Urban Morphometrics: “neighborhood” potential

Metric Mean Depth computes the average (topological road network) distance from each street to each other street within the predetermined radius. Metric centralities differ from Angular centralities in that metric ‘shortest path’ computations do not consider turning angle.
Metric Mean Depth is adept at identifying inherent neighborhood structures. At smaller scales there are no meaningful structures of clustering in the area, while in the first instance, at radius of 1.600 meters, significant clusters of neighborhoods begin to emerge.
At radius of 3.500 meters the Depth Clusters are large and correspond well with city and town centres in the area. At scale of 5.000 meters, some clusters dissolve and merge with other clusters.