Arctic Workshop Logo 2002       INSTAAR logo

Program

View Abstracts

Transportation

Registration Submit Abstract Accommodations
Home Student Support Local Info
     

32nd Annual Arctic Workshop Abstracts
March 14-16, 2002
INSTAAR, University of Colorado at Boulder

Previous | Abstract Index | Next



ESTIMATING BIOPHYSICAL CHARACTERISTICS OF ARCTIC TUNDRA VEGETATION USING SOIL-ADJUSTED VEGETATION INDICES AND MULTI-RESOLUTION REMOTE SENSING DATA

AUTHORS

LAIDLER, GITA J. Queen's University.
Treitz, Paul M. Queen's University.

The Normalized Difference Vegetation Index (NDVI) is the most commonly applied spectral vegetation index (VI) in arctic biophysical remote sensing analysis. However, due to the low stature and patchy nature of vegetation cover in the dwarf-shrub tundra sub-zone, it is important to investigate the utility of VIs that compensate for high degrees of soil reflectance. Therefore, the objectives of this study are to explore the relationships between soil-adjusted VIs, percent cover, and vascular plant biomass, in a tundra environment where exposed soil and gravel till have significant influence on the spectral response, and hence, the characterization of vegetation communities. In response to previous studies suggesting that high spatial resolution remote sensing data (i.e., <10m) is required for adequate delineation and classification of Arctic tundra vegetation communities, an IKONOS multispectral image (i.e., 4m spatial resolution) was acquired for a study area within the Lord Lindsay River watershed on Boothia Peninsula, Nunavut. Coincident with image acquisition, extensive field data (e.g. percent cover, above-ground biomass, surface spectral characteristics) were collected to determine community composition. These data were then used in an evaluation of the performance of an unsupervised classification and spectral vegetation indices (NDVI, SAVI, MSAVI) in describing vegetation community characteristics. This poster will present a comparison of the application of IKONOS and Landsat 7 ETM+ data for delineating tundra vegetation communities and biophysical properties characterized by small-scale variations in moisture and topographic gradients. Results suggest that high spatial resolution satellite data provide detailed information on vegetation community structure and biophysical properties, while soil-adjusted VIs provide seemingly more realistic estimates of plant cover and vigor. As the Arctic tundra is believed to be quite sensitive to minor changes in temperature and/or precipitation, estimates of vegetation community composition, distribution, and biomass are essential to determining a baseline for monitoring or modeling future changes that may follow trends of global climate change.

 

Previous | Abstract Index | Next

Copyright © 2001 INSTAAR, Univ. of Colorado