Sponsored by the Northwest Fire Science Consortium
Presented by Jessie Dodge, University of Idaho
This study sought to evaluate the effect of spatial scale in remotely assessing ground cover components measured one-year after the 2007 Egley Fire Complex of eastern Oregon, in a predominately ponderosa pine (Pinus ponderosa) forest. Spearman’s rank correlations were used to correlate the Normalized Differenced Vegetation Index (NDVI) derived from 2007 Quickbird (0.6 m resolution) and 2007 and 2008 Landsat 5 TM images (30 m resolution) to surface variable covers (overstory tree canopy cover, understory green, non-photosynthetic vegetation (NPV), rock, soil, and char) collected in the summer of 2008. Results demonstrate the importance of matching pixel size of remote sensing images to the scale of field data collection, confirm the difficulty in obtaining ground reference data with an accuracy matching images with sub-meter resolution and the effects of tree canopy cover in obscuring surface cover components.