A historical population and densification analysis of Flanders and Brussels (1986-2013) to calibrate an activity-based land-use change model

Several land-use models have been developed to evaluate the impact of land-use changes and to monitor the ongoing growth of urban sprawl. Cellular automata (CA) models can provide spatially explicit simulations of urban growth for large regions. However, in areas with strongly mixed land uses, like Belgium, different types and intensities of human activity occur within a single dominant land use. A CA model normally does not directly simulate these activities. Recently, White et al. (2012) and Crols et al. (2015) proposed a multiple activity-based CA model that also projects population and employment at the model resolution. To produce the required historical land-use and activity data, we combined high resolution data of today with socio-economic statistics of the past and a soil sealing time series derived from remote sensing (Vanderhaegen and Canters, in press), and hypothesise a local relationship between soil sealing, built-up land uses and population. We analysed this hypothesis and developed historical maps for 1986 and 2001. The population and sealed surface maps provide interesting data on the urban sprawl phenomenon in the past decades. By computing a densification index we observe that most urban areas witness a recent population density increase while in several rural areas the built-up area per inhabitant is still growing. The resulting maps of 2001 are currently used in a historical calibration of the model of Crols et al. (2015) for the period 2001-2013. The calibration enables a thorough analysis of the model components and the parameter choices, and may lead to an improved activity-based CA model.