CONNECTIVITY ORDINANCES IN OREGON MUNICIPALITIES IMPACTS OF STREET LAYOUT REGULATION IN RESIDENTIAL SUBDIVISIONS Brandon Pike Professional Paper In fulfillment of a Master of Community and Regional Planning University of Oregon | Spring 2018 Faculty Committee Chair: Marc Schlossberg, PhD Second Reader: Yizhao Yang, PhD This page intentionally left blank. Brandon Pike 1 Table of Contents Acknowledgements ............................................................................................................... 4 Abstract .................................................................................................................................... 5 Keywords ......................................................................................................................................... 5 Chapter 1: Introduction & Previous Study ........................................................................ 6 Purpose ............................................................................................................................................ 6 Previous Study ................................................................................................................................ 7 Background ..................................................................................................................................... 9 Research Questions ..................................................................................................................... 10 Chapter 2: Methodology ..................................................................................................... 11 Answering the Research Question ........................................................................................... 11 Measurement & Data .................................................................................................................... 13 Community Profiles ..................................................................................................................... 15 Summary .................................................................................................................................................. 15 Housing .................................................................................................................................................... 15 Income ...................................................................................................................................................... 15 Limitations ..................................................................................................................................... 16 Chapter 3: Findings ............................................................................................................. 18 Findings Summary ....................................................................................................................... 18 Connectivity Ordinances & Intersection Density ................................................................... 18 Street Connectivity & Transit ..................................................................................................... 21 Intersections .................................................................................................................................. 23 .......................................................................................................................................................... 23 Land Use Zoning & Street Connectivity ................................................................................... 24 Effects of 2007-2008 Financial Crisis on Oregon Residential Development ..................... 25 Chapter 4: Conclusions & Recommendations ............................................................... 26 Conclusion Summary .................................................................................................................. 26 Connectivity ordinances of the three study cities ................................................................. 26 Zoning, Population Density, & Intersection Density .............................................................. 27 Future Research ........................................................................................................................... 27 Recommendations & Evaluations ............................................................................................. 29 Appendices ............................................................................................................................ 31 Appendix A: Definitions .............................................................................................................. 31 Brandon Pike 2 Appendix B: Residential Subdivisions ..................................................................................... 32 Appendix C: Connectivity Ordinance By City ......................................................................... 33 Appendix D: Paths & Trails in Study Cities ............................................................................. 36 Appendix E: Residential Zones by City, Arranged by Similar Housing Density Levels . 37 Appendix F: Notes on Study Cities’ Layout ............................................................................ 38 Appendix G: Topography ............................................................................................................ 38 Appendix H: Census Data for Study Cities.............................................................................. 40 Appendix I: Population of Study Cities Over Time ................................................................ 42 Appendix J: References .............................................................................................................. 43 Brandon Pike 3 ACKNOWLEDGEMENTS I would like to offer a sincere thank you to Marc Schlossberg, PhD and Yizhao Yang, PhD for guiding me through this process as my committee members. I would also like to thank the many staff members at the City of Beaverton, City of Bend, City of Hillsboro, Washington County, Deschutes County, and Oregon Metro for their assistance in gathering the data that made this project possible. Thank you to my wife, Laura, for being patient and supportive as I have pursued higher education. Finally, I wish to thank the members my Master of Community and Regional Planning 2018 cohort for their continued support and advice. To Seth Thompson for his hospitality and Rachel Hiller for her GIS assistance: I may not have made it through this project, let alone this master’s program, without your support. If you have questions about this research or would like to discuss my findings, feel free to contact me at brandonrjpike@gmail.com. Brandon Pike 4 ABSTRACT Street connectivity ordinances influence development practices and the built environment within cities, dictating street layout for the foreseeable future. Cities in Oregon, a state with a robust statewide planning program and stated goals which include urban growth boundaries that regulate development in urban areas, utilize a number of strategies to regulate street layout and connectivity. This study examines both the effects those strategies have had on the built environment and how effective or ineffective they have been over time in three Oregon cities: Beaverton, Bend, and Hillsboro. Bend adopted new connectivity ordinances in 2006, offering a chance to research street connectivity before and after that point. This study’s findings indicate that, after adoption, Bend’s new ordinances worked to moderately increase intersection density, one of the most widely-used metrics for measuring street connectivity. This occurred alongside intersection density levels which decreased in the other two cities over the study period. This study also addresses the greater context of what those policies and their outcomes mean to urban areas and their residents long-term. Finally, the study presents a handful of other findings that appeared in the research related to transit and street connectivity, as well as zoning and street connectivity. KEYWORDS Street connectivity; block size; urban mobility; street network; street layout; active transportation; walkability Brandon Pike 5 CHAPTER 1: INTRODUCTION & PREVIOUS STUDY PURPOSE Street connectivity influences active transportation levels in residential neighborhoods (Berrigan, Pickle, and Dill, 2010), which, in turn, influences greenhouse gas emissions (GHGs) in urban areas. Street connectivity also impacts transit availability and access to social services (Badia, Estrada, & Robusté, 2016). Neighborhoods with low levels of street connectivity have also been shown to lead to populations that lead more sedentary lifestyles than neighborhoods with street networks that have high connectivity, translating into additional health problems for those populations (Koohsari, et al. 2017). Oregon cities remain among the top housing-constrained cities in the US, with some real estate firms placing Portland and Eugene in the 10 cities with the highest housing shortages (Pan, 2017). This occurs while Oregon maintains high growth rates compared to the US as a whole (US Census, 2017; Population Research Center, 2017), and added over 310,000 people between 2010 and 2017—an 8.1 percent increase (State of Oregon, 2017). The form that additional development takes, then, will have lasting affects in the region for a growing number of people. The State of Oregon has an active interest in limiting suburban sprawl with their use of urban growth boundaries; accordingly, it is important that policy-makers and planners in the state know what their decisions lead to in terms of residential development and viable transportation infrastructure for all types of travel. As more people move into the region, cities and neighborhoods that have high street connectivity will benefit economically due to their ability to support flexible uses and nearby destinations (Ellickson, 2012). Additionally, efficient use of space is likely a top priority for cities who have seen rapid growth and increased traffic. In light of these long-term environmental, social, economic, and public health-related factors that are related to street connectivity, studies which examine policies that influence connectivity of city streets are valuable to professionals in a variety of fields, from city planning, transportation engineering, and public administration to private developers and affordable housing advocates. Knowing more about the long-term effects of connectivity ordinances can allow policy makers to make more informed decisions regarding future development. Brandon Pike 6 PREVIOUS STUDY This section summarizes findings of literature related to street connectivity. As previously mentioned, past research points to a number of benefits that arise from areas with well-connected street systems, such as public health and equity, and environmental, economic, and transportation-related impacts. Some existing literature focuses on street connectivity as it relates to social equity. Van der Kloof, Bastiaanssen, and Martens (2014) found that, while bicycle access can increase mobility of those most likely to experience transport-related social exclusion and accessibility barriers, it does not necessarily translate to users who are able to access the services they need. They conclude that having the proper infrastructure in place can help bridge that gap. High levels of street connectivity, then, can increase access to services for the most vulnerable members of society. Areas with high levels of street connectivity can also help lower obesity rates for vulnerable populations (Wang, Wen, and Xu, 2013). One particular marginalized group that can benefit from well-connected street layouts are people with disabilities. Those with mobility and visual impairments often rely on public transportation for their daily transport. Public transit systems are usually more economically viable and better able to serve their citizens in areas with high street connectivity (Badia, Estrada, and Robusté, 2016). Once people with mobility and visual impairments arrive in their destination’s immediate area, they deserve infrastructure that safely supports them, such as well-maintained sidewalks, and audible and tactile cues at crosswalks. Thompson (2013) found that maintaining the infrastructure in neighborhoods with traditional development and well-connected streets is significantly cheaper for municipalities over time. All this suggests that people with disabilities stand to benefit from having neighborhoods with well-connected street networks, and that those neighborhoods are more affordable to maintain over time. Well-connected street layouts also lead to more available destinations within walking distance (Koohsari, et al., 2017; Ozbil, Peponis, and Stone, 2011). One of the key benefits of a well-connected street layout, then, is the increased appeal of walking and bicycling for transport, which becomes easier since travel distances are shorter and routes tend to be less complicated in terms of wayfinding (Kulash, Anglin, and Marks, 1990). Along with pedestrian and bicycle-related impacts, street connectivity influences transit access and viability. Partially as a result of the drop in fuel prices and recent increased usage of ride-hailing technologies like Uber and Lyft, major transit agencies are having to rethink their routes to remain viable and better serve their constituents (Gunda & Atluri, 2017), especially those who traditionally have not had the level of transit service Brandon Pike 7 of cities such as New York City or Chicago. Badia, Estrada, and Robusté (2016) found that neighborhoods with grid patterns (and, consequently, more connected street networks), were more likely to experience success when redesigning their bus networks. Similarly, in order for a corridor to support transit, it should have at least eight housing units per acre (Dunham-Jones and Williamson, 2009). Since traditional development and traditional street patterns often leads to higher housing units per acre, areas of new development that boast higher street connectivity can lead to higher housing density, creating more areas that can support transit. The environmental impacts related to street connectivity stem from the lower vehicle miles travelled (VMT) that results from cities with high connectivity (Koohsari, Owen, Cerin, Giles-Corti, & Sugiyama, 2016). Street networks that provide more direct paths between destinations lead to shorter trips, and, therefore, fewer GHGs released into the atmosphere (2016). Finally, measuring street connectivity can prove challenging. Within the literature, it is often measured using the street connectivity index or intersection density, which is calculated by dividing the number of intersections in a location by its area unit (Tresidder, 2005; Handy, Patterson, and Butler 2003). Additionally, there are metrics that gauge walkability, such as the Pedestrian Catchment Area (PCA) and the Center for Disease Control and Prevention’s Walkability Audit—the latter being an qualitative analysis of street features, while the former creates a buffer around a given point, and uses network analysis to record nodes and length of street segments. Tresidder (2005) presents a variety of connectivity metrics, including intersection density, street density, connected node ratio (CNR), average block length, and the Gamma and Alpha indices. With the data available and the scope of this study, it was decided intersection density would be the most appropriate and efficient metric available for analysis. Brandon Pike 8 BACKGROUND In light of increased research related to the benefits of well-connected street networks, many cities have begun adopting connectivity ordinances in recent years that require developers to meet minimum standards for things like block length and perimeter size (Stangl, 2015). Of the three study cities (see Methodology), Beaverton and Hillsboro have had some form of connectivity ordinances in their codes since at least the early 1990s. Bend adopted block length and block perimeter size ordinances in late 2006. This provides windows in which to examine street connectivity in Bend adopted block length and Bend both before and after they adopted connectivity ordinances, block perimeter size ordinances while comparing those results with in late 2006. To measure the the same analyses of Beaverton impact of those ordinances, this and Hillsboro, whose ordinances study looks at development in remained the same during the two windows of time: ten years study’s timeframe. Accordingly, before and ten years after were this timeframe is broken into two 10-year periods: 1997 to 2006, and put into place. 2007 to 2016. Urban growth boundaries (UGBs) were instituted by the Oregon State legislature in 1973, stemming from the landmark land-use legislation SB 100. Since that time, various state agencies have played roles in limiting development to urban areas in the state, as opposed to developing on fertile agriculture land far away from urban centers. This is worth noting, as research has shown UGBs can limit the amount of suburban development that occurs altogether (Song & Knaap, 2004), and, in turn, UGBs can influence the street layout of new development. Rather than strictly limiting development to achieve a certain level of street connectivity, cities often regulate other aspects of street layout and design in order to influence connectivity (Duany and Talen, 2002). These policies, often called connectivity ordinances, which can also have the intended effect of providing sufficient emergency vehicle access, often include regulations of block length, block perimeter size, and the presence and/or length of cul-de-sacs. In order to understand how connectivity ordinances influence the built environment in Oregon cities, this study aims to answer the following questions. Brandon Pike 9 RESEARCH QUESTIONS Have Bend’s street connectivity ordinances in residential subdivisions led to an increase in street connectivity? And, more broadly, what effects have these ordinances had on the built environment in Oregon cities? Brandon Pike 10 CHAPTER 2: METHODOLOGY ANSWERING THE RESEARCH QUESTION The broad goal of this study is to explore the effectiveness of street connectivity policies in Oregon. To narrow the scope, the 10 largest Oregon cities by 2016 population were selected (Population Research Center, 2017), then cities of a similar size which saw increases in population during the same periods were chosen (see appendix I). This produced a group of four Oregon cities: Beaverton, Bend, Hillsboro, and Medford. Finally, the goal was to examine cities which took different approaches to regulating street connectivity in recent decades. Hillsboro appeared to take the most stringent approach, Beaverton the least, and Bend and Medford, who take relatively similar approaches, fell somewhere in the middle, with their block length and cul-de-sac regulations being virtually identical. Medford did not have their data collected in a way that worked in the study’s analysis, and since their approach to regulating street connectivity is similar to Bend, they were omitted from the final study group. Table 1 shows a brief summary of these cities’ street connectivity ordinances. See appendix C for a detailed list of the study cities’ connectivity ordinances. It should first be noted that, while the policies examined in this study are commonly found in other Oregon cities’ development codes, there are other types of street connectivity policies that may be more effective. For various reasons—they may be considered too drastic or too hindering to development in some cities, for instance— these policies are not as common as the policies analyzed in this study. Perhaps the most obvious approach to maintaining a certain level of street connectivity in new development is to require developers to build a street network that adheres to a street connectivity level of a certain value, or, more simply, to adhere to a grid layout. Further study could to be conducted to discover why exactly cities hesitate to establish such policies outside central business districts (see Future Research section), though it is not hard to imagine the potential public pushback that could occur if attempts were made to limit residential development in such ways. With that in mind, the following paragraph summarizes the general connectivity ordinances found across the state. There are three main types of policies related to street connectivity commonly found in development codes, with any combination of the three in a given city’s code, plus other less common ordinances. First, cities can regulate block length and block perimeter size. They tend to have different requirements depending on the type of zoning in place—commercial, residential, and industrial, for example. For the purposes of this study, only residential block length requirements were examined. This is because other types of zones, such as commercial and industrial, have widely-varying ordinances Brandon Pike 11 which are often different than residential zones. The decision was made to focus solely on residential areas so that valid comparisons could be made. Second, cities can regulate cul-de-sacs—namely, whether they are permitted and/or their length. Third, cities can regulate new subdivisions based on their internal street connections to existing development. This can be implemented in a variety of ways, as cities can be as strict or as lenient as they wish to be. For example, Hillsboro requires developers to outline how their development will connect to existing streets, and deviation from a well- connected network requires a comprehensive explanation from the developer for why that layout is needed. Beaverton, on the other hand, only requires developers to include an accessway for pedestrians and bicyclists between their development if block lengths exceed 600 feet. Portland and Eugene, two of Oregon’s three largest cities, employ a handful of perhaps more stringent methods in regulating street connectivity. Instead of focusing on those cities’ methods for regulating street connectivity, which have both been well-researched (Tresidder, 2005; Handy, Paterson, and Butler, 2003; Metropolitan Service District, Street Design Work Team, et al., 1997), it was decided that this study would examine how small to mid-sized cities in Oregon regulate street connectivity. This is in hopes that similarly-sized cities that are facing an increase in development activity could be able to use the information presented in this study to make informed decisions regarding connectivity ordinance adoption. Table 1 | Summary of Street Connectivity Ordinances by City Beaverton Bend Hillsboro Block Length 600* 660 530 to 600** Block Perimeter None 2000 1800 to 2750** Cul-de-Sacs Accessway may be required Discouraged Discouraged Cul-de-Sac: Length None None 450 Brandon Pike 12 Connectivity Analysis No No Yes Required? * See appendix C for exceptions ** Depending on proximity to transit infrastructure The process of isolating cities in Oregon that saw development during the same eras was intended to control for market forces that affect street connectivity—in other words, subdivisions built during the 1950s, for instance, tend to have different street layouts than those from the 1980s, oftentimes without the influence of street connectivity policies (Handy, Paterson, and Butler, 2003). This process isolated three cities that took different approaches to street connectivity in new residential development, with the intent to show how effective each city’s approach has been over time. The years examined in this study are 1997 to 2017—ten years before and ten years after there were changes in Bend’s development codes related to connectivity ordinances. Both Beaverton and Hillsboro’s ordinances have remained the same over the study period, both implementing their policies prior to 1997. Beaverton and Hillsboro, then, provide a control group for observing changes in Bend street connectivity over this time period. MEASUREMENT & DATA Street connectivity analysis for the three study cities was conducted in two ways: 1. Conducting a policy review, which examined the ordinances related to cul-de- sacs, block lengths, and block perimeter sizes in the three cities, and 2. Measuring intersection density. The policy review revealed that, as previously stated, Hillsboro had the strictest connectivity ordinances over the study period, while Bend’s ordinances (adopted 2006) were moderately strict, and Beaverton’s were the least strict. Along with stricter ordinances related to block length, perimeter, and cul-de-sacs than the other cities, Hillsboro also requires developers to submit what they call a connectivity analysis for developments with proposed internal streets (see table 1). This adds another level of oversight by the City, which is likely resource-intensive for both the developer and the planning agency to prepare and review, respectively. Next, data was obtained in order to measure the effectiveness of the ordinances over time—data such as subdivisions, zoning, and city limits from the appropriate city, county, and regional government agencies. This allowed for the ability to examine each Brandon Pike 13 subdivision in the study cities, as well as what land use zone it belonged to and when it was developed. The next step involved locating each intersection in the study cities. Using street layer data from the State of Oregon, line intersection analysis was performed, producing nodes at each street intersection and intersections between streets and paths. During intersection density analysis, roundabouts were treated as a standard single intersection. The only exception was if the center of the roundabout contained a destination with access points (see Compass Park in Bend for an example). Next, the street connectivity was measured inside subdivisions that fell within the bounds of the study’s timeline. Some judgement was required when sifting through the subdivision data. This study intended to examine street connectivity of typical residential subdivisions, so certain parcels and developments were eliminated from the analysis, such as the Broken Top Club golf course neighborhood and Mount Bachelor Village Resort on the western edge of Bend. While zoned single-family (RS) and platted within the timeframe of the study’s bounds, those uses necessitate entirely different street layouts than standard residential neighborhoods and are outright prohibited in many residential zones. This decision was to allow for this study to be applied to a typical residential neighborhood. Other than a small handful of similar situations, every residential subdivision platted between 1997 and 2016 for Beaverton, Bend, and Hillsboro were included in the study. Additionally, partitions and small subdivisions that did not include new street creation were excluded from the study, unless they were part of a larger subdivision taking place over time. This was accomplished through data analysis that required partitions and small subdivisions to meet two criteria in order to be included in the study: 1. The partition or small subdivision’s name needed to closely match that of the larger subdivision—“Arbor Roses” and “Arbor Roses No.2”, for an example. 2. The partition or small subdivision needed to lie directly adjacent to the larger subdivision. The data from all three cities needed significant attention, as duplicate subdivisions needed to be deleted or consolidated. Consideration was taken for when the area was originally platted, so that each subdivision was placed into the year it was first on record as being platted with Washington County (for Beaverton and Hillsboro) or Deschutes County (for Bend). This process of filtering the data created a grouping of all residential subdivisions in the study cities which occurred between 1997 and 2016, did not occur within the larger bounds of a golf course or resort-type development, and included the creation of streets. Brandon Pike 14 COMMUNITY PROFILES Summary Two United States Censuses were conducted during this study’s timeframe, occurring during the fourth year of both 10-year periods of the study—the 2000 Census during the 1997-2006 period and the 2010 Census during the 2007-2016 period. This provides detailed snapshots in data form of Oregon, Beaverton, Bend, and Hillsboro during the study. See appendix H. All study cities saw considerable population and housing growth over the study period, but Bend’s growth was most pronounced. The following subsections summarize changes in the cities and Oregon as a whole over the study’s timeline. Note: All data presented in this section are from the 2000 and 2010 US Census and the 2012-2016 American Community Survey 5-Year Estimates, unless otherwise noted. Housing Between 2000 and 2016, Beaverton added around 7,700 housing units, increasing 19.3 percent from 32,500 to 40,267. Bend saw the highest rate of housing growth and added close to 15,000 units, increasing from 22,507 to 37,406—just shy of a 40 percent increase in total units over a 16-year period. Hillsboro added over 11,000 units, going from 27,211 to 38,495—a 29.3 percent increase. Table 2 summarizes the change in housing units over time for the study cities. Table 2 | Change in Housing Units Over Time Oregon Beaverton Bend Hillsboro 2000 1,452,709 32,500 22,507 27,211 2010 1,675,562 39,500 36,110 35,487 2016 (Estimate) 1,706,290 40,267 37,406 38,495 Percent Change 14.9% 19.3% 39.8% 29.3% Income Hillsboro residents make around $10,000 to $17,000 more per year than Beaverton, Bend, and Oregon residents as a whole (see table 3). However, the median home price in Hillsboro is at least $30,000 less than Beaverton or Bend, and $3,500 more than Oregon as a whole. Consequently, while Beaverton and Bend’s ratio of income to Brandon Pike 15 median home price were both around 20 percent in 2016, Hillsboro’s was just over 29 percent. This shows that Hillsboro residents must spend less on their housing all while making more per year than Beaverton or Bend residents. Table 3 | Income & Housing Statistics Oregon Beaverton Bend Hillsboro Median Household Income 2000 $40,916 $47,863 $40,857 $51,737 2010 $49,260 $54,885 $53,006 $60,695 2016 (Estimate) $53,270 $59,620 $55,625 $70,180 Percent Change Median Home Price $237,300 $286,200 $271,300 $240,800 Ratio of Income to Median Home Price 22.4% 20.8% 20.5% 29.1% (2016) Sources: US Census: Selected Economic Characteristics; Livability.com LIMITATIONS Criticisms of using intersection density to gauge street connectivity and walkability exist (Haynie, 2016; Knight & Marshall, 2015). However, it remains perhaps the most widely- used method of measuring connectivity, since the scale at which it can be conducted is vast compared with analyses that are site-by-site based, for instance, and is less time- intensive than other large-scale connectivity metrics. The following paragraph explores some of the key criticisms of intersection density analysis. When using intersection density to measure street connectivity, one concern stems from the fact that new developments cannot entirely claim responsibility for intersections along routes that already exist. Unless it is a greenfield development with no existing street network, developers are likely to need to include one or more existing streets in their plans. In other words, it does not make sense to judge a new development by the streets or roads that already may exist within or adjacent to its bounds due to preexisting infrastructure. While not the developer’s choice in design, existing streets in the development instantly have a guaranteed length of roadway, plus whatever new Brandon Pike 16 streets the development includes. This potentially increases the likelihood for developments to include more intersections. However, this criticism, while worth noting, is often rebuffed by researchers (Tresidder, 2005) since developers are still chiefly responsible for the development’s final layout of new streets within the mandates of the regulatory agencies. Additionally, the argument can be made that this potential conflict is cancelled out through regulation, as rules for access and easements are the same regardless of if the current developer has to design around existing street infrastructure or if they have an empty parcel of land (assuming, of course, that they are in the same jurisdiction in either scenario). One potential limitation to this project is the scale at which development occurred during the study’s timeframe. While research is limited on the subject (Morris, 2009), larger developments tend to have different street layouts than small developments, often due to the flexibility in design and layout that comes with developing larger pieces of land. In light of the 2007-2008 economic recession which affected housing development considerably in US cities, it could be posited that street connectivity in Beaverton, Bend, and Hillsboro post-2008 was affected by economic factors as well as the connectivity ordinances in place. To see the drastic changes in the number and sizes of residential subdivision developments in the study cities before and after the Recession, see “Effects of 2007-2008 Financial Crisis on Oregon Residential Development” in Findings. Consequently, along with connectivity ordinances, this drop in development activity could have played a role in these cities’ street connectivity levels over the study period. There is a key distinction to make between street connectivity and walkability and other forms of active transportation. Not all neighborhoods with high levels of intersection density or, more broadly, street connectivity, will be conducive to active transportation. While street connectivity paves the way for that to be possible, the proper infrastructure, such as sidewalks, streetlights, and bike lanes, need to be in place for a neighborhood to be walkable and bike-friendly. A neighborhood with very high street connectivity yet lacking adequate sidewalk or bicycle infrastructure, will not encourage active modes of transportation. Further analysis that includes sidewalk and bicycle lane data—if available—could help correct for this (see Future Research). Finally, the data used in the study had its own limitations, as certain adjustments and calculations needed to be made for each city, including that of subdivision size in the case of Hillsboro and Beaverton. While these calculations were checked for accuracy and confirmed to be within 0.01 acres of actual size, this should be noted. Brandon Pike 17 CHAPTER 3: FINDINGS FINDINGS SUMMARY This chapter presents the findings of the study. A brief summary is listed below: • Bend’s connectivity ordinances adopted in 2006 seem to have increased intersection density in the second 10-year period. See Connectivity Ordinances & Intersection Density. • Hillsboro’s stricter connectivity ordinances have not led to significantly higher levels of intersection density citywide. See Connectivity Ordinances & Intersection Density. • Cities with more high density zoning did not have higher intersection density over the second 10-year period. See Land Use Zoning & Street Connectivity. • While Hillsboro’s ordinances that aim to increase street connectivity in areas close to the MAX light rail corridor have influenced some areas with high connectivity—notably the Orenco new urbanist transit-oriented development (TOD)—this study found numerous examples of developments outside the MAX transit stop buffer zone that achieved the same or even higher levels of intersection density. See Street Connectivity & Transit. • Beaverton’s ordinance that encourages mid-block accessways may have led to development with high internal street connectivity but low external street connectivity (Song and Knaap, 2004). See Connectivity Ordinances & Intersection Density. • The connectivity analysis Hillsboro requires developers to complete does not appear to have led to significantly higher levels of external street connections (Song and Knaap, 2004). See Connectivity Ordinances & Intersection Density. • The 2007-2008 financial crisis had a significant impact on residential development in Oregon, and may have influenced street connectivity indirectly. See Effects of 2007-2008 Financial Crisis on Oregon Residential Development. CONNECTIVITY ORDINANCES & INTERSECTION DENSITY After Bend adopted connectivity ordinances in 2006, intersection density levels in residential subdivisions increased 0.12 intersections per acre during the second 10-year period, going from 0.33 to 0.45. During that period, the other cities saw a decrease in intersections per acre, going from 0.34 to 0.30 in Beaverton and from 0.40 to 0.36 in Brandon Pike 18 Hillsboro. This seems to suggest that Bend’s ordinances were effective in raising connectivity, as the other two cities’ connectivity ordinances remained the same during the study timeline and both of their connectivity levels decreased in the second period, while Bend made changes to their ordinances and their connectivity level increased. Table 4 presents the intersection density findings (see appendix A for intersection definitions). Beaverton and Hillsboro’s connectivity ordinances remained the same during the study periods and their intersection density decreased, while Bend adopted new connectivity ordinances and their intersection density increased. Hillsboro’s strict connectivity ordinances have not led to significantly higher levels of intersection density than the other cities, with Bend’s somewhat more relaxed ordinances in place during the second 10-year period producing a higher level of intersection density. See figure 2 for maps showing all intersections included in the study. Beaverton’s ordinance that encourages mid-block accessways may have a significant impact: creating areas with high levels of internal street and path connectivity but low levels of external street connectivity. Song and Knaap (2004) found that overall density had increased in Washington County, home to Beaverton and Hillsboro, since the 1960s. Their research also suggests that while internal street connectivity improved from the 1990s to the time of their study, external street connectivity decreased during that time. This could be a product of the path system throughout Beaverton, which winds through the residential portions of the city but does not have very many connections to commercial areas (see appendix D). Table 4 | Summary of Residential Subdivisions & Intersections Beaverton Bend Hillsboro Total Residential Subdivisions 1997-2006 75 403 204 2007-2016 19 114 62 Percent Change -74.7% -71.7% -69.6% Brandon Pike 19 Residential Subdivision Area (Acres) 1997-2006 416.34 5560.05 1192.72 2007-2016 96.93 1067.31 294.68 Average Residential Subdivision Area (Acres) 1997-2006 5.55 13.80 5.85 2007-2016 5.10 9.36 4.75 Intersections Street-Street Intersections 1997-2006 121 1817 418 2007-2016 27 476 93 Street-Street & Street-Path Intersections 1997-2006 142 1853 482 2007-2016 29 480 106 Street-Street, Street-Path, & Path- Path Intersections 1997-2006 172 1871 562 2007-2016 39 481 113 Intersections Per Acre Street-Street Intersections Per Acre 1997-2006 0.29 0.33 0.35 2007-2016 0.28 0.45 0.32 Street-Street & Street-Path Intersections Per Acre 1997-2006 0.34 0.33 0.40 2007-2016 0.30 0.45 0.36 Street-Street, Street-Path, and Path-Path Intersections Per Acre Brandon Pike 20 1997-2006 0.41 0.34 0.47 2007-2016 0.40 0.45 0.38 STREET CONNECTIVITY & TRANSIT Of the three study cities, Hillsboro is the only one that has connectivity ordinances that change as subdivisions get closer to transit stops. The regional light rail system, MAX, passes through the city, bisecting Hillsboro from east to west almost perfectly in half. Hillsboro requires subdivisions that occur within ½ mile of MAX stops to adhere to slightly stricter development practices than in some other parts of the city. Within those transit stop buffer zones, residential block perimeter sizes drop 35 percent, from 2750 to 1800 feet, and block lengths drop from 600 to 530 feet. Figure 2 shows subdivisions and intersections alongside the MAX system, with half-mile buffers from the MAX stops shown in purple. Many of the stops adjoin areas that were Figure 1 | Hillsboro Light Rail 7-:.. already heavily Hillsboro Residential Subdivisions, developed before this Intersections, & Light Rail Jones Farm • City Limits study’s timeframe, - Streets • Subdivisions Built 1997 to 2006 such as downtown • Street-Street Intersections: 1997 to 2006 • Subdivisions Built 2007 to 2016 Hillsboro. The Orenco • Street-Street Intersections: 2007 to 2016 - MAX neighborhood was developed at a later date—after Hillsboro’s transit-specific connectivity ordinances were established. Orenco lies along the MAX line and includes the New Urbanist transit- Orenco oriented development (TOD) Orenco Brookwood Station, and has an understandably high intersection density of Brandon Pike 21 0.54 intersections per acre (see figure 1). What is somewhat surprising, however, is that other developments (namely those to the northwest and south of Orenco) that are nowhere near the MAX stop buffer zones and, therefore, have less strict connectivity ordinances in place have levels of intersection density that are as high or even higher than Orenco. The Jones Farm neighborhood in northwest Hillsboro, for example, has 0.56 intersections per acre—slightly higher than the Orenco neighborhood even though Jones Farm is well outside the MAX buffer. Likewise, the Arbor Roses development in southwest Hillsboro has 0.54 intersections per acre. Sixteen of the Brookwood development’s 46 intersections are street-path intersections, as the developer chose to include a number of pathways through the neighborhood. This led to a very high relative density of 0.92 intersections per acre inside Brookwood. Even without including street- path intersections, Brookwood has an intersection density of 0.60 intersections per acre. Table 5 summarizes these neighborhoods/developments. Table 5 | Intersection Density of Neighborhood Examples in Hillsboro Orenco* Jones Farm Arbor Roses Brookwood Intersections 99 55 34 46 Acres 181.7 98.8 62.8 49.9 Street-Street & Street-Path 0.54 0.56 0.54 0.92 Intersections Per Acre *Includes Orenco Station, Orenco Gardens, and Orenco Meadows Brandon Pike 22 Hillsboro Residential Subdivisions & Intersections INTERSECTIONS • City Limits - Streets • Subdivisions Built 1997 to 2006 • Subdivisions Built 2007 to 2016 • Street-Path Intersections: 1997 to 2006 Street-Path Intersections: 2007 to 2016 • Street-Street Intersections: 1997 to 2006 Street-Street Intersections: 2007 to 2016 Figure 2 | Residential Subdivisions & Intersections Bend Residential Subdivisions & Intersections • City Limits Beaverton Residential - Streets Subdivisions & Intersections • Subdivisions Built 1997 to 2006 • Street-Path Intersections: 1997 to 2006 • City Limits • Street-Street Intersections: 1997 to 2006 - Streets • Subdivisions Built 2007 to 2016 • Subdivisions Built 1997 to 2006 Street-Path Intersections: 2007 to 2016 • Street-Path Intersections: 1997 to 2006 Street-Street Intersections: 2007 to 2016 • Street-Street Intersections: 1997 to 2006 • Subdivisions Built 2007 to 2016 Street-Path Intersections: 2007 to 2016 • Street-Street Intersections: 2007 to 2016 ===-~"""-- LAND USE ZONING & STREET CONNECTIVITY It is worth briefly examining the zoning strategies employed by each city, since development type is largely influenced by the zoning in place. Figure 3 shows the amount of land in each residential zone by city. Gray colors represent very low density zones, red colors represent low density, yellow colors represent medium density, and green colors represent high density zones. The darker the shade, the higher the density within its color category. Figure 3 | Area of Residential Zones Beaverton Bend Hillsboro VERY LOW DENSITY HIGH DENSITY Bend has a very small amount of its residential land zoned as high density residential (1.8 percent), with most of its land zoned as either medium or very low density. Beaverton and Hillsboro, on the other hand, have more of their residential land zoned for high density (20 and 13 percent, respectively), with the remainder of their residential land zoned predominantly low or medium density. Neither Beaverton nor Hillsboro have any land zoned below 3.5 units per acre, while Bend has 11.5 percent of their residential land zoned as what’s categorized as very low density for the purposes of this study: between 1 unit per 2.5 to 10 acres. See Appendix E for a detailed summary of the study cities’ residential zones. While Beaverton and Hillsboro both zone for higher population density, Bend achieved higher levels of intersection density in the second 10-year period than the other two cities while maintaining predominantly very low- to medium-density zones. EFFECTS OF 2007-2008 FINANCIAL CRISIS ON OREGON RESIDENTIAL DEVELOPMENT The impact of the 2007- 2008 financial crisis Figure 4 | Residential Subdivisions Platted Over greatly affected Oregon the Study Periods development, as evidenced through declining development Total Residential Subdivisions activity that occurred immediately afterward. 403 Figure 4 shows the reduction in development in the study cities when comparing the first 10- 204 year period with the second. All three cities 114 saw a reduction of 75 between 70 and 75 62 percent in the total 19 number of residential 1997-2006 subdivisions during the second 10-year period. Beaverton Bend As previously mentioned - - - 2007-2016 Hillsboro in the literature review, subdivision size can affect street layout. Since the number of subdivisions platted as well as their average size decreased during the 2007-2016 period, it is likely that this is due to the 2007-2008 financial crisis. Further research would need to be conducted to confirm this (see Future Research). Brandon Pike 25 CHAPTER 4: CONCLUSIONS & RECOMMENDATIONS CONCLUSION SUMMARY This chapter first summarizes the conclusions that can be made about each of the three cities’ connectivity ordinances based on this study’s data analysis. The chapter then presents policy recommendations for future development, and ideas for future related research. 1. Connectivity ordinances can lead to an increase in street connectivity 2. Population density does not equate to intersection density, and vice versa. 3. Connectivity ordinances that become more strict based on proximity to transit can lead to high levels of connectivity. However, some neighborhoods outside the transit buffer zones perform even higher in terms of intersection density, suggesting there is more to consider than just proximity to transit. 4. High density zoning does not appear to lead to high levels of street connectivity by itself CONNECTIVITY ORDINANCES OF THE THREE STUDY CITIES It appears that Bend’s connectivity ordinances increased intersection density in residential subdivisions after being adopted in 2006. Based on this analysis, the residential subdivisions built between 2007 and 2016 had 0.12 more intersections per acre than the subdivisions built between 1997 and 2006. This change is unlikely the result of mere market forces, since Hillsboro and Beaverton’s ordinances remained the same during the study’s timeframe, and both of their intersection density levels decreased in the second 10-year period. This finding suggests that Bend’s implementation of connectivity ordinances worked to increase street connectivity in residential development. Hillsboro’s strict connectivity ordinances, along with their required connectivity analysis, may have led to slightly higher levels of intersection density in residential subdivisions than Beaverton, who takes a more relaxed approach to connectivity ordinances. In two measures Beaverton achieved slightly higher levels of intersection density than Hillsboro: when street-path intersections were included in the first 10-year period, and path-path intersections were included in the second period. This suggests that Brandon Pike 26 Beaverton’s flexible ordinances that encourage accessway paths can produce relatively high levels of connectivity when including street-path and path-path intersections, yet the low street-street intersection density levels from both periods in Beaverton (lower than the other cities saw in either period) may indicate that their ordinances are too lax to produce high levels of street-street intersection density. While path intersections can lead to walkability, vehicular and road-based transit do not benefit from areas with high levels of path-path intersection density in the same way they benefit from areas with high street-street intersection density. ZONING, POPULATION DENSITY, & INTERSECTION DENSITY While Beaverton and Hillsboro’s developments in the twenty years of the study were significantly more dense in terms of population, the second 10-year period saw Bend’s developments achieve a higher level of intersection density while maintaining a much lower population density—Bend has approximately 2,322 people per square mile, Beaverton has 4,795, and Hillsboro has 3,833 (see appendix H). This leads to an important recognition for planners: population and/or housing density does not equate to street connectivity, and more broadly, walkability and transit viability/access. In the same way, high levels of street connectivity do not necessarily equate to population density. Instead, the two can complement each other, with high street connectivity in places of high density working together to benefit the area’s economy, transportation network, and so on. Likewise, a densely-populated area without an accompanying connected street network is unlikely to lend itself to multimodal transportation options. Density * Connectivity A noteworthy conclusion: Population density does not equal street connectivity, and vice versa. FUTURE RESEARCH This study has far-reaching connections between the built environment and the policies that affect it. Accordingly, there are a number of research questions that were prompted Brandon Pike 27 by the research in this report. Table 6 summaries the potential opportunities for continued study. Table 6 | Opportunities for Future Research Research Question Potential Method(s) Potential Data Source(s) Why did Bend’s moderately- Interview planners, developers, strict ordinances work better and various stakeholders from This study’s dataset; people than Hillsboro’s strict ordinances Beaverton, Bend, and Hillsboro involved with planning and to promote street connectivity? to explore why this was the case development in the study cities Using cities from different states with different UGB policies, How does the presence of urban compare cities with both similar growth boundaries influence and contrasting street State, regional, county, and local street connectivity? connectivity policies. This could planning agencies control for both state/regional policy and street connectivity policies themselves. In addition to conducting a similar connectivity analysis, researchers could compare How did the 2007-2008 financial economic factors both leading Municipalities’ economic crisis impact street connectivity up to and immediately after the development departments; US in Oregon cities? financial crisis. These factors Census Economic Data; could include building councils of government (COGs) applications and sizes and types of developments. This could take an approach How likely is it for street stubs to similar to this study; updated one day connect to the greater street and path data would need State, regional, county, and local street network? to be obtained at some point in planning agencies the future and analyzed for a given location(s). What types of businesses exist in these cities? Do the industries Conduct a land-use mix analysis US Census Economic Data, present in a given city affect its of the cities, alongside a review street connectivity? of relevant literature COGs Measure walkability and bike- How accurate is intersection ability in the same subdivisions This study’s dataset; sidewalk density in predicting walkability used in this study, comparing and bicycle infrastructure data and bike-ability in these cities? those findings with the findings from city, regional, county, and from the study. state agencies Brandon Pike 28 Compare findings from this How much does zoning study against the specific This could be accomplished influence street connectivity? zones—low, medium, and high using the same dataset used in density, for instance—that this study. development took place. How does street connectivity in Conduct a similar analysis of the cities used in this study cities in other states and/or Varied, depending on locations compare to cities in other states countries, and compare with chosen, plus this study’s dataset and countries? these results RECOMMENDATIONS & EVALUATIONS Based on the findings and conclusions presented within this paper, the following are two policy recommendations for cities interesting in adopting connectivity ordinances. 1. Cities with high rates of growth should adopt connectivity ordinances aimed at increasing street connectivity in new development. 2. Cities should consider the strengths and weaknesses of adopting policies similar to Hillsboro’s connectivity analysis, and should not assume those policies will be effective in all cases. These recommendations are based upon the following observations: • Bend adapted to high levels of growth by adopting connectivity ordinances, and those ordinances seem to have worked to increase street connectivity in new development. • Hillsboro has had moderate success with their more strict connectivity ordinances, but likely has had to devote more resources to achieve their street connectivity levels than Bend, who had even higher levels during the second 10- year period than Hillsboro. While it is not possible to say with certainty that these findings can provide concrete ways to encourage higher intersection density in residential development, it is possible Brandon Pike 29 to offer a critique of the policies analyzed within this study. Finally, table 7 outlines the connectivity ordinances used by the three cities and their potential strengths, weaknesses, and impacts. Table 7 | Evaluation of Connectivity Ordinances Analyzed in this Study REGULATORY A STRENGTH(S) WEAKNESS(ES) IMPACTS PPROACHES MODERATELY BLOCK PERIMETER, EFFECTIVE WITH LOW OPERATING COSTS EFFECTIVENESS CAN LEAD TO AN BLOCK LENGTH, FOR PLANNING LIKELY DEPENDS ON INCREASE IN STREET AND CUL-DE-SAC AGENCIES; CAN BE HOW STRICT THE CODE CONNECTIVITY. SEE ORDINANCES RELATIVELY IS WRITTEN BEND AS AN EXAMPLE. FLEXIBILITY FOR DEVELOPERS MAY LEAD TO AREAS A WITH HIGH INTERNAL LLOWS DEVELOPERS STREET FLEXIBILITY IN DESIGN; MID-BLOCK CREATES FEWER CONNECTIVITY, BUT CREATES MORE A ROADWAY-BASED LOW EXTERNAL CCESSWAYS PEDESTRIAN AND TRAVEL ROUTES CONNECTIONS (SONG BICYCLE AND KNAAP, 2004). CONNECTIONS SEE BEAVERTON AS AN EXAMPLE. COULD LEAD TO COMPREHENSIVE SLIGHT INCREASE IN ANALYSIS AND CONNECTIVITY, BUT CONNECTIVITY RESOURCE-INTENSIVE POTENTIAL INFLUENCE MAY NOT BE WORTH A FOR BOTH PRIVATE NALYSIS OVER SUBDIVISION THE RESOURCES AND PUBLIC SECTORS CONNECTIVITY ON A REQUIRED. SEE CASE-BY-CASE BASIS HILLSBORO AS AN EXAMPLE. Brandon Pike 30 APPENDICES APPENDIX A: DEFINITIONS § Block Length – “The distance along a street between the centerline of two intersecting through streets from lot line to lot line.” (Bend Development Code Chapter 10-10 1.2, 2006). § Block Perimeter – “The distance to travel once completely around the block, ending at the starting point as measured from the centerline of the street.” (Bend Development Code Chapter 10-10 1.2, 2006). § Cul-de-Sac – “[A] short street having one end open to traffic and terminated by a circular vehicle turnaround. Cul-de-sacs shall include partial cul-de-sac bulbs or "eyebrows" designed and developed according to City standards” (Bend Development Code Chapter 10-10 1.2, 2006). § Circuit – “A finite, closed path starting and ending at a single node” (Tresidder, 2005). § Dangle node – “The endpoint of a link that has no other connections. A dead-end or cul- de-sac” (Tresidder, 2005). § Development Code – “Development codes are ordinances implementing a local government’s comprehensive plan. They include two components: a zoning ordinance and a subdivision ordinance, which may be adopted and published as separate documents under their own titles. In some cases the sections pertaining to subdivision of land may be included in the zoning ordinance” (University of Oregon Libraries). § Link – “A roadway or pathway segment between two nodes. A street between two intersections or from a dead end to an intersection” (Tresidder, 2005). § Node – “The endpoint of a link, either a real node or a dangle node” (Tresidder, 2005). § Path-Path Intersection – An intersection between two paths and/or trails that are used by pedestrians and/or cyclist § Real node – “The endpoint of a link that connects to other links. An intersection” (Tresidder, 2005). § Street Intersection – Any junction of two streets or roadways, as defined by ORS 801.320 (Legislative Counsel Committee, 2017). Additionally, merging lanes of highways do not meet the definition for this study. § Street-Path Intersection – An intersection between a street and a path or trail that is used by pedestrians and/or cyclists § Street-Street Intersection – An intersection between two or more streets § Street Stub – Usually temporary dead-end streets that do not abut existing development that would inhibit future transportation connections. Brandon Pike 31 Hillsboro Residential APPENDIX B: Subdivisions RESIDENTIAL SUBDIVISIONS RESIDENTIAL SUBDIVISIONS Bend Residential Subdivisions • City Limits Beaverton Residen - Streets • Subdivisions Built 1997 to 2006 Subdivisions • Subdivisions Built 2007 to 2016 • City limits Streets • Subdivisions Built 1997 to 2006 • Subdivisions Built 2007 to 2016 --~= =='- t\. APPENDIX C: CONNECTIVITY ORDINANCE BY CITY Beaverton (Adopted Prior to Bend (Adopted 2006) Hillsboro (Adopted Prior to 1997) 1997) “Unless exempted under paragraph 4 below, full street connections spaced "In any block that is longer than 600 “The block lengths […] shall not not more than 530 feet apart shall be feet as measured from the near side exceed the following standards as provided in all contiguous vacant right-of-way line of the subject street measured from centerline to and/or underdeveloped sites 5.0 Block Length: to the near side right-of-way line of centerline of through intersecting gross acres or larger planned or Ordinance the adjacent street, an accessway streets. zoned for residential or mixed-use shall be required through and near development.” “Within 1/2 mile of the middle of the block." Beaverton 660 feet block length […] in all existing neighborhood activity Development Code 60.55.25 Residential zones.” Bend centers or transit stops, maximum Development Code (2006) 3.1.200 B block lengths shall be 600 feet.” Hillsboro Development Code (2007) 12.50.520 “ 14. Street and Bicycle and "An exception may be granted to the Pedestrian Connection Hindrances. maximum block length in “Full street connections are not Street, bicycle, and/or pedestrian conformance with the Class C connections are not required where Variance criteria in Chapter 5.1.400 required where barriers prevent their construction or require different street one or more of the following for Transportation Improvement conditions exist: Requirements. The applicant must connection spacing. Such barriers demonstrate that the block length include the following: Block Length: A. Physical or topographic cannot be satisfied due to a. Topography; Exception(s) conditions make a general street, topography, natural features, existing bicycle, or pedestrian connection development or other barriers. When b. Railroad right-of-way; impracticable. Such conditions a variance is granted, the land include but are not limited to the division or site plan shall provide c. Freeway right-of-way; alignments of existing connecting blocks divided by one or more d. Pre-existing development streets, freeways, railroads, slopes in walkways or access ways, in excess of City standards for conformance with the provisions of patterns; maximum slopes, wetlands or other Section 3.1.300; Pedestrian Access and Circulation, below. Walkways Brandon Pike 33 bodies of water where a connection shall be located to minimize out-of- e. Streams, wetlands or waterways could not reasonably be provided; direction travel by pedestrians and regulated under Metro UGM shall be universally designed to Functional Plan Title 3; and/or B. Existing buildings or other accommodate full access to development on adjacent lands bicyclists and pedestrians alike, f. Significant Natural Resources physically preclude a connection now regardless of disability." Bend regulated under Section 12.27.200.” and in the future, considering the Development Code (2006) 3.1.200 B Hillsboro Development Code (2007) potential for redevelopment; or, 12.50.520 C. Where streets, bicycle, or pedestrian connections would violate provisions of leases, easements, covenants, or restrictions written and recorded as of May 1, 1995, which preclude a required street, bicycle, or pedestrian connection.” Beaverton Development Code (2005) 60.55.25.14 Standard Zones: “Except where precluded by the barriers listed in Subsection 4, above, maximum block lengths “The block […] perimeters shall not between local and Collector streets exceed the following standards as shall be 1000 feet, and the maximum measured from centerline to perimeter of blocks formed by local centerline of through intersecting and Collector streets shall be 2750 Block Perimeter – streets. feet.” […] 2,000 feet block perimeter in all Residential zones.” Bend Development Code (2006) 3.1.200 B Light Rail and Mixed-Use Zones: “Maximum block perimeter lengths created by the street and alley pattern shall be 1600 feet.” (Ord. 6120 § 1, 2015) Brandon Pike 34 "A cul-de-sac street shall only be used when the applicant "The City may require an accessway demonstrates that environmental or to connect from one cul-de-sac to an topographical constraints, existing Cul-de-Sacs adjacent cul-de-sac or street." development patterns, or compliance Only permitted when approved by Beaverton Development Code with other standards in this code Review Authority and City Engineer 60.55.25.9.A. preclude street extension and through circulation. " Bend Development Code (2006) 3.4.200 N "Connectivity Analysis Required. Land use applications on sites with proposed internal street systems shall include a connectivity analysis describing how the proposed internal street, pedestrian and bicycle network provides safe and convenient access to the following: Connectivity a. Adjacent residential Analysis – – developments and transit stops; b. Adjacent undeveloped property likely to be developed in the future; and c. Neighborhood activity centers, major transit routes and other transit facilities within one-half mile of the site." Hillsboro Development Code 12.50.520 Brandon Pike 35 APPENDIX D: PATHS ~~ Hillsboro Paths - Subd1v1s1ons Built 2.0:,0~! igci;to 9 & TRAILS IN STUDY CITIES 0 2016 PATHS & TRAILS • CStitrye eLtism its 2006 Sttrreeett--:~rteh e~~~~~~:e~~ions 2007 to 2016 -;;; Subd1v1s1ons Bu,111997 tot 997 to 2006 Paths I\ t • Street-Path lntersect,~~~s 1997 to 2006 I ~ • Street-Street lntersec' " ~ IN STUDY CITIES Bend Paths City limits ; §~~~~isions Built 19.9~!? f887 to 2006 • Street-Path lntersect1f·onS· 1997 to 2006 • Street-Street lnt~rsei~ to 2016 • Subdivisions Built 2~·ons· 2007 to 2016 : §~~==t~i~el~~~r:~~tionS: 2007 to2 016 • Slope of 25% or More .. Paths Brandon Pike 36 APPENDIX E: RESIDENTIAL ZONES BY CITY, ARRANGED BY SIMILAR HOUSING DENSITY LEVELS Beaverton Bend Hillsboro Units Units Units Zone Zone Zone /Acre /Acre /Acre — — Area Reserve District (UAR) 0.1 — — — — Suburban Low Density Residential (SR 2 1/2) 0.4 — — — — Low Density Residential (RL) 1.1 - SFR-10 Single Family 3.5 to 4.0 Residential 4.35 Urban Low Density 4.4 — — SFR-8.5 Single Family 4.0 to Single Family (R10) Residential 5.0 Urban Standard Density Single Family 6.2 Standard Density Residential 4.0 - SFR-7 Single Family 5.0 to (R7) (RS) 7.3 Residential 6.25 — — — — SFR-6 Single Family 6.0 to Residential 7.5 Urban Standard Density Single Family 8.7 — — SFR-4.5 Single Family 8.0 to (R5) Residential 10.0 Urban Medium Density Single Family 10.9 — — — — (R4) SCR-OTC Station — — — — Community Residential 6.0 to Orenco Townsite 12.0 Conservation — — — — SCR-LD Station Community 9.0 to Residential Low Density 14.0 — — Medium Density Residential 6.0 - MFR-1 Multi-Family 11.0 (RM and RM-10) 21.7 Residential to 16.0 SCR-DNC Station — — — — Community Residential 9.0 to Downtown Neighborhood 23.0* Conservation Urban Medium 17.0 Density Multi-Family 21.8 — — MFR-2 Multi-Family (R2) Residential to 21.25 SCR-MD Station Community 18.0 — — — — Residential – Medium to Density 23.0 23.0 — — — — MFR-3 Multi-Family Residential to 28.75 — — — — SCR-HD Station Community 24 to Residential – High Density 30 Urban High Density Multi-Family (R1) 43.6 High Density Residential 21.7 - (RH) 43 — — APPENDIX F: NOTES ON LAYOUT OF STUDY CITIES Beaverton’s city limits resemble a tapestry with a few large holes cut out of the upper portion. Neighborhoods and census designated places such as Cedar Hills, West Slope, and Marlene Village have maintained their position technically outside the boundary of the city of Beaverton, though Beaverton surrounds them in all directions (see appendix B). Most of Beaverton’s residential development lies to the south of the central business district (CBD). Conversely, Bend and Hillsboro have layouts and city limits that may be considered more traditional. Bend’s CBD sits directly in the middle of the city, with the remaining development radiating out in an oval shape. Other than its CBD falling in the western portion of the city, Hillsboro has a layout more similar to Bend than to Beaverton. These distinctions are worth noting when considering the effect city layout can have on development patterns, and, indirectly, street connectivity. APPENDIX G: TOPOGRAPHY Topography is often used as a reason to build winding streets with low connectivity, often with developers and cities citing environmental hazards or degradation as a reason to not build connected streets on hilly terrain. Perhaps a better question that municipalities should ask themselves when it comes to topography and street connectivity policy, is should they allow residential development on hilly terrain at all? If the answer is yes, why not encourage connected street networks? At what point do hills become so steep that they can no longer support connected streets, but can still somehow support suburban development with low street connectivity? To some of the most prominent examples of cities in the western United States with high street connectivity, the answer to that question is almost never. Cities like Seattle and San Francisco, both water-adjacent and built up into hillsides, offer examples of cities that adhered to a strict pattern when first developing. The time period that those cities’ cores were developed were in times when active transportation modes such as walking and Brandon Pike 38 cable cars were the standard forms of urban transportation. It should be said that, no matter the slope development occurs upon, mitigation of environmental degradation should take place, and building into a hillside poses particular challenges when it comes to things like storm water runoff and maintaining the health of the watershed (Goldshleger, Karnibad, Shoshany, and Asaf, 2012). However, a trade-off can occur when the positive environmental effects of not building into a hillside are outweighed by the negative ones of building automobile-centric cities that consume large amounts of energy. Bend, while surrounded Bend Residential Subdivisions, Intersections, & Slope by buttes and City Limits - Streets mountains, is actually • Subdivisions Built 1997 to 2006 • Street-Path Intersections: 1997 to 2006 relatively flat in terms of • Street-Street Intersections: 1997 to 2006 • Subdivisions Built 2007 to 2016 • Street-Path Intersections: 2007 to 2016 topography. This map • Street-Street Intersections: 2007 to 2016 • Slope of 25% or More shows all land within the city with a slope of 25 percent or greater. This amounts to just under 7 percent of Bend’s total land area, and much of this land is undevelopable— namely in riparian zones and Pilot Butte Neighborhood Park. Beaverton and Hillsboro, similarly, are developed on land that is relatively flat based on examination of topographic data, though slope data was not readily available to conduct the same Datasource(s): OeschutesCoonty, CityolBend Mapereat&dbyBrandooPike analysis in those cities. Brandon Pike 39 APPENDIX H: CENSUS DATA FOR STUDY CITIES Oregon Beaverton Bend Hillsboro Land Area (Acres) 2000 61,437,888.00 10444.79 20492.78 13804.79 2010 61,432,268.81 11987.19 21126.38 15295.99 Population 2000 3,421,399 76,129 52029 70,186 2010 3,831,074 89,803 76639 91,611 2016 (Estimate) 3,982,267 94,865 84416 100,462 Housing Population Density (People/Square Mile) 2000 35.6 4,664.5 1,624.8 3,253.80 2010 39.9 4,795.1 2,322.0 3,833.30 Housing Density (Units/Square Mile) 2000 15.1 1,991.3 702.9 1,261.5 2010 17.5 2,109.1 1,094.0 1,484.9 Housing Units 2000 1,452,709 32,500 22,507 27,211 2010 1,675,562 39,500 36,110 35,487 2016 (Estimate) 1,706,290 40,267 37,406 38,495 Percent Change 14.9% 19.3% 39.8% 29.3% Vacancy 2000 8.2% 5.2% 6.4% 7.8% 2010 9.3% 5.8% 12.0% 6.2% 2016 (Estimate) 9.4% 5.1% 8.9% 6.2% Owner Occupied 2000 64.30% 47.7% 62.9% 52.3% 2010 62.2% 49.7% 57.9% 54.5% 2016 (Estimate) 61.4% 47.6% 58.9% 55.6% Renter Occupied 2000 35.7% 52.3% 37.1% 47.7% 2010 37.8% 50.3% 42.1% 45.5% Brandon Pike 40 2016 (Estimate) 38.6% 52.4% 41.1% 44.4% Income Median Household Income 2000 $40,916 $47,863 $40,857 $51,737 2010 (Estimate) $49,260 $54,885 $53,006 $60,695 2016 (Estimate) $53,270 $59,620 $55,625 $70,180 Percent Change 23.2% 19.7% 26.5% 26.3% Population for Whom Poverty Status Is Determined 2016 (Estimate) 15.7% 13.4% 12.4% 12.9% Transportation Means of Transportation to Work (2016 Estimate) Drove alone 71.4% 68.3% 75.1% 73.2% Carpooled 10.3% 11.2% 7.5% 11.3% Public transportation (excluding taxicab) 4.4% 10.0% 0.6% 6.7% Walked 3.9% 3.4% 3.3% 2.4% Bicycle 2.4% 1.1% 3.1% 1.6% Taxicab, motorcycle, or other means 1.1% 0.9% 1.0% 0.9% Worked at home 6.4% 5.1% 9.4% 3.9% Means of Transportation to Work (2010 Estimates) Drove alone 72.0% 71.3% 78.6% 73.2% Carpooled 10.8% 9.2% 7.5% 11.1% Public transportation (excluding taxicab) 4.2% 7.9% 0.6% 7.2% Walked 3.9% 4.3% 2.9% 2.7% Bicycle 2.1% 1.1% 2.2% 1.3% Taxicab, motorcycle, or other means 1.0% 1.4% 0.7% 0.8% Worked at home 6.1% 4.6% 7.5% 3.7% Means of Transportation to Work (2000) Drove alone 73.2% 72.5% 74.6% 73.4% Carpooled 12.2% 10.6% 12.7% 13.8% Brandon Pike 41 Public transportation (including taxicab) 4.2% 8.3% 1.4% 6.5% Walked 3.6% 3.1% 2.8% 2.2% Other Means 1.9% 0.6% 2.8% 1.2% Worked at home 5.0% 4.5% 5.7% 3.0% Sources: 2012-2016 American Community Survey 5-Year Estimates, US Census 2010 Demographic Profile, US Census 2000 Demographic Profile; US Census 2000 Summary File 1; US Census 2010 Summary File 1; Census 2000 Summary File 3 (SF 3) - Sample Data APPENDIX I: POPULATION OF STUDY CITIES OVER TIME Census Beaverton Bend Hillsboro 1880 - - 402 1890 - - 1,246 1900 249 - 980 1910 386 536 2,016 1920 580 5,415 2,468 1930 1,138 8,848 3,039 1940 1,052 10,021 3,747 1950 2,512 11,409 5,142 1960 5,937 11,936 8,232 1970 18,577 13,710 15,365 1980 31,962 17,263 27,664 1990 53,310 20,469 37,598 2000 79,277 52,029 70,187 2010 89,803 76,639 91,611 2016 (Estimate) 97,590 91,122 105,164 Brandon Pike 42 APPENDIX J: REFERENCES Andreou, E. (2014) The effect of urban layout, street geometry and orientation on shading conditions in urban canyons in the Mediterranean. Renewable Energy, 63, 587-596. Badia, H., Estrada, M., Robusté, F. (2016). Bus network structure and mobility pattern: A monocentric analytical approach on a grid street layout. Transportation Research Part B: Methodological, 93, Part A, 37-56. Ben-Joseph, E. (1995). Changing the residential street scene: Adapting the shared street (woonerf) concept to the suburban environment. American Planning Association. Journal of the American Planning Association, 61(4), 504. Bernhardt, R. (2006). One step at a time UDOs in Nashville. Planning, 72(1), 20-23. Berrigan, D., Pickle, L., Dill, J. (2010). Associations between street connectivity and active transportation. International Journal of Health Geographics 9,20 1-18. City of Beaverton. (2005). Beaverton Development Code. City of Bend. (2006). Bend Development Code. City of Hillsboro. (2007). Hillsboro Development Code. Duany, A., Talen, E. (2002). Making the good easy: The smart code alternative. Fordham Urban Law Journal 29(4), 1445-1468. Dunham-Jones, E., Williamson, J. (2009). Retrofitting suburbia: Urban design solutions for redesigning suburbs. John Wiley & Sons. Ellickson, R. (2012). The law and economics of street layouts: How a grid pattern benefits a downtown. John M. Olin Center for Studies in Law, Economics, and Public Policy Research Paper No. 459. Frazier, L. (24 March 2016). 2015 Census estimates: Oregon population growth nearly hits pre- recession rate. The Oregonian. Goldshleger, N., Karnibad, L., Shoshany, M., & Asaf, L. (2012). Generalising urban runoff and street network density relationship: A hydrological and remote-sensing case study in Israel. Urban Water Journal, 9(3), 189-197. Guo, J., Elhat, C., & Copperman, R. (2007). Effect of the built environment on motorized and nonmotorized trip making - Substitutive, complementary, or synergistic? Transportation Research Record, 2010(2010), 3-13. Gunda, R., & Atluri, M. (2017). Implementing Houston's First Bus Rapid Transit System. Institute of Transportation Engineers. ITE Journal, 87(4), 34-38. Haldeman, B. (1914). The Street Layout. The Annals of the American Academy of Political and Social Science, 51, 182-191. Handy, S., Paterson, R. G., Butler, K. (2003). Planning for street connectivity: Getting from here to there. American Planning Association. Haynie, Dawn. (November 2016). Examining the measures of street connectivity in the American city and their interdependencies as applied in practice. 56th ACSP Conference, Portland, OR. Brandon Pike 43 Hooper, P., Giles-Corti, B., & Knuiman, M. (2014). Evaluating the Implementation and Active Living Impacts of a State Government Planning Policy Designed to Create Walkable Neighborhoods in Perth, Western Australia. American Journal of Health Promotion, 28(3), S5-S18. Legislative Counsel Committee. (2017). CHAPTER 801—General Provisions and Definitions for Oregon Vehicle Code, https://­www.­oregonlegislature.­gov/­bills_laws/­ors/­ors801.­html (2017) (last accessed Mar. 30, 2018). Knight, P., & Marshall, W. (2015). The metrics of street network connectivity: Their inconsistencies. Journal of Urbanism, 8(3), 241-259. Koohsari, M. J., Owen, N., Cerin, E., Giles-Corti, B., & Sugiyama, T. (2016). Walkability and walking for transport: Characterizing the built environment using space syntax. The International Journal of Behavioral Nutrition and Physical Activity, 13, 121. Koohsari, M. J., Owen, N., Cole, R., Mavoa, S., Oka, K., Hanibuchi, T., & Sugiyama, T. (2017). Built environmental factors and adults' travel behaviors: Role of street layout and local destinations. Preventive Medicine, 96, 124-128. Koohsari, M. J., Sugiyama, T., Shibata, A., Ishii, K., Liao, Y., Hanibuchi, T., Owen, N., & Oka, K. (2017). Associations of street layout with walking and sedentary behaviors in an urban and a rural area of Japan. Health & Place, 45, 64-69. Kulash, W., Anglin, J., & Marks, D. (July/August 1990). Traditional Neighborhood Development: Will the Traffic Work? Development, 21-24. Maclennan, C., Norquist, J., & Harp, T. (2004). Smart Growth for Community Development. The Journal of Law, Medicine & Ethics, 32(4), 27-31. Marshall, W., & Garrick, N. (2010). Effect of Street Network Design on Walking and Biking. Transportation Research Record, 2198(2198), 103-115. Matthews, J. W., & Turnbull, G. K. (2007). Neighborhood Street Layout and Property Value: The Interaction of Accessibility and Land Use Mix. Journal of Real Estate Finance And Economics, 35(2), 111-141. McCann, B. A. and Rynne, S. (2010). “Chapter 7: Creating Complete Streets: Design Principles and Features” in Complete streets: best policy and implementation practices. Chicago, American Planning Association, 79-99. Metropolitan Service District, Street Design Work Team, et al. (1997). “Chapter 2: Goals,” and “Chapter 3: Design Guidelines” Creating livable streets: street design guidelines for 2040. City of Portland, Metro Regional Services, 5-55. Morris, M. (2009). Smart codes: Model land-development regulations (Report (American Planning Association. Planning Advisory Service) ; no. 556). Chicago, Illinois: American Planning Association. Narvaez, L. (2012). City rules: How regulations affect urban form. Urban Design International, 17(4), 351-352. Orvell, M. (2012). The death and life of Main Street : Small towns in American memory, space and community. Chapel Hill: University of North Carolina Press. Ozbil, A., Peponis, J., & Stone, B. (2011). Understanding the link between street connectivity, land use and pedestrian flows. Urban Design International, 16(2), 125-141. Brandon Pike 44 Pan, Y. (17 January 2017). Sold Out: These 10 U.S. cities have the biggest housing shortages. Realtor.com Population Research Center. (April 2017). Table 6. Rank of incorporated cities by July 1, 2016 Population Size. Portland State University. Robinson, C. (1914). The Sociology of a Street Layout. The Annals of the American Academy of Political and Social Science, 51, 192-199. Song, Y., & Knaap, G. (2004). Measuring urban form: Is Portland winning the war on sprawl? Journal of the American Planning Association, 70(2), 210-225. Stangl, P. (2015). Block size-based measures of street connectivity: A critical assessment and new approach. Urban Design International, 20(1), 44-55. Stangl, P., & Guinn, J. M. (2011). Neighborhood design, connectivity assessment and obstruction. Urban Design International, 16(4), 285-296. State of Oregon. (December 2017). Oregon’s demographic trends. State of Oregon: Department of Administrative Services, Office of Economic Analysis. Thompson, D. (October 2013). Suburban sprawl: Exposing hidden costs, identifying innovations. University of Ottawa: Sustainable Prosperity Institute. Tresidder, M. (2005). Using GIS to measure connectivity: An exploration of issues. Portland State University, Field Area Paper. United States Census Bureau. (2017). Annual estimates of the resident population: April 1, 2010 to July 1, 2017. University of Oregon Libraries. (n.d.). Scholars’ Bank: Oregon Local Documents. Van der Kloof, A., Bastiaanssen, J., Martens, K. (2014). Bicycle lessons, activity participation and empowerment. Case Studies on Transport Policy, 2(2), 89-95. Wang, F., Wen, M., Xu, Y. (2013). Population-adjusted street connectivity, urbanicity and risk of obesity in the U.S. Applied Geography, 41, 1-14. West, J., & Lowe, A. (1997). Integration of transportation and land use planning through residential street design. Institute of Transportation Engineers. ITE Journal, 67(8), 48. Xiaomin, X., Zhen, H., Jiasong, W. (2006). The impact of urban street layout on local atmospheric environment. Building and Environment, 41, 10. 1352-1363. Brandon Pike 45