Land-Cover Model (LCM)

The objective of the Land-Cover Model (LCM) is to simulate global land-use and land-cover changes by reconciling the land-use demand with the land potential. The basic idea of the model is to keep changing gridded land cover within different world regions until the total demands for this region are satisfied. The LCM, which is part of the Terrestrial Environmental System (TES), is driven by changes in the demand for food and feed as computed by the Agricultural Economy Model (AEM) and by changes in the potential vegetation as simulated by the Terrestrial Vegetation Model (TVM). The output of the LCM is used by the Terrestrial Carbon Model (TCM), the Land-Use Emissions Model (LUEM) and the Land Degradation Model (LDM).

Model input	Land cover previous time step
	Demand for food, feed, biofuel crops and timber products
	Potential vegetation
Assumptions	Reduced potential productivity of crops
	Population density
	Management factors and cropping intensity
	Initial land-cover map from 1970 (see Initial State Indicators)
Model output	Updated land cover map (0.5 by 0.5 degree grid)

Five steps can be distinguished within the model:

1. Adaptation of natural vegetation
   • Potential migration zones are calculated using maximum dispersal distances and migration rates
   • Potential and natural vegetation are compared at the grid-cell level
   • If differences are encountered:
   • The ability of those cells to adapt is assessed (i.e., for adaptation cells must be within the potential migration zone)
   • Cells that are able to adapt will convert from the original to a new vegetation type using assumptions for transition periods.
2. Treating unsuitable land and extensive grassland
   • Extensive grassland is defined as land with 'grass and fodder species' with potential productivity less than 25% of the theoretical maximum potential
   • Extensive grasslands are fixed over time
   • Agricultural land, not being extensive grassland, is considered unsuitable or too marginal for agriculture if potential productivity drops below 10% of the theoretical maximum potential
   • Unsuitable land reverts back to its natural vegetation.
3. Extraction of timber
   • Agricultural land, regrowth forest and protected reserves are excluded
   • Preferences of grid cells for timber extraction are based on:
   • Minimal distance to agricultural land, regrowth forest, and large rivers and other bodies of water
   • Forest coverage within a cell
   • Random preference
   • Timber is extracted from cells with the highest preference values until regional timber demand is satisfied
   • Vegetation regrows to its original state after exploitation, unless it will be converted into agricultural land.
4. Abandoning and reallocating existing agricultural land
   • Agricultural grid cells are sorted according to their crop productivity
   • Starting with the grid cell having the highest crop productivity from the previous time step, a certain amount of the area within each cell is allocated to a particular crop on the basis of the 'local' potential productivity for that crop, as well as changes in regional demand
   • Agricultural land that is not allocated is taken out of production. This land is either not needed to meet crop demands or has become unsuitable to meet these demands
   • After being abandoned land reverts back to its natural vegetation.
5. Expanding agricultural land
   The following procedure is applied if additional agricultural land is needed to cover the demand:

• Agricultural land and protected bioreserves are excluded
• Preferences of grid cells for expansion of agricultural land are based on:
  • Minimal distance to agricultural land and large rivers and other bodies of water
  • Potential productivity of crops
  • Population density
  • Random preference
• Expansion of agricultural land starts with grid cells with highest preference values, until regional crop demands are satisfied or until all suitable land is used
• Crops are allocated over these cells using potential productivity and remaining crops demand.

For a full description of the land-cover model, see Alcamo et al. (1998).

Built-up area

Built-up or urban area is derived from the DIScover database (Belward and Loveland, 1995). This database has been adopted by the International Geosphere-Biosphere Programme Data and Information System office (IGBP-DIS). The original database can be obtained from the USGS NASA Distributed Active Archive Center. The fraction of built-up area for each 0.5 by 0.5 degree grid cell in IMAGE 2.2 is calculated by dividing the number of original 1 x 1 km cells classified as 'urban and built-up' by the total number of 1 by 1 km cells included in the 0.5 by 0.5 degree cell considered.

In the land-cover model the built-up area is assumed to be not available for any other type of land use. The size of the built-up area is fixed, which means that IMAGE 2.2 does not simulate changes in the size of the built-up area.