Spatial Signatures

Dynamic classification of
the built environment
Dani Arribas-Bel
Martin Fleischmann
[@darribas]
[@martinfleis]

How we arrange “stuff” in cities matters…

Source: A map of every building in America (New York Times)

… it matters a lot

Urban Form

What does it look like?
“Physical structure and appearance of cities”

What do we talk about…

… when we talk about urban morphology?

buildings
streets
plots
open spaces

How can we describe it…

… numerically?

Urban morphometrics

“quantitative analysis of urban form”

All about measuring.

Measuring

dimension
shape
spatial distribution
intensity
connectivity
diversity

Why?

Because we (finally) can!

Data

Tools

OSMnx
PySAL
foot
momepy

momepy

Urban Morphology Measuring Toolkit

momepy.org

Few examples

… before we move on.

longest axis length
dimension
equivalent rectangular index
shape

distance to neighbours
spatial distribution
covered area ratio
intensity

closeness centrality
connectivity
Simpson’s diversity of tessellation area
diversity

Evolution of urban patterns

42 places, 6 historical periods

400 m buffer

Scale peaked in modernism

We forgot how to make a grid

Classification

Spatial Signatures

A characterisation of space based on form and function designed to understand urban environments

A characterisation of space based on form and function designed to understand urban environments

A characterisation of space based on form and function designed to understand urban environments

A characterisation of space based on form and function designed to understand urban environments

Form

What does it look like?
“Physical structure and appearance of cities”

Qualitative, quant. case studies, morphometrics, remotely sensed

Function

What is it used for?
“Activities that take place within an environment”

Geography, Economics, Sociology, Environmental Sciences, Tansport studies…

Form & Function

  • Richer picture
  • Topography, history, technology, cultural values…
  • More robust representations

Opportunities

We don’t have good ways of measuring form & function in cities

Opportunities

Opportunities

  • Fragmented understanding

  • Physical classification - ??? - Geodemographics

Enclosed Tessellation
Embedding form & function
Spatial Signatures

Enclosed Tessellation
Embedding form & function
Spatial Signatures

Characters

Form
  • Building
  • ET cell
  • Street segment
  • Cell context
  • …
Function
  • Population
  • Land use/cover
  • Access to uses
  • Activity densities
  • …

Context

Every piece of information is considered within its spatial context

Enclosed Tessellation
Embedding form & function
Spatial Signatures

Benefits

  • Data-driven & (multidiscipline) theory-informed
  • Granular & scalable
  • Flexible deployment

Illustration

The Setup

Five different cities

  • Historical environments
  • Geography & culture
  • Data landscapes

The Cities

The Data

Form
  • Local open data
  • OpenStreetMap
  • Satellite-derived
Function
  • Local open data
  • Land use/cover
  • Global grids
  • Nightlights
Clustered with K-Means + clustergram

The Signatures

Barcelona

Dar es Salaam

Houston

Medellin

Singapore

Barcelona

Houston | Medellin

Singapore | Dar es Salaam

To take away

  1. Urban form and function matter
  2. Spatial Signatures: F&F for cities, in detail, at scale
  3. Good measurement → better understanding

Spatial Signatures

Dynamic classification of
the built environment
Dani Arribas-Bel
Martin Fleischmann
[@darribas]
[@martinfleis]