Is there an ideal city dimension for walking?

While in an urban design workshop on Changzhou Island organized by the Guangzhou Lingnan Architecture Research Center, I continued my fascination with ideals grid for walkability. Practitioners often speak of a 10-minute neighborhood, or 800m walking distance. Worldwide, 80m is used for grid spacing. Ten 80m blocks is 800m. Is there magic here?

This blog post proposes research to find out.

—Michael King

Blocks and intersections in two historic Chinese cities compared to idealized 80m grid structure. Excerpt from report prepared for Guangzhou Lingnan Architecture Research Center, www.gzlarc.com, in 2018.

City grids, like any construct, do not presage success, they merely set the stage for it. Consider the garden as analogy. The grid of a garden is soil, water and sun. With this gardens can flourish. Without it, nothing will grow. Similarly, grids create the conditions for cities to succeed. Without a good grid, cities flounder.

Cities consists of grids, made up of streets and blocks. In laying out cities and subdivisions, or in regenerating an existing neighborhood, one must choose the block and street size as well as signal spacing, crossing location, and a host of other variables which affect mobility. Historically, a number of methods have been used, with no system dominant. Accordingly, the world’s streets and blocks exist in a multitude of measurements, from narrow lanes to superblocks.

Is there an ideal dimension? To put a finer point on it, is there an ideal dimension to maximize walkability?

Researchers have used various parameters to understand street and block dimensions. Sevtsuk (2016) analyzes plot frontages, plot depths, street widths, and block lengths. King (2014) compared block size to traffic signal and crossing spacing. Ewing (2010) found land use diversity, intersection density, and the number of destinations within walking distance to be relevant. These studies and others evaluate existing blocks and streets, which have been compromised by or designed to accommodate cars. It seems that if one were to derive an ideal dimension to maximize walkability, one would need to evaluate street networks that existing before or apart from car traffic.

Research proposal

To answer this question I propose the following research exercise:

Collect ~100 pre-car digital maps of cities around the world.
Measure block length, intersection density, street widths, link/node ratio, direct/actual travel distance, block size, and other variables. Limit study to the grid itself, not other factors such as land use, destinations, and density.
Calculate statistically relevant attributes for each variable. Describe specific attributes which are co-related to specific patterns.
Conclude an ideal dimension per variable.

The results of this research will help to counter the existing practice of designing cities first for drivers, and then for other users. With an understanding of an empirically-derived walkability measure, professionals would be able to push for more walkable cities. This overlaps with transit-oriented development and other efforts to counter the effects of climate change.