A Graph theory approach for geovisualization of land use change: An application to Lisbon
- A Graph theory approach for geovisualization of land use change: An application to Lisbon
Urban sprawl and growth has experienced increased concern in geographic and environmental literature. Preceding the existence of robust frameworks found in regional and urban planning, as well as urban geography and economics, the spatial properties of allocation of urban land use are still far from being completely understood. This is largely due to the underlying complexity of the change found at spatial level of urban land use, merging social, economic and natural drivers. The spatial patterns formed, and the connectivity established among the different subsets of land-use types, becomes a complex network of interactions over time, helping to shape the structure of the city. The possibility to merge the configuration of land-use with complex networks may be assessed elegantly through graph theory. Nodes and edges can become abstract representations of typologies of space and are represented into a topological space of different land use types which traditionally share common spatial boundaries. Within a regional framework, the links between adjacent and neighboring urban land use types become better understood, by means of a Kamada-Kawai algorithm. This study uses land use in Lisbon over three years, 1990, 2000 and 2006, to develop a Kamada-Kawai graph interpretation of land-use as a result of neighboring power. The rapid change witnessed in Lisbon since the nineties, as well as the availability of CORINE Land Cover data in these three time stamps, permits a reflection on anthropogenic land-use change in urban and semi-urban areas in Portugal’s capital. This paper responds to (1) the structure and connectivity of urban land use over time, demonstrating that most of the agricultural land is stressed to transform to urban, gaining a central role in future. (2) Offer a systemic approach to land-use transitions generating what we call spatial memory, where land use change is often unpredictable over space, but becomes evident in a graph theory framework, and (3) advance in the geovisual understanding of spatial phenomena in land use transitions by means of graph theory. Thus, the structure of this combined method enables urban and landscape to have a better understanding of the spatial interaction of land-use types within the city, promoting an elegant solution to rapid geovisualization for land-use management in general.