Development of a Consistent Methodology to Derive Land-Cover Information on a European Scale from Remote Sensing for Environmental Modelling (PELCOM)

End: 03.2000

For many environmental and agricultural studies on the various scales, up-to-date and reliable information on land use and land cover (LULC) is urgently needed. The development of a sound environmental policy relies intensely on the outcomes of environmental models, which themselves are significantly influenced by the areal and spatial accuracy of LULC data. For that reason the PELCOM project has been formulated. It aims at the establishment of a 1-km pan-European land cover database that can be updated frequently and which is based on the integrative use of multispectral and multitemporal 1-km resolution NOAA-AVHRR satellite data and ancillary data.
Different methodologies will be assessed to map and monitor land cover of the European continent by using low resolution satellite data. The main objective is to arrive at a consistent methodology for the establishment and regular update of a 1-km pan-European land cover database to be used as an input for environmental impact studies and climate research. The main goals of the project are:

	to reach consensus on a land cover nomenclature for the European continent;
	to develop a consistent classification methodology, based on NOAA-AVHRR data and ancillary data, suitable for mapping the land cover of the European continent;
	to establish a Central Project Information Server (CPIES) and a meta-information system;
	to arrive at a methodology for monitoring land cover changes at a European scale using NOAA-AVHRR data;
	to apply the classification methodology to various regions of the European continent;
	to apply the established land cover database in different environmental and climatic models.

In the final phase of the project the following environmental modelling approaches will be tested on the established land cover data base:

	large scale inventories of biogenic emissions from forest (ARCS)
	ecosystem modelling (RIVM)
	improvement of boundary conditions for meteorological models (CNRM)

Results

Results comprise definition of and agreements on data sources and ancillary data sets and the development of an improved stratified and integrated classification methodology to map major land cover types for pan-Europe using NOAA-AVHRR satellite and additional geographic data sets. Both multi-temporal NDVI profiles and multi-spectral AVHRR scenes are used as input in the classification procedure exploiting the advantages of both data types. Due to the limited accuracy of identifying forests on basis of unsupervised clustering of NDVI time-series, the identification of forests was implemented on basis of thresholding the synthesis of the visible reflectance of channel one. In addition, urban areas and inland water were masked on basis of ancillary data. The classification scheme has been used to establish a land cover database of pan-Europe with a 1 km spatial resolution that fulfils the needs of the environmental models.
In addition a new technique for change detection has been tested. It is based on fraction images of multitemporal NOAA/AVHRR data derived from spectral mixture analysis. A direct comparison of these fraction images, calculated from satellite images recorded at different dates, makes it possible to highlight those areas where the proportion of the various land cover types has changed significantly.
Three case studies have been performed applying the land cover data base for environmental modelling including a biodiversity study, meteorological and emission modelling.
A CD-ROM has been produced that contains the pan-European land cover data base, the final report and additional information on the project. The CD-ROM is available from the project co-ordinator.

Project Home-Page

The PELCOM landcover data base may be downloaded by pressing here 

For any further comments please contact Florian Kressler.