The Phare CORINE Inventories in CEE: A support System for Environmental Decisionmakers

George Büttner
Phare CORINE Land Cover Project
FÖMI Remote Sensing Centre, Bosnyák tér 5, Budapest, Hungary
e-mail: Buttner@rsc.fomi.hu

THE CORINE PROGRAMME

From 1985 to 1990, the European Commission founded and funded the CORINE programme. The objectives of CORINE (Coordination of Information on the Environment) were to gather environmental data, coordinate related activities, and improve the consistency of information on the state of environment in the European Union. These objectives have been pursued and have led to two main results:

The CORINE Information System includes several databases (European Commission, 1994a):

In June 1991, during the Conference of European Environment Ministers at the Dobris Castle (Czech Republic), it was decided to extend the CORINE methodology to the countries of Central and Eastern Europe as a first step towards the integration of environmental information systems throughout Europe. The Phare Programme has provided the framework for the implementation of the three most important components of the CORINE Programme: inventory of air emissions, biotopes and land cover. In 1994, the European Environment Agency started its activities from Copenhagen (Denmark), taking over the maintenance and use of the CORINE Information System.

Corinair

Concern regarding atmospheric emissions in Europe has increased in recent years with the recognition that air pollution may not only be a health hazard, but also may be implicated in environmental problems, such as acid deposition and forest damage. In an attempt to control these problems, the EC adopted a range of legislation on atmospheric emissions. To support this legislation, and as a basis for further policy development, it is essential to have information on atmospheric emissions throughout Europe. Corinair, developed in collaboration with UNECE and OECD, has provided a European method for atmospheric emission inventories (European Commission, 1994b). After the creation of a database in 1985 covering the EC Member States, this inventory was extended to most countries in Europe. Results are available for 1990 (Corinair90), covering 30 countries (including the Phare countries). These include emissions of eight main pollutants (SO2, NOx, N2O, NH3, CH4, CO2, CO, NMVOC) in seven main categories (residential and industrial combustion, oil refining, production processing, solvent evaporation, road transport and nature) and more than 200 sub-types of sources. Large point sources (polluting plants) are also included in the database. Corinair95 has been completed for the EU countries, while the database for Phare countries for 1994 will soon be available. From 1995, the EEA has decided to update Corinair on a yearly basis in all EU and Phare countries (Kutas, 1998).

CORINE Biotopes

An inventory of sites considered to be worthy of nature conservation measures at the EC level has been completed for the CORINE Biotopes Programme. The inventory constitutes a major tool for the implementation of harmonised environmental policies in the countries of Europe. The selection of sites for inclusion in the register is based on an agreed set of criteria. Some of these sites may already have protection status under national legislation, but many are unprotected. The inventory includes the following data: site name, site location, area, altitude, habitat type and cover, motivation, human activities, list of species (flora and fauna) and their number, site description (character, quality, vulnerability, designation, ownership, documentation) (European Commission, 1991). Together with the Phare countries, this database contains more than 10,000 site descriptions. CORINE Biotopes has been completed for Bulgaria, the Czech Republic, Estonia, Hungary, Latvia, Poland, Slovak Republic and Romania. It is ongoing in Lithuania, and will start soon in Albania, Bosnia-Herzegovina, Macedonia and Slovenia (Evrard, 1998).

CORINE Land Cover

The aim of the CORINE Land Cover mapping facility is to provide information on the state and changing biophysical coverage of the Earth's surface. The programme was initiated for the EU countries in the 1980's to provide quantitative information on land cover, at a scale 1:100,000. Today the CORINE Land Cover (CLC) database covers 31 countries across Europe and North Africa. CLC is mapped by interpreting satellite images, with the results stored as a database in a Geographic Information System (European Commission, 1993). The database represents a basic tool for studies on the environment, impact assessment and regional planning on national as well as on European levels. More detailed information on CLC is given overleaf, while Table 1 gives the status of CORINE Land Cover in the Phare countries.

Country Area (km2) Contractor Status No. of sheets
Albania 29000 GSC in progress 35
Bosnia-Herzegovina 51000 GZ BiH starting 1998 38
Bulgaria 111000 Ministry of Environment completed, 1996 105
Czech Rep. 79000 GISAT completed, 1996 94
Estonia 45000 EEIC completed, 1998 37
Hungary 93000 FÖMI completed, 1996 84
Latvia 64000 LEDC completed, 1998 44
Lithuania 65000 HNIT-Baltic completed, 1998 42
F.Y.R. Macedonia 25000 Ministry of Urban Planning, Construction and Environment starting 1998 22
Poland 313000 IGIK completed, 1996 297
Romania 238000 IGR completed, 1996 196
Slovak Rep. 49000 IG SAS completed, 1996 55
Slovenia 20000 GZS completed, 1998 20
Total: 13 1182000     >1069

Table 1: Status of CORINE Land Cover in the Phare countries (1998).

Topic Centres

The European Environment Agency (EEA), established by the European Commission, manages the European Environment Information and Observation Network (EIONET). Institutions or organisations across Europe have been contracted as European Topic Centres to execute particular tasks identified in the Agency's multi-annual work programme. Three of these ETCs have direct links to the above CORINE databases: The European Topic Centres on Land Cover (ETC/LC), Nature Conservation (ETC/NC) and Air Emissions (ETC/AEM). The main objective of the ETCs is to produce, provide and manage specific information for environmental policy development and implementation in Europe.

Since 1997, ETCs have been extended with support from the EC DGIA Phare Programme towards Central and East European countries with the creation of the so-called Phare Topic Links (PTL). With the set-up of three PTLs, this European network for land cover, nature conservation and air emissions information exchange has now been extended to about 30 countries. Figure 1 shows how these initiatives are inter-linked.

figure 1

Figure 1: Institutional connections within the CORINE Information System

TECHNOLOGY OF CORINE LAND COVER (CLC) MAPPING

The basic aim of the CORINE Land Cover Programme is to provide an inventory of the Earth's surface features. Computer assisted visual interpretation of satellite images has been the approach chosen to serve as the mapping methodology. The choice of scale (1:100,000), with a minimum area to be mapped (25 hectares) and a minimum width of linear elements (100 metres) represents a trade-off between cost and detail in land cover information. The nomenclature includes 44 classes in five groups: artificial surfaces, agricultural areas, forests and semi-natural areas, wetlands and water bodies.

Landsat's Thematic Mapper data has been the most widely used satellite imagery during the course of the Programme because it provides good area coverage, and sufficient thematic and geometric details at reasonable cost. Recent, new high-resolution satellites with a middle infra-red channel (IRS-1C/D, SPOT-4) certainly increases the possibilities. Figure 2 provides the workflow of CLC mapping while Table 2 reviews the most important technical parameters of the database.

Land Cover Mapping at the 1:50,000 Scale

Standard CLC maps are especially useful at the European level. To satisfy regional or local needs better, more details are needed both in terms of geometry as well as in terms of the thematic content. Several initiatives exist to extend the CLC methodology to a working scale of 1:50,000 and even 1:20,000 (ETC/LC, 1997a). Within the framework of the Phare Programme, an experimental project has been executed at the scale of 1:50,000 for 120 map sheets in four countries: the Czech Republic, Hungary, Poland and Slovak Republic. Four hectares was the size of the minimum mapping unit as a consequence of using integrated SPOT PAN and Landsat TM satellite data (compared with 25 hectares at a scale 1:100,000). International teams of experts have extended the standard nomenclature with level-4 classes representing the landscape conditions of the above four countries (Feranec et al., 1995). The number of level 4 classes was about twice that of the level 3 classes.

figure 2

Figure 2: Flowchart of CLC mapping.

  • Method: Computer assisted satellite photo-interpretation
  • Ancillary data: topographic maps, vegetation maps, land use statistics, etc.
  • Working scale: 1:100,000
  • Minimum mapping unit: 25 hectares
  • Minimum linear feature width: 100 m
  • Nomenclature: hierarchically structured in 3 levels: 44 classes for level 3
  • Cartographic projection: Lambert azimuthal equal area
  • Various projections in country databases
  • Implementation: national teams
  • Supervision: Land Cover Technical Unit
  • Overall classification reliability: better than 85 percent
  • Geometric accuracy compared to topographic map: <100 meters RMS

Table 2: Basic characteristics of the CLC database.

The experimental project proved that CLC mapping could take place on a larger scale. One of the ongoing activities of PTL/LC is to further extend the level-4 nomenclature to include all Phare countries. A new version of the CLC nomenclature at a scale of 1:50,000 for the Phare countries was finished in May 1998, within the framework of PTL/LC activities. This nomenclature includes 104 land cover classes. The main benefits of the extended nomenclature and the 4 hectare minimum mapping unit compared to the standard CLC mapping are:

UPDATING AND CHANGE DETECTION

Updating of data is a central question to any database collection which includes features that change in time. CLC can fulfil its aims only if the database is updated regularly. The proposed average frequency of update for the CLC database is once every 10 years. This doesn't rule out changes that may take place more rapidly in certain areas (e.g. urbanisation). With the accessibility of land cover data for more than a single date, one has the possibility to analyse land cover changes and to make predictions for the future. CLC mapping is a labour intensive methodology, requiring skilled photo-interpreters. Because of the nature of nomenclature and the rules of interpretation, updating also can not be automatic. Due to the fact, however, that land cover changes are generally slow, there is no need to repeat the interpretation during the course of updating, only to recognise changes that have taken place between one update and another. With the right hardware, this process is more than evident to a photo-interpreter who is familiar with the CLC methodology. Therefore updating would be significantly cheaper than re-producing the basic database.

The updating process is based on computer-assisted photo-interpretation (CAPI) technology, with simultaneous use of the basic CLC data, the corresponding satellite image map, and the new satellite image map (Perdigao and Annoni, 1997). The updating procedure usually reveals errors in the original database that first should be corrected, in order to avoid detection of false changes (PTL/LC, 1998; Büttner et al., 1998).

Once we have produced the CLC database for two different dates, change detection is an automatic procedure. The change database can be visualised by printing evolution maps and can be summarised using change indicators (ETC/LC, 1997b).

One of the tasks of PTL/LC is to produce a retrospective CLC database dating back to the 1970's for six Phare countries (Bulgaria, Czech Republic, Hungary, Poland, Slovak Republic and Romania) and to analyse the changes which have taken place during the twenty year period (PTL/LC, 1998).

APPLICATION OF CORINE LAND COVER DATA

The examples in Table 3 illustrate the most relevant applications of the CLC Hungary database. We can conclude that CLC data, when combined or integrated with other data sets in a GIS environment, constitutes geo-referenced data of basic importance to environmental analysis, catchment modeling, regional planning and crop monitoring.

Application Domain User Uses
Administration Somogy and Szolnok counties County base maps in GIS format
Agrometeorology Meteorological Service Database of arable lands
Catchment management Office of the Prime Minister GIS base layer map of the Lake Balaton catchment area
Catchment scale phosphorus pollution modeling (similar studies in Slovakia and Romania) Phare Multi-Country Programme in cooperation with the Phare Danube River Basin Programme Modeling run-off and diffuse phosphorus pollution in the Zagyva river catchment
Crop mapping and yield forecast Ministry of Agriculture Masking out non-agricultural areas
Environmental modeling (science) IACR Rothamsted (UK) Carbon cycle modeling
Environmental modeling (practical) Ministry of Environment Modeling effects of acid rain
Floodplain management Water Research Institute GIS base layer map
Regional planning Institute for Regional Development (VÁTI) Base map
Planning for sustainable agriculture (in preparation for EU accession) Ministry of Agriculture Elements of a multi-layered GIS

Table 3. Examples of CORINE Land Cover applications in Hungary (1997-98).

CONCLUSIONS

The results of the CORINE Programme in Central and Eastern Europe have been integrated into the environmental information system of the European Union. Due to the clearly defined, relatively simple technologies and good project management, uniform and harmonised databases on air emissions, biotopes and land cover have already been compiled for most of the Phare countries.

Examples of the applications of Land Cover data have shown that this data can also be used at a national level and can foster multi-country cooperation on various problems related to the environment. There is a clear and declared interest from the Hungarian government to apply the CLC methodology further, compiling a recent database at a scale of 1:50,000 and producing a retrospective database for period of the 1980's.

REFERENCES

Büttner, G., Maucha, G., Bíró, M., 1998. Land Cover change detection using the CORINE methodology. To be presented on ECO'BP 98 Symposium (ISPRS Commission VII), Sept 1-4, 1998.

ETC/LC, 1997a. Assessment of the existing experiences of the 4th and 5th level CORINE Land Cover nomenclature. Report prepared for the European Environment Agency.

ETL/LC, 1997b. Report on updating. Prepared for the European Environment Agency.

European Commission, 1991. CORINE Biotopes Manual, Office for Official Publications of the European Communities.

European Commission, 1993. CORINE Land Cover, Technical Guide, Office for Official Publications of the European Communities.

European Commission, 1994a. CORINE Information Systems, Summary Data Dictionary, EEA Task Force.

European Commission, 1994b. Corinair Handbook, Office for Official Publications of the European Communities.

Evrard, E. (1998) personal communication

Feranec, J., Otahel, J., Pravda, J., 1995. Proposal for a methodology and nomenclature scale 1:50,000 CORINE Land Cover project, Final Report, Institute of Geography, Slovak Academy of Sciences, Bratislava.

Kutas, T. (1998) personal communication

Perdigao, V., Annoni, A., 1997. Technical and methodological guide for updating CORINE Land Cover Data Base. JRC/EEA.

PTL/LC, 1998. Methods of Computer-Assisted Photo-interpretation and Land Cover Changes Detection. Prepared for the European Environment Agency.

ACKNOWLEDGEMENTS

The comments and suggestions of the following experts were highly appreciated in the production of this paper: C. Steenmans, Project Manager, EEA, Copenhagen, Denmark; M. Bossard, Leader of LCTU, IGN FI, Paris, France; J. Feranec, Member of LCTU, IG SAS, Bratislava, Slovak Republic; and J. Kolár, Leader of PTL/LC, GISAT, Prague, Czech Republic.

REC * PROGRAMS * TELEMATICS * DETERMINE

HOME PAGESEARCH