Document Actions
Monitoring of Surface Sealing with Satellite and Aerial Imagery
in Potsdam
One of the major environmental concerns of the German Government is to reduce new surface sealing from currently over 100 to 30 hectares per day in 2020. Federal State and local authorities in Germany need up-to-date information about surface sealing as an input for modelling (urban climate, water balance) as well as for evaluation processes (soil protection).
Remote sensing enables a fast and area-wide survey of the soil surface. Satellite sensors with high or very high geometric resolution or hyperspectral sensors as well as aerial images allow analyses of the soil surface at any time. Thus change detection and monitoring of urban indices, e.g. surface sealing, is possible.
In case of the following presented study the local authorities of Potsdam, a city in Eastern Germany near Berlin, required a change detection of the city since the political change in 1989 using a standardised method that furthermore allows a monitoring in the future. The available imagery - aerial images of 1992, Indian Remote Sensing Satellite (IRS) data of 1998 and QuickBird satellite data of 2004 - were analysed through regression tree modelling. In addition to the remote sensing data geo data such as land use and biotope type maps, topographic and building data (ATKIS) and population data were used.
Furthermore surface sealing data including buildings of the Energy and Water Company Potsdam (EWP) for a range of economical relevant areas as well as surface sealing data of all public streets were used as references in the model. The block geometry of the land use and biotope type map served as reference areas.
Since the developed method has to be transferable and repeatable it uses not only the radiometrically standardized imagery and the absolute spectral reflectance values but relative values and textures. Therefore the method assures the analysis of different sensors of satellite and aerial imagery as well as different regions on small- or large-scales.
.
