Using spatial graph

To capture patterns in urban form, we have to analyze morphometric characters within a spatial context, taking into the account neighbouring elements. A simple example could be a mean height within 100 metres (measured as a location-based character for each building). momepy is using libpysal spatial graph to capture the relationship between elements, in this case, buildings, in the form of a binary matrix (1 = neighbours, 0 = not neighbours). Spatial graphs are an essential part of momepy and are used on many occasions and in various forms.

This section covers:

• Generating spatial graphs
  • Queen contiguity
  • Distance
• Examples of usage
  • First order contiguity
  • Higher order / distance
• Measuring diversity
  • Queen contiguity
  • Range
  • Theil index
  • Simpson’s diversity index
• Using two spatial graphs
  • Building adjacency
  • Mean interbuilding distance