Statistical processing

Source data

The compilation of statistics on land cover is based on a several maps on land coverage. The maps are all produced outside Statistics Denmark and are used for a variety of administrative purposes. Sources are:

• GeoDanmark - a Danish topographical map, maintained by the municipalities and the Agency for Data Supply and Infrastructure.
• The Danish Cadastral Map from the Danish Geodata Agency
• Fields and field blocks, an agricultural map from the Danish Agriculture Agency.
• Management plans state forests
• Management plans Danish Defense
• Map of Natura2000 habitat types
• Map of protected habitat types

For land use, the sources used are:

• The Danish Cadastral Map from the Danish Geodata Agency
• The Danish register of buildings and dwellings
• The Danish register of businesses, as well as the register of agricultural establishments
• The civil registry
• Municipal plans
• The Danish Tax authorities register on properties

Number of inhabitants is from FOLK1A as of 1 Jan.

Frequency of data collection

Statistics Denmark does not collect primary data for these accounts. The sources for the land cover accounts have their own, separate update frequencies, independent of the statistical use. Land use is based on registers/data already in Statistics Denmark.

Data collection

No separate collection of primary data for this statistics. Data has been downloaded from other public authorities.

Data validation

Land cover: As the source maps are put together and overlaid, a number of corrections and filterings are done to remove inconsistent geometrical features. For instance when different sources have slightly different mappings of the same lake or other feature. Also small 'slivers' created in the overlaying of maps are removed. All the corrections and manipulations of the source maps are documented in details and with illustrations in the technical report.

Land use: Data has been validated throughout the process of developing the method for classification of land use based on registers. This includes plausibility checks of results for land where use is well-known, as well as checks of relative sizes of different uses, in relation to each other as well as in relation to employment.

Data compilation

Land cover: First, data are collected, i.e. downloaded from a number of sources (se 'sources'). Then the many source maps are put together through GIS techniques. This part of the processing is done by DCE, Århus University and documented in a technical report. Statistics Denmark receives data as one consolidated map of land cover. Then the map data are aggregated to tables by land cover category and geographical regions. Finally, calculations of shares (percentages) and areas per capita by regions are done.

Land use: Data on land use from the registers have been linked to land parcels through coordinates of addresses, and in some cases other GIS-techniques. The result is a data set where each parcel is assigned one or more land uses - a number of parcels have no uses registered. In the cases of more than one use for one parcel, the area is distributed between different uses (e.g. dwelling and business, or two different kinds of business) using information from the buildings register (when applicable, otherwise, equal distribution is assumed). The largest parcels (>5 km²) are classified in a manual process. Finally, data are aggregated from the detailed use information to the land use classes used for publication.

Adjustment

No corrections are made besides what is described under statistical processing and data treatment.