INPUT - European Population Map 2006

Population figures are usually collected by national statistical institutes at small enumeration units (e.g. census tracts or building units). However, still for many countries in Europe, data are distributed at coarser geographical units like municipalities. This level of resolution is insufficient for analysis in many fields. In addition, the heterogeneity of the size of the geographical units causes great distortions in analysis, i.e. the Modifiable Areal Unit Problem (MAUP) (Openshaw 1984). Dasymetric mapping techniques have long been applied world-wide to derive finer (and MAUP-free) depictions of the population distribution. These techniques disaggregate population figures reported at coarse source zones into a finer set of zones using ancillary geographical data. Previous attempts to map the European population at high resolution have used CORINE Land Cover (CLC) as the main source of ancillary data. In this article we test new geographical datasets to produce an updated and improved European population grid map. It is tested whether using more detailed ancillary data in the dasymetric mapping significantly yields higher accuracies. As final outcome of this cartographic exercise, a European population grid map for the reference year of 2006, with a spatial resolution of 100 x 100 meters, is presented and validated against reference data. Resident population reported at commune level, a refined version of CLC and information on the soil sealing degree are used as the main inputs to produce the final map.

Available land use/cover maps differ in their spatial extent and in their thematic, spatial and temporal resolutions. Due to the costs of producing such maps, there is usually a trade-off between spatial extent and resolution. The only European-wide, consistent and multi-temporal land use/cover dataset available is the CORINE Land Cover (CLC) map. Despite the value and usefulness of CLC, its minimum mapping unit of 25 hectares considerably limits its applications at large scales of analysis. Our objective was to improve the spatial detail of CLC 2006 by incorporating land use/cover information present in finer thematic maps available for Europe such as the CLC change map, Soil Sealing Layer, Tele Atlas® Spatial Database, Urban Atlas, and SRTM Water Bodies Data. Relevant data from these datasets were extracted and prepared to be combined with CLC in a stepwise approach. Each step increased the level of modifications to the original CLC. This process generated a newly refined version of the CLC 2006 map with an improved minimum mapping unit of 1 hectare for all types of artificial surfaces and inland waters, while keeping constant the original 100 meter cell size. A validation of the new map was carried out using LUCAS 2006 sample points (European Land Use/Cover Area frame Statistical survey). The observed increase in spatial detail was done, however, at the cost of the cartographic consistency.