ANNEX  V

Marginal Urban Densities

Introductory notes

As is already mentioned, due to large number of people settled in the City center, even a considerable drop in density at the outer city rings is not immediately reducing the overall density. But when we look at the marginal densities, this fall is much more visible. Thus we may get a better idea of the real size of urban hinterland.

This picture gets even clearer when we notice the occasional spikes at a considerable distance from the city center. They obviously indicate “annexation” of some larger city. For instance in Tokyo Metro Area such spikes are obvious already at the 1,500 km² threshold, when city of Kawasaki is included and particularly at 4,000 km² when the second largest city, Yokohama is added. In Seoul Metro Area such spike appear at the 2,000 km² threshold when the port of Incheon is included, in Cairo such spike happens when the neighbouring city of Gizah is included etc.

Sources

Since the data on densities were derived from the ones on population and the area, the sources are the same as the one quoted in annexes I and II (the local, the national, and the supra national statistic offices, as well as the Wikipedia and city population websites).

Comments

Contrary to the absolute densities, the marginal ones drop bellow the 100 person per hectare mark already at the 1,000 km² threshold (with the exception of Manila, Delhi, Jakarta, Seoul and Tehran). While at the 500 km² threshold there are over 30 cities with absolute densities in three digits, little more than half of them have marginal densities of 100 people per hectare or more. By the 2,000 km² threshold, the marginal density falls bellow 50 people per hectare level with very few exceptions, which indicate the annexation of some other larger city.

Therefore the marginal densities indicate that there is rarely a city whose hinterland is stretching beyond 3,000 km², without addition of some other large city. Without such “injections” the density beyond 3,000 km² is reduced to single digits, so we may consider that at this point ends some Larger Urban Zone of a single city, and that any urban “unit” beyond this threshold should be regarded as a conurbation.