Urban land cover mapping, using open satellite data. Case study of the municipality of Thessaloniki.

Abstract

The use of open satellite data has become a valuable tool for monitoring the urban environment due to high temporal, spectral and spatial resolution. Planners and stakeholders can rely on these data since they offer an alternative point of view of a very complex urban scenery. The processing of satellite images with the combined use of different types of data (statistical, in situ etc) can extract information for the present situation, in urban and suburban areas. This study use image data from the Sentinel satellite mission, in order to extract urban land cover characteristics in Thessaloniki city. The Sentinel mission started the image acquisition on 2014. The data set are freely available for downloading. The use of high resolution satellite images (e.g IKONOS) in urban studies, has been proved very useful all these years. On the other hand the use of a medium size spatial resolution satellite, as Sentinel-2, with 10m pixel size, is a very promising optical earth-observation satellite, which is tested in this study. Image processing techniques used to produce classified map of urban characteristics and land cover classes (built up areas, vegetation etc). The contribution of a radar image of Sentinel-1, in the classification process, is also tested. All the above processing is done with an automatic procedure, giving additional value to the proposed methodology. The final classified map will give planners a management tool for decision-making concerning the sustainable urban development and will be the basis for implementing hydrology studies, planning, land use change detection and other.