Göttlicher, D.; Obregón, A.; Homeier, J.; Rollenbeck, R.; Nauss, T. & Bendix, J. (2009): <b>Land-cover classification in the Andes of southern Ecuador using Landsat ETM+ data as a basis for SVAT modelling</b>. <i>International Journal of Remote Sensing</i> <b>30</b>, 1867-1886.
Resource Description
Title:
Land-cover classification in the Andes of southern Ecuador using Landsat ETM+ data as a basis for SVAT modelling
Short Name:
Land-cover classification in southern Ecuador
FOR816dw ID:
449
Publication Date:
2009-04-01
License and Usage Rights:
Resource Owner(s):
Individual:
Dietrich Göttlicher
Contact:
email:
dietrich.goettlicher <at> staff.uni-marburg.de
Faculty of Geography
Deutschhausstr. 10
Laboratory for Climatology and Remote Sensing
Philipps University of Marburg
35032 Marburg
Germany
Individual:
André Obregón
Contact:
email:
webmaster <at> tropicalmountainforest.org
Individual:
Jürgen Homeier
Contact:
email:
jhomeie <at> gwdg.de
Faculty of Resource Management
University of Applied Sciences and Arts (HAWK)
37077 Göttingen
Germany
Individual:
Ruetger Rollenbeck
Contact:
email:
rollenbe <at> staff.uni-marburg.de
Laboratory for Climatology and Remote Sensing
Faculty of Geography
Philipps University of Marburg
Deutschhausstr. 10
35032 Marburg
Germany
email:
bendix <at> staff.uni-marburg.de
Faculty of Geography
Deutschhausstraße 10
Philipps University of Marburg
Laboratory for Climatology and Remote Sensing
35032 Marburg
Germany
Abstract:
A land-cover classification is needed to deduce surface boundary conditions for a<br/>
soil?vegetation?atmosphere transfer (SVAT) scheme that is operated by a<br/>
geoecological research unit working in the Andes of southern Ecuador. Landsat<br/>
Enhanced Thematic Mapper Plus (ETM + ) data are used to classify distinct<br/>
vegetation types in the tropical mountain forest. Besides a hard classification, a<br/>
soft classification technique is applied. Dempster?Shafer evidence theory is used<br/>
to analyse the quality of the spectral training sites and a modified linear spectral<br/>
unmixing technique is selected to produce abundancies of the spectral<br/>
endmembers. The hard classification provides very good results, with a Kappa<br/>
value of 0.86. The Dempster?Shafer ambiguity underlines the good quality of the<br/>
training sites and the probability guided spectral unmixing is chosen for the<br/>
determination of plant functional types for the land model. A similar model run<br/>
with a spatial distribution of land cover from both the hard and the soft<br/>
classification processes clearly points to more realistic model results by using the<br/>
land surface based on the probability guided spectral unmixing technique.
Literature type specific fields:
ARTICLE
Journal:
International Journal of Remote Sensing
Volume:
30
Page Range:
1867-1886
Publisher:
Taylor & Francis
Metadata Provider:
Individual:
Dietrich Göttlicher
Contact:
email:
dietrich.goettlicher <at> staff.uni-marburg.de
Faculty of Geography
Deutschhausstr. 10
Laboratory for Climatology and Remote Sensing
Philipps University of Marburg
35032 Marburg
Germany
