Transit Oriented Development

Using Public Transit to Create More Accessible and Livable Neighborhoods

~~~~~~~~~~~~~~

TDM Encyclopedia

Victoria Transport Policy Institute

~~~~~~~~~~~~~~~~~~~~

Updated 18 August 2008

This chapter describes Transit Oriented Development (TOD), which refers to residential and commercial districts located around a transit station or corridor with high quality service, with good walkability, parking management and other design features that facilitate transit use and maximize overall accessibility.

 

 

Description

Transit Oriented Development (TOD) refers to residential and Commercial Centers designed to maximize access by Transit and Nonmotorized transportation, and with other features to Encourage Transit Ridership. A typical TOD has a rail or bus station at its center, surrounded by relatively high-density development, with progressively lower-density spreading outwards one-quarter to one-half mile, which represents pedestrian scale distances. It includes these design features (Morris, 1996):

 

·       The neighborhood is designed for Cycling and Walking, with adequate facilities and attractive street conditions.

 

·       Streets have good Connectivity and Traffic Calming features to control vehicle traffic speeds.

 

·       Mixed-use development that includes shops, schools and other public services, and a variety of housing types and prices, within each neighborhood.

 

·       Parking Management to reduce the amount of land devoted to parking compared with conventional development, and to take advantage of the parking cost savings associated with reduced automobile use (NJDOT, 2007).

 

·       Transit Stops and Stations that are convenient, comfortable and Secure, with features such as comfortable waiting areas, real time vehicle arrival information, venders selling refreshments and periodicals, washrooms, and information.

 

 

Transit Oriented Development generally requires at least 6 residential units per acre in residential areas and 25 employees per acre in Commercial Centers, and about twice that for premium quality transit, such as rail service (Pushkarev and Zupan, 1977; Ewing, 1999; Cervero, et al, 2004; Reconnecting America and the CTOD, 2008). These densities create adequate transit ridership to justify frequent service, and help create active street life and commercial activities, such as grocery stores and coffee shops, within convenient walking distance of homes and worksites. However, other factors are also important beside simple density. Transit ridership is also affected by factors such as employment density and Clustering, demographic mix (students, seniors and lower-income people tend to be heavy transit users), transit pricing and rider subsidies, Parking Pricing and Road Tolls, the quality of transit service, the effectiveness of transit Marketing, walkability, and street design. A particular density may be inadequate to support transit service by itself, but becomes adequate if implemented with a variety of Transit Encouragement and Smart Growth strategies. The assumption that transit cannot be effective except in large cities with high population densities can be a self-fulfilling prophecy, because it results in transport and land use decisions that favor automobile travel over transit.

 

Is It Really TOD? (Patrick Siegman, in Tumlin and Millard-Ball, 2003)

 

What’s the difference between a true transit-oriented development, which will deliver promised social and economic benefits, and a transit-adjacent development? A true TOD will include most of the following:

 

• The transit-oriented development lies within a five-minute walk of the transit stop, or about a quarter-mile from stop to edge. For major stations offering access to frequent high-speed service this catchment area may be extended to the measure of a 10-minute walk.

 

• A balanced mix of uses generates 24-hour ridership. There are places to work, to live, to learn, to relax and to shop for daily needs.

 

• A place-based zoning code generates buildings that shape and define memorable streets, squares, and plazas, while allowing uses to change easily over time.

 

• The average block perimeter is limited to no more than 1,350 feet. This generates a fine-grained network of streets, dispersing traffic and allowing for the creation of quiet and intimate thoroughfares.

 

• Minimum parking requirements are abolished.

 

• Maximum parking requirements are instituted: For every 1,000 workers, no more than 500 spaces and as few as 10 spaces are provided.

 

• Parking costs are “unbundled,” and full market rates are charged for all parking spaces. The exception may be validated parking for shoppers.

 

• Major stops provide BikeStations, offering free attended bicycle parking, repairs, and rentals. At minor stops, secure and fully enclosed bicycle parking is provided.

 

• Transit service is fast, frequent, reliable, and comfortable, with a headway of 15 minutes or less.

 

• Roadway space is allocated and traffic signals timed primarily for the convenience of walkers and cyclists.

 

• Automobile level-of-service standards are met through congestion pricing measures, or disregarded entirely.

 

• Traffic is calmed, with roads designed to limit speed to 30 mph on major streets and 20 mph on lesser streets.

 

 

Transit Oriented Development is a particular category of Smart Growth, New Urbanism and Location Efficient Development. It can do more than simply shift some car trips to transit: it also increases Accessibility and Transportation Options through land use Clustering and mix, and nonmotorized transportation improvements. This reduces the distance required for car trips, allows a greater portion of trips to be made by walking and cycling, and allows some households to reduce their car ownership, which together can result in large reductions in vehicle travel (Land Use Impacts on Transport). This reduces total transportation costs and helps create a more Livable community, in addition to supporting TDM objectives.

 

High-quality transit supports the development of high-density urban centers, which can provide accessibility and agglomeration benefits (efficiencies that result when many activities are physically close together), while automobile-oriented transportation conflicts with urban density because it is space intensive, requiring large amounts of land for roads and parking facilities (Voith, 1998; Boroski, et al, 2002). Large scale Park & Ride facilities tend to conflict with Transit Oriented Development, since a rail station surrounded by large parking lots and arterials with heavy traffic is unlikely to provide a good environment for residential development or pedestrian access. It is therefore important that such facilities be properly located, designed and managed to minimize such conflicts (CLF, 2001).

 

Transit Oriented Development location is a valuable and scarce resource, similar to waterfront property. It tends to increase property values 5-15%, reflecting the direct benefits to residents and businesses of having diverse transportation options, and resulting automobile and parking cost savings (Diaz, 1999; Weinberger, 2001; RICS, 2002; Smith and Gihring, 2003). As a result, such projects can often be funded through “value capture” strategies, in which the costs of improvements are paid through the additional tax revenue or a special Local Improvement District (LID) tax assessment in the affected area (Smith and Gihring, 2003). The development industry is finding that TODs tend to be profitable investments (Reconnecting America, 2004; Hoban, 2005) Improving transit stations and their neighborhoods can be a catalyst for economic development and urban renewal. Railway station surroundings are the “shop window” of a town, a place where many people see what the community has to offer. It is therefore important that such areas be attractive and inviting to visitors.

 

Table 1            Transit Density Requirements (based on Pushkarev and Zupan, 1977)

Mode

Service Type

Minimum Density

(Dwelling Units Per Acre)

Area and Location

Dial-a-Bus

Demand response serving general public (not just people with disabilities.

3.5 to 6

Community-wide

“Minimum” Local Bus

1/2-mile route spacing, 20 buses per day

4

Neighborhood

“Intermediate” Local Bus

1/2-mile route spacing, 40 buses per day

7

Neighborhood

“Frequent” Local Bus

1/2-mile route spacing, 120 buses per day

15

Neighborhood

Express Bus – Foot access

Five buses during two-hour peak period

15

 

Average density over 20-square-mile area within 10 to 15 miles of a large downtown

Express Bus – Auto access

Five to ten buses during two-hour peak period

15

Average density over 20-square-mile tributary area, within 10 to 15 miles of a large downtown

Light Rail

Five minute headways or better during peak hour.

9

Within walking distance of transit line, serving large downtown.

Rapid Transit

Five minute headways or better during peak hour.

12

Within walking distance of transit stations serving large downtown.

Commuter Rail

Twenty trains a day.

1 to 2

Serving very large downtown.

This table, based on research by Pushkarev and Zupan (1977), indicates typical residential densities needed for various types of transit service. Such requirements are variable depending on other geographic, demographic and management factors.

 

 

Table 1 summarizes residential densities required for various types of transit services. These thresholds are guidelines that reflect “average” conditions and are highly variable depending on various factors, such as:

 

  • Service quality. Improved Transit Service Quality (more comfortable vehicles and waiting areas, more frequent service, better user information, HOV Priority) increases ridership and reduces density requirements.

 

  • Transit service pricing. Lower fares and wider distribution of passes (for example, by neighborhood UPass programs, through which all residents pay for a transit pass through their property taxes) increases ridership and reduces density requirements.

 

  • Demographics. Lower-income, students, seniors and disabled populations ride transit more than average and so reduces density requirements.

 

 

 

  • Walkability. More Walkable neighborhoods and commercial centers increase the area conveniently accessible to transit and therefore reduce density requirements.

 

  • Marketing. Targeted Marketing can increase transit ridership and reduce density requirements.

 

 

For example, Light Rail service may normally require a density of 9 units per acre within 1/4-mile of the rail line, but this may be reduced to 5 units per acre if the area is very walkable, a major portion of employed residents have Commute Trip Reduction Programs at their worksites that include financial incentives (such as priced parking or significantly subsidized transit passes), transit service quality is high, and if the transit agency applies affective marketing programs.

 

 

How It Is Implemented

Transit Oriented Development can consist of new urban transit lines and stations, new suburban neighborhoods designed around public transit stations, and incremental changes to existing urban neighborhoods that have public transit. PBQD (1996) and Robert Cervero, et al, 2004 describe Transit Oriented Development planning practices. Morris (1996), ARC (2001), Nelson/Nygaard (2002) and Dittmar and Ohland (2004) describe specific changes to zoning laws and policies to encourage TOD. Christopher (2007) describes land use policies to support bus transit.

 

 

Travel Impacts

Successful Transit Oriented Development can significantly reduce per capita motor vehicle travel, as discussed in the chapter on Land Use Impacts. See Kittleson & Associates (1999), Rood (1999), Cervero, et al. (2004), Tumlin, et al (2005), Evans and Pratt (2007) and Gard (2007) for additional information on how TOD affects travel patterns. 

 

Dill (2006) found that 30% or more of Portland area Transit Oriented Development residents commuted by MAX (the regional light rail system) at least once a week and 23-33% used transit as their primary commute mode. This compares to less than 10% of workers in the automobile-oriented suburbs of Hillsboro and Beaverton, and 15% of Portland workers. Overall, transit commuting increased when people moved to TODs. Nearly 20% of the commuters switched from non-transit to transit modes and 4% did the opposite, for a net of about 16%. Evans and Pratt (2007) summarize extensive research on the effects of TOD on travel. They found:

·       In Portland, Oregon, as of 1995, the average central area TOD transit share for non-work travel was roughly four times that for outlying TODs, which in turn had over one-and-two-thirds times the corresponding transit share of mostly-suburban, non-TOD land development.

·       In Washington DC, work-commute transit mode shares decline from 75% at downtown office buildings right at Metrorail stations to just over 10% on average at office buildings within roughly 1/2-mile of a station but located in the suburbs outside of the Capital Beltway. Transit mode shares along the Washington Metro system were found to decrease by 7 percentage points for every 1,000 feet of distance from a station in the case of housing and by 12 percentage points in the case of office worker commute trips.

·       A 2003 California TOD travel characteristics study found TOD office workers within 1/2 mile of rail transit stations to have transit commute shares averaging 19% as compared to 5% regionwide. For residents, the statewide average transit share for TODs within 1/2 mile of the station was 27% compared to 7% for residences between 1/2 mile and 3 miles of the station.

·       TOD residents are generally associated with lower automobile ownership rates. For example, auto ownership in three New Jersey “Transit Village Areas,” for example, averaged 1.8 vehicles per household compared to 2.1 outside the transit villages.

 

 

A parking and traffic generation study of Portland, Oregon transit oriented developments recorded 0.73 vehicles per housing unit, about half the 1.3 value in the ITE Parking Generation Handbook, and it recorded 0.15 to 0.29 vehicle trips per dwelling unit in the AM period and 0.16 to 0.24 vehicle trips per dwelling in the PM period, about half the 0.34 AM and 0.38 PM values in the Trip Generation Handbook (PSU ITE Student Chapter 2007).

 

Using a regression model that accounts for various demographic and geographic factors, Bailey (2007) found that household located within ¾-mile of high-quality public transportation service average of 11.3 fewer daily vehicle-miles (a 26% reduction), regardless of land use density and vehicle ownership rates. The table below shows how land use affects vehicle ownership, daily mileage and mode split in the Portland, Oregon region. Transit-Oriented Neighborhoods, which have both good transit and mixed land use, have far lower vehicle ownership and use, and far higher rates of walking and public transit than other parts of the region.

 

Table 2            Land Use Impacts on Vehicle Ownership and Travel (Ohland and Shelley Poticha, 2006)

Land Use Type

Auto Ownership

Daily VMT

Mode Split

 

Per Household

Per Capita

Auto

Walk

Transit

Bike

Other

Good transit/Mixed use

0.93

9.80

58.1%

27.0%

11.5%

1.9%

1.5%

Good transit only

1.50

13.28

74.4%

15.2%

7.9%

1.4%

1.1%

Remainder of county

1.74

17.34

81.5%

9.7%

3.5%

1.6%

3.7%

Remainder of region

1.93

21.79

87.3%

6.1%

1.2%

0.8%

4.0%

Residents of transit-oriented neighborhoods tend to own significantly fewer motor vehicles, drive significantly less, and rely more on walking and public transit than residents of other neighborhoods.

 

 

Community design features of TODs also affect non-commute travel mode choice. There were significant differences between respondents in the different neighborhoods in the share that walk and take transit to non-commute destinations. However, few respondents take transit to non-commute destinations on a regular basis. In most cases, less than ten percent of the respondents used transit to non-commute destinations on a weekly basis.

 

These higher rates of transit and walking travel may partly reflect self selection. Many of the residents of the TODs, particularly those that commute by transit, placed a high importance on transit and walking accessibility when choosing their home. Many also prefer walking and transit to driving and agree with “pro-environment” statements. Even if self-selection explains a large share of the effects on mode choice, this should not detract from the finding that these developments are providing a desired housing option that facilitates such choices.

 

Kuby, Barranda and Upchurch (2004) evaluate the effects of local station conditions on light rail transit ridership in U.S. cities. They find that local accessibility factors are important, including employment, population, portion of renters, bus lines, airports, park-and-ride spaces and centrality. They calculate that, on average, each 100 jobs leads to 2.3 daily boardings, each 100 residents to 9.3 daily boardings, each 100 park-and-ride spaces leads to 77 boardings, each bus to 123 boardings, and an airport to 913 boardings. Similarly, Cervero, et al. (2004) develop a model for predicting the effects of increased residential and commercial density, and improved walkability around a station on transit ridership. For example, increasing residential density near transit stations from 10 to 20 units per gross acre increases transit commute mode split from 20% to 24%, and up to 28% if implemented with pedestrian improvements. Krygsman, Dijst and Arentze (2004) and Pushkarev and Zupan (1997) identify the distribution of access/egress trip times (which average about 6 minutes for bus and tram, and 10 minutes for trains), indicating acceptable TOD service areas.

 

Renne (2005) found that in major U.S. metropolitan regions transit commuting decline dramatically during the last three decades (from 19% in 1970 to 7.1% in 2000), but in the 103 TODs within those regions transit commuting increased from 15% in 1970 to 17% in 2000, an 11% growth rate. The percentage of transit commuting was over three times higher in TODs compared to averages for maturing – heavy rail regions and over twice as much for TODs in new start – light rail regions. TODs in Portland, Oregon, and Washington D.C., which have aggressive policies to promote transit, have experienced even greater ridership growth (58% for both). Households in TODs also owned fewer vehicles: only 35% of TOD households own two or more vehicles compared with 55% in metropolitan regions overall, although TOD residents have slightly higher average incomes.

 

A study of neighborhoods around SkyTrain rail transit stations in Vancouver, BC found that households located within 300 metres of a station owned about 10% fewer vehicles on average than households located more than 1,000 meters from the station, and that average household vehicle ownership is 31% lower than at suburban locations a few miles away (Bunt and Joyce, 1998). Of course, this could partly reflect self-selection (households that own fewer than average automobiles choose to live in such areas), but there is evidence that many residents actually reduce their vehicle ownership when they move to such areas. A study of Orenco Station, a New Urbanist community on Portland's Westside MAX light rail line found that 22% of the residents commute by public transit, far higher than the 5% average for the region, and 69% use public transit more often than they did in their previous community (Podobnik, 2002). Bento, et al, (2003) find that, in cities with rail transit services, a 10% reduction in the average distance between homes and rail transit stations reduces VMT about 1%.

 

Beaton (2006) found that in the Boston region, rail transit zones (areas within a 10-minute drive of commuter rail stations) had higher land use density, lower commercial property vacancy rates, and higher transit ridership than other areas. Regional transit ridership declined during the 1970s and 80s (it has rebounded since 1900), but declined significantly less in rail zones, indicating that TOD increases ridership compared with what would otherwise occur. In 2000, transit mode split averaged 11-21% for rail zone residents, compared with 8% for the region overall. Areas where commuter rail stations closed during the 1970s retained relatively high transit ridership rates, indicating that the compact, mixed land use patterns that developed near these stations has a lasting legacy. Land use density did not increase near stations built between 1970 and 1990, but did increase near stations build after 1990. This can be explained by the fact that the value of smart growth development (using land use policies to create more compact, mixed, multi-modal land use) only became widely recognized in the 1990s, and much of the research and literature on transit oriented development is even more recent (Cervero et al, 2004).

 

Lund, Cervero and Willson (2004) found that residents living near transit stations in various California cities are around five times more likely to commute by transit as the average resident worker in the same city. Various factors influence transit ridership rates. TOD residents are more likely to use transit if there is less of a time benefit for traveling via highways (compared to transit), if there is good pedestrian connectivity at the destination, if they are allowed flexible work hours, and if they have limited vehicle availability. TOD residents are less likely to use transit if the trip involved multiple stops (or “trip chaining”), if there is good job accessibility via highways, if they can park for free at their workplace, and if their employer helps to pay vehicle expenses (such as tolls, fuel, etc.). Physical design factors such as neighborhood design and streetscape improvements show some influence in predicting project-level differences, but have relatively minor influences on transit choice among individual station area residents.

 

Reconnecting America (2004) studied demographic and transport patterns in “transit zones,” defined as areas within a half-mile of existing transit stations in U.S. cities. It found that households in transit zones own an average of 0.9 cars, compared to an average of 1.6 cars in the metro regions as a whole. These lower rates of car ownership near transit may be by choice rather than poverty: car ownership rates near Metro stations in Arlington County are much lower than in the region as a whole, while average household income is higher than the regional average.

 

This study also found that automobile travel is also much lower in transit zones. Only 54% of residents living in transit zones commute by car, compared to 83% in the regions as a whole. More residents commute by car in the regions with small and medium-sized systems (72% and 77%, respectively) than in the large and extensive systems (65% and 49%, respectively). The regions with the lowest percentage of residents commuting by car are New York (36%), Washington D.C. (54%), and Seattle (54%). The regions with the highest percentage of residents commuting by car are Memphis (86%), Dallas (86%), Tampa (79%) and Sacramento (89%) — all systems with newer, smaller fixed-guideway transit networks. The size of the transit system seems to be a significant determinant of whether or not residents commute by car, with more transit ridership in cities with larger rail transit systems.

 

Schlossberg, et al. (2004) describe methods of evaluating transit oriented development, taking into account urban form, pedestrian accessibility, transit usage, and socio-demographic change before and after transit-oriented development in two U.S. cities. They find that many transit stations are not optimally located to maximize pedestrian access, and that automobile-oriented streets (wide, with heavy and fast traffic) can create a significant barrier to walking. 

 

One major study predicted that Transit Oriented Development would reduce single-occupant vehicle commuting by 22.5%, increase transit and nonmotorized travel by 27%, and reduce congestion 18% compared with increasing highway capacity (1000 Friends, 1997). Another study predicts that TOD reduces automobile travel by 20-25% compared with conventional development (Cambridge Systematics, 1992). The table below indicates how land use design features typically reduce per capita vehicle trips in an area.

 

Table 3            Travel Impacts of Land Use Design Features (Dagang, 1995)

Design Feature

Reduced Vehicle Travel

Residential development around transit centers.

10%

Commercial development around transit centers.

15%

Residential development along transit corridor.

5%

Commercial development along transit corridor.

7%

Residential mixed-use development around transit centers.

15%

Commercial mixed-use development around transit centers.

20%

Residential mixed-use development along transit corridors.

7%

Commercial mixed-use development along transit corridors.

10%

Residential mixed-use development.

5%

Commercial mixed-use development.

7%

 

 

Land use patterns at both origins and destinations affect travel behavior. Employees who work in areas with high employment densities, good pedestrian conditions and attractive urban environments with shops and restaurants nearby are more likely to commute by transit and rideshare use (Davidson, 1994; Evaluating Nonmotorized Transport).

 

Table 4            Travel Impact Summary

Travel Impact

Rating

Comments

Reduces total traffic.

3

Reduces per capita vehicle travel.

Reduces peak period traffic.

2

“

Shifts peak to off-peak periods.

0

 

Shifts automobile travel to alternative modes.

3

Encourages transit and nonmotorized travel.

Improves access, reduces the need for travel.

3

Increases density and land use mix.

Increased ridesharing.

0