Representing Urban Phenomena in Their Context and at Different LoD: from Raw Data to Appropriate LoD

Abstract

More than half of the world population lives in cities today. This proportion rises to 80% in developed countries. The density of urban population causes environmental troubles such as noise, urban heat waves, and chemical pollutions or magnetic pollution. Sensors and models are used to improve knowledge related to these phenomena particularly in cities. The aim of our research is to propose methods to view these phenomena in contextualised ways and at different levels of details. In the context of data exploration, we wish to generate from the initial phenomena data other levels of detail to allow the visual perception of the information at different visual scale. We also propose symbols that resist as well as possible to scale change and without excessive covering the other information such as streets, buildings or names. The first solutions presented in this paper are implemented and illustrated through two examples: nocturne temperature in Paris with very sparse initial data and concentration of chlorine with very dense initial data.