e-GEOS Production Site produced this product by satellite analyses in the context of the Copernicus Global land Hot Spot Mapping (C-GL-HSM) framework.

Data and products are based on medium to high and very high resolution satellite images (from approximately 1 to 30m spatial resolution) with a change assessment frequency between 1 to 20 years. The Image data sources used for mapping are Landsat 7 and 8. The validation process made use of Spot-6 and Sentinel-2 images as reference data.

Images temporal range: 1997-2003

It is the time frame that has been accepted to collect the satellite images useful to produces the vector data.

The Reference year is included in this time frame and correspond to mean year considering all the image’s year used. It is the year on which the majority of the used images are.

The classification scheme follows the Land Cover Classification System (LCCS) developed by the United Nations Food and Agriculture Organization (FAO).

Since LCCS is a hierarchical system, the modular legend can be aggregated to the dichotomus legend.

The FAO LCCS handbook which describes each class in detail, can be downloaded here: http://www.fao.org/3/a-i5232e.pdf

This LCCS Land Cover map includes the following land cover classes (associated raster code in [ ]):

A11 - Cultivated and Managed Terrestrial Area(s) [3]

A12 - Natural And Semi-Natural Primarily Terrestrial Vegetation [4]

A23 - Cultivated Aquatic or Regularly Flooded Area(s) [6]

A24 - Natural And Semi-Natural Aquatic or Regularly Flooded Vegetation [7]

B15 - Artificial Surfaces and Associated Area(s) [0]

B16 - Bare Area(s) [11]

B27 - Artificial Waterbodies, Snow and Ice [13]

B28 - Natural Waterbodies, Snow and Ice [14]

The produced and independently validated Land Cover and Land Cover Change maps and statistics are available to global users.

The report and the validation file can be downloaded from the link section.

Basic image processing: Cloud/Shadow masking, Data Selction (based on occlusion and sesonality considerations), Atmospheric correction (TOA) of satellite data, Coregistration.

Automatic classification: Feature extraction from Dense Multitemporal Time Series (D MTS), statistics generation, automatic classificaiton (ROI based or decision tree) and labeling according to the required output LCCS legend schema.

Visual inspection and refiment: check and refinement of the LCCS product generated through the automatic procedure in order to corrected classification errors and to refine borders where necessary.

Internal validation: independent validation of the LCCS product based on external reference data (where available) and on other datasets for intercomparison. The scope of the internal validation is to make a qualitative and quantitavie check of the declared Thematic and Positional accuracies.