Simulating regional urban development under land use planning constraints

Corentin Fontaine

    Research output: Contribution to conferenceAbstractpeer-review


    The primary objective of our Regional “Urban” Growth model (RUG) is to simulate, at a regional scale, the spatial evolution of urban and semi-urban land uses according to
    socio-economic scenarios that sketch individual preferences and regional planning options adopted within different future socio-economic contexts. The model considers urban
    land uses to be not only a function of location accessibility but also to depend on
    [hypoth. 1] location preferences, i.e. local externalities (Caruso et al., 2005), and
    [hypoth. 2] land use planning choices (Shackley and Deanwood, 2003).
    We built RUG as a residential model with no direct consideration of commercial, industrial, services or leisure land uses. In this framework, the term urban should be understood as any human settlement whether in cities, suburbs or rural zones. It is important to note that RUG does not aim to understand why people go to a particular location, but instead represents where people go when they make choices according to a particular set of goals and preferences. The three main steps followed to develop our model are : [1] the choice of the mathematical functions, [2] the tests and validation of the model structure and [3] the calibration with observed data in two regions of England . Subsequently, the RUG output is going to be used within the RegIS2 tool, to assess impacts on four regional sectors.
    This research mainly investigates the mathematical functions that best represent the mechanisms of residential location choice. The core of RUG is a combination of
    [1] an accessibility component, based on the well-known distance decay function,
    [2] a measure of individual preferences, based on local amenities, after Caruso et al. (2005) and [3] an original representation of the regional planning constraints that we propose. The result is a non-linear model based mainly on different types of exponential functions to combine the variables and their parameters.
    Monte Carlo-based uncertainty and sensitivity analysis are performed on the model, using SIMLAB 2.2, a public domain software developed by the Joint Research Center of the European Commission (Crosetto and Tarantola, 2001).
    Regional space is seen as a regular grid of cells of 1 km resolution. Both spatial and non-spatial data are used. Spatial data include [1] population location at baseline time ,
    [2] measures of location accessibility , [3] percentage of urban and semi-urban land uses and [4] location of potentially restricted areas . Non-spatial data include [1] the regional population at time t and [2] the occupancy rate (average number of people per household) at time t.
    First results give good confidence in the methodology, e.g. the modification of the local externality parameters returns various urban development patterns that are plausible.
    The resulting output product of RUG has been formatted in such a way that it can also be used as an input to the meta-models of the RegIS2 project, a scientific partnership of the University of Louvain and five British Universities . The objective of RegIS2 is the production of a computer tool which can be used by non scientists (planners and other stakeholders) to evaluate future climate and socio-economic change impacts on agriculture, biodiversity, hydrology and coastal zones in two regions of England, namely East Anglia and the North West. In this context, RUG has been built as a stand-alone model (within MATLAB7) to allow a user to set most parameters through an interface; the choices being made either from personal knowledge or from expert judgment (predefined socio-economic scenarios). The RegIS2 contract does not include a proposal to model urban land use change but the previous RegIS project highlighted the major interests of regional stakeholders in urban land use planning. It was considered to be both appropriate and useful, therefore, to examine what could be achieved within the RegIS2 framework with a simple urban land use model, such as RUG, that allows exploration by planners of the environmental consequences of alternative urban development pathways.
    Original languageEnglish
    Publication statusPublished - 2005
    EventEuropean colloquium on Theoretical and Quantitative Geography - Tomar, Portugal
    Duration: 9 Sep 200513 Sep 2005

    Scientific committee

    Scientific committeeEuropean colloquium on Theoretical and Quantitative Geography


    • Planning
    • Regional scale
    • Urban model
    • Mathematical development
    • Urban and semi-urban land use
    • Local externalities
    • Uncertainty analysis
    • Sensitivity analysis


    Dive into the research topics of 'Simulating regional urban development under land use planning constraints'. Together they form a unique fingerprint.

    Cite this