PARENT SESSION
Poster Session # 5: GIS and Remote Sensing.
Monday, August 4 Presentation from 5:00 PM to 6:30 PM. SITCC Exhibit Hall B.
Land cover estimates in interior Alaska across classifications and resolutions.
Calef, Monika*,1, Verbyla, David1, Epstein, Howard2, Shugart, Herman2, 1 University of Alaska, Fairbanks, AK2 University of Virginia, Charlottesville, VA
ABSTRACT- Modeling the effects of climate change in Alaska at regional scales requires vegetation data sets that accurately identify vegetation types across the landscape. This study was interested in the compatibility and accuracy of currently existing classifications for Interior Alaska and the effect of scale. Three 1km-resolution based land cover classifications were compared with each other and with a 25m-resolution TM image. Total overall agreements between vegetation classifications range from 59% to 96%. Agreement between the four 1km-resolution classifications and the TM image range from 40% to 52%. Kappa values are very low, corroborating that much of the agreement between different classifications can be attributed to random chance. The resolution of the vegetation classification affects the representation of vegetation types: the major vegetation types eliminate the smaller types with increasing coarseness. The land cover classifications were developed on state- and continental scale and do not accurately represent the vegetation for a small area in interior Alaska.
Key words: land cover classification, Alaska, scale