Urban Form and Walking Activity: Analyzing their Relation within De-facto Boundaries for Walkers ( poster )

Abstract

Background Many studies have evaluated various aspects of built-environments that affect people’s walking at different geographic scales. The findings in these studies imply that, if the scale is too small, some attributes of physical-environments are unobserved despite of their significant impact on walking activity. Conversely, when the scale is too large, attribute measures are too aggregated to sufficiently reflect their characteristics. Aim The objectives are to define a spatial unit that can capture characteristics of physical-environments related to walking, and to unveil the relation between those characteristics and pedestrian volume. Method A boundary is defined as a group of blocks surrounded by expressways and arterials based on findings in previous studies; i) Expressway and arterial are the roadway functional classifications that mainly serve mobility of vehicles, not pedestrians; ii) Massive single-use roadways play a role of border in cities; iii) The roadways including arterials where many vehicles pass at a high speed decrease the frequency of people crossings; iv)Walk trips tend to concentrate on narrow streets not on wide arterials. A spatial regression was performed with a daily pedestrian volume within a spatial unit as a dependent variable, and independent variables categorized by urban form and accessibility to transit and amenities. The independent variables include two key variables – tare and mesh – that quantitatively characterize urban form. Tare is the area between a pair of blocks that have negative correlation with coverage (footprint area divided by total area). An area with a greater value of tare offers poor usability and diversity. Mesh implies the average block size that have negative correlation with network density. An area with a greater value of mesh provides poor connectivity and permeability to walkers. Results The coefficient estimates for ten variables in the model were statistically significant. Tare and mesh are negatively related to pedestrian volume, and variables that measure mixed building-use and accessibility are positively associated. Conclusion The findings of this study suggest a new way of defining a spatial unit for walking analysis and establish a groundwork to design urban environments especially for an area composed of local streets and collector roads.