Complex network modeling of a river basin: an application to the Guadalquivir River in Southern Spain

In this paper, complex network analysis is used to analyze the structure of a whole river basin. A classification of nodes and edges that allow the river basin to be represented as a directed network is given. Both global and local characterization metrics are presented. Thus, centralization indexes, average path length, diameter, network efficiency, edge length, degree and strength distributions can be computed and interpreted. The closeness and betweenness centrality as well as the PageRank index of the different nodes can also be calculated and provide information on the relative position of each node in the network in terms of their average distance to upstream and downstream nodes, their reachability from other nodes, etc. In addition, a new centrality index based on the environmental impact of a potential spill or contaminant release on each node of the network has been applied. This type of centrality index is particularly useful in hydrologic networks and helps to identify key nodes from a contamination propagation point of view. The application of the proposed approach to the Guadalquivir River basin, in Southern Spain, is presented. Apart from validating the environmental impact centrality index, the most important finding is the scale-free character of the network and its consequences.