Detecting cities with high intermediacy in the African urban network

Abstract: Cities play different roles depending on their location within the transport network. Two cities of similar size might have distinct characteristics if one is located on a corridor between two capitals and the other is near a barrier, such as a mountain range. The level of intermediacy is a property of cities that characterises their position in the urban network. We measure the level of intermediacy of African cities by constructing the road infrastructure network obtained from OpenStreetMap. The infrastructure network allows defining city metrics such as degree and centrality. A proxy for the number of journeys that flow through each network edge is approximated using a mathematical model based on the level of attraction or gravity between all pairs of cities. Our model considers the extra time of crossing an international border as a parameter that enables us to proxy the cost of having fragmented regions with costly political barriers. Our results show that small cities have a wide range of intermediacy. We detect a phase transition where cities with less than one million inhabitants have a centrality that depends on the size and degree. For cities above one million inhabitants, centrality tends to be larger and depending primarily on city size rather than degree.