Using quadtree representations in building stock visualization and analysis

Abstract

Many public facilities, institutions, and private domains in Germany are still limited to using administrative zones when visualizing densities of statistical data. When dealing with geospatial information the following question is becoming increasingly more important: „How much data (persons/buildings) must be aggregated to consider regulations of data protection?”. This article presents a quadtree representation to provide building data management on a small scale. In recent years, German buildings were fully geocoded as a building polygon. Regarding statistical and cartographic analysis this allows the use of grids of arbitrary size and any administrative units. The city of Hamburg and several surrounding communities (Hamburg conurbation) are used as an example to demonstrate the principle of a quadtree. It is also possible to apply the general model to other sensitive data and to other regions (e.g. cities or rural areas). This includes the possibility of using regular grids in any projection useful for European cooperation (INSPIRE). The building representation, using varying cell sizes is not to be understood as a pure visualization as it serves as a new structure on a small scale to carry out cell-based analyses in the future.