Uma abordagem baseada em grafos para visualização do comportamento de águas pluviais

Abstract

Graph Theory is a branch of Mathematics that investigates objects and ther relationships. Graphs are structures composed of two sets: one of vertices (or nodes) and another of edges that connect them. These structures are widely used in various applications, such as social networks, communication networks, and transportation systems, among others. Another promising application of graph-based approaches is the representation of landscapes in Geomorphology. Thus, this work proposes an approach for visualizing the behavior of rainwater runoff by employing a graph-based representation and applying well-known Graph Theory algorithms to explore which of them may lead to meaningful insights. The methodology involved bibliographic research and the analysis of tools in order to achieve a deeper understanding of the subject and to select software appropriate for the development of the study. The QGIS software was used for the manipulation and analysis of geospatial data, Python was employed for data processing and treatment through scripts developed in this study, and Gephi was used for the visualization and analysis of the generated graphs. A representative dataset was selected for this proposal and subsequently subjected to a data treatment process, including the removal of duplicated data and less relevant columns, using a script developed in this study. The processed data were then imported into Gephi, resulting in a directed acyclic graph that represents the landscape. Based on this graph, it was possible to apply algorithms and identify areas with higher potential risk associated with natural phenomena, especially those related to rainwater runoff.