Skip to content

Land Use Scanner

Jump to bottom
Jip Claassens edited this page Feb 1, 2023 · 36 revisions

introduction

VU University Amsterdam and Object Vison have a long-standing experience in integrated land use modelling. Research is done in both a national and an international context. Applications include studies into the future of agriculture in the Netherlands, deforestation in Surinam, climate impacts and adaptation measures. Most of this research was carried out for and with an operational land use model that integrates urban and non-urban types of land use, the Land Use Scanner.

model outline

Land Use Scanner is a geographical information system (GIS)-based model, developed by Hilferink and Rietveld (1999) as an operational tool for the integrated simulation and evaluation of future land-use patterns. Depending on the application, between 10 and 20 land-use classes are distinguished, typically ranging from agricultural and natural on the one hand, to urban (industrial, residential, et cetera) on the other. The model input is specified per land-use class and consists of: (i) local suitability maps, usually per 1-hectare grid cells; and (ii) regional land-use demand. The model produces projections of future land use, per grid cell. In addition to local conditions (e.g., soil type), the suitability maps may also contain reference to the wider spatial context (e.g., land use in neighbouring grid cells or the density of specific facilities within a specified radius). The regional demand for different types of land use is usually derived from external, sector-specific models. The model can apply different procedures to allocate land to the most suitable locations depending on the objective of the study at hand. When land use is described in a discrete manner (with only one type of dominant use per cell) the allocation ensures that: (i) aggregated (i.e., for the study area at hand) suitability is maximised; and (ii) regional land-use demand is met (Koomen et al., 2008). The model is also able to deal with a continuous description of land use per grid cell (with different fractions devoted to individual uses) and simulate the competition between different land-use types for limited space. This process is based on the discrete choice theory formulated by McFadden (1978) that explains the choice behaviour between mutually exclusive alternatives from an economic perspective. An extensive description of the different model versions is provided elsewhere (Koomen et al., 2011a).

One of the main strengths of the model is the option to generate ‘what-if’ type simulations to explore different future scenarios or policy interventions and to visualise and communicate expected outcomes. It can do so more or less interactively. As such, the model is not merely a modelling tool, but more importantly, a communication tool stretching different domains that connects analysts with policy makers (Koomen et al., 2011a). See, for example, future land-use projections in Flanders (De Moel et al., 2012), simulations of residential land use in the Elbe River Basin (Hoymann, 2010), assessments of flood risk along the river Rhine (Te Linde et al., 2011), and climate adaptation measures in the European Union (Verburg et al., 2012).

Furthermore, note that Land Use Scanner in most applications relies strongly on external model results as inputs, while its outputs are regularly used in dedicated impact assessment models. The model is used, for example, in all land-use related impact assessment studies by PBL Netherlands Environmental Assessment Agency that study, for example, future flood risk scenarios (Rijken et al., 2013) and potential urban transformation processes (van Duinen et al., 2016). Moreover, the Joint Resource Centre of the European Commission uses a very similar approach in their ‘Land-Use-based Integrated Sustainability Assessment’ (LUISA) modelling platform, which is based on Land Use Scanner (EC-JRC, 2016; Lavalle et al., 2011). Figure 1 provides for a graphical depiction of the structural design of the Land Use Scanner modelling framework.

The model integrates the best available knowledge and insights from the various fields involved. The output from impact models can be iteratively incorporated in Land Use Scanner to, for example, update local suitability or regional land demand, or both. Another feature of the model is that it integrates developments in all types of land use, making it different from the many models that mostly focus on developments in urban land use, e.g., UrbanSim by Waddell (2002) or SLEUTH by Clarke et al. (1997) or rural types of use, e.g., ProLand by Möller and Kuhlmann (1999). In most applications, Land Use Scanner simulates transitions from one distinct type of use to another, while UrbanSim, for example, also models changes within specific land-use classes and their function. Land Use Scanner is thus better equipped to simulate more drastic transitional processes, whereas UrbanSim can be used, for example, to model more gradual transformation processes such as locational choices of households within urban areas. To some extent, these differences can be classified as focussing on land cover and land function respectively, as was discussed extensively by (Jacobs-Crisioni et al., 2017). This distinction will be further blurred, however, as ongoing model development work results in a Land Use Scanner version that incorporates more economic logic and better replicates the land market (Borsboom-van Beurden and Zondag, 2011; Koomen et al., 2015).

This section is an excerpt from Claassens, J., Koomen, E., & Rijken, B. (2020). Linking socio-economic and physical dynamics in spatial planning. In S. Geertman, & J. Stillwell (Eds.), Handbook of Planning Support Science (pp. 383-396). Edward Elgar Publishing.

development

This integrated GIS-based land use model was first developed in close co-operation with the Netherlands Environmental Assessment Agency (PBL), Geodan and the Agricultural Economics Research Institute (LEI). It has been under constant development since its conception, and most recently by the Netherlands Environmental Assessment Agency (PBL) and Object Vision for several urbanisation studies (e.g. PBL, 2021).

The GeoDMS framework underlying Land Use Scanner is developed and maintained by the Object Vision company and also applied in other spatially explicit models. Examples include a land-use modelling framework for the European Commission that covers all member states of the European Union at a 100-metres resolution (Lavalle, et al., 2011) and the 2UP model that simulates urban development and population change around the globe (Koomen, et al., 2023). A full account of the original Land Use Scanner is provided by Hilferink and Rietveld (1999), whereas recent applications are documented in a book by Koomen and Borsboom-van Beurden (2011).

This website provides more information on the outline and characteristics of the model. A fully functional demonstration version of the model can be downloaded here.

variants

Over the decades, many versions of the Land Use Scanner have been developed, such as:

Land use modelling documentation

© Object Vision BV

Clone this wiki locally