Many land use models have been developed to evaluate the impact of land use changes on the environment and on the socio-economic system. Recently, White et al. (2012) proposed a multiple activity based cellular automata (CA) model that not only predicts abstract, dominant land-use states for each model cell, but also several socio-economic activities (e.g. population and employment). They included long-distance interactions directly in the CA influence rules by using a variable grid structure. Crols et al. (in review) updated this model in order to make the influence rules dependent on transport network travel time between variable grid cells instead of Euclidean distance. At present, this update was applied to Belgium, but has not yet been evaluated with a historical calibration due to a lack of historical land use and activity data at a high spatial resolution. Therefore, we wish to simulate these historical data with an approach that combines the available high resolution data of today, socio-economic statistics of the past, and high resolution proxy data of the past derived from remote sensing. As proxy data, we obtained the estimated percentage of soil sealing per pixel for the regions of Flanders and Brussels for the years 1986, 2001 and 2013 at a resolution of 30 m, developed in an extended version of the work of Vanderhaegen et al. (2015). We then hypothesise a local relationship between soil sealing, built-up land uses and population. In this presentation, we will analyse and validate the proposed hypothesis, and use several initial conditions to run the activity based CA model from 1986 to 2013. This approach leads in a next step to the historical calibration of the model of Crols et al. (in review), to evaluate its components and parameter choices.
|Status||Published - 2015|
|Evenement||19th European Colloquium of Theoretical and Quantitative Geography (ECTQG 2015) - MAIB, Valenzano, Bari, Italy|
Duur: 3 sep 2015 → 7 sep 2015
|Conference||19th European Colloquium of Theoretical and Quantitative Geography (ECTQG 2015)|
|Periode||3/09/15 → 7/09/15|