Forest crown cover estimation in northern boreal and temperate European forest

    Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientific

    Abstract

    A method for forest crown cover estimation using high resolution optical earth observation data was developed and tested at four study sites in Europe. Crown cover was estimated using the probability estimation method of VTT and Image2006 data. The accuracy of the crown cover predictions was assessed using reference data that were collected by visual interpretation of very high resolution aerial and space borne imagery. The average crown cover values in the reference data varied from 17 % to 86 % and in the predictions from 18 % to 80 %. The absolute root mean square error of the crown cover predictions varied between 14 % and 33 %. The results of the study showed that it is possible to map forest crown cover with twenty to thirty meter spatial resolution optical earth observation data using the single pixel values. However, understanding the variable results at different sites requires further investigation.
    Original languageEnglish
    Title of host publicationProceedings of the First Sentinel-2 Preparatory Symposium
    PublisherEuropean Space Agency (ESA)
    Number of pages5
    Editioncd-rom
    ISBN (Print)978-929092271-1
    Publication statusPublished - 2012
    MoE publication typeB3 Non-refereed article in conference proceedings
    Event1st Sentinel-2 Preparatory Symposium - Frascati, Italy
    Duration: 23 Apr 201227 Apr 2012

    Publication series

    SeriesESA Special Publication
    NumberSP-707
    ISSN0379-6566

    Conference

    Conference1st Sentinel-2 Preparatory Symposium
    CountryItaly
    CityFrascati
    Period23/04/1227/04/12

    Keywords

    • remote sensing
    • optical
    • forest
    • density
    • crown cover
    • boreal
    • temperate

    Fingerprint Dive into the research topics of 'Forest crown cover estimation in northern boreal and temperate European forest'. Together they form a unique fingerprint.

    Cite this