Multiscale comparison of ERS -1/2 tandem forest growing stock maps with existing landcover products

Reik Leiterer*, Johannes Reiche, Oliver Cartus, Maurizio Santoro, Christiane Schmullius, Zengyuan Li

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference paperAcademicpeer-review

Abstract

The purpose of this poster is to show a useful approach for determining the quality of forest growing stock (6SV) maps derived from ERS-1/2 tandem data. For the validation of the large-area maps, a special cross-comparison design based on freely available Earth Observation land cover (LC) products had to be developed due to unavailability of in situ measurements. A multiscale comparative assessment design, linking quality flags and LC products such as GlobCover, MODIS VCF, GLC2000 and AVHRR LC, has been applied. The sampling design, based on the FAO FRA2010 sampling design and the Degree Confluence Project, uses a 1 degree sampling grid with 10 × 10 km sample plots. The method provides appropriate information about plausibility and quality of the ERS forest GSV maps. Since similar global data sets will be available in the future (e.g. within the framework of ESA Sentinel programme), the developed approach is well suitable for scale and thematic independent accuracy assessment applications.

Original languageEnglish
Title of host publicationProceedings of the Symposium - Dragon 2 Programme Mid-Term Results 2008 - 2010
Publication statusPublished - 2010
Externally publishedYes
Event2010 Dragon 2 Mid Term Results Symposium - Guilin City, China
Duration: 17 May 201021 May 2010

Publication series

NameEuropean Space Agency, (Special Publication) ESA SP
Volume684 SP
ISSN (Print)0379-6566

Conference/symposium

Conference/symposium2010 Dragon 2 Mid Term Results Symposium
Country/TerritoryChina
CityGuilin City
Period17/05/1021/05/10

Fingerprint

Dive into the research topics of 'Multiscale comparison of ERS -1/2 tandem forest growing stock maps with existing landcover products'. Together they form a unique fingerprint.

Cite this