Forest type, Turkey 2012 (Source: GeoVille/EEA)
This product provides the classification of different types of vegetation (e.g. grassland, shrubs, peat land, forest, steppe and savannah) based on optical satellite imagery. This product can also focus on the classification of different types of bushes and forest, as well as the quantification of any changes that may have occurred between the acquisitions of at least two satellite images, as forest monitoring has become more and more important in applications such as REDD+ monitoring.
Forest type maps:
Forest type maps are in-depth examinations of the forest categorizing the forest into deciduous/broadleaved, coniferous forests and mixed forest. Besides this generic classification, individual tree species such as pine, bamboo, or palm trees can be mapped. Base / surficial geology and elevation are important additional determinants of forest type.
Known restrictions / limitations
In tropical rain forest areas frequent cloud cover can be an issue for the production of the maps, but can be mitigated by combing radar and optical satellite images.
Level of detail in forest type class often demanded by foresters/ecologists requires significant ground data collection.
Volume classes typically quite broad. It is challenging to determine forest biomass/volume for tropical forests.
Lifecycle stage and demand
Pre-licensing & Exploration:
Development, Production & Decommissioning:
Geographic coverage and demand
Demand and coverage is global.
OTM:029 Pre-licensing site selection
Input data sources
Optical: VHR1, VHR2, HR1, HR2
Radar data supporting optical: VHR1, VHR2, HR1, HR2 (radar)
Spatial resolution and coverage
Spatial resolution: 1 - 30 m pixel size
Minimum Mapping Unit (MMU)
The MMU is dependent on the input data resolution, the mapped objects and the accuracy to be achieved. Monitoring forest stands, typically hectares to km² at a time.
For optical satellite data with 4 m spatial resolution a MMU of 256 m² can be achieved for example.
Accuracy / constraints
Thematic accuracy: 80-90%
Spatial accuracy: The goal would be one pixel, but depends on reference data
Accuracy assessment approach & quality control measures
Stratified random points sampling approach utilizing VHR reference or other geospatial in-situ data. Statistical confusion matrix with user’s, producer’s accuracy and kappa vegetation/forest type.
Frequency / timeliness
Observation frequency: Produced locally or regionally, normally on a 3 to 5 year basis (frequency can be lower or higher depending on demand)
Timeliness of deliverable: Dependent on size of the mapped area, resolution, MMU, and number of mapped classes.
Freely available or commercially acquired depending on the sensor selected.
Delivery / output format
|Peer Reviewer:||Hatfield Consultants|
Maria Lemper; Jan Militzer
# of Pages:
Internal – Project consortium and science partners
External – ESA
This page has no comments.