The GIS spatial data structure is a raster data structure because these data are typically derived from scanners mounted on aircraft or satellite platforms. Raster data structures tessellate space and assign each spatial element a unique value, thus providing explicit information for each location. The most common raster structure is a square lattice whose values are stored as two-dimensional arrays on the computer which includes many advantages as:

• simplicity,
• easiness of image display and processing,
• easiness of data aggregation and data overlay,
• uniform cell size and shape for multidimensional spatial analysis and spatial simulation modeling.

The maps module contains some regular tessellation maps (512 512 pixels).

The manner in which geographic information is represented is linked with spatial analysis and the modeling of geographic phenomena. The maps used here are represented as a matrix of cells.

First, GEODE laboratory had carried out a simple photo-interpretation of three series of aerial photography taken at 1942, 1962 and 1993, to determine forms and boundaries of the forestry regions at the specified years. Next, the resulting maps were digitized and geometrically corrected to make them superposable and represented in a raster form. The data model used here is a series of "snapshot" images. This type of representation employs a sequence of images which are a sequence of n thematic maps of the region area analyzed.