Análisis del sistema hidrográfico nacional mediante redes complejas

Resumo

The goal of this thesis is the analysis of the structure and functionality of the national river basins system using complex network analysis tools and river basin as the key elements of the system. From this, more specific objectives have followed, like studying the connectivity among the different river basins and its effect on the rainfall imbalances in a climate change scenario, modelling a whole river basin analysing the centrality of its elements from a hydrological and environmental perspective or studying the proliferation of groundwater abstraction wells and its impact on the groundwater mass bodies to help prevent their overexploitation. The methodology used is based on complex network analysis (CNA) and uses global and local characterisation measures and both standard and application-specific centrality indexes. This research has been carried out at three levels of analysis: national, river basin and aquifer. Thus, at the national level, CNA has allowed modelling and characterising the Spanish river basin system, with its interbasin transfer infrastructure, identifying out-star nodes and surplus and deficit basins and analysing the degree and strength distribution, the disassortativity and the community structure. This provides a holistic perspective of the system that reveals a large-diameter, low-efficiency, sparse network. This is a consequence of the high costs of establishing interbasin transfer connections. The network has a community structure where there are some nodes with high betweenness centrality, thus reflecting the hydrographic imbalances of the country. In this regard, the analysis may shed light on how and where could additional interbasin transfer connections may be required in order to mitigate the effects of climate change. In a second study, the Guadalquivir river basin has been modelled in detail, leading to a sparse network, with small world characteristics, large diameter and low efficiency. The degree distribution follows a power law distribution, characteristic of scale free networks. The node centrality indexes allow, among others, to identify critical elements in flood events, which are more frequent and more severe due to climate change. In addition, a new environmental impact centrality index based on the pollutant propagation, which allows the identification of critical nodes in this regard. Thirdly, the groundwater abstraction wells in the area of Doñana National Park has been modelled as a weighted undirected network. The importance of its component structure as well as the strength distribution and eigenvector centrality, among others, has been studied. Two new application-specific centrality indexes have also been proposed, measuring the total impact of a well on neighbouring wells and, vice versa, the total impact of neighbouring wells on a given well. This provides a tool that may contribute to a more rational management of the aquifer.