I. Table of Contents
Page 1: Executive Summary
Page 2: Goals of the Study
Page 2: Peer Corridor Descriptions Charlotte
Denver
Phoenix
Sacramento
Salt Lake City
Page 5: Economic Indicators Property Values Business Development Transportation Times and Costs
Environmental Indicators Air Quality Indexes (AQI)
Carbon Monoxide Concentration Trends
Ozone Concentration Trends
Health Indicators Asthma Rates Traffic Fatalities
Results and Conclusion

II. Executive Summary

This paper serves as an addendum to the previously existing Transit Oriented Development in the Triangle Region report compiled by the Triangle J Council of Governments. It analyzes the impacts of light rail transit in regions that are comparable to the Triangle Region (as defined by the region surrounding Duke to NC State to UNC). This study focuses on the effects of Vehicle Miles Traveled (VMT) as a result of light rail implementation. Additionally, we examine how light rail impacts the following sub-categories: environment, economic, and health indices. These indices shape our assessment of whether or not light rail is a worthwhile investment in the Triangle Region. In order to enable politicians to make the best decisions regarding light rail, we provide a comprehensive analysis of peer corridors that are similar to the Triangle Region. Based on population growth projections for the Triangle Region, we find it necessary to offset the corresponding increases in traffic rates and the detrimental health effects that are associated with this growth of automobile usage. We found that building a light rail transit system generates additional tax revenues from increased property values and spending at commercial centers - associated with increased access to these businesses. However, there are clear economic costs associated with implementing light rail line system; aside from the initial costs, these include: decreased revenue from gasoline purchases (on a per capita basis) and home values as well as maintenance costs. Based on our peer corridor research, we find that there are positive environmental impacts although they cannot be directly attributed to the introduction of light rail. However the association between light rail and vehicle miles travelled is unclear and the environmental indicators we researched are indirectly tied to these findings.

III. Goals of the Study
- Client
- Triangle J Council of Governments is a voluntary organization of municipal and county governments in North Carolina's Region J (Chatham, Durham, Johnston, Lee, Moore, Orange and Wake counties). It is one of 17 regional councils established in 1972 by the General Assembly to aid, assist, and improve the capabilities of local governments in administration, planning, fiscal management, and development.

- Our assignment - Triangle J has asked us to quantify the benefits of building a new transit system in a way that will be easily understood by local politicians such that they can shape bills to maximize expected results. To accomplish this task, we have been asked to do an analysis of Vehicle Miles Traveled (VMT), examining how they can best be reduced, and by how much. Five identified peer cities, chosen by Triangle J for their similarities to the Triangle region, will be examined for the environmental, health, and economic impacts of introducing light rail. This information will be included in the currently incomplete Section IV of the report “Transit-Oriented Development in the Triangle Region.”

IV. Peer Corridors

A. Charlotte

The light rail system in Charlotte, called LYNX, opened on November 24, 2007. Before light rail implementation, bus transportation was provided by the Charlotte Department of Transportation under the branding known as Charlotte Transit, which was in existence from 1976 to 1999. Due to the relatively new nature of the light rail, it is difficult to quantify many of the benefits and costs of the project. However, because of the city’s similarities to the Triangle, it remains a valuable city to examine as the light rail continues to be expanded.

B. Denver

The Regional Transit District of Denver runs a variety of transit infrastructures including bus lines, airport shuttles, and special services for senior citizens. It opened the first of five light rail lines in 1994 and made additions in 2000, 2002, and 2006. The newest lines are set to open in 2013 and 2016. The rail lines run through eight distinct transit corridors and include over 35 stops, both in the city center and suburbs. Below is a graph of Denver’s Vehicle Miles Traveled from 2000 to 2008, which does not include the system’s original completion date but shows a slow, steady increase then leveling off of VMT for the period of the rail system’s expansion.

C. Phoenix

The Phoenix metropolitan area opened a light rail transit system in December, 2008 which cost the city approximately $1.4 billion to build. The system, known as the Valley Metro, serves a daily ridership of approximately 45,000 people along 32 stations and 20 miles of rail in the Maricopa County. On Saturdays, ridership is around 28,000 people and declines to 20,000 on Sundays and holidays. Valley Metro has seen ridership growth increase by 12% annually. Today the system incorporates a ride-sharing platform whereby commuters are matched with other commuters if they would prefer to carpool.

D. Sacramento

In 1987, Sacramento built the first pieces of its light rail transit system. This project was 18.3 miles long and served a small portion of the Sacramento population. The transit lines expanded in 1998, 2003, 2004, 2005, and 2006. Currently, the city operates a total of 36.9 miles of light rail, almost double what it had in 1987. Sacramento appears to be increasing in population density. Whether or not this is a result of the light rail lines cannot be verified with certainty. In 2000, there were 4,189 people per square mile. By 2010, that number increased to 4,660 people per square mile. Below is a graph of VMT since 1980 for the Sacramento region. VMT is the y-axis. Of particular interest are the drops in VMT in certain years. Since 1980, VMTs have been continuously increasing at a fairly steady rate. However, when the light rail lines were built in 1987, there was a drop in VMT in 1988. The same drop occurred after expansions to the system in 1998 and 2006. These drops are only short-term, but the long-term effect is an overall curb in the rate of growth of VMT. The red dots show the dates of expansions to the light rail system.
E. Salt Lake City

The Utah Transit Authority (UTA) integrates three transit systems—commuter rail, light rail and bus service. The light rail system TRAX began operation in 1999. The TRAX light rail provides service in electricity powered train cars that follow along fixed guide ways. The first light rail line travels from Salt Lake City to Sandy. The second line connects downtown to the University of Utah and provides an extension to the University of Utah Hospital. Currently, a third line, an extension from the Sandy Line is in construction. According to a Performance Audit by the UTA, the light rail system is currently 19 miles. The American Public Transportation Association estimated that as of 2011, the estimated ridership was 43,500 thousand daily boards. Comparing the second quarter of the 2011 and the second quarter of the preceding year, 2010, ridership increased by 6.84%. The high level of ridership on the light rail has allowed the light rail to maintain a low cost per passenger in comparison to alternate forms of transit in the region. According to the audit performed in 2008, the TRAX line has an operating cost per vehicle mile of $8.18. Vehicle Miles of Travel (VMT) by County by Road Ownership. 2005. Utah Department of Tranportation (UDOT), Utah.

The annual VMT data from 1988- 2005 represents an increase in VMT in Salt Lake county. A possible explanation for the leveling off of VMT around year 2001 could be attributed to the completion of the second line of the TRAX railway.

V. Economic Indicators

VI. Environmental Indicators

A. Air Quality Indices (AQI)

As the name implies, the Air Quality Index (AQI) is a quantitative method of calculating air quality and health. The EPA calculates the AQI for five major air pollutants regulated by the Clean Air Act: ground-level ozone, particle pollution (also known as particulate matter), carbon monoxide, sulfur dioxide, and nitrogen dioxide. According to the EPA, when AQI values are above 100, air quality is considered to be unhealthy-at first for certain sensitive groups of people, then for everyone as AQI values get higher.

Since Charlotte and Phoenix both had their light rails implemented at the end of these AQI trends, 2007 and 2008 respectively, these graphs are more useful to observe as natural trends without light rail. Both cities see a peak in days above 100 AQI in 1998-1999. Both cities also decrease their number of days above an AQI level of 100 after these dates. By the time the light rail was implemented, the AQI had drastically decreased to below 20 days per year. This indicates that in the absence of light rail, AQI had generally decreased since 1998.

The number of days with an AQI level above 100 in Denver had begun to drop before its implementation in 1994. Since the light rail was established, however, this decrease has continued at a steadier rate. Similar to Charlotte and Phoenix, whose number of days above an AQI level of 100 decreased before light rail was implemented, Denver’s decrease may be natural rather than attributed to the addition of light rail.

Salt Lake City’s number of days above an AQI level of 100 appears to have stabilized before the implementation of light rail in 1998. This is similar to Charlotte and Phoenix, whose number of days above an AQI level of 100 decreased before light rail was implemented.

Sacramento’s number of days above an AQI level of 100 are more variable than the other peer corridors. Due to the several additions to the light rail system since its implemetation in 1987, its is difficult to attribute any changes in AQI to the light rail.

Conclusion: These AQI indexes show that the number of days with an AQI lindex over 100 decreased and leveled out before the implementation of most light rail systems. Most cities were experiencing a downward trend that began years before most of their light rail lines were built. Denver is the exception. When it’s light rail system began, the number of days with an AQI index over 100 decreased at an even faster rate than before. Ultimately, however, light rail seems to have a minimal effect on this indicator. The decreases may be due to more large scale policy changes, potentially at the national level, that affect pollution in cities such as these.

B. Carbon Monoxide (CO) Concentrations

Increases in carbon monoxide are due to a variety of factors, the largest of these being car exhaust. Reductions in yearly vehicle miles travelled should cause a decline in CO concentrations at the same time. However, after inspecting changes in carbon monoxide concentrations, it will be easier to see if this is truly the case. If it is not, CO concentrations may not be as correlated with VMT as originally anticipated. All of the carbon monoxide was gathered from a U.S. Environmental Protection Agency website.# Charlotte and Phoenix implemented their light rail systems in recent years. Each of these cities have been experiencing declining CO concentrations since 1990 despite the fact that their light rail lines were not built until 2007 and the end of 2008, respectively. These declines are probably due to other factors, such as amendments to the Clean Air Act in 1990. The graphs for each of these cities are below. CO Air Quality, 1990-2009 (Based on the Annual 2nd Maximum 8-hour Average) Charlotte-Gastonia-Rock Hill, NC-SC SITE=371190038 POC=1 & SITE=371190041 POC 1

Salt Lake City’s light rail lines opened in 1999. As seen from the graph below, CO concentrations appear to have been declining since 1991. However, upon further inspection, CO concentrations only declined from about 6.5 to about 5.5ppm between 1990 and 1999. After the transit system was built, CO concentrations decreased from 5.5ppm to about 2ppm in 2004. After 2004, the concentrations hovered at about 2ppm. What is most important here is that in the ten years prior to the rail lines, it decreased by only 1ppm. After the rail line was built, it decreased about 3.5ppm over 5 years. Therefore, it is highly likely that the implementation of a light rail system caused a dramatic (and speedy) decline in ozone concentrations.

Denver built its first pieces of its light rail system in 1994. The effects of these rail lines on CO concentrations show a very similar pattern to what happened in Salt Lake City. Between 1990 and 1994, Denver’s carbon monoxide concentrations declined from 7ppm to about 6ppm. This is roughly the same rate as the decline without transit in the Salt Lake City area. However, after the light rail lines were built, CO concentrations declined from 6ppm to about 2ppm in 2002. That is a 4ppm decline over 8 years. Between 2002 and 2009, CO concentrations lingered around 2ppm. This information shows that the Denver transit lines had a significant impact in reduction of CO concentrations in a short amount of time.

Sacramento’s light rail transit began in 1987, which is unfortunately before these records begin. It can be assumed, however, that Sacramento’s CO concentrations were above the national standard (whether or not they were declining cannot be determined from this information). However, between 1990 and 2008, Sacramento experienced an extreme decline in CO concentrations from about 10ppm to 2ppm. This decline could be a result of the transit system, but it could also be a result of public policy factors, such as amendments in the Clean Air Act in 1990.

While the table below is very incomplete, the information that it portrays is clear. The Triangle area has a carbon monoxide concentration that has decreased to about 2ppm. As shown in previous charts, CO concentration tends to level off at 2ppm, indicating that the existence of light rail may only have the potential to decrease CO concentration to that extent. Therefore, the Triangle’s light rail proposal may not decrease CO concentration in any major way.

Conclusion: Carbon monoxide concentrations trends between the years 1990 and 2009 indicate that the largest and fastest decreases in CO ppm occur at the same time that most transit systems are built. Therefore, CO ppm appears to be correlated to the implementation of light rail systems. This is the case in all of our peer corridors, and would hold true for the Triangle area if a light rail line was created in the future. However, the potential to decrease CO concentrations appears to halt at about 2ppm. It seems likely that 2ppm is the lowest to which CO concentration can be diminished without further environmentally friendly policy changes. Because the Triangle already has a CO concentration of 2ppm, a light rail system should not decrease the CO ppm in any substantial manner. Lastly, it is important to note that CO concentrations were decreasing before most of the light rail lines were built, so the decline could be due in part to other factors, such as national policy actions (i.e. Clean Air Act changes in 1990). Nevertheless, after light rail systems were finished, CO usually decreased at a much faster rate than their previous course.

C. Ozone benefits

Ozone is a pollutant that is often caused by high amounts of automobile exhaust and other sources of pollution. Theoretically, if ozone amounts decrease, the local environment should improve. Additionally, if less people drive and take transit instead, ozone amounts are expected to decrease. This would result in a healthier environment for all residents and workers of the area. Our analysis of this indicator shows if ozone actually decreases with the implementation of a light rail system. All ozone data was collected from an EPA source.# Charlotte’s ozone concentration has been consistently above the national average since before 1990. The graph below shows the O3 concentration in the years between 1990 and 2009. Charlotte’s light rail system was built in 2007. Between 2007 and 2009, Charlotte experienced its first major decrease in ozone concentration since before these records begin. It is highly probable that this decrease is due to the existence of the light rail system. Ozone Air Quality, 1990-2009 (Based on Annual 4th Maximum 8-Hour Average) Charlotte-Gastonia-Rock Hill, NC-SC Site=37119034 POC=1 & Site=37119041 POC=1

Because Phoenix’s light rail transit is the newest of the five peer corridors, its data is likely to be somewhat inconclusive. The data below shows that for the past 20 years, Phoenix’s ozone concentration has stayed right around the national average. We see a small decline in the concentration between 2008 and 2009 (the year its light rail system was built), but it is uncertain whether or not this decline will continue or what factors may have caused it.

Salt Lake City’s ozone concentration has remained right at the national standard of 0.075ppm since 1990, with minor fluctuations. The creation of the city’s light rail system in 1994 seems to have had no positive impact on ozone concentrations. It may not have had an impact of any type.

Sacramento’s ozone levels have also hovered around the national standard since before 1990. We do not have data before the light rail system was built, but we do know that ozone concentration has changed little (if at all) since the beginning of the light rail lines.

Until recently the Triangle region had an ozone concentration that had been consistently above the national standard for many years. It has only seen a minimal decline in this trend over the course of the past twenty years. Currently, the ozone ppm is right at the national average.

Conclusion: What these ozone trends indicate is that light rail has little to no impact on ozone concentrations between 1990 and 2010. This is the case in every peer corridor that we have analyzed, and it seems fair to say that the Triangle would see the same trends if a light rail system was built in this area.

VII. Health Indicators

A. Traffic Fatality Rates

Fatality rates are a convenient statistic to measure for this study because they are measured in terms of VMT. Because fatality rates are constantly improving with improvements in car safety features and road quality, the rate at the projected completion date of the Triangle rail system will likely be lower than the current rate. Figure 1 shows the steady decline in North Carolina fatality rates over the past fifteen years, from almost 2 traffic deaths per 100 million VMT in 1994 to 1.26 in 2009. Although the steep decline in this graph suggests that the fatality rate will soon approach zero, a look at the national trends from the 1950s to today (Figure 2) reveals a logarithmic curve that appears likely to level out at just under 1 fatality per 100 million VMT. North Carolina’s fatality rate is consistently higher than the national average, which was 1.13 in 2009, but there is nothing to suggest that the North Carolina rate does not follow the same roughly logarithmic trend.

Based on these data, we assume that the fatality rate in North Carolina will eventually hover at or just under 1 per 100 million VMT. This figure multiplied by the number of VMT prevented by the rail system per year, which is not known at this time, gives the annual number of traffic fatalities prevented by the rail system. For example, if the rail system eliminates 200 million VMT per year after it opens, this translates to about two lives saved per year.

However, this potential is only half of the story. Transit systems can also be hazardous and have fatality rates, the result of crashes between rail cars and motor vehicles or pedestrians. These rates are more variable than traffic fatality rates and depend on the safety features of individual transit systems. If planners of the Triangle rail project do not want the rail to result in a net increase in the number of traffic deaths in the region, their safety engineers should aim to keep the system’s fatality rate under 1 per 100 million miles traveled (preferably zero, of course). Proper road signage marking rail tracks, loud signals to alert cars and pedestrians that a train is approaching, and a public safety campaign alerting drivers not to take risks when driving near the rail line could help keep the rail’s fatality rate low.
Figure 1. North Carolina Traffic Fatality Trends.# Figure 2. Graph of national fatality rates from the 1950s to present.#

B. Asthma

Studies show that asthma is associated with several different air pollutants found in car exhaust, especially ozone, NO2, and larger airborne particulates. Schildcraut, et al. found that A 1 ppb change in CO and a 20 ppb change in NO2 is associated with symptom odds ratios of 1.08 and 1.09 respectively.#

A study in Atlanta during the 1996 Olympic Games found that “acute asthma care events” dropped by over 40% and pediatric when traffic counts dropped by 22.5%. The lack of car traffic led to a 13% reduction in ambient ozone concentrations (Friedman, et al).#

An eighteen-year study by Moore, et al. in Southern California found that asthma hospitalizations among 0-19 year-olds increased by 4.6% for every 10-ppb increase in the quarterly average 1-hour ozone concentration.# Inversely, a 10-ppb reduction in 1-hour ozone concentration could lead to a 4.4% decrease in asthma discharges.

Because we have EPA ozone data for several of the Triangle area’s peer cities, we focused on the Moore study to determine whether the Triangle could expect a change in asthma rates after the implementation of the rail system. Figure 3 shows the changes in ozone concentration in the time period from two years before the system opened to two years after. It shows that ozone concentration did not decrease uniformly after a light rail system opened. Suburban testing sites all reported a mild decrease in ambient ozone, but two of the three urban areas sampled saw an increase in ozone concentration.

Figure 3. Change in ozone concentration after rail implementation in peer cities.#

Based on the Moore study, we would expect pediatric asthma hospitalization rates to decrease in Charlotte by approximately 4.7% from 2005 to 2009 and those in Salt Lake City to decrease by about 1.9% from 1997 to 2001. Hospitalizations in Salt Lake City did fall by 0.9% in the four year period measured and by 11.2% the following year.# However, Charlotte’s Mecklenburg County actually saw a 28.4% increase in asthma hospitalizations among 0-14 year-olds and a 2.4% increase in total asthma hospitalizations in the measured time period.#

This unexpected increase in Charlotte could be due to any number of factors, including an increase in an asthma irritant other than ozone. It also could be that ozone decreased at the data collection site but increased in Mecklenburg county overall, leading to higher county-wide asthma hospitalization rates. Also worth noting is that construction on rail lines continued in Charlotte and the other peer cities after the first line opened. Construction can be a source of pollutants and may have masked some of the environmental and health benefits of the rail systems.

Overall, the asthma results were inconclusive and it is not clear whether the addition of a rail line would change the Triangle area’s asthma rates. Our data suggests that rail decreases ozone concentration and asthma rates in suburban areas, but effects in city centers are less clear. Further research with smaller sample areas for asthma data in the vicinity of environmental data collection sites would be helpful.

VIII. Results and Conclusion

After conducting our research and analyzing our findings, we have found that a light rail system may not be as beneficial as many people have hoped. In our peer corridors, light rail did not change ozone concentrations nor did it have a large impact on the air quality index of most peer corridors. It did, however, decrease carbon monoxide rates, but only down to 2ppm, and then it leveled out. Because the Triangle already has a CO concentration at 2ppm, implementation of a light rail system may have little impact in this regard. Our research was inconclusive regarding asthma rates; asthma rates may decrease in suburban areas, but we cannot know for sure what its effects are on city centers. Additionally, light rail will likely result in an annual fatality reduction of 1 per 100 million VMT. While this is beneficial, there will only be a net fatality reduction if the rail system’s fatality rate does not add more fatalities than it reduces.

Light rail systems did positively benefit vehicle miles traveled in our peer corridors. Specifically in Sacramento, our data showed a dip in VMT directly after the implementation and expansion of light rail lines. VMT would then begin to climb again, until another expansion occurred. Therefore, we can conclude that light rail decreases VMT in the short term, and slows its growth in the long term. Furthermore, it is possible that light rail systems will affect the shape and development of cities in the future, as its structure can centralize around transit hubs. This would increase access to transit and its ridership while decreasing VMTs at the same time.