the Metrorail was the first effort to develop

modern rail transit in the city. The consideration of Miami’s experience in developing a transit system where none had existed at that scale is important. Specifically, land use and built environment outcomes can be evaluated as successful or unsuccessful in

cultivating transit ridership.

Finally, the Miami region was selected because of high population growth over time. Population growth suggests that development pressure will also be high within at

least some neighborhoods. This growth allows for consideration of where development

occurs, what proximity to transit exists, and how ridership is impacted. Further, in some

metropolitan areas, declining population could have unexpected impacts that may

negatively impact ridership models or otherwise distort neighborhoods near rail facilities.

Within the context of Miami, a growing population and stagnating ridership might

suggest certain deficiencies in the system. The objective to evaluate a city in its success

in developing a transit culture is intended to demonstrate lessons learned for other cities

that may lack such a transit culture, a supportive built environment, and a responsible

tax base leveraging existing transportation assets.

Qualitative Analysis: Content

This case study relies on a mixed-method approach for analyzing ridership on the Miami Metrorail. The qualitative approach requires obtaining reports and newspaper articles that discuss the design and construction of the Miami Metrorail. Within these documents, the research will seek to understand what alternatives were proposed, how the selected alternative was chosen, and what objectives that alternative was intended to satisfy. Beyond these general concepts, the researcher will identify why specific configurations of the route were chosen, how ridership and demand models were developed, and what expectations existed for growth along the corridor. These

evaluations specifically consider the built environment and land use patterns in proximity of certain stations. These documents, while qualitative in nature, will explain the vision for the system and the decisions rendered for promoting or preventing

development in certain areas.

To complement the official documents outlining project objectives, a content

analysis will be performed to review articles published about the Miami Metrorail

between 1976 and 1989 to understand how local residents received information about

the system. The content analysis will identify key subjects discussed in each article and

categorize the article as positive, negative or neutral in support of the Metrorail or the

Metrorail’s operations. The content analysis can show how reporting might have

impacted public perception, as well as consider how public input may have shaped

article content.

To identify key subjects, the researcher will read newspaper articles and record

the primary subject of the article, as well as any secondary or tertiary subjects. The

intent of this recording is to highlight key narratives about Miami Metrorail. The articles will be identified using Newsbank to search for articles that include “Miami Metrorail” in the lead paragraph. Articles selected were in English, excluded public notices, and excluded articles that ran in multiple editions. The content analysis focuses on system design, construction, and operation and generally excludes articles where the Metrorail is cited, but is not the primary focus of the article. By way of example, articles discussing road closures in proximity to Metrorail stations were not included. Because of

the date range from 1976 through 1989, the Miami Herald had the largest number of

archived articles in Newsbank.

Articles were ranked based on the presentation of information to the reader.

While subjective in nature, articles were ranked based on how a reader would perceive

the information in the article. If an article presents both positive and negative

perspectives, then it will typically be categorized as neutral. Articles where strictly

factual information is presented, such as ridership counts, will typically be categorized as neutral. If information presented is skewed more positively or more negatively, then

the article will be categorized accordingly. If factual information is presented but will

likely be perceived as positive or negative, such as safety concerns about the system,

then the article will be classified accordingly. The ranking of articles is not intended to provide a census of all articles during this time, but rather a barometer for how the system was covered and the range of topics discussed during the early period of the system.

The qualitative assessment will utilize government documents, early research performed by Pickrell (1989, 1991), articles from the Miami Herald, articles from smaller newspapers where available, and archived project documentation. The qualitative assessment, and the content analysis in particular, will demonstrate how the system was designed, received, and constructed within the city. The qualitative assessment will seek to answer the following questions:

• Why was the system constructed?

• What were the economic development goals of the Miami Metrorail?

• How did neighborhoods within the study area support or hinder development near stations?

• Was the new rail system embraced or opposed by residents?

Quantitative Approach: The Built Environment

Where the qualitative analysis evaluates the history and purpose of Miami’s

Metrorail, a Built Environment survey will evaluate the neighborhoods around six

stations of the system. The survey considers the interconnectivity and appeal of station

catchment areas for non-motorized travel modes that may enhance ridership. The six

stations will be chosen based on ridership data collected between 2016 and 2019 and

reported by Miami Dade County. These data have been assembled from the monthly

ridership technical reports and cross-verified for alignment with the National Transit

Database (NTD). The six stations chosen are: Government Center, Dadeland South,

Brownsville, Santa Clara, University, and Overtown. Government Center and Dadeland

South were selected because they have the highest ridership totals throughout the period. Brownsville and Santa Clara were selected because they have the lowest ridership totals throughout the period. University and Overtown were selected because they are the only stations that did not experience noteworthy declines in ridership between 2016 and 2019. The full Miami station list is depicted in Table 3-1.

The built environment survey will be conducted using Google Streetview and

Google Earth to evaluate streets within a 0.25 mile radius of the selected stations. Using a Geographic Information System (GIS), the study area will be defined and a “walk” performed along each of the streets. During this analysis, various urban design characteristics will be identified and documented. The intent of this research is to understand the accessibility of the station for pedestrians and users of other non- motorized sources such as bicycles or scooters. Using the MIUDQ as a basis, several criteria have been selected that align to the categories of imageability, enclosure, human scale, and transparency. These criteria focus on connectivity. The checklist will

focus on intersections for each street within the radius of the transit station. This data, while quantitative in nature, is also intended to be a qualitative evaluation to determine the general connectivity of the neighborhood surrounding the rail station to determine

FLM accessibility. The built environment survey will evaluate the variables identified in

Table 3-2.

Quantitative Analysis – Demographic and Parcel-Level Data

Additional quantitative data will be collected from large data sets that describe demographic and parcel-level data. These datasets are published by the US Census

Bureau, the National Transit Database, Miami Dade County, and the State of Florida.

These sources of information will focus on demographic, economic, and ridership data to provide insight into different characteristics in the station catchment areas. These data will also be considered for the overall Metropolitan area. Understanding trends across locations and years yields two benefits in understanding how ridership is impacted by key characteristics. First, ridership can be analyzed to determine whether any of these variables have a statistically significant impact on ridership in Miami, FL over time. Second, statistically significant variables from earlier studies can be compared to variables that may be statistically significant in this research.

A combination of datasets will be compiled to analyze variables that may impact ridership on the Miami Metrorail. These datasets will provide data in GIS shapefile formats, which can be exported to Microsoft Excel for further analysis. The specific data that will be considered focuses on neighborhood-level characteristics that exist within the station catchment areas, as well as in the Metropolitan area as a whole. Unlike the built environment survey, this data can be more easily aggregated across large areas for comparative purposes. Because of the variety of data being collected, different data

ranges will be analyzed for different sources depending on availability of data. For the

United States Census, data will be considered at the block level from 1990, 2000, and

2010, with specific categories of interest including population, population density, percent of minorities, and age characteristics. The NTD will be used for ridership data

for the Miami Metrorail for 2001 through 2019. Additional ridership data and public

parking utilization data will be compiled at the station level from Miami Dade County’s

monthly reports from 2016-2019. The reports from Miami Dade are consolidated from

the monthly PDF to an Excel spreadsheet.

In addition to the spreadsheet data, several files can be evaluated using GIS.

These data will be reviewed using GIS shapefiles, with the opportunity to extract data

tables into a spreadsheet. These sources include Florida Department of Revenue

(FDOR) data from 2019, which will report parcel uses, vacancy rates, development

year, and valuations. The Florida Geographic Data Library (FGDL) maintains parcel

data since 2012 that also shows parcel use, zoning category, and date of sale

information that can be analyzed. Finally, the Florida Department of Transportation

(DOT) maintains two shapefiles that can be accessed through FGDL that show Public

Transit Station Facilities and Road Classifications. The latter shapefile also includes the

road type, speed limits, and average daily traffic counts for each road. The sources of

data are identified in Table 3-3, while the variables considered are listed in Table 3-4.

Statistical & GIS Analyses

To evaluate the variables identified in the previous sections, a statistical analysis

will be performed using the Ordinary Least Squares (OLS) method. This method

assumes a linear model to predict an outcome based on certain input variables, also

known as explanatory variables. The input variables will align with the criteria identified

in the previous sections to determine how they impact ridership, the dependent variable in this case. The null hypothesis for this statistical method assumes that the chosen

variables will have no impact on ridership. For the OLS, a spatial join will be performed

between the Census Block shapefile and the FDOR Shapefile for the preferred year of

each data set.

As an alternative, a Geographically Weighted Regression (GWR) Model will also

be used to perform a statistical analysis. This model will require the same dependent

variables, but will smooth the statistical model to address distance from the station

center point. This modeling technique is an extension of OLS and is appropriate for

spatial analysis. Since this research considers spatial data within a radius of rail transit

stations, a GWR model is appropriate if data is spatially defined. This GWR analysis

can be performed within ArcGIS, as can the OLS analysis. Similarly, the null hypothesis

remains that the explanatory variables have no impact on the dependent variable,

station ridership. Like the OLS, a spatial join will be performed between the Census

Block shapefile and the FDOR Shapefile for the preferred year of each data set.

Given the spatial components of this research, GIS will provide a useful tool for

analyzing changes over time and across locations. Six stations will be compared with

the mapping tool to show differences among the built environments, especially as it

pertains to road types. Dummy variables have been introduced for specific land use

typologies, including Industrial Use, Office Use, Mixed Use Developments, Multifamily

Residential, Single Family Home Residential, and Vacant Properties. The risk of multi-

collinearity will be evaluated due to variables that may have correlation among

themselves, such as number of renters and number of multi-family units, average

household size and percent of minority residents, population density and multi-family

housing, and other similar potential relationships. Managing multi-collinearity will consider the Koenker Statistic and the Variance Inflation Factor (VIF).

Built Environment Survey data will be entered into GIS where appropriate to evaluate differences across these stations. The quantitative data can also be presented in GIS to visually demonstrate how census blocks or parcel-level data vary across the stations or across the metropolitan area. This visualization provides deeper analysis

where population growth occurred in comparison to where transit services were

established. Other significant variables will be displayed in GIS to visually depict how

they differ across station locations as appropriate.

Summary

Based on these three data approaches, the intent is to provide a multi-part analysis that evaluates the quantitative data within a qualitative framework for each of the six station areas. Because research often focuses on the quantitative aspects of

Census or Origin/Destination pairings, qualitative characteristics of neighborhoods may be overlooked relative to how residents experience the transit system. Simliarly,

potential biases for stations may be ignored by simply considering data. This

methodology considers the context of qualitative elements for the rail line development

and the present changes to the built environment that make it appealing or dissuading

for transit use. This case study will provide a deeper analysis for six stations that have

significant variances in ridership and ridership trends. Demographic and economic data

will be helpful in comparing this research to the findings of other research as a

comparative effort in understanding the similarities and differences of the Miami rail

system to those of other metropolitan areas.

Table 3-1. This Table shows station boardings for each station in the Miami Metrorail system between 2016 and 2019. Selected stations for study highlighted. Calculations for percent change and rank orders for number of boardings and percentage change of boardings is also shown. Stations Station Boardings Calculation Rank Orders of Boardings/% Chg Stations 2016 2017 2018 2019 % Change Rank 2016 Rank 2019 Rank % Dadeland South 2,069,411 2,094,920 2,092,696 1,946,138 -5.96% 2 2 3 Brickell 1,937,900 1,822,347 1,886,867 1,760,387 -9.16% 3 3 4 Civic Center 1,689,096 1,564,876 1,546,657 1,516,450 -10.22% 5 5 5 Government Center 3,431,153 3,180,692 3,024,331 2,799,298 -18.42% 1 1 15 Dadeland North 1,894,000 1,696,112 1,721,199 1,597,159 -15.67% 4 4 11 Douglas Road 1,165,543 1,061,467 1,039,834 985,098 -15.48% 6 6 10 Overtown/Arena 587,257 555,831 587,738 584,137 -0.53% 11 10 2 University 572,968 541,649 573,256 616,278 7.56% 13 12 1 MIA Intern'l Airport 629,472 591,553 581,720 543,058 -13.73% 9 9 9 Allapattah 659,792 624,468 589,736 551,439 -16.42% 8 8 13 South Miami 1,026,337 933,059 892,746 818,992 -20.20% 7 7 17 Coconut Grove 588,230 516,659 528,562 481,817 -18.09% 10 14 14 Earlington Heights 582,962 553,798 525,964 473,926 -18.70% 12 11 16 Palmetto 446,625 422,721 412,374 391,351 -12.38% 17 17 7 Vizcaya 413,478 381,530 381,276 366,132 -11.45% 20 19 6 Culmer 445,077 430,415 398,307 374,850 -15.78% 18 16 12 Northside 563,111 518,941 478,460 427,293 -24.12% 14 13 21 Brownsville 297,252 314,661 281,935 257,854 -13.25% 22 22 8 Hialeah 530,125 459,335 423,971 371,865 -29.85% 15 15 23 Dr. MLK Jr. 450,978 414,315 370,088 335,810 -25.54% 16 18 22 Tri-Rail 423,427 352,900 364,471 332,926 -21.37% 19 21 19 Okeechobee 405,234 378,718 327,769 321,992 -20.54% 21 20 18 Santa Clara 285,901 269,593 252,516 218,835 -23.46% 23 23 20 Station boardings compiled from Miami-Dade County Monthly Technical Reports (2016-2019). Calculations and rank orders computed by Joseph Dever.

Table 3-2. This table shows Built Environment variables to be considered in this research. Variable Criteria Method Source Calculation Calculation by Number of intersections from Shapefile Researcher Researcher Create Count by Google Type of Intersection Shapefile Researcher Earth/Streetview Crossing type for Create Count by Google intersection Shapefile Researcher Earth/Streetview Number of signalized Create Count by Google intersections Shapefile Researcher Earth/Streetview Create Count by Google Points of Interest Shapefile Researcher Earth/Streetview Accessible courtyards, plazas, parks, and Create Count by Google gardens Shapefile Researcher Earth/Streetview Create Count by Google Number of street trees Shapefile Researcher Earth/Streetview

Table 3-3. This table shows data sources, file formats, and relevant information contained in the dataset. Source File Format & Available Data within Set Periods US Census Excel (1990-2010) Population, Demographics, Income, Mode Choice National Transit Excel (2001-2019) Ridership on Metrorail & Metromover Database Miami Dade PDF/Excel (2016- Ridership on Metrorail & Public Parking Data County 2019) by station and month Florida Geodatabase Parcel Use, Mill Rate, Valuation Department of (2019) Revenue FGDL Parcel Geodatabase Zoning Type, Parcel Use, Address Data (2019) Florida Shapefiles (2020) Road Classifications, Speed Limits, Road Department of Type Transportation Florida Shapefile (2019) Location of Transit Stations Department of Transportation

Table 3-4. This table shows demographic and parcel variables to be considered in this research. Variable Field Variable Name Source Description Total Residents in 2010 2010 US Census POP_2010 Total Residents in 1990 1990 US Census POP_1990 Density Under Age 18 1990, 2010 US Census DENUNDER18 Minority Residents 1990, 2010 US Census MINORITY Percentage of Minority 1990, 2010 US Census PCT_MNRTY Residents Residents Under Age 18 1990, 2010 US Census POPUNDER18 Residents over Age 65 1990, 2010 US Census AGE_65_UP Average Household Size 1990, 2010 US Census AVE_HH_SZ Number of Renter-Occupied 1990, 2010 US Census RENTER Housing Units. Florida Dept. of Revenue Just Value of Property JV (2019) Florida Dept. of Revenue Improvement Value IMPROVVAL (2019) Florida Dept. of Revenue Land Value LNDVAL (2019) Value Per Acre Calculated VAL_ACRE Florida Dept. of Revenue Effective Year Built EFFYRBLT (2019) Florida Dept. of Revenue Number of Acres of Parcel ACRES_1 (2019) Number of Households 1990, 2010 US Census HOUSEHOLDS Florida Dept. of Revenue Number of Residential Units NORESUNITS (2019) Dummy Variable - Single Calculated (FDOR '19) DV_SFH Family Houses Dummy Variable - Office Calculated (FDOR '19) DV_OFFICESPACE Space Dummy Variable - Multifamily Calculated (FDOR '19) DV_MULTIFAMILY Residences Dummy Variable - Industrial Calculated (FDOR '19) DV_IND Use Dummy Variable - Mixed Use Calculated (FDOR '19) DV_MIXEDUSE Dummy Variable - Vacant Florida Dept. of Revenue DV_VACANT Parcel (2019)

CHAPTER 4 4. DATA

Content Analysis

The content analysis considered how reporting on Miami Metrorail’s system might have influenced readers or reflected how readers felt at that time. 209 articles were reviewed between 1976 and 1989 to evaluate the content of articles prior to the opening of the system until five years after the system was opened. The data considered how changes in opinion might occur over time, as well as evaluated the categories and sub-categories mentioned most frequently. The sub-categories were documented based on the importance of topics discussed in each article. While many categories might be applicable, the purpose of the categorization was to identify the primary topics for each article. In some cases, only one topic was applicable. In other cases, two or three subjects might be addressed. The highest number of subjects discussed at length was four, with that article being a lengthy expose on the Metrorail system. A percentage breakdown of summary categories is shown in Figure 4-1.

Of the 209 articles reviewed, 96 were considered to be neutral, while 76 showed a negative bent and 37 a positive bent. Within the 209 articles, 345 sub-categories were identified as the primary, secondary, or tertiary subjects of different articles. The 209 reviews included articles reporting the news, as well as op-ed pieces. 192 articles were reviewed from the Miami Herald, which was the largest newspaper in Miami during the period of this content analysis. Some articles were reviewed from sources outside South

Florida to better understand alternative perspectives of the Miami Metrorail system. All sources with more than one article reviewed had a negative disposition towards the

Metrorail.

Sub-categories were consolidated into broader summary categories to better understand how often certain subjects appeared, as well as identify differences in the frequency between summary categories. Table 4-1 shows the summary categories and frequencies for all reviews. Notably, the majority of categories had a negative outlook towards the Metrorail. Only Design, which includes subjects like concept planning, station design and future service possibilities, and Development, which includes development proposals, redevelopment applications and other land use changes, had more articles considered positive than negative. Both of those summary categories had the majority of their articles classified as neutral.

The remaining 7 categories had more negative articles classified than positive

articles. Categories concerning construction, operations, and social impact showed a

particular negative skew. The Construction category included articles about cost, bid

packages, and schedule. The operations category included articles discussing service

operations, planned service operations, as well as other miscellaneous articles about

the operational activities of the system. Finally, the social impact category included

subjects like community activism, public meeting participation, and social engagement

for the Miami Metrorail. These three categories had the majority of their total number of

articles classified as negative, with two-thirds of the Construction category’s articles

classified as negative. Given the cost and schedule delays for opening the system, the

construction category being negative was not surprising. The Operations category being

so skewed negative was surprising, though the system experienced a rail car crash

prior to the system opening and documented numerous complaints because of late-

night service, special event service, and the eventual service connection to Tri-Rail. The

Funding category was slightly negative, while the categories for mobility and Other were close to neutral, with less than a 2-topic differential.

The coverage for mobility concerns showed balance. Mobility associated with

personal vehicles included subjects like traffic, parking, and road expansion. Non-

automotive mobility included bus feeder discussions, walkability, FLM problems, and

support of bicycle services at stations. These articles tended to balance the

conversation between topics such as congestion potentially being alleviated by

Metrorail, or that parking would be required to feed the system. Similarly, positive

subjects pertaining to bus expansion were tempered with negative associations related

to cost concerns. In many cases, the balance between public transit and vehicle use

brought some of this balance, as did a focus on creating alternative travel modes that

would not require as much gasoline because of the high fluctuations in gas prices

during this time (Zaldivar, 1984).

The sub-categories include more detailed topics mentioned in various articles.

Within these sub-categories, only three topics had a positive skew of more than 3

artricles, while 8 topics had a negative skew of more than 3 articles. Concept Design

had the greatest positive skew, with 6 more articles being positive than negative. Cost

had the greatest negative skew, with 12 more articles being negative than positive.

Construction, operations, and ridership all had 11 more articles being negative than

positive. The topic of taxes had a positive skew, primarily driven by articles that

supported tax increases to fully-fund the project for capital completion and operational

services. Additional funding requirements were often conflated with cost-overruns,

meaning there was not a consistent approach by writers in connecting the complicated

mechanisms for how these topics were related. The issue of walkability had a negative

skew, with the negative articles focusing on reduced sidewalk coverage due to cost

overruns, publicity surrounding dangerous crosswalks due to construction, reducing

funding for bicycle amenities, and poor maintenance of walkways and vegetation under

Metrorail tracks. Density changes, land use, growth and traffic issues received balanced

coverage.

The content analysis spanned the years 1976 through 1989, with the Metrorail

system becoming partially operational in 1984. The bar charts show clearly that the

number of neutral articles exceeded positive or negative articles, with a peak of

reporting in the years preceding operations. Negative article mentions were consistent

by year, but tapered in frequency after 1986. Positive sub-category mentions exhibited a

decline over time, with a peak of positive articles in 1979. This year marked funding

milestones and construction initiation, as well as numerous articles about the

anticipation of the system. The longitudinal analysis shows that negative references

exceeded positive references in every year except 1979 and 1987. Consistent for

positive, negative, and neutral articles, the number of articles and the sub-categories

mentioned show a decline after the system became operational in 1984. The outliers

were negative articles in 1985 and 1986, which frequently discussed cost overruns,

ridership disappointments, and safety concerns of the system.

Demographic and Parcel Data

The first model to examine explanatory variables to determine ridership used the

Ordinary Least Squares (OLS) model to examine twelve variables related to Census

and Parcel data within ¼ mile of the six stations selected. The two datasets that

contributed data, the 2010 Census and the 2019 Department of Revenue parcels,

provided variables that show strong correlation within the ¼ mile study area of the six stations selected. These two datasets were combined in ArcGIS Pro using a spatial join.

The station identification and 2019 boardings at each station were added to the attributes table of the spatial join output file. The results from the GIS OLS report are shown in Table 4-3. Land uses that lacked any use category and intentionally un-zoned parcels were excluded from this analysis.

Within the model, 9 of the 12 explanatory variables were considered statistically significant. Seven of the variables were defined by the Census and FDOR datasets, while two of the explanatory variables were binary dummy variables that explored the

relationship among specific land uses. Prior to running the OLS model, several

variables were excluded due to multi-collinearity and missing data. Age categories from

18-21 and 22-19 were excluded due to multi-collinearity with renters. Median Age,

Household Size, and Average Family Size were also excluded due to multi-collinearity.

The dummy variable for Residential housing was excluded due to multi-collinearity errors with the single-family housing variable and general residential categories. The

Just Value and Effective Year Built variables were included after controlling for these

other multi-collinearity constraints. Both Just Value and Effective Year Built variables

had higher numbers of null values caused by vacant and city-owned properties where values were not assessed or where no structure was present to identify a “Year Built”.

Parcels were included if any part of the parcel was intersected by the ¼ mile radius.

Of the 9 variables considered statistically significant, five had a positive

correlation to ridership. The sources of explanatory variables with a positive correlation

were mixed and included population types, parcel land values, and land use category.

The explanatory variables with a positive correlation were number of residents over age

65, number of renters, just value, effective year built, and office space & professional services land use. All five of these variables were statistically significant at the 95% confidence level, with all of them being statistically significant at the 99% confidence level. The R-Squared value for this model was 0.70, a value exhibiting a good fit. The

Robust Probability was used to determine significance because of the Koenker Statistic having a low probability. The VIF value for all categories was below the target maximum threshold of 7.5.

Of the 9 variables considered statistically significant, four had a negative impact

on ridership. The sources of the explanatory variables with a negative correlation were

also mixed across data sources. The variables with a negative correlation were the

percentage of minority residents, average household size, parcel size (in acres), and an

industrial land use classification. All four of these variables were statistically significant at the 95% confidence level. The values with the highest statistical significance were average household size, parcel size (in acres), and an industrial land use classification, with all being statistically significant beyond the 99% confidence level. The Robust

Probabilities value was used for each of the explanatory variables because the Koenker

Statistic was significant. All VIF values were within the maximum target threshold of 7.5.

The second model replicates the first, but expands the study area to a ½ mile

radius around each station. When using the same variables for the half-mile radius input, an error message was received for all variables due to high multi-collinearity. To correct the model, some adjustments were made to the original model. The Actual Year

Built and the Effective Year Built variables were removed due to high number of parcels

missing data and station overlap at the half-mile radius band for Government Center and Overtown stations. Parcels were included if any part of the parcel was intersected by the ¼ mile radius. Due to the overlap between these two stations, parcels were

selected for the closest station to the parcel centroid.

The revised model considers 11 explanatory variables for 2019 ridership on

Miami’s Metrorail. Four of the eleven have positive correlations that are considered

statistically significant, while seven of the eleven explanatory variables have negative

correlations that are statistically significant. Explanatory variables with a positive correlation to ridership were number of minority residents, number of residents over age

65, number of renters, and office space land use classification. These variables had a

statistically significant correlation at the 99% confidence level. The R-Squared value for

this revised model was 0.44, an acceptable fit for the model. As seen previously, the

Koenker Statistic probability necessitated the statistical test to compare against the

Robust Probabilities statistic. VIF values were within acceptable amounts for all

variables except the number of minority residents and number of renters, which were

17.83 and 15.57, respectively. The complete results are presented in Table 4-4.

Of the 11 variables considered statistically significant, seven had a negative

impact on ridership. All seven of these variables were statistically significant at the 99%

confidence level. The explanatory variables were density of residents under age 18,

average household size, parcel size (in acres), single family home land use

classification, industrial land use classification, mixed use land classification, and vacant

land classification. The results are graphically depicted in Figure 4-4. As seen

previously, the Koenker Statistic probability necessitated the statistical test to compare

against the Robust Probabilities statistic. VIF values were within acceptable amounts, except as identified for number of minority residents and number of renters.

Built Environment Data

The parcel-level data provides demographic and land use information that

highlights statistically significant variables. Another area of interest within this study is to

compare ridership at the six rail station catchment areas to certain built environment

criteria that impact how pedestrians and motorists might perceive those areas. To

evaluate these data, an OLS Regression was performed for six built environment

criteria that included the following explanatory variables: intersection type, intersection

signalization, through-traffic allowance, number of street trees, number of parks, and

total number of intersections in the catchment area. A new feature class was created in

ArcGIS Pro that contained information at each intersection for the selected built

environment characteristics. An intersection was considered to include all public streets

that connected to the road. Driveways, including those for shopping centers, were

excluded. Because the sample size was too small, a GWR was not considered

appropriate for statistical analysis.

Of the six variables considered, only three were considered correlated to

ridership at a statistically significant level. All three had a positive correlation to

ridership. These variables included the number of signalized intersections, the total

number of intersections, and the number of street trees within the catchment area.

Despite the statistical significance, the R-Squared value for the model was only 0.21,

which suggests a weak model. Unlike the previous models, the Koenker Statistic was

not statistically significant so the Robust Probability was not required for use. The VIF

for each explanatory variable was below the maximum desired threshold of 7.5,

suggesting that risks of multi-collinearity and heteroskedasticity are low. The full results are presented in Table 4-5. A map of the data is displayed in Figure 4-5.

Newspaper articles by Summary Category

11% 8% 1% Automobile Mobility 11% Non-Automotive Mobility Construction 18% Design

13% Development Funding Operations 10% 8% Other Social Impact 20%

Figure 4-1. Percentage of summary categories identified in the 209 articles reviewed for the content analysis.

Table 4-1. Summary categories showing the number of sub-categories classified as Negative, Neutral, or Positive.

Grand Summary Category Negative Neutral Positive Total Mobility – Personal Vehicles 6 17 5 28 Mobility – Public or Active 10 20 8 38 Construction 30 9 6 45 Design 3 15 9 27 Development 13 41 16 70 Funding 13 15 6 34 Operations 33 19 11 63 Social Impact 20 15 1 36 Other 2 2 4 Grand Total 130 153 62 345 Categories with high negative skew highlighted in red, while categories with high positive skew highlighted in green.

Table 4-2. Summary Categories showing the number of sub-categories classified as Negative, Neutral, or Positive. Grand Sub-categories Negative Neutral Positive Total Community Engagement 12 10 1 23 Concept Design 3 15 9 27

Connectivity 3 10 7 20 Construction 13 6 2 21 Cost 16 3 4 23 Density 1 5 2 8 Equity 1 3 4 FLM 3 4 1 8 Funding 12 14 2 28 Growth 8 23 9 40 Land Use 3 9 3 15 Operations 19 11 8 38 Other 2 2 4 Parking 3 8 2 13 Redevelopment 1 4 2 7 Ridership 14 8 3 25 Safety 7 2 9 Schedule 1 1 Taxes 1 1 4 6 Traffic 3 9 3 15 Walkability 4 6 10 Grand Total 130 153 62 345 Categories with high negative skew highlighted in red, while categories with high positive skew highlighted in green.

45 1976 40 1977

35 1979 1980 30 1981 25 1982 1983 20 1984 category Mentions 15 1985 Sub - 10 1986 1987 5 1988 0 1989 Negative Neutral Positive

Figure 4-2. This chart shows the distribution of positive, negative, and neutral sub- categories mentioned by year.

Table 4-3. Results from Ordinary Least Squares Model at the ¼ mile radius with Land Value and Land Improvement Values removed. t- Variable Coefficient StdError Probability Robust_t Robust_Pr Statistic Intercept 1560417.7460 17795.9752 87.6837 0.000000* 31.3070 0.000000* DENUNDER18 -15183.1301 2699.8594 -5.6237 0.000000* -2.1708 0.029989* PCT_MNRTY -2580.9146 541.7466 -4.7641 0.000003* -2.6510 0.008053* AGE_65_UP 2830.9018 647.8752 4.3695 0.000016* 3.6397 0.000290* AVE_HH_SZ -233569.2513 15322.5150 -15.2435 0.000000* -8.7742 0.000000* RENTER 3897.5533 60.4597 64.4653 0.000000* 76.3014 0.000000* JV 0.0157 0.0012 12.6235 0.000000* 3.7307 0.000206* EFFYRBLT 5290.2953 524.5667 10.0851 0.000000* 5.7226 0.000000* ACRES_1 -138173.7538 4391.9096 -31.4610 0.000000* -4.9276 0.000002* DV_IND_USE -879353.3035 72780.3080 -12.0823 0.000000* -9.2244 0.000000* DV_OFFICESPAC 387989.6132 38380.5627 10.1090 0.000000* 5.6707 0.000000* E DV_MIXED_USE -69490.1053 129314.9220 -0.5374 0.5911 -0.2799 0.7796 DV_MULTIFAMIL -84133.2653 29503.8114 -2.8516 0.004378* -1.4868 0.1372 Y Multiple R-Squared]: 0.70261 Adjusted R-Squared: 0.701707 Joint F-Statistic: 785.528433 Prob(>F), (12,3990) degrees of freedom: 0.000000* Koenker (BP) Statistic: 2079.605394 Prob(>chi-squared), (11) degrees of freedom: 0.000000*

Figure 4-3. Figure shows each ¼ mile station catchment area with parcel data showing standard residuals from the OLS regression model.

Table 4-4. Results from Ordinary Least Squares model using a ½ mile radius and with Just Value and Effective Year Built variables removed. Coefficie t- Probabilit Robust_P Variable StdError Robust_t nt Statistic y r DENUNDER18 -6340.7078 1088.83985 -5.823361 0.000000* -6.664862 0.000000* MINORITY 455.88829 75.06558 6.0732 0.000000* 4.168614 0.000036* 2409.1746 AGE_65_UP 233.513357 10.317074 0.000000* 8.224802 0.000000* 7 - AVE_HH_SZ -308240.5 5926.62999 0.000000* -45.756401 0.000000* 52.009406 1641.9931 RENTER 100.210725 16.385403 0.000000* 11.165621 0.000000* 4 - ACRES_1 -41437.383 2046.7382 0.000000* -17.096093 0.000000* 20.245571 DV_SFH -642890.53 20561.2134 -31.26715 0.000000* -33.65739 0.000000* - DV_IND -1204057.2 48936.1284 0.000000* -21.663128 0.000000* 24.604669 368396.04 DV_OFFICE 22397.3243 16.448217 0.000000* 17.001882 0.000000* 4 DV_MIXEDUS -827354.67 91559.4024 -9.036261 0.000000* -7.902424 0.000000* E - DV_VACANT -659955.56 21996.413 0.000000* -25.506483 0.000000* 30.002872 Multiple R-Squared]: 0.70261 Adjusted R-Squared: 0.701707 Joint F-Statistic: 785.528433 Prob(>F), (12,3990) degrees of freedom: 0.000000* Koenker (BP) Statistic: 2079.605394 Prob(>chi-squared), (11) degrees of freedom: 0.000000*

Figure 4-4. Figure shows each ½ mile station catchment area with parcel data showing standard residuals from the OLS regression model. This figure shows the adjusted model identified in Table 4-4.

Table 4-5. Results showing Ordinary Least Squares Model using a ¼ mile radius to evaluate specific built environment criteria using a physical survey of each intersection in the station catchment areas. Variable Coefficient StdError t-Statistic Probability Intercept -880913.3295 625613.86 -1.408078 0.160842 INTERSECTION_TYPE 67586.4921 143705.43 0.470313 0.638711 SIGNAL 628616.8572 177154.7 3.548406 0.000505* THROUGH 854.110455 241211.3 0.003541 0.997178 TREES 39914.72253 9640.8977 4.140146 0.000058* INTEREST_POINTS -212366.8041 275444.77 -0.770996 0.441712 INTERSECTION_TTL 35963.9646 7651.9299 4.699986 0.000006* Multiple R-Squared]: 0.233229 Adjusted R-Squared: 0.20767 Joint F-Statistic: 9.125107 Prob(>F), (6,180) degrees of freedom: 0.000000* Koenker (BP) Statistic: 10.1528 Prob(>chi-squared), (11) degrees of freedom: 0.118363

CHAPTER 5 5. RESULTS & DISCUSSION

Station Identities & Context

This section will provide an overview of the different stations and provide relevant data and historical backgrounds for those neighborhoods. First, Brownsville Station is located approximately 4 miles Northwest of the central business district of Miami. The station is characterized by lower density than might be expected so close to a major city. Fenced parcels of land and single-family homes are common throughout the ¼ mile radius of the station. The station is located on NW 27th Avenue, which has a North-

South orientation. The elevated nature of the Metrorail, combined with a wide highway with a curbed-median and left-turn lanes throughout the length of the catchment area, creates a community that is split in half. Limited crosswalks and pedestrian shelters prevent improved pedestrian and cyclist connectivity between these two halves.

Furthermore, many developments along NW 27th Ave are set back from the road.

Street trees are infrequent throughout the area and not well maintained. According to

2010 Census data, 1964 residents lived within ¼ mile of the station and 87% of those residents were African American.

Historically, Brownsville was settled by African Americans after the I-95 and I-395 interstates were built through Overtown. More African American residents left Overtown for the Brownsville neighborhood following the 1980 riots in Overtown (Zaldivar, 1981).

During the construction of Metrorail, distrust among the community led to pushback against increased density, with many residents feeling that the number of apartments was already too high. Distrust sowed because of past experiences with the interstate system in Overtown and concern for gentrification and other changes to the community

led to residents fighting to protect against high density and transit-oriented development

(Baird, 1982). Low ridership has persisted at this station since the construction of the system. Brownsville had only 1 major residential development with more than 150 units constructed. It occurred in 2011 and consisted of 401 units (FDOR Data, 2019).

Santa Clara shares many similarities to Brownsville, but is located only 2 miles

Northwest of the CBD. Like Brownsville, the station catchment area is cleaved in two

halves, bisected by NW 12th Ave. Santa Clara has several large warehousing and

industrial land uses within ¼ mile of the station. Several city offices, including the Miami

Parks and Recreation Division and the Miami Solid Waste Division are located in close

proximity to the rail station. Both offices have large parking lots that front major roads

and fencing that reduces walkability and imageability. Of note, Jackson Memorial

Hospital is located in the Southeast quadrant of the catchment area, though the land is

predominantly parking lots and access roads to the hospital lack sidewalks in some

areas. Jackson Memorial Hospital is more centrally connected with Civic Center

Metrorail station, though some improvements to the fringe properties could enhance the

fabric of the community towards Santa Clara station. Overall, the station catchment area

suffers from long blocks, numerous driveway accesses, poor street tree coverage and

wide roads to cross.

Santa Clara, like Brownsville, has experienced a change in demographics. In

1990, approximately half of all residents were non-hispanic whites, with the the

remaining half split approximately equally between African American and Hispanic

residents. Since 1990, more Hispanic residents have moved into the neighborhood, and

now comprise 58% of the population, while the percentage of African Americans has

remained approximately the same. Low ridership has persisted at this station since the

construction of the system. Santa Clara had only 1 major residential development with

more than 150 units constructed. It occurred in 2003 and consisted of 412 units (FDOR

Data, 2019).

Residents of the area attempted to prevent gentrification and higher intensity

development during the construction of the transit system (Zaldivar, 1981). For a

neighborhood with such proximity to downtown, the number of residents is far lower

than would be expected and the number of vacant parcels is much higher than would be

expected in developing a transect of the urban geography. These vacant parcels

include parking lots and may be used for land banking in advance of new construction

occurring. The percentage of parcels developed since 1983 in the Santa Clara station

catchment area is the lowest of the six stations studied at 10% (FDOR Data, 2019).

University Station is located approximately 7 miles Southwest of the Miami CBD.

Between 2016 and 2019, ridership remained essentially flat, diverging from 21 other

stations that saw ridership decline by an average of 14%. During this time, attendance

at the University of Miami grew slightly (University of Miami, 2020). Notable new

construction during this ridership analysis includes the Lennar Foundation Medical

Center that opened across the street from University station in December 2016. That

facility employs 400 staff and sees over 1,000 patients per day (Cueto, 2016) Despite

new construction in the ¼ mile catchment area, the number of residents declined from

4848 to 3252 between 1990 and 2010 according to the US Census.

The station area is characterized by poor connectivity and the station is

surrounded by large parking lots for both transit users and students on campus.

Connectivity is considered poor because of the lower number of intersections and the meandering nature of many roads. Block lengths are longer as well. Like the other station areas, a major highway splits the station catchment area in two halves, with one half consisting primarily of the University of Miami’s campus and the other half including more strip mall development and single-family homes. A pedestrian bridge does cross

US Highway 1 to connect University Station with campus.

While the station primarily serves the University, there is a strong mix of uses and points of interest within the catchment area. University facilities, including academic buildings, sports facilities, and residences are in close proximity. There is a mix of stores, medical centers and office buildings close to the station. Finally, there are numerous residences, including a mix of single family and multifamily dwellings in a neighborhood with limited through traffic. While lacking sidewalks, there are numerous destinations accessible without the need for a car. The quality of street trees is strong, though the width of roads and number of surface parking lots diminishes some of the qualities that might reduce the urban heat island effect and improve the quality of a walk.

Further Southwest, Dadeland South occupies the southern terminus of the

Metrorail. This station is approximately 10 miles from the CBD. The area has consistently been one of the stations with the highest ridership, ranking second only to

Government Center since rail operations began in 1984. Large parking lots and a high number of points of interest have maintained Dadeland as a major activity center. Large condominium buildings, hotels, and offices provide additional mixed-use support to the activity center within the station catchment area. The neighborhood has a high number

of sidewalks, but frequently appears to have constructed its sidewalks as a novelty. In numerous instances, sidewalks end before they get to building entrances, have large street trees obstructing pedestrians on the sidewalk, and have auxiliary elevated connections between bridges so that people do not need to leave a building complex.

These are illustrated in Figure 5-2 and 5-3. South Dixie Highway is a large highway that

requires any pedestrians to cross 6-8 lanes of traffic to continue a trip on foot, though

pedestrian traffic is not well-supported since most businesses are auto-oriented and the

neighborhoods southeast of the station exhibit poor connectivity.

Unlike the three station catchment areas discussed previously, Dadeland South

has changed dramatically since the construction of Metrorail. Nearly 96% of the parcels

that now exist within the catchment area have been developed since 1984. Intense

development pressure was ultimately supported by local planners and developers in

what was once nicknamed “deadland”, though local homeowners organizations initially

attempted to slow or prevent development due to traffic concerns (Capuzzo, 1979;

Cottman, 1982). Traffic concerns existed 5 years prior to the completion of the Metrorail

station, primarily due to high demand for Miami’s southern suburbs as the city rapidly

grew. Unlike other stations, advertisements cited its desirable location and proximity to

Metrorail as engrained in the neighborhood’s ethos even if the connectivity and

accessibility of many destinations did not provide an easy means of travel (Herald Staff,

February 22, 1982). In 2010, more than 50% of residents were Hispanic, which

demonstrated a shift from 1990 when more than 50% of residents were non-Hispanic

whites.

The four stations presented are located outside of the Central Business District.

Two stations studied are located in the CBD, with the Southernmost parcels of

Overtown and the Northernmost parcels of Government Center overlapping at the ¼ mile radius. These two catchment areas have similar characteristics, but differ in some ways. The Overtown neighborhood has better traditional urbanism than the other stations, with better demarcation of sidewalks, wider sidewalks, more signalized intersections, and fewer long-distance building setbacks. The street trees in this area generally provided better shade coverage than seen at other station locations and did not inhibit sidewalk travel. Despite the pedestrian-friendly nature of many streets, the

Overtown area has a noticeable level of vacant parcels and surface parking lots that are frequently fenced. Several parcels had walls surrounding the property, reducing the quality of imageability for the area and eliminating pedestrian connections. Like the other stations, the Overtown station catchment area is surrounded by wide streets that have wide corners allowing for faster motorized vehicle turns. Between 2016 and 2019,

Overtown was the only station that experienced ridership growth.

The Overtown neighborhood has a proud African American heritage that was fractured by the Interstate Highway system. The riots of 1980 galvanized the decline for the neighborhood, leading to discussions for “redevelopment” (Zaldivar, 1981). These redevelopment conversations ranged across a variety of solutions, including the consideration of new slum clearance, the rebuilding of damaged neighborhoods from the riots, the incorporation of middle income housing units, and the development of large scale special projects. The Miami Metrorail provided a transit tool that could be used for any of these redevelopment ideas at a large scale, though remaining residents

of the neighborhood expressed doubt, fear, and concern over proposed changes to the neighborhood, particularly in the idea that wealth generation would bypass existing residents in favor of large developers (Baird, 1982). Given this backdrop of prior infrastructure projects that had decimated the neighborhood, reducing the population from 40,000 to approximately 10,000 in the early 1980’s, consensus for redevelopment projects was difficult to achieve. Inability to achieve consensus led to Federal grant opportunities being missed for acquiring parcels for redevelopment and ultimately inhibiting some developments (Herald Staff, February 21, 1988). Regardless, the proximity to the CBD encouraged redevelopment of 95% of its parcels (including multi- unit condominiums) since 1984.

Government Center station is located on the Western periphery of the CBD,

South of Overtown. Government Center, as its name suggests, is located in close proximity to numerous government complexes. This station has maintained the highest ridership of any station since the construction of the system (Bivins, 1984). Further,

Government Center is connected to the Metromover, which is one of the only operational people movers in the country. The Metromover circulates passengers in a loop of the downtown CBD and was expanded to include two legs that connect the neighborhoods of Brickell and Southern part of Edgewater. Government Center is also located in proximity to Miami Central Station, which provides Brightline high speed rail service. When considering bus connectivity and these other rail facilities, Government

Center has the highest accessibility to other passenger transportation systems.

The neighborhood around Government Center has a grid pattern of smaller

blocks. These blocks have quality street trees, appropriate crosswalks, and wider

sidewalks. Most buildings front the road and most do not have walls or facades that obstruct a pedestrian’s viewshed. Numerous parking garages exist so there are curb cuts with automobile traffic that can obstruct the sidewalk. Towards the Southern and

Western parts of the catchment area, this pattern is broken by large highways and numerous ramps. Underneath these highways, large surface parking lots create an unappealing environment for walking.

As part of the CBD, Metrorail leaders anticipated and encouraged higher development uses with increased density. Government Center has experienced an influx of residents since 1990, increasing by more than 1000 residents according to the

US Census data in 1990 and 2010. More than half of residents are Hispanic, while 27% are non-Hispanic whites. The number of jobs in this neighborhood also increased and redevelopment projects revitalized many parts of the station catchment area. Since

1984, nearly 77% of the parcels have been developed or redeveloped. Despite the high percentage of developed parcels since Metrorail began operations, the station catchment area has the oldest structures of the 6 stations studied, with the oldest structure dating to 1905. Several other parcels have structures developed in the early

1900’s that pre-date the oldest structures in any of the other catchment areas studied.

As such, Government Center exhibits strong historical qualities, while also embracing large-scale change. Similarly, the area exhibits strong walkable qualities in certain locations while also exhibiting weak walkable qualities in other locations.

The difficulty in classifying stations can be seen in the TOD, TAD, and Hybrid designations that were assigned by Renne and Ewing (2013) to each of the stations mentioned. Four stations were designated as TOD, while one each was identified as

Hybrid and TAD. Despite the rigor in assigning station designations for each station, the mixed histories and built environment qualities for each station area do not seem to compute with ridership in a meaningful way. The two highest-performing stations,

Dadeland South and Government Center, are considered TOD, but so are Overtown, which ranked 10th out of 23 stations in ridership, and Santa Clara, which ranked last in ridership. Brownsville ranked 22nd in ridership (second to last), but was considered a

Hybrid station with emerging TOD qualities. University was assigned a TAD designation, which indicates the lowest orientation towards transit development, and it

ranked 12th out of 23 stations. These designations are shown in Table 5-1.

Content Analysis

Looking holistically at the results from the content analysis, the more positive

categories appear to be those that are aspirational, such as concept design and future

development possibility. General categories with which people are familiar, such as

driving, walking, and riding the bus have a more balanced reporting. Those categories

where people feel that they are being misled or where there is a principal-agent problem

exhibit higher negative tendencies. Of particular note within this last statement, the

Funding category, which included sub-categories for Federal funding and increased

taxation, had a less negative outlook than the construction, operations, and safety

categories, even though operations occurred as planned following the late opening of

the system and safety issues were limited to a handful of negative, high-profile events.

These skewed outcomes suggest some bias in the reporting and perception of what

subjects are newsworthy and how those subjects are reported.

Expected negative articles, such as increasing taxes to fund the system’s

operations, were actually supportive of higher taxation in order to preserve the system’s

operational capacity. Tension seems to exist between significant cost overruns and the preference to maintain the system in a limited functional state. This point seems to confuse how capital and operational budgets are established through Federal, state and local funding mechanisms. At no point were the finer distinctions of tradeoffs seen in the articles being reviewed, though the Miami Herald did publish several editorials identifying preferences for certain sources. When considering capital cost overruns as part of a broader debate in funding the completion of the capital work or funding full operational capacity of the system, this single category would be skewed negative. The nuance between funding and taxes, however, creates a situation where tax financing was considered positive because of its implicit support of the system.

The sub-categories with the largest negative skew, construction, cost, operations, and community engagement repeatedly reported on the same problem over time. The construction delays and cost overruns were undeniably bad, as were the ridership forecasts that were hidden by UMTA and project leaders for two years.

However, the system that was constructed was the one that was promised to residents of Miami. Despite the late delivery of the project, there was no special event or mandate that necessitated the opening of the system at any given time. While the outcomes of the project may be criticized, there was scant coverage on whether the Metrorail system that was constructed did not fulfill its promise for Miami Dade County. Additional rail expansions may have better distributed riders, but the rail system that was promised, including the downtown people mover, was delivered.

Possibly more important, coverage on community engagement frequently highlighted negative interactions between the community and county officials at public

meetings. The general orientation of these negative interactions involved a desire to maintain neighborhoods, prevent gentrification, or eliminate density increases. Some articles mentioned the goal of Transit Oriented Development for improving ridership outcomes of the Metrorail system, but these goals were not covered to the same depth as organized community groups. The effectiveness of community groups differed, but residents in Brownsville prevented higher density development and lack unity in

Overtown prevented the awarding of a Federal grant to develop a large parcel in that neighborhood. Overtown, due to its proximity to the CBD, did have numerous large- scale projects constructed following the completion of Metrorail, but the success in blocking the Community Development Block Grant occurred because of a lack of vision for the area (Herald Staff, June 15,1980). Dadeland South and Brickell also had active community groups seeking to reduce development pressure, but residents in both communities ended up capitulating and accepting higher density along the corridor

(Dibble, 1983). These outcomes show that some minority neighborhoods can

successfully advocate for their goals, though some redevelopment may be good for

stimulating jobs and vitality for a neighborhood.

The content analysis cannot prove a correlation between ridership or system

outcomes, but it does provide an opportunity to consider how residents perceive their

city. Miami constructed the Metrorail for numerous reasons, including a desire to be

seen as a world-class city. That goal, coupled with an interest in providing alternative

transportation in a time where gas prices wildly fluctuated, suggests that the intent of

the system was one developed for outsiders and for use as a last resort. The widening

of highways running parallel to the Metrorail alignment and the byzantine structure of

the overall transportation system created a transit system that was unlikely to be used.

Adding lanes galvanized driving as the preferred mode. Creating a transit system that

required three modes—bus to rail to people mover—to arrive downtown ensured a

complicated network to understand and a difficult number of transfers to navigate effectively. The decision to not fully construct the original rail proposal with extensions to western suburbs, to the airport and to primary activity centers along the coast further eroded the ability for residents to live without a car. Finally, the lack of bus connectivity among these places and the failure to adequately feed the rail system likely increased trip times (Feldstein Soto, 1985). Unfortunately, the reviewed articles for the content analysis exclude a detailed account of these issues. Instead, most of these outcomes were gleaned from individual articles and constructed in totality in this discussion. While the content analysis cannot prove correlation, the content analysis suggests that the subjects of news writers did not educate readers on what would create a world-class

transportation system, rather choosing to reinforce accepted expectations.

Quantitative Analysis

The statistical results in this study align with prior literature seeking to model

demographic, land use and built environment characteristics. The R-Squared value for

the quarter-mile radius was particularly strong, which makes sense in the context of the

spatial features of the data. Stronger explanations for ridership would occur in closer

proximity to each station. The R-Squared value for the built environment survey was

weak, though it did demonstrate some important qualities that align to built environment

literature regarding the importance of smaller blocks and quality FLM experiences.

These affirmative findings are important, but it needs to be stressed that every city and

every transit station is spatially and geographically unique. The findings in this study are

relevant to Miami. While findings can be applied to other cities, additional comparisons should be performed to ensure validity across spatial and geographic realities.

In Miami, these variables are reinforced by the selection of specific stations and the characteristics of those places. Each of the stations selected has a unique identity different from each of the others. Dadeland South has an extensive array of commercial stores and hotels; University has a major university and multiple types of activity centers like academic buildings, laboratories, and even venues; Government Center has numerous large office towers; Overtown, directly adjacent and overlapping Government

Center at the ½ mile radius, shares many of the same office characteristics, but has more recent construction; Santa Clara has multiple government offices and numerous industrial facilities mixed with numerous open lots; Brownsville has predominantly residential construction on smaller lots with a high proportion of vacant lots. In many ways, the most important aspect of this study is reaffirming what the visual reality of

each station area is.

Because of these unique characteristics and the small sample size of the stations

selected, the possibility for station selection bias exists. While additional stations could

improve the model, the addition of stations might also have the opposite impact,

reducing the explanatory power of the model. The six stations selected show strong

demographic, land use and built environment characteristics that exhibit where station

catchment areas succeed and where they falter. While a more robust station count

might improve modeling capability, the specific features of successful and less

successful rail station catchment areas might be less clear. In fact, the differences

across each of the stations is stark, allowing a clearer analysis of how Miami might

improve its station areas if it seeks to enhance ridership numbers in the future. In general, the findings in the data are summarized in Table 5-2 and are aligned with the following:

• The models developed most strongly suggest that professional services and office space, the effective year built, the number of signalized intersections, and the number of street trees most positively impact station ridership.

• The models also suggest that parcel size and industrial land use are correlated most significantly to reduced transit ridership.

• Demographic variables were found to be correlated, but were impacted by multi- collinearity and are likely impacted by other socio-economic variables. These other variables include number of renters, percentage of minority residents, residents over age 65, and average household size.

• Other variables were found to have no correlation or an ambiguous relationship.

In developing the models, several rounds of data were evaluated, with a number of potentially useful variables being discarded. These data included the single-family housing land use, general residential land use, population under the age of 18, population densities, and Value Per Acre. These variables were excluded due to errors associated with multi-collinearity that prevented an adequate model from being developed. In general, the need to exclude many of these variables aligns with known trends for race and age. One example of collinearity occurred among the number of renters and the number of multi-family housing. Since many renters occupy multi-family units, the explanatory variable for multi-family housing units exhibited characteristics of multi-collinearity. For districts where rental housing pre-dominated, this trend might extend to many residential units within a station catchment area.

Perhaps the greatest variable that was excluded from the final model were the designations for vacant lots, parking lots, and municipal owned lots. Because these

parcels do not have taxable structures in many cases, no effective year-built variable could be evaluated. This constraint caused some multi-collinearity violations, as well as amplified the VIF values in the earlier models. In one early model, rates of vacancy were found to be statistically significant, but the R-Squared Value of the model was slightly weaker than the final model chosen at the ¼ mile radius.

Available datasets further limited the analysis that could be performed in two ways. First, station-level data was only available for the years 2016 through 2020.

Because of the Covid-19 pandemic, 2020 ridership data could not be utilized. The

Census data was available for 1990, 2000, and 2010. However, ridership data was only available through the National Transit Database beginning in 2001. This combination of muddled data prevented a better longitudinal analysis from being performed. Upon the release of the 2020 Census data, a connection between that data and the 2019 parcel data would be useful to compare against the findings from this study. Analysis would be

expected to confirm or reject the findings identified within this model.

The choice to join 2019 Parcel Data with 2010 Census data may be criticized.

The decision to choose these data to join resulted from two key considerations. First,

the Department of Revenue parcel data was first available in 2012 from FGDL. This

2012 data set excluded differentiation among parcels for condo units. Therefore,

approximately 2400 condo units were excluded as separate units. The 2019 data set

included this condo distinction in the parcel description field, which was used in this

study to examine multi-family units. The second consideration involved the period during which this study was performed. The 2020 Census was anticipated for release in

early 2021. By preparing the 2019 data, there was an expectation that 2020 Census

data might be available prior to submission of this research. The 2019 and 2020 data

combined would provide a timely review of other possible independent variables. A

comparison of the Ordinary Least Squares results using the 2012 Parcel Data and the

2010 Census Data showed similar outcomes with the 2019 data set. This data is

available in Appendix C.

When looking at spatial data, a GWR regression is considered appropriate. This

study was unable to include a GWR due to the small sample size of six selected

stations. The datasets and ridership data provided too little variability to assess results.

Attempts at increasing the number of data points created coincident locations that

generated an error that prevented analysis. Changes to the allowable coincident

locations subsequently resulted in errors of multicollinearity or in too little variability.

Limited data especially restricted the available information for the built

environment survey. This difficulty was anticipated from the start of this study, though

there was an expectation that sidewalk data would be more readily available. After

performing the built environment survey using Google Earth, the sidewalk metric was

eliminated due to incomplete sidewalk networks and surprising numbers of sidewalks

for each station. A better metric for future evaluation would be the number of crosswalks

and connected linear footage of sidewalks. An example of the issues was a sidewalk

that ended at a Trader Joe’s parking lot adjacent to a major crosswalk across seven

lanes of traffic on US Highway 1 near the Dadeland South Metrorail Station The

terminus of this sidewalk occurred 372’ from the entrance to the second busiest rail

station in Miami (Figure 5-2). A second example of poor connectivity also occurred at

Dadeland South, where several sidewalks were predominantly blocked by gated entrances or street trees (Figure 5-3).

Beyond the sidewalk connectivity question, the statistically significant correlation among street trees is likely stronger than presented in the model. Santa Clara and

Brownsville were the first stations where the researcher conducted the built environment survey. There were so few trees in the study areas for Brownsville and Santa Clara that almost every street tree in the study area was included in the dataset. Several trees

were included despite being very small or occurring in the middle of the block. For the

built environment surveys at the other stations, there were many large trees and

significantly more trees occurring in the middle of the block. Because of the number of

trees, the count that occurred focused predominantly at each intersection, extending

approximately 100’-150’ from the intersection curb. Trees planted in large pots or street

planters were excluded. Therefore, the coefficient for street trees is considered to be

weaker than if a full tree audit were performed (Figure 5-4). To improve this analysis,

tree canopy coverage using raster data may be more meaningful for future research.

Among the more surprising findings for this research was the correlation between

signalized intersections and ridership. Some prior research has shown that the number

of intersections is associated with higher ridership (Ewing & Cervero, 2010). This study

shows a weak correlation between total intersections and ridership, but did not find any

significance between the type of intersection and ridership. This distinction primarily

considers the idea that a grid network would optimize connectivity for non-motorized

transit users. However, while the total number of intersections in the station catchment

area was correlated, neither the type of intersection (3 or 4 way) nor the prevention of motorists from making certain turns proved to have any correlation.

Future Research & Summary

Future research could consider the orientation of 3-way intersections and might also consider pedestrian safety in crossing intersections where through-traffic is prevented. The interest in these “non-through travel” intersections is driven by their prevalence in Brownsville and Santa Clara, which both had the lowest ridership. In several configurations, traffic was not capable of making turns due to a fully-divided median. The inability to make certain turns using an automobile could also be found along some of the larger arterial highways, but few were as physically separated as those seen in Brownsville. In all cases, these non-through intersections were decidedly unfriendly to pedestrians, due to the size of the roads, the lack of crosswalks, the lack of street trees, and the atypical road configurations that prevented those automobile maneuvers. While not statistically significant, pedestrian access and safety might have some relevant findings within this built environment context.

In applying these findings, the City of Miami, in conjunction with Miami Metrorail, can improve neighborhoods by focusing on the 5 explanatory variables that can be most easily controlled through land use and built environment changes. First, connectivity of its neighborhoods can be improved by enhancing intersection features like adding signalized intersections and crosswalks, planting proper street trees, and focusing on changing the diversity of land uses near transit stations. This latter recommendation focuses on high numbers of industrial sites that exhibit lower ridership. Consequently, higher density residential and greater office density could improve ridership by creating a better land use mix, while increasing activity centers. Finally, the effective parcel size

should be reduced. While this change likely has many issues given the large parcel sizes in each station catchment area, architectural breaks can be utilized to provide mixed-uses and the reduction of single-use properties. For pedestrians, the architectural

break can improve the scale of the walk to make it more interesting and allow for

sidewalks to be activated (Ewing & Clemente, 2013).

Many of these changes would work in tandem with other city initiatives, such as

climate goals that stipulate greater levels of street trees. Furthermore, both Brownsville

and Santa Clara are located Northwest of Miami, making them less susceptible to rising

oceans. If done conscientiously to limit impacts of gentrification, land use and built

environment changes could achieve multiple goals for the city, as well as enhance

levels of ridership and farebox recovery. One final aspect of the built environment that

was not previously discussed is the inclusion of bicycle lanes and other micromobility

options. The built environment survey identified few designations for dedicated bicycle

or scooter lanes. In all cases, these lanes may not be feasible because of the high-

speed roadways located in proximity of all stations studied. However, reductions in

speed limits and increased safety initiatives may allow for stronger connections for

these modes at the half-mile radius by expanding the walkshed of stations.

Qualitative Assessment, History & Discussion

The discussion of micromobility and the FLM problem should be considered

within the context of the overall transportation system. During the conceptual design

stage of the Metrorail system, acquisition of right of way for a 200’ wide pedestrian mall

along the length of the rail right of way was eliminated because the “mall was not

considered feasible or appropriate from an urban design standpoint” (UMTA, 1980).

While this consideration also included concerns over cost, all of the stations in this study

exhibit built environment challenges with large roads or highways that reduce pedestrian connectivity and make FLM solutions more difficult.

Based on available project documentation and primary sources, the Miami

Metrorail was constructed to accomplish several goals. The specific goals of the

transportation system were to support transportation needs, land use needs, and

economic development needs (UMTA, 1980). Transportation improvements were

considered necessary due to fluctuating gas prices, increasing congestion along the

Dixie Highway and Interstate 95, as well as constrained parking in the CBD. Buses were

intended to feed the rail system to increase ridership and reduce car dependency. Land

use changes were considered a foregone conclusion with downtown employment

expected to increase by 29,500 jobs between 1975 and 1985. Finally, economic

development potential was considered necessary to support Latin American tourists who shopped extensively in the CBD and were expected to double in number between

1975 and 1985 (UMTA, 1980). Notably, in an enumerated list of goals, improving CBD transportation services ranked 3rd, behind promoting development goals and

maximizing cost effectiveness (UMTA, 1980).

With the goals for development set, Miami focused on a multi-faceted

transportation system that would use large park and ride stations, bus feeder systems,

and a downtown circulator to move residents. The expectations may have been

justified, but the reality is that Miami constructed a system that was disjointed and

required numerous mode transfers to travel to a final destination. Comparing the current

built environments shows that even if riders disembarked at a particular station that the

FLM problem remains. Brownsville and Santa Clara have not become meaningfully

denser, while Government Center still maintains several blocks of surface parking lots in the heart of downtown. More importantly, even as the public transportation vision came to fruition, parking lots and wider highways reduced incentives for using the Metrorail. In looking at Miami Herald articles during the construction phase of the Metrorail, parking concerns dominated the discussion across many articles.

The fact that the UMTA and Metromover focused on reducing congestion suggests an automotive focus and not a human focus. Ridership was a tangential goal from the standpoint that it would allow cars to more quickly access downtown. Prior to opening the system in 1984, bicyclists sought to have bike lockers at stations and to have the right to bring the bicycles on board the train. While some stations eventually constructed bike lockers, the attitude at the time disputed the cyclists’ right for these amenities. In addition to some of these disputes, the built environment survey showed that, even after nearly forty years, many stations lacked sidewalks and crosswalks within ¼ mile of stations. Dedicated bike lanes were not a prominent feature in any of the station catchment areas. In other places, sidewalks abruptly end, causing pedestrians or cyclists to cross the street or travel with automotive traffic.

In the years preceding the opening of the Miami Metrorail, two neighborhoods are worth discussing for their attempts to encourage or discourage rail travel. Even four decades after completion, each community’s efforts still have significant impacts in both

Dadeland South and Brownsville. In many ways, these two stations are emblematic of

Miami’s consideration for its Metrorail system. Dadeland was once known as

“Deadland” before the luxury mall and transit station were developed (Cottman, 1982).

A far-flung suburban area South of Miami, Dadeland embraced development near its

station, breaking ground on numerous complexes in the early 1980’s prior to the rail system becoming operational (Hirsch, 1982). The developments included residential, commercial, hospitality, and parking services. As seen in Table 5-3, 96% of parcels in

Dadeland South’s quarter-mile catchment area were developed after 1983. The area

includes many sidewalks, narrower streets, and high numbers of street trees. Despite

these amenities, auto-oriented development pervades the catchment area, including large portions of the area cutoff by highway US-1, parking lots that front most roads, and sidewalks obstructed by potted plants and street trees. While auto-centric development predominates, an attempt to obscure that reality creates a more human- focused sense of place.

Where Dadeland South embraced development, Brownsville sought to prevent apartments from being constructed along the route (Baird, 1982). Many residents feared displacement and hoped that social programs to assist in homeownership would be available. Many residents also felt that absentee landlords that already owned many of the properties in Brownsville would further erode the quality of life (Herald Staff,

November 30, 1981). These expectations may have been justified based on the

displacement that occurred for many of these residents forced from Overtown during the

1940’s and 1950’s during the freeway expansion (Lowe & Ferguson, 1983). Regardless

of the root cause of the distrust, Brownsville residents sought to limit denser

developments, while also fighting proposals for concrete curbing that would effectively

create one-way streets. In the latter battle, residents were partially successful,

convincing builders to allow for left turns and U-turns at more frequent intervals and

avoiding the fate that more severely plagues Santa Clara’s station (Baird, 1983). Unlike

Dadeland South’s high percentage, only 29% of Brownsville’s parcels have been built after 1983. This percentage masks that 25% of the parcels in Brownsville are vacant or comprised of surface parking. Not only is this percentage the highest, but it is also the highest absolute number of all the station catchment areas. Brownsville may have been

successful in maintaining community character, but these decisions may have lost

investment in the community near downtown.

Figure 5-1. Overview of the geographic locations for the six selected stations.

Table 5-1. Table showing station name, designation by Renne and Ewing (2013) as TOD/TAD/Hybrid, and the 2019 ridership ranking from the Miami-Dade Technical Ridership Reports (2019). Ridership Station Designation Rank (2019) Brownsville Hybrid 22 Dadeland South TOD 2 Government Center TOD 1 Overtown TOD 10 Santa Clara TOD 23 University TAD 12

Table 5-2. Table showing Correlated variables categorized by impact.

Correlation Impact Variable Coefficient Probability Robust_Pr Yes Positive SIGNAL 628616.8572 0.000505* 0.000250* Yes Positive TREES 39914.72253 0.000058* 0.000001* Yes Positive EFFYRBLT 5290.2953 0.000000* 0.000000* Yes Positive DV_OFFICESPACE 387989.6132 0.000000* 0.000000* Yes Positive INTERSECTION_TTL 35963.9646 0.000006* 0.000000* Yes Negative ACRES_1 -138173.7538 0.000000* 0.000002* Yes Negative DV_IND_USE -879353.3035 0.000000* 0.000000* Yes Possible PCT_MNRTY -2580.9146 0.000003* 0.008053* Yes Possible AGE_65_UP 2830.9018 0.000016* 0.000290* Yes Possible AVE_HH_SZ -233569.2513 0.000000* 0.000000* Yes Possible RENTER 3897.5533 0.000000* 0.000000* Yes Possible JV 0.0157 0.000000* 0.000206* No Unclear INTERSECTION_TYPE 67586.4921 0.638711 0.614801 No Unclear THROUGH 854.110455 0.997178 0.99704 No Unclear INTEREST_POINTS -212366.8041 0.441712 0.429122 No Unclear DENUNDER18 -15183.1301 0.000000* 0.029989* No Unclear DV_MIXED_USE -69490.1053 0.5911 0.7796 No Unclear DV_MULTIFAMILY -84133.2653 0.004378* 0.1372 For some demographic variables impacted by multi-collinearity, a designation of “Possible” has been identified as the impact.

Figure 5-3. An image from Google Streetview captured by the researcher showing obstructed sidewalk connectivity due to a private development.

Figure 5-5. Five policy recommendations for improving the built environment located near the selected stations.

Table 5-3. Table showing percentage of parcels developed after 1983 (excluding vacant and unbuilt lots) and the boarding rank of the 6 stations studied. Percent 2019 Parcels Built Boarding Station after 1983 Rank Government Center 76.55% 1 Dadeland South 95.93% 2 Overtown 95.82% 3 University 45.36% 4 Brownsville 29.29% 5 Santa Clara 9.76% 6 Department of Revenue Parcel Data 2019 & Calculations by Joseph Dever.

CHAPTER 6 6. CONCLUSION

This study performed a mixed-method approach to evaluate both qualitative and quantitative data to understand ridership at six stations on the Miami Metrorail system.

These six stations were chosen based on ridership between 2016 and 2019 as documented by Miami Dade County in their Metrorail monthly reports. The six chosen facilities were the two stations with the highest ridership, the two stations with the lowest ridership, and the two stations that had the best ridership percentage change between

2016 and 2019. For the latter two stations chosen based on percentage change in ridership, these were the only two stations that did not have a significant decrease during the period. The intent of choosing these stations was to understand the variables that impact how ridership fluctuates from station to station and make informed decisions for how to better integrate stations within their neighborhoods.

A content analysis was performed on 209 articles written between 1976 and

1989. The Miami Herald published 192 of the articles, which included both op-eds and traditional articles. Some national newspapers, such as the Atlanta Journal Constitution, published articles that were reviewed to provide depth in the overall perception for the system. The analysis cataloged articles for positive or negative skew, categories of subjects discussed, and sub-categories discussed. Nine categories and twenty-one sub-categories were identified as being discussed within articles. Given the ability for an article to include multiple topics, a total of 345 sub-categories were defined in the 209 articles. The sub-categories were subsequently classified within the relevant categories.

The content analysis found a significant negative skew in articles across time, category, and sub-category. The skew was greater for negative categories than the

skew for positive categories. Categories identified as having more favorable coverage were Design and Development. Categories identified as having more negative coverage were Construction, Funding, Operations, and Social Impact. Of the negative categories, three of them had more than half of all articles written with a negative perception. None of the positive articles had more than a third of all articles written with a positive perception. The positive sub-categories were Concept Design, Connectivity, and Taxes.

The negative sub-categories were Community Engagement, Construction, Cost,

Funding, Operations, Ridership, Safety, and Walkability. The negative coverage shows a disconnect between the voter support of the Metrorail system in the 1972 Decade of

Progress bond and the negative coverage for the realization of the rail transit goal promised in that bond.

The statistical analysis used the Ordinary Least Squares regression model to evaluate Census and Parcel data at ¼ and ½ mile intervals of selected stations. The selected six stations were chosen based on ridership attributes that made them unique

within the Miami Metroail system context. Government Center and Overtown stations

overlapped in their radii slightly at the ¼ mile interval and moderately at the ½ mile

interval. No other stations overlapped and centroid selection prevented the overlap from

being repeated within the data sets. The OLS regressions utilized data from the 2010

Census and also separate parcel data from both 2012 and 2019. Beyond the quantitative data selected for the OLS regression, a built environment survey was

performed to document the number of intersections, type of intersections, number of

street trees, and other relevant information for a ¼ mile radius. This built environment

OLS regression was performed at the ¼ mile radius only.

The data used in this study included multiple sources across a variety of time periods. The primary sources of data used in this study were from the US Census, the

Florida Department of Revenue, Florida Department of Transportation, and Google

Streetview. US Census data from 2010 allowed for an evaluation of demographic

characteristics. Florida Department of Revenue data from 2012 and 2019 identified

parcel uses, parcel values, and other characteristics related to the built environment.

Florida Department of Transportation data, while not used in regressions, provided an

analysis of the built environment, road classifications, and Metrorail station locations.

Finally, Google Streetview, while not a dataset, allowed the researcher to compile data

through surveying some built environment characteristics focusing on network

connectivity and walkability.

The data required cleanup to ensure continuity in running regressions, but was

not otherwise manipulated. Standard cleanup practices involved calculating Value per

Acre, assigning zero values instead of null values for effective year built, number of

residential units, and similar null characteristics. Perhaps the greatest cleanup of

datasets involved overlap between Government Center and Overtown stations that

required an evaluation of centroid location to ensure parcels were assigned to the

closest station. Similarly, at the ½ mile radius, parcels were removed from datasets if

their centroids fell outside of the study area. Due to proximity to the station, parcels at

the ¼ mile radius were included as long as they intersected the study area. A spatial

join was used for census and parcel data in order to run regressions in an appropriate

manner.

A limitation in the findings of this study is the lack of data supported by a GWR regression. GWR has traditionally provided useful statistical analysis for spatially related data. In the case of this study, not enough data points were developed for analysis. The primary limitations preventing a GWR regression were too many spatially coincident features and too little variation in the ridership results due to low station count. Efforts were made to run the GWR at the ¼ and ½ mile radii for the six selected stations using

2010 US Census data combined with either 2012 FDOR parcel data or 2019 FDOR parcel data. None of the regression attempts were successful for three reasons: coincident features located at the same coordinates, low variation in the ridership results, and multi-collinearity among variables. Attempts to add variables for greater variation led to issues of multi-collinearity; Attempts to increase the variation among the dependent variables led to higher numbers of coincident features, and the study of the six stations led to simple lack of variability. The number of condos in Miami led to high numbers of coincidental features. Data sources limited the ability to analyze datasets because Census data was only available for 1990, 2000, and 2010, while Department of

Revenue data was only available for 2012 through 2019. Ridership data only allowed analysis at the station level from 2016-2020, and only at the system level from 2001 through 2020.

This study found that five variables were most significantly correlated to positive ridership. Two variables involved land use or building characteristics and three involved built environment characteristics. The effective year built and office space were both positively correlated at the 99% confidence level. The coefficient for effective year built was 5,290, meaning an additional 5,290 riders for each year newer. This finding aligns

with newer developments near Government Center, Overtown, University, and

Dadeland South which increased the number of residential units through the construction of condos and apartments. The dummy variable for office space had a coefficient of 387,989, which was so high because of the binary nature of the dummy variable. The high proportion of offices and professional services in Government Center,

Overtown, and Dadeland South likely impacted this outcome, since those three stations

had the highest ridership of the six studied.

The built environment survey indicated that the number of signalized

intersections, the number of street trees, and the total number of intersections within a

¼ mile radius were positively correlated. The number of signalized intersections had a

coefficient of 628,616. While this value is high, the relative importance of density and

through traffic are important to consider because signalized intersections may be more

appropriate in these locations. Government Center, Overtown, and Dadeland South have the highest ridership and have the highest number of signalized intersections. The

more important coefficient to consider is 35,963, which is the coefficient for the total

number of intersections. This variable demonstrates that connectivity of the local street

network matters, while the number of signalized intersections may be a proxy for a

density boost. The coefficient for the number of street trees is 39,914 and suggests that

a tree canopy could be seen as an amenity to shelter pedestrians and cyclists from the

sun and rain. These findings of street connectivity and number of street trees are

consistent with prior literature from Cervero and Ewing.

Two independent variables had significant negative impacts on ridership. The

size of a parcel of land (in acres) had a negative coefficient of -138,173. This outcome

suggests that larger parcel sizes reduce ridership, an idea that aligns with the concept that walkable distances diminish as spatial extent expands. The second independent variable with a negative coefficient was the dummy variable for industrial land use classification. This variable had a low coefficient of -879,353. Industrial uses may have larger footprints and diminish the ability to walk to those locations. Industrial areas may also prevent walkability due to lack of foot traffic from other visitors like tourists and

shoppers.

The six station locations chosen for this research demonstrate unique qualities

for each location. They also demonstrate the stark contrast between the higher-

performing stations and lower-performing stations. The density of both Government

Center and Dadeland South may not appear from population, but the building height

and density separate them from the other stations. Notably, Brickell, which is the

densest station area from a population standpoint, does not have the highest Metrorail

ridership. Even though Dadeland South is far from the Central Business District, its land

uses and setbacks are consistent with those seen closer to a traditional CBD. Despite

Dadeland South’s density, the built environment survey suggests that sidewalks are frequently cut off or obstructed in proximity to the station, suggesting the existence of large parking structures may also support higher ridership. Brownsville and Santa Clara struggle to attract ridership. Single-family homes, and sprawling, lower density residential units pervade these communities, along with expansive office campuses and industrial facilities that have high levels of surface parking. Street trees are scant across the station catchment area of ¼ mile. Vitality seems lacking. The effective year built

shows that nearly 75% of parcels were constructed before or have not been significantly renovated since 1983, the year before the station went into operation.

Many residents may consider these qualities good from a gentrification

standpoint, but all communities require investment to maintain viability over time and

provide bountiful places to live. Brownsville sought to limit development that was

originally mandated with the rail line due to concerns about gentrification. Unfortunately,

the access to downtown and the vitality that could be enhanced by having proximity to a

major infrastructure asset seems missing. Overtown and University have avoided the

decline in ridership seen elsewhere and new developments in these locations suggest

those investments may partially support the ridership gains. However, Overtown missed

an opportunity for a redevelopment grant in the 1980’s because of an inability for the

community to agree on how redevelopment should occur.

All of the station catchment areas studied have major flaws in how the stations

are integrated within their built environments. The number of wide roadways that split

the station catchment areas are significant and often create a barrier for significant

portions of the ridership catchment area. The number of parking garages near Dadeland

South, Government Center, and Overtown prevent more dynamic streetscapes from

existing. The lack of street trees in most station catchment areas causes concern given

the typically hot days in Miami. The lack of bicycle lanes across each station area was

readily apparent. For Miami to create a more successful system, small investments in

local station areas may provide benefits, while also enhancing the local community.

These investments could involve increasing street trees, better connecting intersections,

particularly near Brownsville and Santa Clara where four-way intersections were

eliminated with a paved, raised median. Other immediate benefits might be to redevelop

Miami Metrorail property with office or high-density residential land uses.

This study reiterates earlier research with a combination of variables that might

be related to ridership. These variables include office and industrial land uses, number

of intersections and signalized intersections demonstrating network connectivity, parcel

size and building age, and the number of street trees. Unlike prior research, this study

considers the age of Miami’s Metrorail as an indicator for how well neighborhoods could

integrate with the large public rail infrastructure that was constructed. All stations can

improve their connectivity to rail transit by revising local land use classifications,

improving FLM connections such as bike or scooter share, planting street trees to better

protect pedestrians and cyclists from the sun and rain, and improve the number and the

visibility of crosswalks at specific intersections. Most stations have wide highways

passing in proximity to the station so road diets may be important to enhance both

safety and micromobility in those contexts. These changes could increase ridership at

the station level and potentially catalyze other neighborhood investments.

Miami Metrorail was constructed with high hopes. The rail line that was

constructed met the specifications for Federal funding and was considered the first step

towards creating a world class transit system in Miami. Despite service additions from

Tri-Rail, the system that Miami desired was never constructed. The East-West rail line,

the connection to Miami Beach, and the Northeast rail corridor were never constructed.

More importantly, the bus systems intended to feed riders to Metrorail were

underfunded. Highway widening projects along the Metrorail corridor and falling gas

prices ultimately maintained the automobile as the preferred mode of transportation.

Despite Pickrell’s criticisms in 1989 and 1991, Metrorail probably could have succeeded if zoning changes were made in accordance with original plans, if the system were fed riders as planned, if driving remained less appealing, and if Metrorail executed its system plans. Even without these large-scale improvements, however, Miami has failed to connect its rail system in a manner that suits non-motorized modes.

For these reasons, the Miami Metrorail has experienced a decline in ridership

since 2016, and generally experienced a stagnation in ridership since 2001. The original

projections for ridership have still not been accomplished even as Miami’s population

has increased in the interceding decades. This research finds that Miami has not

succeeded in creating neighborhoods that support the transit system that was

developed. From a qualitative standpoint, the competing goals of the system to reduce

congestion (decrease car trip travel times) and increase density in proximity to stations

were not achieved through the construction and operation of Metrorail. Despite

constructing the exact rail system approved by voters in the 1972 Decade of Progress

bond, the content analysis demonstrated that more negative articles were written than

positive articles, suggesting either a regret by the public for supporting transit initiatives

or inconsistent coverage that sensationalized potentially negative outcomes. Despite

these criticisms, Metrorail operates a useful service that can be enhanced by better

connecting its adjacent neighborhoods through First-Last Mile improvements and land use changes that would preference non-motorists by focusing on street safety and

improved accessibility.

APPENDIX A A. INDEPENDENT VARIABLE DESCRIPTIONS

Table A-1. Independent Variable Descriptions. Source Variable Variable Name Shapefile Description Field Descripti on 2010 US Census POP_201 Total Residents in 2010 This field counts the total population 0 identified in the 2010 US Census. 1990 US Census POP_199 Total Residents in 1990 This field counts the total population 0 identified in the 1990 US Census. 1990, 2010 US DENUND Density Under Age 18 This field provides the density of the Census ER18 population under 18 years of age. 1990, 2010 US MINORIT Minority Residents This field counts the "Minority Census Y Population", determined by the Population that is not white. 1990, 2010 US PCT_MN Percentage of Minority This field divides the "Minority Census RTY Residents Population", determined by the Population that is not white, by the total Population. The output provides the percent of the minority population. 1990, 2010 US POPUND Residents under Age 18 This field counts the number of Census ER18 residents under the age of 18. 1990, 2010 US AGE_65_ Residents over Age 65 This field counts the number of Census UP residents over the age of 65. 1990, 2010 US AVE_HH_ Average Household Size This field calculates the average Census SZ household size by dividing the total population by the total number of households. 1990, 2010 US RENTER Number of Renter- This field counts the number of renter Census Occupied Housing Units. occupied housing units. Florida Dept. of JV Just Value of Property Total Just Value (land just value plus Revenue (2019) building value plus special feature value) of the property. Factors to be considered in determining just value are: present cash value; use; location; quantity or size; cost; replacement value of improvements; condition; income from property; and net proceeds if the property is sold. The net proceeds equal the value of the property minus 15% of the true market value. This accounts for the cost of selling the property.

Table A-1. Continued. Source Variable Field Variable Name Shapefile Description Description Florida Dept. of IMPROVVAL Improvement Value The value of the improvements Revenue (2019) on the parcels. Florida Dept. of LNDVAL Land Value The value of the land on the Revenue (2019) parcel. Calculated VAL_ACRE Value Per Acre The value per acre calculation identifies the Total Just Value of the parcel divided by the number of acres of the parcel. The number in this calculation are derived from the Department of Revenue dataset. Florida Dept. of EFFYRBLT Effective Year Built This field identifies the Effective Revenue (2019) Year Built. This year can be either the original construction date for a structure or a more recent year when major improvements were completed. Florida Dept. of ACRES_1 Number of Acres of This field identifies the Acreage Revenue (2019) Parcel of the parcel. 1990, 2010 US HOUSEHOLDS Number of This field counts the number of Census Households Occupied housing units. Florida Dept. of NORESUNITS Number of This field identifies the Number Revenue (2019) Residential Units of Residential Units associated with a particular parcel. Calculated DV_SFH Dummy Variable - This field is a dummy variable (FDOR '19) Single Family Houses that identifies only parcels with Single Family Homes located on the parcel. The field is determined by use classifications from Department of Revenue data. Calculated DV_OFFICESPACE Dummy Variable - This field is a dummy variable (FDOR '19) Office Space for all parcels with office space or professional services located in the structure on the parcel. The field is determined by use classifications from Department of Revenue data. Calculated DV_MULTIFAMILY Dummy Variable - This field is a dummy variable (FDOR '19) Multifamily for all residential parcels that Residences have multiple units located on the parcel. The type of multifamily housing is not specified and can include condos, apartments, or other types of rental units. The field is determined by use classifications from Department of Revenue data.

Table A-1. Continued. Source Variable Field Variable Shapefile Description Description Name Calculated DV_IND Dummy This field is a dummy variable that (FDOR '19) Variable - identifies all parcels that are designated as Industrial Use having mixed-uses on site. This variable does not breakout or examine the types or densities of uses on site. The field is determined by use classifications from Department of Revenue data. Calculated DV_MIXEDUSE Dummy This field is a dummy variable that (FDOR '19) Variable - identifies all parcels that are designated as Mixed Use having mixed-uses on site. This variable does not breakout or examine the types or densities of uses on site. The field is determined by use classifications from Department of Revenue data. Florida Dept. of DV_VACANT Dummy This field is a dummy variable that Revenue (2019) Variable - identifies all parcels that are listed as Vacant Parcel vacant. The vacant parcels will include all parcels that are unbuilt, unused, or otherwise categorized as vacant structures. The field is determined by use classifications from Department of Revenue data.

APPENDIX B B. VARIABLE DESCRIPTIVE STATISTICS

Table B-1. This table shows the summary statistics for the analysis of FDOR parcel data within ¼ mile of the six selected stations.

Standard Range of Median Deviation DV DV DV Effective Effective Effective Year Residential Industrial Office DV Mixed DV DV Vacant Residential Station Name Frequency Just Value Year Built Year Built Built Units Acres Use Space Use Multifamily Parcels Use Brownsville 508 $114,048,062 95 1963 23.6 1186 110.0 6 7 4 39 125 333 Dadeland South 1539 $1,090,558,745 64 2007 8.7 2455 711.2 1 185 1 1213 35 1272 Government Center 781 $1,818,982,793 112 2008 21.8 886 819.3 5 116 5 506 50 506 Overtown 522 $1,398,762,395 98 1989 9.5 1075 1614.3 1 11 1 416 52 416 Santa Clara 129 $494,376,440 92 1955.5 19.7 548 143.8 63 4 3 1 33 1 University 485 $621,789,710 68 1970 16.1 633 1079.8 0 31 0 357 19 406

Table B-2. This table shows the summary statistics for the analysis of FDOR parcel data within ½ mile of the six selected station.

Standard Range of Median Deviation DV Single DV DV Effective Effective Effective Year Residential Family Industr DV Office DV Mixed DV DV Vacant Residential Station Name Frequency Just Value Year Built Year Built Built Units Acres House ial Use Space Use Multifamily Parcels Use Brownsville 1923 $382,682,566 97 1959 24.4 3255 406.4 1198 19 20 13 1923 419 1350 Dadeland South 3671 $2,348,901,850 68 1971 19.2 5321 19054.5 320 3 211 3 3671 67 3309 Government Center 9256 $7,764,176,863 113 2008 13.6 13785 6881.3 17 17 779 19 9256 230 7908 Overtown 2600 $3,313,956,704 102 2008 15.3 4977 3561.0 6 16 65 16 2600 348 2021 Santa Clara 1030 $1,620,331,522 98 1956 21.9 3939 476.0 238 186 22 8 1030 147 580 University 1276 $1,599,793,519 93 1970 19.4 2044 1742.8 529 0 42 3 1276 65 1076

Table B-3. This table shows the summary statistics for the analysis of 2010 Census data within ¼ mile of the six selected stations.

Housing 2010 White Minority Renter Residents Station Frequency Units Population Residents Residents Households Households 65 and Up Brownsville 37 810 1964 14 1950 457 661 170 Dadeland South 25 1563 1905 636 1269 684 1062 170 Government Center 74 972 1109 310 799 532 643 14 Overtown 42 944 3727 693 3034 582 822 116 Santa Clara 30 413 1424 284 1140 400 400 251 University 26 499 3252 1838 1414 249 443 116

Table B-4. This table shows the summary statistics for the analysis of 2010 Census data within ½ mile of the six selected stations. Housing 2010 White Minority Renter Residents Station Frequency Units Population Residents Residents Households Households 65 and Up Brownsville 117 2869 7075 86 6989 1655 2483 916 Dadeland South 95 3644 6189 1763 4426 1817 2899 647 Government Center 170 8750 10365 2404 7961 4974 6063 1127 Overtown 141 4140 8063 1420 6643 2490 2970 457 Santa Clara 113 3179 8097 546 7551 2555 2896 1393 University 97 1292 5920 3134 2786 538 1178 377

APPENDIX C C. SUPPLEMENTAL DATA ANALYSIS FOR 2012 PARCELS

This appendix shows the results from spatially joining the 2010 Census data with the 2012 Florida Department of Revenue parcel shapefile. This spatial join is being included as an appendix to demonstrate that parcel characteristics and Ordinary Least

Squares results are consistent for both 2012 and 2019 parcel data. The inclusion of this appendix addresses any concerns that the 2010 Census data does not align with conditions from 2019. The 2019 parcel data was selected because it considered individual condo units as separate parcels, whereas individual condo units were combined at the single parcel level in the 2012 data. For that reason, analysis of number of multifamily units and percentage of newly created units was easier for 2019 data.

Table C-1. This table shows the ArcGIS Pro statistical output. Start Time: Saturday, March 13, 2021 12:02:27 AM

WARNING 000642: Problems reading 510 of 1295 total records.

WARNING 000848: Features with bad records (only includes first 30): OBJECTID = 16, 17, 18, 19, 27, 31, 32, 33, 34, 36,

39, 42, 43, 56, 60, 63, 77, 78, 95, 96, 97, 100, 101, 102, 103, 105, 106, 107, 108, 109.

Summary of OLS Results Coefficient t- Probability Robust_Pr VIF Variable [a] StdError Statistic [b] Robust_t [b] [c] ------Intercept 5212750.565 2609701.882 -1.997 0.046119* -1.915 0.056 - JV 0.004 0.002 1.934 0.054 1.528 0.127 1.426 EFFYRBLT 3493.055 1330.349 2.626 0.008813* 2.515 0.012087* 1.189 ACRES -32027.002 7487.280 -4.278 0.000025* -5.019 0.000001* 1.315 DENUNDER18 -12822.327 9467.708 -1.354 0.176 -1.178 0.239 2.193 DEN_MNRTY -4736.554 1453.657 -3.258 0.001184* -6.289 0.000000* 1.747 AGE_65_UP -10913.906 3186.966 -3.425 0.000663* -4.777 0.000003* 1.809 AVE_HH_SZ -299428.568 20110.857 -14.889 0.000000* -10.527 0.000000* 1.603 RENTER 5647.819 836.796 6.749 0.000000* 6.430 0.000000* 2.087 - DV_IND_USE 1207346.293 102419.906 -11.788 0.000000* -13.553 0.000000* 1.225 DV_OFFICESPACE 460756.770 103962.524 4.432 0.000014* 3.913 0.000108* 1.534 DV_MIXED_USE -279655.607 179278.428 -1.560 0.119 -1.153 0.249 1.387 DV_MULTIFAMILY -323180.176 106674.204 -3.030 0.002541* -3.845 0.000140* 1.068

OLS Diagnostics Input Features: Parcels_12_Clip_QTR_Intersec1 Dependent Variable: BOARD19 Number of Observations: 785 Akaike's Information Criterion (AICc) [d]: 23416.373127 Multiple R-Squared [d]: 0.435842 Adjusted R-Squared [d]: 0.427073 Joint F-Statistic [e]: 49.700956 Prob(>F), (12,772) degrees of freedom: 0.000000* Joint Wald Statistic [e]: 579.824669 Prob(>chi-squared), (12) degrees of freedom: 0.000000* Koenker (BP) Statistic [f]: 189.696031 Prob(>chi-squared), (12) degrees of freedom: 0.000000* Jarque-Bera Statistic [g]: 20.131117 Prob(>chi-squared), (2) degrees of freedom: 0.000043* Notes on Interpretation * An asterisk next to a number indicates a statistically significant p-value (p < 0.01). [a] Coefficient: Represents the strength and type of relationship between each explanatory variable and the dependent variable.

[b] Probability and Robust Probability (Robust_Pr): Asterisk (*) indicates a coefficient is statistically significant (p < 0.01); if the Koenker (BP) Statistic [f] is statistically significant, use the Robust Probability column (Robust_Pr) to determine coefficient significance. [c] Variance Inflation Factor (VIF): Large Variance Inflation Factor (VIF) values (> 7.5) indicate redundancy among explanatory variables. [d] R-Squared and Akaike's Information Criterion (AICc): Measures of model fit/performance. [e] Joint F and Wald Statistics: Asterisk (*) indicates overall model significance (p < 0.01); if the Koenker (BP) Statistic [f] is statistically significant, use the Wald Statistic to determine overall model significance. [f] Koenker (BP) Statistic: When this test is statistically significant (p < 0.01), the relationships modeled are not consistent (either due to non-stationarity or heteroskedasticity). You should rely on the Robust Probabilities (Robust_Pr) to determine coefficient significance and on the Wald Statistic to determine overall model significance. [g] Jarque-Bera Statistic: When this test is statistically significant (p < 0.01) model predictions are biased (the residuals are not normally distributed).

Table C-2. This table shows the ArcGIS Pro statistical output. Multi- Res. House- Renter Ind Office Mixed family Vacant Station Frequency Units holds No. LU LU LU LU LU Brownsville 1512 3591 23226 12510 12 21 12 114 297 Dadeland South 513 2469 37419 24114 3 78 6 12 63 Government Center 639 1608 26901 22275 12 99 27 0 129 Overtown 480 2937 3084 2223 24 24 6 9 123 Santa Clara 429 1248 2400 2400 216 18 15 3 54 University 300 747 7860 4059 0 12 0 18 54

APPENDIX D D. CENSUS DATA MD COUNTY

2010 Census Data within ¼ mile of the six selected stations is shown in Table B-3.

2010 Census Data within ½ mile of the six selected stations is shown in Table B-4.

Table D-1. This table shows the summary statistics for the analysis of 1990 Census Data within ¼ mile of the six selected stations. White 1990 Housing (NH) Minority House Renter Age Station Population Units Residents Residents holds Households 65+ Brownsville 2171 766 5 2166 677 470 208 Dadeland South 711 331 386 325 309 206 66 Government Center 74 37 3 71 37 37 7 Overtown 1129 860 175 954 509 506 223 Santa Clara 940 5 435 505 4 3 435 University 4848 490 3496 1352 453 239 152

Table D-2. This table shows the summary statistics for the analysis of 1990 Census Data within ½ mile of the six selected stations. Minority 1990 Housing White (NH) Resident House Renter Age Station Population Units Residents s holds Households 65+ Brownsville 7198 2685 25 7173 2449 1642 907 Dadeland South 4222 2262 2414 1808 2068 1552 574 Government Center 3552 1719 293 3259 1627 1410 680 Overtown 5313 3199 286 5027 2170 2148 705 Santa Clara 6342 1984 988 5354 1893 1586 1346 University 6665 1181 4503 2162 1115 460 400

Table D-3. This table shows the summary statistics for the analysis of 1990 Census Data within ½ mile of the six selected stations.

FIPS White State 1990 Housing (NH) Minority Renter Age Code Population Units Residents Residents Households Households over 65 12011 8,262 7,351 7,322 940 4,663 827 4,846 12021 13 5 2 11 4 1 2 12025 1,936,968 771,209 585,510 1,351,458 692,309 316,430 270,794 12087 72 46 65 7 38 12 9

Table D-4. This table shows the summary statistics for the analysis of 2010 Census data for Miami Dade County. Between 1990 and 2010, the FIPS State Code for Miami and Dade Counties was consolidated. This change is the reason for the FIPS codes in the following table being different for those in Table D-3. FIPS White State 1990 Housing (NH) Minority Renter Age Code Population Units Residents Residents Households Households over 65 12011 1,923 2,011 895 1,028 1,018 248 649 12086 2,495,998 989,245 383,251 2,112,747 867,197 383,448 351,961 12087 18 35 16 2 10 4 5

APPENDIX E E. SHAPEFILE & RIDERSHIP DATA SOURCES

Table E-1. GIS Datafiles & Shapefiles GIS Datafiles & Shapefiles - Sources File Name Year Originator Description Florida Department of Revenue (Hosted on the Florida 2019 parcel shapefile for the state of Florida. Clipped to Geographic Data identify parcels within Miami-Dade County and near PARCELS_2019 2020 Library website) station areas. Florida Department of Revenue (Hosted on the Florida 2012 parcel shapefile for the state of Florida. Clipped to Geographic Data identify parcels within Miami-Dade County and near PARCELS_2012 2012 Library website) station areas. Minnesota Population Center (Hosted on the Florida Geographic 1990 Census data for the state of Florida. Clipped for CENBLK1990 1991 Data Library website) Miami-Dade County and for station catchment areas. U.S. Census Bureau (Hosted on the Florida Geographic 2010 Census data for the state of Florida. Clipped for CENBLK2010 2011 Data Library website) Miami-Dade County and for station catchment areas. University of Florida GeoPlan Center (Hosted on the Florida Geographic CNTBND_SEP15 2016 Data Library website) County boundaries for the state of Florida. U.S. Department of Transportation, Bureau of Transportation Statistics (Hosted on 2018 shapefile for intermodal stations in the state of the Florida Florida. Filtered to identify Miami Metrorail stations Geographic Data only. Buffer used to identify the radius of specific BTS_IPCD_2018 2019 Library website) station catchment areas. Florida Department of Transportation 2019 shapefile for major roads in the state of Florida. (Hosted on the Clipped to show major roads in Miami-Dade County and Florida Geographic used in conjunction with Google Streetview to analyze MAJRDS_OCT19 2019 Data Library website) station catchment areas.

100

Table E-2. Miami-Dade County Ridership Technical Reports Ridership Technical Reports - Sources File Name Year Originator Description Monthly report from Miami-Dade County on Miami-Dade County ridership and parking statistics for Metrorail, 2019-01- (https://www.miamidade Metrobus, and Metromover. These files are Ridership- .gov/global/transportatio regularly removed from the website as new data Technical- n/ridership-technical- becomes available. This report is available in Report 2019 reports.page) PDF format upon request to the researcher. Monthly report from Miami-Dade County on Miami-Dade County ridership and parking statistics for Metrorail, 2019-02- (https://www.miamidade Metrobus, and Metromover. These files are Ridership- .gov/global/transportatio regularly removed from the website as new data Technical- n/ridership-technical- becomes available. This report is available in Report 2019 reports.page) PDF format upon request to the researcher. Monthly report from Miami-Dade County on Miami-Dade County ridership and parking statistics for Metrorail, 2019-03- (https://www.miamidade Metrobus, and Metromover. These files are Ridership- .gov/global/transportatio regularly removed from the website as new data Technical- n/ridership-technical- becomes available. This report is available in Report 2019 reports.page) PDF format upon request to the researcher. Monthly report from Miami-Dade County on Miami-Dade County ridership and parking statistics for Metrorail, 2019-04- (https://www.miamidade Metrobus, and Metromover. These files are Ridership- .gov/global/transportatio regularly removed from the website as new data Technical- n/ridership-technical- becomes available. This report is available in Report 2019 reports.page) PDF format upon request to the researcher. Monthly report from Miami-Dade County on Miami-Dade County ridership and parking statistics for Metrorail, 2019-05- (https://www.miamidade Metrobus, and Metromover. These files are Ridership- .gov/global/transportatio regularly removed from the website as new data Technical- n/ridership-technical- becomes available. This report is available in Report 2019 reports.page) PDF format upon request to the researcher. Monthly report from Miami-Dade County on Miami-Dade County ridership and parking statistics for Metrorail, 2019-06- (https://www.miamidade Metrobus, and Metromover. These files are Ridership- .gov/global/transportatio regularly removed from the website as new data Technical- n/ridership-technical- becomes available. This report is available in Report 2019 reports.page) PDF format upon request to the researcher.

101

Table E-2. Continued.

Ridership Technical Reports - Sources File Name Year Originator Description Miami-Dade County Monthly report from Miami-Dade County on ridership (https://www.miamid and parking statistics for Metrorail, Metrobus, and ade.gov/global/trans Metromover. These files are regularly removed from portation/ridership- the website as new data becomes available. This report 2019-07-Ridership- technical- is available in PDF format upon request to the Technical-Report 2019 reports.page) researcher. Miami-Dade County Monthly report from Miami-Dade County on ridership (https://www.miamid and parking statistics for Metrorail, Metrobus, and ade.gov/global/trans Metromover. These files are regularly removed from portation/ridership- the website as new data becomes available. This report 2019-08-Ridership- technical- is available in PDF format upon request to the Technical-Report 2019 reports.page) researcher. Miami-Dade County Monthly report from Miami-Dade County on ridership (https://www.miamid and parking statistics for Metrorail, Metrobus, and ade.gov/global/trans Metromover. These files are regularly removed from portation/ridership- the website as new data becomes available. This report 2019-09-Ridership- technical- is available in PDF format upon request to the Technical-Report 2019 reports.page) researcher. Miami-Dade County Monthly report from Miami-Dade County on ridership (https://www.miamid and parking statistics for Metrorail, Metrobus, and ade.gov/global/trans Metromover. These files are regularly removed from portation/ridership- the website as new data becomes available. This report 2019-10-Ridership- technical- is available in PDF format upon request to the Technical-Report 2019 reports.page) researcher. Miami-Dade County Monthly report from Miami-Dade County on ridership (https://www.miamid and parking statistics for Metrorail, Metrobus, and ade.gov/global/trans Metromover. These files are regularly removed from portation/ridership- the website as new data becomes available. This report 2019-11-Ridership- technical- is available in PDF format upon request to the Technical-Report 2019 reports.page) researcher. Miami-Dade County Monthly report from Miami-Dade County on ridership (https://www.miamid and parking statistics for Metrorail, Metrobus, and ade.gov/global/trans Metromover. These files are regularly removed from portation/ridership- the website as new data becomes available. This report 2019-12-Ridership- technical- is available in PDF format upon request to the Technical-Report 2019 reports.page) researcher.

102

Table E-2. Continued.

Ridership Technical Reports - Sources File Name Year Originator Description Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report was removed in March 2021, 2018-01-Ridership- but was saved by the researcher in PDF format and is Technical-Report 2018 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report was removed in March 2021, 2018-02-Ridership- but was saved by the researcher in PDF format and is Technical-Report 2018 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report was removed in March 2021, 2018-03-Ridership- but was saved by the researcher in PDF format and is Technical-Report 2018 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report was removed in March 2021, 2018-04-Ridership- but was saved by the researcher in PDF format and is Technical-Report 2018 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report was removed in March 2021, 2018-05-Ridership- but was saved by the researcher in PDF format and is Technical-Report 2018 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report was removed in March 2021, 2018-06-Ridership- but was saved by the researcher in PDF format and is Technical-Report 2018 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report was removed in March 2021, 2018-07-Ridership- but was saved by the researcher in PDF format and is Technical-Report 2018 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report was removed in March 2021, 2018-08-Ridership- but was saved by the researcher in PDF format and is Technical-Report 2018 Miami-Dade County available upon request.

103

Table E-2. Continued.

Ridership Technical Reports - Sources File Name Year Originator Description Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report was removed in March 2021, 2018-09-Ridership- but was saved by the researcher in PDF format and is Technical-Report 2018 Miami-Dade County available upon request. Miami-Dade County Monthly report from Miami-Dade County on ridership (https://www.miamid and parking statistics for Metrorail, Metrobus, and ade.gov/global/trans Metromover. These files are regularly removed from portation/ridership- the website as new data becomes available. This report 2018-10-Ridership- technical- is available in PDF format upon request to the Technical-Report 2018 reports.page) researcher. Miami-Dade County Monthly report from Miami-Dade County on ridership (https://www.miamid and parking statistics for Metrorail, Metrobus, and ade.gov/global/trans Metromover. These files are regularly removed from portation/ridership- the website as new data becomes available. This report 2018-11-Ridership- technical- is available in PDF format upon request to the Technical-Report 2018 reports.page) researcher. Miami-Dade County Monthly report from Miami-Dade County on ridership (https://www.miamid and parking statistics for Metrorail, Metrobus, and ade.gov/global/trans Metromover. These files are regularly removed from portation/ridership- the website as new data becomes available. This report 2018-12-Ridership- technical- is available in PDF format upon request to the Technical-Report 2018 reports.page) researcher. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report contained the year to year comparison so 2016 ridership was also gathered from this report. This report was removed in March 2021, but 2017-01-Ridership- was saved by the researcher in PDF format and is Technical-Report 2017 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report contained the year to year comparison so 2016 ridership was also gathered from this report. This report was removed in March 2021, but 2017-02-Ridership- was saved by the researcher in PDF format and is Technical-Report 2017 Miami-Dade County available upon request.

104

Table E-2 continued.

Ridership Technical Reports - Sources File Name Year Originator Description Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report contained the year to year comparison so 2016 ridership was also gathered from this report. This report was removed in March 2021, but 2017-03-Ridership- was saved by the researcher in PDF format and is Technical-Report 2017 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report contained the year to year comparison so 2016 ridership was also gathered from this report. This report was removed in March 2021, but 2017-04-Ridership- was saved by the researcher in PDF format and is Technical-Report 2017 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report contained the year to year comparison so 2016 ridership was also gathered from this report. This report was removed in March 2021, but 2017-05-Ridership- was saved by the researcher in PDF format and is Technical-Report 2017 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report contained the year to year comparison so 2016 ridership was also gathered from this report. This report was removed in March 2021, but 2017-06-Ridership- was saved by the researcher in PDF format and is Technical-Report 2017 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report contained the year to year comparison so 2016 ridership was also gathered from this report. This report was removed in March 2021, but 2017-07-Ridership- was saved by the researcher in PDF format and is Technical-Report 2017 Miami-Dade County available upon request. Monthly report from Miami-Dade County on ridership and parking statistics for Metrorail, Metrobus, and Metromover. This report contained the year to year comparison so 2016 ridership was also gathered from this report. This report was removed in March 2021, but 2017-08-Ridership- was saved by the researcher in PDF format and is Technical-Report 2017 Miami-Dade County available upon request.

105

Table E-2 continued.

106

LIST OF REFERENCES

Boarnet, M. G., Giuliano, G., Hou, Y., & Shin, E. J. (2017). First/last mile transit access as an equity planning issue. Transportation Research Part A, 103, 296–310. https://doi.org/10.1016/j.tra.2017.06.011

Caughy, M. O., O, C. P. J., & Patterson, J. (2001). A brief observational measure for urban neighborhoods. Health and Place, 7(3), 225–236. https://doi.org/10.1016/S1353-8292(01)00012-0

Cervero, R. (1994). Rail-oriented office development in California: How successful? Transportation Quarterly, 48(1), 33.

Cervero, R. (2000). Growing Smart by Linking Transportation and Urban Development. Virginia Environmental Law Journal, 19(3), 357–374.

Cervero, Robert, et al. (2004). Transit Oriented Development in America: Experiences, Challenges, and Prospects. Washington, DC: National Academy Press, Transportation Research Board, TCRP Report 102.

Cervero, R. (2007). Transit-Oriented Development's Ridership Bonus: A Product of Self- Selection and Public Policies. Journal of Environment and Planning A: Economy and Space, 39 (9), 2068-2085. https://doi.org/10.1068%2Fa38377

Cervero, R., & Murakami, J. (2010). Effects of Built Environments on Vehicle Miles Traveled: Evidence from 370 US Urbanized Areas. Environment and Planning A, 42(2), 400–418. https://doi.org/http://epn.sagepub.com/content/by/year

Cueto, Isabella (November 22, 2016). Lennar Foundation Medical Center opens. The Miami Hurricane. Retrieved from: https://www.themiamihurricane.com/2016/11/22/lennar-foundation-medical- center-opens/

Dawkins, C. & Moeckel, R. (2016) Transit-Induced Gentrification: Who Will Stay, and Who Will Go?, Housing Policy Debate, 26:4-5, 801-818, DOI: 10.1080/10511482.2016.1138986

Ewing, R. H., & Clemente, O. (2013). Measuring urban design. [electronic resource] : metrics for livable places. Island Press.

Ewing, R., & Cervero, R. (2010). Travel and the Built Environment. Journal of the American Planning Association, 76(3), 265–294. https://doi.org/10.1080/01944361003766766

Filion, P. (2001). Suburban Mixed-Use Centres and Urban Dispersion: What Difference Do They Make? Environment and Planning A, 33(1), 141–160. https://doi.org/http://epn.sagepub.com/content/by/year

107

Flyvbjerg, B., Holm, M. K. S., & Buhl, S. L. (2013). Underestimating Costs in Public Works Projects: Error or Lie? https://doi.org/10.1080/01944360208976273

Gan, Z., Yang, M., Feng, T., & Timmermans, H. J. P. (2020). Examining the relationship between built environment and metro ridership at station-to-station level. Transportation Research Part D, 82. https://doi.org/10.1016/j.trd.2020.102332

Giuliano, G., & Hanson, S. (2017). Chapter 1: Introducing Urban Transportation. The geography of urban transportation (Fourth edition.). The Guilford Press.

Giuliano, G., & Hanson, S. (2017). Chapter 9: Land Use Impacts of Transportation Investments. The geography of urban transportation (Fourth edition.). The Guilford Press.

Goldwyn, E., Levy, A., & Ensari, E. (2020). The Boston Case: The Story of the Green Line Extension. NYU Marron Institute of Urban Management. Retrieved from: https://marroninstitute.nyu.edu/uploads/content/Boston-Case_The-Story-of-the- Green-Line-Extension.pdf

Kim, S., Ulfarsson, G. F., & Todd Hennessy, J. (2007). Analysis of light rail rider travel behavior: Impacts of individual, built environment, and crime characteristics on transit access. Transportation Research Part A, 41(6), 511–522. https://doi.org/10.1016/j.tra.2006.11.001

Kuby, M., Barranda, A., & Upchurch, C. (2004). Factors influencing light-rail station boardings in the United States. Transportation Research Part A, 38(3), 223–247. https://doi.org/10.1016/j.tra.2003.10.006

Mackett, R., & Edwards, M. (n.d.). The impact of new urban public transport systems: Will the expectations be met? TRANSPORTATION RESEARCH PART A- POLICY AND PRACTICE, 32(4), 231–245.

Parsons Brinckerhoff Quade & Douglas, Inc., 1996. Transit and Urban Form. National Academy Press.

Pickrell, D. (1989). Urban Rail Transit Projects: Forecast versus Actual Ridership and Costs. Transportation Systems Center, United States Department of Transportation.

Pickrell, D. H. (1992). A desire named streetcar: fantasy and fact in rail transit planning. Journal of the American Planning Association, 58, 158–176. https://doi.org/10.1080/01944369208975791

Purciel, M., Neckerman, K. M., Lovasi, G. S., Quinn, J. W., Weiss, C., Bader, M. D. M., Ewing, R., & Rundle, A. (2009). Creating and validating GIS measures of urban design for health research. Journal of Environmental Psychology, 29(4), 457– 466. https://doi.org/10.1016/j.jenvp.2009.03.004

108

Renne, John L. and Ewing, Reid, "Transit-Oriented Development: An Examination of America’s Transit Precincts in 2000 & 2010" (2013). UNOTI Publications. Paper 17. https://scholarworks.uno.edu/unoti_pubs/17

Renne, J., Tolford, T, Hamidi, S. & Ewing, R. (2016) The Cost and Affordability Paradox of Transit-Oriented Development: A Comparison of Housing and Transportation Costs Across Transit-Oriented Development, Hybrid and Transit-Adjacent Development Station Typologies, Housing Policy Debate, 26:4-5, 819-834, DOI: 10.1080/10511482.2016.1193038

Ryan, S., & Frank, L. F. (2009). Pedestrian Environments and Transit Ridership. Journal of Public Transportation, 12(1), 39–57.

Shoup, D. C. (2005). The high cost of free parking. Planners Press, American Planning Association.

Taylor, B. D., Kim, E. J., & Gahbauer, J. E. (2009). The Thin Red Line: A Case Study of Political Influence on Transportation Planning Practice. Journal of Planning Education and Research, 29, 173-193. doi:10.1177/0739456X09344718

Taylor, B., & Morris, E. (2015). Public transportation objectives and rider demographics: are transit’s priorities poor public policy? Transportation, 42(2), 347–367. https://doi.org/10.1007/s11116-014-9547-0

UMTA, Urban Mass Transit Administration (1980). Miami downtown people mover : environmental impact statement.

Venter, C. J. (2020). Measuring the quality of the first/last mile connection to public transport. Research in Transportation Economics, 83. https://doi.org/10.1016/j.retrec.2020.100949

Vuchic, V. R. (2007). Urban transit systems and technology. [electronic resource]. John Wiley & Sons.

Zuo, T., Wei, H., Chen, N., & Zhang, C. (2020). First-and-last mile solution via bicycling to improving transit accessibility and advancing transportation equity. Cities, 99. https://doi.org/10.1016/j.cities.2020.102614

109

LIST OF DATA SOURCES

Miami-Dade Ridership Technical Reports (2016-2019). https://www.miamidade.gov/global/transportation/ridership-technical-reports.page

National Transit Database (2001-2020). Monthly Module Adjusted Data Release | FTA (dot.gov)

University of Miami, Institutional Research and Strategic Analytics (2020). https://www.irsa.miami.edu/facts-and-information/fact-book/index.htmlLIST OF REFERENCES

LIST OF NEWSPAPER REFERENCES

Advertisement. (1982, February 22). Miami Herald , p. 87. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1647C2AE7AE4C2BA%402445023-1646D59A42589FF0%4086- 1646D59A42589FF0%40.

Anderson, P. (1982, December 26). Miami Herald , p. 4. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16044F2EA5CA0D5D%402445330-1604400686CF4BF2%403- 1604400686CF4BF2%40.Caughy, M. O., O, C. P. J., & Patterson, J. (2001). A brief observational measure for urban neighborhoods. Health and Place, 7(3), 225–236. https://doi.org/10.1016/S1353-8292(01)00012-0

Arnold, J. (1979, August 20). Miami Herald , p. 21. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1602968026324D5A%402444106-16024A0B5F0635D2%4020- 16024A0B5F0635D2%40.

Auchmutey, J. Staff Writer. (1985, December 1). U.S. transit systems heading off the track. The Atlanta Journal and The Atlanta Constitution, p. A/1. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB7C1D443F04F4D.

Baird, L. (1982, August 1). RESIDENTS DON'T WANT APARTMENTS ALONG ROUTE OF RAPID-TRANSIT LINE. Miami Herald, The (FL), p. 5. Available from NewsBank: Access World News – Historical and Current:

110

https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB359714000FDEB.

Baird, L. (1983, March 20). BROWNSVILLE CURBS TO COME OUT. Miami Herald, The (FL), p. 3. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35BB61E0AE6E8.

Balmaseda, L. and Bivins, L. (1984, May 8). REPORT: NEGLIGENCE A FACTOR IN RAIL CRASH. Miami Herald, The (FL), p. 1D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35FB5A5186448.

Bass, A. (1981, September 16). Miami Herald , p. 4. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=imag e/v2%3A114CF48AE24B9638%40WHNPX-1603A37717693D29%402444864- 16029B63F2464A5B%403-16029B63F2464A5B%40.

Betancourt, M. (1982, September 23). DEMOCRATS LASH OUT AT EACH OTHER IN RUNOFF. Miami Herald, The (FL), p. 2. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB359D74CE1D859.

Birger, L. (1984, December 31). METRORAIL WILL SAVE US TIME, FRUSTRATION. Miami Herald, The (FL), p. 3BM. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3616E085A40EC.

Bivins, L. (1984, May 4). 2 METRORAIL CARS COMPLETE LOSSES AFTER ACCIDENT. Miami Herald, The (FL), p. 2C. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35F9C4F933AFB.

Bivins, L. (1984, May 30). RAIN CONVERTS MANY TO METRORAIL. Miami Herald, The (FL), p.8A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35FABB0E629EF.

Bivins, L. (1984, June 2). METRORAIL CARRYING MORE RIDERS THAN EXPECTED. Miami Herald, The (FL), p. 3B. Available from NewsBank: Access World News – Historical and Current:

111

https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35FBCA51A131F.

Blanchard, B. (1982, September 10). CITY CREATES TAX DISTRICT FOR METRORAIL LOOP. Miami Herald, The (FL), p. 2C. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB359AD0C6C2FC1.

Blanchard, B. (1982, September 16). TRAFFIC FEARS DOMINATE MEETING ON SOUTHPARK. Miami Herald, The (FL), p. 12. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB359B998A488CB.

Blanchard, B. (1983, July 31). SOUTHSIDE BLOCK NAMED 'PEDESTRIAN BYWAY'. Miami Herald, The (FL), p. 12. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35D1477E3CA77.

Bloch, J. (1981, November 19). Miami Herald , p. 141. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603FC5FEE8492AD%402444928-1603EF19D2ABFBCF%40140- 1603EF19D2ABFBCF%40.

Bluh, R. (1981, February 19). Miami Herald , p. 6. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603E361E0559435%402444655-16038A058F35C5C8%405- 16038A058F35C5C8%40.

Browning, M. (1981, March 15). Miami Herald , p. 37. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16039F8543B7F451%402444679-16038A703699844D%4036- 16038A703699844D%40.

Burnside, S. (1982, June 5). KENDALL DENSITY TOO MUCH ALREADY. Miami Herald, The (FL), p. ?????. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35929D0775E7A.

Capuzzo, M. (1979, August 23). Miami Herald , p. 224. Available from NewsBank: Access World News – Historical and Current:

112

https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16038718819DF7B4%402444109-16024A1723FCF57E%40223- 16024A1723FCF57E%40.

Capuzzo, M. (1979, August 26). Miami Herald , p. 268. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -160389B873817E8E%402444112-160248CD6140A8E4%40267- 160248CD6140A8E4%40.

Capuzzo, M. (1979, September 16). Miami Herald , p. 204. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601A0AFF232DF8B%402444133-160197C3AD041333%40203- 160197C3AD041333%40.

Capuzzo, M. (1979, September 23). Miami Herald , p. 276. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601E2EAF377246A%402444140-160196B5DF3D8A86%40275- 160196B5DF3D8A86%40.

Capuzzo, M. (1979, September 27). Miami Herald , p. 151. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -160232B6077C475B%402444144-1601B025EEDD6AAB%40150- 1601B025EEDD6AAB%40.

Capuzzo, M. (1979, October 14). Miami Herald , p. 298. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -160284AE454801E2%402444161-1601F174D6379F73%40297- 1601F174D6379F73%40.

Capuzzo, M. (1980, March 30). Miami Herald , p. 290. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -160284E0995754B8%402444329-1601F110CEACD774%40289- 1601F110CEACD774%40.

113

Capuzzo, M. (1980, May 18). Miami Herald , p. 237. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601972ADF9344E3%402444378-160008375177FCEB%40236- 160008375177FCEB%40.

Capuzzo, M. (1980, August 28). Miami Herald , p. 230. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601E64E9A9E9DD0%402444480-16019535C3FD87C5%40229- 16019535C3FD87C5%40.

Chardy, A. (1985, April 14). HAVANA EYES DADE'S METRORAIL, BUT WILL PROBABLY BUILD A SUBWAY. Miami Herald, The (FL), p. 26A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3624AB7F7C7D3.

Colon, Y. (1985, February 14). NOISE COMPLAINTS STILL FOLLOW METRORAIL LINE. Miami Herald, The (FL), p. 4. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB361CC49695E3A.

Cottman, M. (1982, October 21). IN 20 YEARS, DADELAND LED KENDALL OUT OF WILDERNESS. Miami Herald, The (FL), p. 1D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35A0D88A382B4.

Coya, A. (1985, April 12). PEOPLE MOVER. Miami Herald, The (FL), p. 1C. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB362478EA6AAAA.

Coya, A. (1985, December 5). IN THE SHADOW OF METRORAIL. Miami Herald, The (FL), p.4. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB36402206D7A69.

Daugherty, J. and Kamin, B. (1979, August 6). Miami Herald , p. 3. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX

114

-16039F4380FD9B35%402444092-160249C214670024%402- 160249C214670024%40.

Davies, F. (1980, February 3). Miami Herald , p. 346. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601F1B430566FB9%402444273-16018F9C8597D9DD%40345- 16018F9C8597D9DD%40.

Davies, F. (1983, January 23). SOUTHSIDE VICTORY JUST THE BEGINNING. Miami Herald, The (FL), p. 25. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35B177B116DA2.

Dibble, S. (1982, October 24). S. DADELAND COMPLEX TO RISE HIGH BY 1984. Miami Herald, The (FL), p. 12. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35A1415A3B371.

Dibble, S. (1983, October 9). NEIGHBORS ACCEPT DEVELOPMENT PLANS – WITH RELUCTANCE. Miami Herald, The (FL), p. 2. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35DB9E19469B5.

Dibble, S. (1984, February 23). PLANNERS PROPOSE CHANGES FOR SOUTHSIDE. Miami Herald, The (FL), p. 8. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35EE3A7D2783D.

Dibble, S. (1984, April 29). ZONING CHANGES ADVANCE FOR BRICKELL STATION AREA. Miami Herald, The (FL), p. 8. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35F6B4F45BC3A.

Dunlop, B. (1980, July 13). Miami Herald , p. 197. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16023EE851249247%402444434-1601F14F891415F8%40196- 1601F14F891415F8%40.

Dunlop, B. (1980, July 13). Miami Herald , p. 200. Available from NewsBank: Access World News – Historical and Current:

115

https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16023EE851249247%402444434-1601F14F8B455EA8%40199- 1601F14F8B455EA8%40.

Dunlop, B. (1985, April 7). POSSIBILITY, DANGER MARK OVERTOWN PLAN. Miami Herald, The (FL), p. 2L. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3623E45070922.

Dugger, C. (1986, June 29). FIGHT ERUPTS OVER NEW TRAM LEGS EXTENSIONS CALLED KEY TO DOWNTOWN. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3653E22D4C497.

Dugger, C. (1986, December 13). AUTO DEALERS--NOT DADE COUNTY--OFFER 'METRO SALE'. Miami Herald, The (FL), p. 1D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB366505693A7DA.

Editorial Staff. (1981, September 21). Miami Herald , p. 6. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16039F81A69A9ED7%402444869-16029BBF1EB6698E%405- 16029BBF1EB6698E%40.

Editorial Staff. (1981, October 2). Miami Herald , p. 6. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603F739CE9CAFD5%402444880-16038927EFBCD337%405- 16038927EFBCD337%40.

Editorial Staff. (1981, October 19). Miami Herald , p. 6. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603E13F550BD4C1%402444897-1603A6E6EFC8E1B0%405- 1603A6E6EFC8E1B0%40.

Elder, R.(1985, July 14). TROLLEYS: RIGHT TIME TO START, BUT THE PAYOFF IS FAR DOWN THE LINE. San Jose Mercury News (CA), p. 6P. Available from NewsBank: Access World News – Historical and Current:

116

https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB7247E6225A8A5.

Feldstein Soto, L. (1985, January 11). METRORAIL'S FUTURE MAY RIDE ON DISCOUNTS, POLL FINDS. Miami Herald, The (FL), p. 3C. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB36172DA8DD43B.

Fernandez, J. (1989, April 2). MIAMI LOSES PARKING REVENUE AS HEAT FANS USE METRORAIL. The Palm Beach Post, p. 22A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EAF82C9C3D91061.

Fiedler, T. (1980, January 29). Miami Herald , p. 10. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601DFB7FAB402BA%402444268-16014D2D108DAAC6%409- 16014D2D108DAAC6%40.

Fields, G. (1988, June 13). THE REVIVAL DOWNTOWN MIAMI IS BACK -- ALMOST. Miami Herald, The (FL), p. 2RE. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3382CE5D69CDD.

Fisher, M. (1980, June 26). Miami Herald , p. 122. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601A8ED55ADC62D%402444417-16019042D0D28F77%40121- 16019042D0D28F77%40.

Fisher, M. (1981, October 8). Miami Herald , p. 110. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603DB921F1438A7%402444886-1603892204435CD0%40109- 1603892204435CD0%40.

Fisher, M. (1984, February 18). IT'S A DYING JUNGLE UNDER METRORAIL TRACKS. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35ED993CD9FC7.

117

Foote Jr., C. (1983, February 27). STUDY ON REVITALIZING NW 36TH STREET NOW UNDERWAY. Miami Herald, The (FL), p. 3. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35B8FABF65D60.

Frank, C. (1983, June 5). METRORAIL: WEST NEXT?. Miami Herald, The (FL), p. 3. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35C98CB4628AA.

Getter, L. (1983, June 16). AREA RESIDENTS SPLIT ON NEED, ROUTE FOR RAIL. Miami Herald, The (FL), p. 2. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35CBC934B5C6F.

Gilbert, C. (1984, May 14). METRORAIL SITES DRAW DEVELOPERS COMPLEXES GROW AROUND STATIONS. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35F854EBD1E1D.

Gilbert, C. (1984, May 21). PARKING NO HEADACHE, AFTER ALL. Miami Herald, The (FL), p. 10A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35F95DE85A1CF.

Gilbert, C. (1984, July 19). OVERPASSES TO LET RAIL RIDERS CROSS U.S. 1. Miami Herald, The (FL), p. 4D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35FFFB387D9A4.

Greene, J. (1979, June 3). Miami Herald , p. 4. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16019B44914A0E93%402444028-16014CF584CB8C73%403- 16014CF584CB8C73%40.

Greene, J. (1979, June 3). Miami Herald , p. 33. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16019B3C5C3D78C9%402444028-16014CF4206F7D8D%4032- 16014CF4206F7D8D%40.

Grogan, B. (1987, April 30). TRICOUNTY RAIL PLANNING EXCURSIONS TO PRO SPORTS AND COMMUNITY EVENTS. Sun-Sentinel, p. 1B. Available from

118

NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document-view? p=WORLDNEWS&docref=news/0EB5D059F9A7F0D2.

Editorial Staff. (1979, June 13). Miami Herald , p. 6. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601A47FCC886C0E%402444038-16014CF02C7591AD%405- 16014CF02C7591AD%40.

Gemoules, C. (1988, September 9). METRO CONSIDERS RESTORING NIGHT-OWL TRAIN SERVICE \. Miami Herald, The (FL), p. 1D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB339354B27095A.

Hancock, D. (1987, July 23). A RAIL OF AN IDEA. Miami Herald, The (FL), p. 10. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB367A1ECE2D831.

Hancock, D. (1988, July 3). MIAMI'S PLAN ENVISIONS MORE RETAIL AREAS. Miami Herald, The (FL), p. 6. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB33879966CC1AF.

Harlow, S. (1979, August 19). Miami Herald , p. 91. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1602967001FAE3B2%402444105-16024A06ABD37A07%4090- 16024A06ABD37A07%40.

Herald Staff. (1977, July 3). Miami Herald , p. 14. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -15FF5EF8AEA3DB94%402443328-15FF522450D6CAD8%4013- 15FF522450D6CAD8%40.

Herald Staff. (1979, September 13). Miami Herald , p. 44. Available from NewsBank: Access World News – Historical and Current:

119

https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601EA6E7058B9D9%402444130-1601977C6E097033%4043- 1601977C6E097033%40.

Herald Staff. (1979, November 25). Miami Herald , p. 46. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16023D27A9339D7C%402444203-1601F1429A805051%4045- 1601F1429A805051%40.

Herald Staff. (1979, December 3). Miami Herald , p. 8. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -160286B1035D8368%402444211-1601F14A9BD5FA52%407- 1601F14A9BD5FA52%40.

Herald Staff. (1981, August 9). Miami Herald , p. 32. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603A442B6EE4520%402444826-16039BF7E3FEBA7C%4031- 16039BF7E3FEBA7C%40.

Herald Staff. (1979, December 23). Miami Herald , p. 343. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16014C05E42E36C5%402444231-16000811BA81C119%40342- 16000811BA81C119%40.

Herald Staff. (1980, June 5). Miami Herald , p. 257. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -160285B803A9CFD7%402444396-16014D60699A1E26%40256- 16014D60699A1E26%40.

Herald Staff. (1980, June 15). Miami Herald , p. 43. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601E434944EDD37%402444406-16014D72D6629D50%4042- 16014D72D6629D50%40.

120

Herald Staff. (1980, June 19). Miami Herald , p. 121. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601E57AF79A0BCF%402444410-16014EF2C4311753%40120- 16014EF2C4311753%40.

Herald Staff. (1980, July 27). Miami Herald , p. 227. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16020664A755D376%402444448-1601F268F827A579%40226- 1601F268F827A579%40.

Herald Staff. (1980, October 23). Miami Herald , p. 297. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16024CA2D0436B06%402444536-1601F2ED3A0966A7%40296- 1601F2ED3A0966A7%40.

Herald Staff. (1980, November 1). Miami Herald , p. 31. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16024F812CEA424F%402444545-16023989A742960E%4030- 16023989A742960E%40.

Herald Staff. (1981, January 4). Miami Herald , p. 35. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16039A2330328199%402444609-1602A0C98C7DC2BE%4034- 1602A0C98C7DC2BE%40.

Herald Staff. (1981, January 29). Miami Herald , p. 262. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603903543C8772F%402444634-16029F4FD9F2758F%40261- 16029F4FD9F2758F%40.

Herald Staff. (1981, October 8). Miami Herald , p. 158. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX

121

-1603DB921F1438A7%402444886-160389222F3BA49D%40157- 160389222F3BA49D%40.

Herald Staff. (1981, November 12). Miami Herald , p. 269. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603F1F906A2729C%402444921-1603E653A682F3AD%40268- 1603E653A682F3AD%40.

Herald Staff. (1981, November 30). Miami Herald , p. 24. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603F693A620CB12%402444939-1603A7E6F7539BD3%4023- 1603A7E6F7539BD3%40.

Herald Staff. (1981, December 30). Miami Herald , p. 23. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -160435FCC351C480%402444969-1603F51032ABFA2B%4022- 1603F51032ABFA2B%40.

Herald Staff. (1982, January 28). Miami Herald , p. 315. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1605DA5966880004%402444998-16049F944AA4A3AD%40314- 16049F944AA4A3AD%40.

Herald Staff. (1982, April 7). Miami Herald , p. 19. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603EBD47186E6E9%402445067-1603A8874EFD1314%4018- 1603A8874EFD1314%40.

Herald Staff. (1982, June 19). Miami Herald , p. 12. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1604DF4097D9AEC9%402445140-16048DC0C107372D%4011- 16048DC0C107372D%40.

Herald Staff. (1982, September 12). Miami Herald , p. 282. Available from NewsBank: Access World News – Historical and Current:

122

https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16048D83D1D8B1BE%402445225-16044C8A6284C3AA%40281- 16044C8A6284C3AA%40.

Herald Staff. (1982, November 14). Miami Herald , p. 36. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16048D4D154D3489%402445288-1604478FA2BD176B%4035- 1604478FA2BD176B%40.

Herald Staff (1982, December 27). POLLUTED AIR MAY COST COUNTIES FEDERAL ROAD FUNDS. Miami Herald, The (FL), p. 2D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35AD0F46BAE99.

Herald Staff. (1983, June 14). STICK TO GAS TAX. Miami Herald, The (FL), p. 16A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35CB52F06CE08.

Herald Staff (1983, June 15). IT WAS REWARDING EXPERIENCE FOR METRORAIL. Miami Herald, The (FL), p. 2B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35CB76EBB0688.

Herald Staff. (1984, May 16). CAR POOLS TO BE ELIMINATED. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35F891AAEC1B7.

Herald Staff (1987, February 12). ROAD EXPERTS TO GIVE UPDATE. Miami Herald, The (FL), p. 9. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB366AC267B9148.

Herald Staff (1988, January 23). TRANSPORTATION PRIORITIES. Miami Herald, The (FL), p. 3D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB336A0212756CF.

Herald Staff. (1988, February 21). ALIVE -- AND GROWING?. Miami Herald, The (FL), p. 2C. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB336FE192E415F.

123

Herald Staff. (1988, October 10). METRORAIL TURNOFF. Miami Herald, The (FL), p. 14A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document-view? p=WORLDNEWS&docref=news/0EB339A31F77E845.

Hernandez, E. (1986, March 18). CONDO BY VIZCAYA STATION APPROVED BY ZONING BOARD. Miami Herald, The (FL), p. 2D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB364A52474859F.

Hernandez, E. (1986, April 10). VIZCAYA STATION CONDO PROJECT MOVES AHEAD. Miami Herald, The (FL), p. 3. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB364C90D484C79.

Hiaasen, C. (1986, July 2). METROMOVER'S NEW LEGS WOULD BE VERY SHAKY. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3654BE0D3959E.

Hirsch, R. (1982, June 20). METRO PLAN COULD PROTECT AGRICULTURAL ACRES. Miami Herald, The (FL), p. 14. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB359388EDA8F2B.

Hirsch, R. (1982, July 9). Miami Herald, p. 19. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1605DCF1F063CF63%402445160-16048DD26EFB119A%4018- 16048DD26EFB119A%40.

Hirsch, R. (1982, November 23). PLAZAS, SHOPS PROPOSED AT TRANSIT STATION. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35A56050DE1A8.

Hirsch, R. (1982, December 10). DADELAND AREA'S METRORAIL-STATION COMPLEX WINS OK. Miami Herald, The (FL), p. 13D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35A956D23C81D.

124

Hirsch, R. (1982, December 26). Miami Herald , p. 63. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16044F27C42E1067%402445330-1604400672DFA08F%4062- 1604400672DFA08F%40.

Hirsch, R. (1983, April 23). U.S. REVIEW TEAM UNERATHS DEFECTIVE METRORAIL COLUMN. Miami Herald, The (FL), p. 2B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35C251B20B696.

Hirsch, R. (1983, April 30). STEEL MISPLACED IN METRORAIL SUPPORTS. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35C3AA7D2B547.

Hirsch, R. (1983, May 11). METRORAIL'S SUPPORTS FIT, TESTS REVEAL. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35C55E030D3A0.

Hirsch, R. (1984, May 1). DEVELOPERS GET OVERVIEW OF OVERTOWN. Miami Herald, The (FL), p. 3B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35F7695BC7786.

Holly, D. (1989, March 3). MORE RIDING RAIL SYSTEMS. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB33BC9B5471CB1.

Holly, D. (1989, March 6). NEW METRORAIL STATION WILL TIE INTO TRI-RAIL. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB33B75F3B460DF.

Holly, D. (1989, December 12). U.S. TRANSIT CHIEF HAILS METRORAIL. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB33EF73498B8A9.

Kaiser, M. (1981, December 28). Miami Herald , p. 30. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX

125

-160434A68920CF38%402444967-1603F52A567DD2B8%4029- 1603F52A567DD2B8%40.

Kleinman, L. (1988, March 17). S. MIAMI METRORAIL GARAGE SEEKS PARKERS TO FILL SPOTS. Miami Herald, The (FL), p. 6. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3375E28CF773B.

Kranish, M. (1983, June 16). BEACH TO METRO: WE WANT A PEOPLE MOVER. Miami Herald, The (FL), p. 3. Available from NewsBank: Access World News – Historical and Current:https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35CBBE34A50AF.

Lasalandra, M. (1986, June 5). RESIDENTS STEAMED OVER TRAINS. Sun-Sentinel, p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB4E9F25947CEF4.

Lasalandra, M. (1986, July 10). TRICOUNTY RAILROAD COULD BE COSTLY TO COUNTIES IF RIDERSHIP IS A FLOP. Sun-Sentinel, p. 6B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB4E9FEF4D3C4BB.

Lasalandra, M. (1989, March 21). VOTE TODAY WILL DETERMINE FUTURE OF DADE'S METRORAIL. The Palm Beach Post, p. 10A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EAF82C539D647D2.

Lasalandra, M. (1989, April 28). SECOND SEXUAL ASSAULT HITS METRORAILATTACK PROMPTS REVIEW OF SYSTEM'S SECURITY. The Palm Beach Post, p. 7A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EAF82D2CA3D92A6.

Lassiter, B. (1987, Aug ust 27). JUDGE DISMISSES CHARGE OF EATING ON METRORAIL. Sun-Sentinel, p. 3B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB5D080ACB2B934.

Lawrence Jr., C. (1982, October 25). MORE SPACES PLANNED, BUT THEY WON'T MATCH DEMAND. Miami Herald, The (FL), p. 32DT. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35A1910845B7B.

126

Lindsay, K. (1982, September 5). CYCLISTS REQUEST METRORAIL FACILITIES. Miami Herald, The (FL), p. 38. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB359A8A5AD61B0.

Lindsay, K. (1982, November 25). METRORAIL SEEKS GRANT FOR BIKE LOCKERS. Miami Herald, The (FL), p. 12D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35A572318B81B.

Lowe, B. and Ferguson, E. (1983, November 29). URBAN RENEWAL, EXPRESSWAYS RIPPED SOUL FROM OVERTOWN. Miami Herald, The (FL), p. 1A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35E292B6B41D6.

Luening, B. (1984, May 21). CROWDS TEST RIDERS' TEMPERS, RAIL SYSTEM'S LIMITS. Miami Herald, The (FL), p. 11A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document-view? p=WORLDNEWS&docref=news/0EB35F94EC5909F3.

Markham, W. (1982, June 20). MIXED-USE PROJECTS COME TO S. FLORIDA. Miami Herald, The (FL), p. 4H. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35936E2F8BF0B.

Matsuda, C. (1979, December 23). Miami Herald , p. 224. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16014C05E42E36C5%402444231-1600081167542F16%40223- 1600081167542F16%40.

McAden, F. (1979, November 16). Miami Herald , p. 49. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601A1E5A07E7BAA%402444194-1601929670EA8764%4048- 1601929670EA8764%40.

McAden, F. (1979, November 22). Miami Herald , p. 5. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601F75124AAB599%402444200-160193496B80F8D1%404- 160193496B80F8D1%40.

127

McAden, F. (1979, November 27). Miami Herald , p. 27. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16023D582ACDD533%402444205-1601F143E25086AE%4026- 1601F143E25086AE%40.

McAden, F. (1979, December 12). Miami Herald , p. 4. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1602415189256EA3%402444220-1601F14070A166EC%403- 1601F14070A166EC%40.

McAden, F. (1979, December 26). Miami Herald , p. 25. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601A3B7D641DBED%402444234-16014D02312630D4%4024- 16014D02312630D4%40.

McAden, F. (1979, December 26). Miami Herald , p. 26. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601A3BB737564E2%402444234-16014D02981CF3DB%4025- 16014D02981CF3DB%40.

McAden, F. (1981, July 28). Miami Herald , p. 15. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16029579396D50A5%402444814-160256183379891F%4014- 160256183379891F%40.

McAden, F. (1981, November 21). Miami Herald , p. 30. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603F4755B26D34B%402444930-1603A7CD0E47A5E0%4029- 1603A7CD0E47A5E0%40.

McAden, F. (1981, December 19). Miami Herald , p. 35. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document-

128

view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1604486EB8332C4D%402444958-1603F503717406E9%4034- 1603F503717406E9%40.

McAden, F. (1981, December 27). Miami Herald , p. 25. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16043468D459F350%402444966-1603F523C65C8FDB%4024- 1603F523C65C8FDB%40.

McAden, F. (1981, December 29). Miami Herald , p. 15. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -160435EDACD6FC3E%402444968-1603F50E027CC8EE%4014- 1603F50E027CC8EE%40.

McAden, F. (1981, December 31). Miami Herald , p. 17. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1604360AA4C2E2EF%402444970-1603F512A74AF165%4016- 1603F512A74AF165%40.

McAden, F. (1981, December 31). Miami Herald , p. 19. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1604360BE9E9DE1C%402444970-1603F51306CBEC1D%4018- 1603F51306CBEC1D%40.

McAden, F. (1982, April 30). Miami Herald , p. 44. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16043E26070BDFC0%402445090-1603E6D6C2DBB3AE%4043- 1603E6D6C2DBB3AE%40.

McAden, F. (1982, August 27). METRORAIL PUTTING THE BRAKES ON SUBURBAN CYCLISTS' DREAM. Miami Herald, The (FL), p. 4C. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3599853F98FF6.

McAden, F. (1982, September 9). OVERTOWN PROJECT IS UNREALISTIC, FEDERAL TRANSIT CHIEF COMPLAINS. Miami Herald, The (FL), p. 1B.

129

Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB359AB3168067C.

McAden, F. (1982, October 5). Miami Herald , p. 8. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1605EF07B45A976B%402445248-1604E70AD49A9FD0%407- 1604E70AD49A9FD0%40.

Melton, E. (1979, December 29). Miami Herald , p. 27. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -160192AFB8E5EC6D%402444237-16014D04E51E5A11%4026- 16014D04E51E5A11%40.

Morgan, E. (1982, March 25). Miami Herald , p. 156. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16043AF0C9D5A5E2%402445054-1604287492CD185E%40155- 1604287492CD185E%40.

Morgan, E. (1982, May 2). Miami Herald , p. 53. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603F98D2A3B7B81%402445092-1603E6DA68027398%4052- 1603E6DA68027398%40.

Morgan, E. (1982, August 29). DEVELOPER TO UNVEIL HOLSUM PLAN. Miami Herald, The (FL), p. 15. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB359A01249CD57.

Mulliken, J. (1986, April 30). GRAHAM BACKS TRICOUNTY COMMUTER RAIL. Sun- Sentinel, p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB4E9E45B625878.

New York Times, T. (1986, April 21). TRANSIT OFFICIAL: METROMOVER KEY TO SYSTEM`S FATE. Sun-Sentinel, p. 9B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB4E9E0CFA15D07.

130

O’neill, J. (1989, January 12). CANOPY TO GIVE RAIL COMMUTERS LONG DRY SPELL. Miami Herald, The (FL), p. 3. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB33AC072557AF9.

Oglesby, J. (1981, October 6). Miami Herald , p. 7. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=imag e/v2%3A114CF48AE24B9638%40WHNPX-1603DB68C42C19C4%402444884- 1603892058DF84E0%406-1603892058DF84E0%40.

Ollove, M. (1981, October 28). Miami Herald , p. 56. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=imag e/v2%3A114CF48AE24B9638%40WHNPX-1603E6EE12033DDA%402444906- 1603A6CC7494FF4C%4055-1603A6CC7494FF4C%40.

Owens, L. (1982, October 21). NORTHEAST DADE MARSHALS VOTES AGAINST THE TAX. Miami Herald, The (FL), p. 3. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35A0E68B349CE.

Panner, M. (1985, May 23). METRORAIL, PARKING GARAGES WILL OFFER BICYCLE LOCKERS. Miami Herald, The (FL), p. 4D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB362A12C03934B.

Peirce, N. (1976, May 10). Miami Herald , p. 7. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -15FF91882BAD28EA%402442909-15FF55D2C8599FD9%406- 15FF55D2C8599FD9%40.

Platte, M. (1985, April 16). BROWARD OFFICIALS RIDE METRORAIL WITH EYE TO FUTURE. Miami Herald, The (FL), p. 2D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3624FF57846EF.

Postlewhaite, S. (1984, May 20). Thousands jam Metrorail trains for opening day. UPI (USA). Available from NewsBank: Access World News – Historical and Current:

131

https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/1568892F335B1C28.

Postlewhaite, S. (1984, May 21). Riders few on Metrorail's first business day. UPI (USA). Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/15688943730C9A80.

PROBLEMS DOWN THE LINE FOR METRORAIL. (1985, April 7). The Dallas Morning News , p. 33a. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0ED3CD514A362DF0.

Roughton Jr., B. Staff Writer. (1987, November 8). Dallas setting its sights on rail system Seeking to emulate MARTA success, officials pushing for '88 referendum. The Atlanta Journal and The Atlanta Constitution, p. A/13. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB7C2FA78387EC9.

Sachs, S. (1982, October 27). SIGN OF THE TIMES: METRO WILL BUILD FEWER PARKS, ROADS. Miami Herald, The (FL), p. 2D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35A1C6B7D4D70.

Smart, T. (1982, August 16). VISIONS OF BULLET TRAIN LURE INVESTORS TO STATE. Miami Herald, The (FL), p. 1A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3598692ADB1AB.

Spreier, S. (1979, July 1). Miami Herald , p. 235. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601A205EE6717B7%402444056-16014CDAA3A4E85F%40234- 16014CDAA3A4E85F%40.

Spring, S. (1982, May 22). METRORAIL PUTS NEIGHBORHOOD OFF TRACK. Miami Herald, The (FL), p. ?????. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35924318FE6BE.

Spring, S. (1982, July 18). DEVELOPER SEEKS FEWER PARKING SPACES. Miami Herald, The (FL), p. 23. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3595D127958DF.

132

Staff Writer. (1986, May 5). `GOOD FAITH` INVESTMENT ESSENTIAL TO RAIL PROJECT. Sun-Sentinel, p. 10A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB4E9E6293A40AA.

Staff Writer (1987, February 20). STATION GARAGE TO EXPAND. Sun-Sentinel, p. 3B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB5D041F7C5D5FE.

Staff Writer. (1989, December 18). LET GO OF SECURITY-BLANKET CARS AND GET ON TRI-RAIL. The Palm Beach Post, p. 2F. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EAF832AE4E654C1.

Steinback, R. (1988, April 17). DADE LEGISLATORS SEEK WAYS TO FUND LOFTY TRANSIT GOALS. Miami Herald, The (FL), p. 3B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3377FB271EC3B.

Tasker, F. and Zaldivar, R.A. (1982, September 14). FEC PLANNING $300-MILLION RAIL GALLERIA. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB359BA34B784CE.

Tasker, F. (1987, August 12). PRESIDENT HAS A TICKET TO RIDE THE RAIL. Miami Herald, The (FL), p. 1D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB367C94B9D3B08.

Thompson, J. (1986, December 25). METRORAIL RIDERSHIP IS INCHING UPWARD. Sun-Sentinel, p. 6B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB4EA4025C65A05.

Thompson, R. (1982, September 5). SENATE DISTRICT 39. Miami Herald, The (FL), p. 11. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB359A8AB4B9EA6.

Viglucci, A. (1986, February 20). THE ROADS WARY OF NEW CONDO. Miami Herald, The (FL), p. 4C. Available from NewsBank: Access World News – Historical and

133

Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB3647C2C0C0573.

Whited, C. (1979, December 7). Miami Herald , p. 37. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -16023AF8F32DAC66%402444215-1601F1381C4EB873%4036- 1601F1381C4EB873%40.

Whited, C. (1980, February 7). Miami Herald , p. 46. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1601DE3D2A627E2E%402444277-16018F0738ADE5C1%4045- 16018F0738ADE5C1%40.

Whited, C. (1981, May 26). Miami Herald , p. 3. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603EC0E056167E8%402444751-16039B92F0EE7C00%402- 16039B92F0EE7C00%40.

Whited, C. (1981, September 30). Miami Herald , p. 17. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603A4DC69C7C43D%402444878-1603891E4C6DE1E8%4016- 1603891E4C6DE1E8%40.

Whited, C. (1982, May 7). AUTO ADDICTS COULD LEARN TO LOVE THE TRAIN. Miami Herald, The (FL), p. 1C. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35918259D29F8.

Whited, C. (1982, August 17). HIGH-SPEED RAIL IS STATE'S TICKET OUT OF BIG JAMS. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35986D190903B.

Whited, C. (1982, October 28). SOUTHPARK IS START OF GROWTH ALONG RAIL LINE. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35A2056C64F25.

134

Whited, C. (1983, March 16). BRICKELL AREA FEELS THE PAIN OF FAST GROWTH. Miami Herald, The (FL), p. 1D. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35BAC98EA44B1.

Whited, C. (1983, September 8). HIGH-SPEED RAIL LINKING 3 CITIES ON RIGHT TRACK. Miami Herald, The (FL), p. 1B. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35D79C4F9BA61.

Williams, D. (1981, February 25). Miami Herald , p. 4. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=imag e/v2%3A114CF48AE24B9638%40WHNPX-16039B8B80A4BC5F%402444661- 16038A07C15A8ABC%403-16038A07C15A8ABC%40.

Willson, E. (1980, May 29). Miami Herald , p. 286. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=imag e/v2%3A114CF48AE24B9638%40WHNPX-1603A066DD6E664B%402444389- 16014D538DE1BD21%40285-16014D538DE1BD21%40.

Willson, E. (1980, November 20). Miami Herald , p. 297. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -160384DB36B58E69%402444564-1602485F48498160%40296- 1602485F48498160%40.

Willson, E. (1981, April 2). Miami Herald , p. 116. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=image/v2%3A114CF48AE24B9638%40WHNPX -1603EE9A9B362472%402444697-16038FB51566F32F%40115- 16038FB51566F32F%40.

Ynclan, N. (1983, January 16). SOUTHSIDE ORGANIZES TO KEEP SCHOOL OPEN. Miami Herald, The (FL), p. 3. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35AF9CD2767FE.

Zaldivar, R.A. (1981, December 28). Miami Herald , p. 29. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document-

135

view?p=WORLDNEWS&docref=imag e/v2%3A114CF48AE24B9638%40WHNPX-16043487469AA235%402444967- 1603F528E25E05F9%4028-1603F528E25E05F9%40.

Zaldivar, R. (1984, September 27). STUDY SEES METRORAIL USE DOWN PROJECTIONS FAR BELOW '78 FIGURES. Miami Herald, The (FL), p. 1A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB36094C115AB1E.

Zaldivar, R. and Lowe, B. (1983, May 26). CONCRETE MAY SMOTHER 'PEOPLE PLACES' OF MIAMI. Miami Herald, The (FL), p. 1A. Available from NewsBank: Access World News – Historical and Current: https://infoweb.newsbank.com/apps/news/document- view?p=WORLDNEWS&docref=news/0EB35C71DCF4E01F.

136

BIOGRAPHICAL SKETCH

Joe Dever is a master’s student at the University of Florida studying Urban and

Regional Planning. During the completion of his degree, he continued his professional work in the construction industry as a cost control manager for the JFK Airport Terminal

1 redevelopment in New York City. Beyond his work at JFK, Joe has experience on numerous large infrastructure projects across multiple sectors including rail transit, manufacturing, energy, and corporate real estate. In these endeavors, Joe has worked for both public and private entities, while having experience in both field and home offices.

Prior to working in the construction industry, Joe received a bachelor’s degree in

Economics from Clemson University’s Honors College in May 2009. While at Clemson, he focused on economic development and regional planning. Joe will finish the MURP program in April of 2021 and expects to continue working in his new role as a Project

Controller. He hopes to use his professional and academic backgrounds to improve the delivery of transit megaprojects and the operations of transit agencies. Improving these areas can help to achieve better outcomes for some of the most challenging issues facing urban areas, including concerns about climate change and equity.

137