Modelling soil sealing density in residential areas for Flanders and the Brussels Capital Region

Frank Canters, Sven Vanderhaegen

Research output: Chapter in Book/Report/Conference proceedingConference paperResearch


This paper presents a modelling approach for estimating the density of sealed surface cover within residential areas, using a set of explanatory variables describing the spatial context of each area, its morphological properties and its
population density. The model is calibrated and applied on Flanders and the Brussels Capital Region using a 30m sealed surface cover fraction map for 2013, obtained through spectral unmixing of four Landsat 8-OLI scenes covering the study area. The work demonstrates that sealed surface cover distribution in
residential areas can be explained by the density of different land uses in an area’s surroundings, the distance to transport features (roads, road crossings, railway stations) and the positioning with respect to urban clusters of different size, in combination with spatial metrics describing the structure of the built-up area or, alternatively, the area’s population density. Models of this kind might be used in combination with land use and population forecasts to obtain more reliable estimates of sealed surface cover density for future time steps.
Original languageEnglish
Title of host publicationProceedings of the 2017 Joint Urban Remote Sensing Event (JURSE)
Place of PublicationDubai United Arab Emirates
Number of pages4
ISBN (Electronic)978-1-5090-5808-2/17
ISBN (Print)978-1-5090-58099
Publication statusPublished - 2017
EventJoint Urban Remote Sensing Event - Ritz-Carlton Hotel Dubai, Dubai, United Arab Emirates
Duration: 6 Mar 20178 Mar 2017


ConferenceJoint Urban Remote Sensing Event
Abbreviated titleJURSE
Country/TerritoryUnited Arab Emirates
Internet address


  • sealed surfaces
  • remote sensing
  • spatial modelling
  • regression trees
  • Flanders
  • Brussels


Dive into the research topics of 'Modelling soil sealing density in residential areas for Flanders and the Brussels Capital Region'. Together they form a unique fingerprint.

Cite this