The Land Use Classification, The Finnish Standard
Statistics Finland has published the Finnish land use classification standard for the use of official statistics. Besides official statistics, the new Land Use Classification is recommended for use in other information systems describing land or area use. The Land Use Classification is also a public administration standard (JHS 186).
Finland has not had a land use classification system for national use that would be widely spread, generally accepted and serve the needs of as many users as possible. Pressures to form this type of classification system and data have come from national and international needs. National needs are such as statistics compilation, regional analyses and planning; international are reporting obligations brought by the EU membership and in general, the supply of land use data to international organisations.
The National Land Survey of Finland started at the beginning of 1997 a project aiming to plan classification and data collection systems of land use and land cover suitable for Finnish conditions. The target was classification data containing national standard classifications for area use, land cover and soil suitable for most users and to which other, separately formed additional elements are as applicable as possible. Another objective was that the classification system satisfies, where possible, both national and international needs. The steering group of the project had experts from many different organisations.
Based on the needs analysis, it was discovered that in addition to land use and land cover, information was also needed about soil and restrictions related to land use as well as data formed with uniform methods from all over the country. In addition, geographic data are needed to help make statistical analyses and monitoring. The data should be in both vector and raster forms.
In connection with the project, it was considered expedient to differentiate the classification systems describing land use, land cover and soil. The system was named SLICES.
The point of departure for attaining an economic end result was that existing data should be utilised as much as possible. Various different national geographic data have already been collected in Finland, which are either as such or slightly modified usable source data for the classification of land use and land cover. The data are meant to be processed so that production would be as automatic as possible. Based on trial work, a suitable production unit is 80 x 80 km².
The introduction of the classification aims to harmonise the definitions and concepts used in land use statistics and thus advance comparability.
The Finnish classification standard differs from international classifications particularly in the respect that the leading classification criterion is land use from the viewpoint of economic activity. Thus the meaning of land area as a resource for economic activity determines to which land use category it is included. Land cover or soil are secondary criteria unlike in many other used classifications which are combined classifications of land use and land cover classifications.
The structure of the Finnish Land Use Classification is hierarchical and has three levels. The main categories, or the top hierarchy level, were formed similarly as many other, e.g. international systems: built-up land, forests (other land) and water areas. Sites classified as built-up land are generally constructions created as a result of human activity and are mainly used for living, working, movement or recreation. They are further divided into categories by the nature of the activity in the area concerned.
The new Finnish land use classification does not apply directly any existing international system. International classifications and Finland's land use classification share certain similarities, such as hierarchical nature and formation of the first level groups. There are differences as well; Finland aimed to produce a classification primarily suited for national use, while the purpose of the examined international systems is to make the same classification universally applicable.
The Finnish land use classification differs from international classification systems particularly in that land use, land cover and the classification of soil are differentiated from one another; land cover is the physical description of the area and land use is its operative description. In international systems categories concerning both land use and land cover are usually always combined. In that case, land use has often been the determining criterion and only such areas are classified by land cover that are considered to be without any special use or whose use according to land cover is self-evident.
A problem with the combined land use and land cover classification is often in that a site with certain cover can have several uses. Similarly, an area in certain use can have several different types of land cover. If a system covering all combinations is formed, the number of categories becomes considerably big. On the other hand, if the number of categories is kept reasonable, information is lost.
Information related to our living environment and land use and land cover classification data are needed on three different levels:
- Locally for operative planning of area use
- Nationally for land use policy and international environmental protection
- Internationally in order to produce comparative analyses and follow global changes.
The majority of use and users are connected to public administration. Potential local and regional users are such as municipalities, regional councils, research institutes and universities. There are considerably fewer users on the national level, such as the environmental administration, the Finnish Forest Research, the defence establishment, Statistics Finland and the Geological Survey of Finland.
Key classification principles and concepts:
Land areas are classified by their use, as well as according to what happens on them either naturally or as a result of human activity. A certain area can belong to only one land use category. The use of an area refers to its actual use for some purpose, not to its potential or planned use. The mode of use is determined by the only or the principal use of the site or such use that excludes all other modes of use.
The site to be classified must have a surface area. Thus, a point-form site or a site with a small area is not taken into account or classified.
The land use category is determined by the primary (principal) use or the economically most important mode of land use. Structures or modes of land use underneath the surface of land are not considered when determining the land use category. Thus, a bomb shelter built inside a rock does not determine the use category of the area “above” it, and a road or railway inside a tunnel is not a traffic area.
Areas under construction are included in their final use category only when the site is ready and taken into use. Until then, the area belongs to its previous land use category (e.g. building permit areas) or the category “Other land” (e.g. large road work land of many years).
Areas left out of permanent use (e.g. empty industrial areas) areas as a rule included in their previous land use category unless the area is taken into other use. If the area cannot, however, be reasonably used for its previous use purpose (e.g. abandoned farms in disrepair, closed mines, closed/landscaped landfill sites, road foundations left out of use, etc.), and the area has not been taken into other use, it is classified under “Forestry land” provided that the production capacity of soil corresponds to the criteria set at least for forest land of low productivity, and a use restriction preventing agricultural use is not directed to the area; otherwise to the category “Other land”, that is, areas without use.
Temporary use of an area for another purpose does not change the permanent land use category of the area (for example, a temporary raw timber storage on a field is still a field). In addition, temporary disuse or operational stoppage does not change the land use category (e.g. a power plant's production stoppage).
The prevailing use of a building refers to the use for which the biggest part of the building's total floor area is used.
The prevailing purpose of use of a building refers to the mode for which the majority of the building's total floor area is meant regardless of the fact whether the building is temporarily empty or in some other use fully or partly.
The surrounding land use refers to the type of land use in which area the site is located or land use in the immediate vicinity of the site. Areas considered minor in land area or significance, such as a barn in a field, are classified by surrounding land use unless otherwise stated in the site model. Surrounding land use is an auxiliary concept not a classification category.
The main groups of the land use classification are indicated with characters. There are nine main groups and they are as follows:
A Residential and leisure areas
B Business, administrative and industrial areas
C Supporting activity areas
D Rock and soil extraction areas
E Agricultural land
F Forestry land
G Other land
H Water areas
I Special use areas
Coordination of land use statistics
When supplying data to international organisations, the data content should be uniform for all countries, otherwise it is difficult to compare data to each other. The formation of the classification system is affected by the use purpose of the classification (only needed data are collected) and the surrounding reality (the system does not include groups that are presumably not present in its use area). In Finland the national land use classification system was developed to be so flexible that international information needs and data materials can be derived from them with a separate calculation. Only the data for which this is not possible are collected separately. Some of the data can be collected from sector authorities as until now.
Concepts and definitions are made more uniform in various coordination projects. However, the development and coordination of different types of land use data and classification methods are fairly new in Europe. Finland is involved via the EU, for example, in international projects (EEA/CORINE Land Cover; Eurostat Land Use), which aim to help develop databases describing national land use and land cover into as comprehensive as possible and on the other hand, make them mutually comparable. CORINE LC is the most concrete one of these.
At the beginning of the SLICES project, it was emphasised that special attention should be paid to the harmonisation of land use statistics with Eurostat's statistics and with the requirements set by the CORINE Land Cover database describing European land use. Thus the research work studied the structure of some international classification systems (CLUSTERS, ECE, Nordisk Areal Klassifikation, CORINE Land Cover) and their suitability as such or modified to Finland.
CLUSTERS (Classification for Land Use Statistics: Eurostat Remote Sensing Programme) is a trial project implemented in connection with Eurostat’s research programme "Remote Sensing and Statistics". The aim was to develop a land use classification that combines both needs related to statistics and land use planning and that could be implemented through remote sensing technology.
CLUSTERS was made uniform with other European classification systems, such as
Agricultural land: CRONOS (Eurostat, Classification Scheme of Surveys on Agricultural Structures)
=- forests: Eurostat's definitions for forest
=- environment: ECE and OECD's classifications
The CLUSTERS nomenclature is hierarchical and has four steps. The classification also includes groups that primarily describe land cover and not land use, such as “Mountain bushes”. The class system is very detailed considering it is a question of a Remote Sensing system. The number of groups on the
On the first level
On the second level
On the third level
On the fourth level
Included are groups that most likely cannot be classified with any map data (e.g. “Trade and financing”, “Heavy industry”). It is possible that the classification is not meant to be taken up to the 4th hierarchy level in all respects.
CLUSTERS includes a considerable number of groups that are not present at all in Finland or their area and significance are marginal. On the other hand, there are quite few classes in Finland that are not in CLUSTERS, but its class division is not the best possible considering our country's circumstances and national needs. Particularly for agricultural land the classification is too detailed.
ECE land use classification
The classification of the Economic Commission for Europe (ECE) under the UN is called the ECE Standard Statistical Classification of Land Use, that is, it was made specifically with regard to statistical aspects. The aim is to provide a “platform” to which national datasets could be gathered for international comparison and analysis. The classification has aimed to take into account the following four areas of use:
- description of present land use (statistics compilation)
- effects and needs of human activity
- environmental protection and ecology
Use possibilities of areas in future (planning).
It is said in the preamble of the classification that it does not offer very good possibilities for monitoring changes in use nor information on “reserving” an area for certain use, by which the spreading of habitation could be followed and assessed, for example.
The ECE classification has in a way two phases: the first step of the hierarchy level describes land cover and levels 2 to 3 land use (except in main group 2 where tree species are the classification principle). The third hierarchy level of the classification is “optimal”; it is therefore not necessarily meant for operative use. The minimum size of the classification is 0.5 ha, which is not, however, the requirement for supplying data.
The ECE classification is used as a basis for data collection of the OECD and FAO relating to land use and the environment.
Nordic classification of area use
In 1981, the meeting of the Nordic Prime Ministers made a proposal to the Nordic environmental statistics committee about the preparation of a joint Nordic classification of area use in Nordic cooperation. The classification was planned in 1981 to 1982 by a joint Nordic working group apart from Iceland; the classification is a Nordic counterpart of the ECE classification. The Meeting of Ministers recommended in 1982 the adoption of the classification in Nordic research use. The classification is particularly meant for comparisons of data between the Nordic countries.
The Nordic area use classification has not been used in Finland in compilation of statistics on land use. The classification has not been established in any other Nordic country either. Studies on the use and position of the Nordic classification of area use during this survey have not been made in other Nordic countries. Statistics Finland provisionally prepared Finland’s national land use classification based on the Nordic classification of area use, but the work was combined into the SLICES project.
CORINE Land Cover
Corine is also a mixed system of land use and land cover as all the other above-mentioned nomenclatures. The nomenclature of Corine has three steps. For Sweden's national needs, a “national Corine” including additional groups 4 to 6 was planned.
Corine is formed from each EEA Member State as a separate project using national topographic maps of scale 1:50 000 to 1:250 000 and satellite images (SPOT, Landsat). Aerial view data are used as reference data for satellite image interpretation, that is, as classification conversion keys by means of which satellite image interpretation is specified. The classification is relatively less detailed. As its smallest area unit was chosen:
- minimum figure 25 hectares
Minimum width (road, river) 100 metres.
The minimum figure of the classification, 25 hectares, means in Finland's small feature natural environment a considerable generalisation. The data are meant to be used in the scale of 1:100 000, when the minimum figure is 0.5 x 0.5 km, on the map 0.5 x 0.5 cm. The generalisation is being changed to the direction that the topography is not preserved, which further weakens the quality of data.
Due to the minimum figure size, the group system is formed so that there are separate groups for “mixed” use areas (similarly as in CLUSTERS). This particularly concerns agricultural areas and built land where land use can be quite minor. The group system includes in all:
On the first level 5 groups
On the second level 15 groups
On the third level 44 groups
Fifteen of the third level groups are such that are not present in Finland at all. In addition, there are also some groups in which Finland does not in practice reach the above-mentioned minimum size (e.g. ‘fruit tree cultivations’ which are very small in size).
The Finnish Land Use Classification was compiled on the basis of international land use classifications and the so-called SLICES project started by the National Land Survey.
In future, Statistics Finland aims to develop more widely Finland’s land use statistics and inventory of land use and produce statistical information on land use at least by main category because land and water areas are the basis of our living environment and all economic activity. Land use data are significant particularly for developing environmental-economic accounts and area use accounts.
In the SLICES project, the so-called stacking order regulates the determination of the land use category:
- When the same area is used for several different purposes (for example, the area is both a field and a power line area) and only one, primary use purpose can be/is to be described in the data for this area.
- Or when a certain use category is ‘inside’ another or goes ‘through’ it (for example, a lake in a park area or a road in a field area).
The stacking order is primarily determined by the significance of area use and secondarily based on the accuracy of source data. A group lower in the stacking order remains ‘below’ a higher one, that is, a lower site is not described if the same area already belongs to a higher hierarchy use category (a stronger hierarchy beats a weaker one: the site is already 'reserved’ for certain use). Overlapping may be caused by the area being actually used for several purposes or by inaccuracies or allocation difficulties in the source data. Those on the same level cannot be overlapping use categories (the same source data) in this stacking system, but the following stacking order is followed.
Site category - Source data
Roads - Road base (Road database/National Land Survey of Finland)
Railways - PerusCD (National Land Survey of Finland)
Private roads - Road base (Road data base/National Land Survey of Finland)
Fields, pastures, other agricultural land - Field database (Farm register; Information Centre of the Ministry of Agriculture and Forestry)
Other traffic areas - Separate digitalisation or PerusCD (National Land Survey of Finland)
Recreation and leisure areas (golf fields, etc.) - Separate digitalisation or PerusCD (National Land Survey of Finland)
Recreation and leisure areas (Park areas) - Separate digitalisation or PerusCD (National Land Survey of Finland)
Excavation areas of ore and other soil materials - Separate digitalisation or PerusCD (National Land Survey of Finland)
Landfill areas - Separate digitalisation or PerusCD (National Land Survey of Finland)
Water areas: inland water areas and sea areas, reservoirs - Finnish Environment Institute (waters)
Fallow land, disused fields - PerusCD (National Land Survey of Finland)
Energy supply areas, power lines - FINGRID/National Land Survey of Finland
Residential, service, industrial and other built areas - Register of Buildings and Dwellings (Population Register Centre)
Forestry forest land, forest land of low productivity and waste land - NFI (National Forest Inventory; Finnish Forest Research Institute)
The list is interpreted as follows, for example:
- Intersection where road, railway and private road intersect: The use category of the area is road.
- Railway in an industrial area and private road on housing plot: Areas are Railway and road are not industrial area/plot.
The stacking order is finally determined by the data available.