LAND-COVER CLASSIFICATION WITH AVHRR MULTICHANNEL COMPOSITES IN NORTHERN ENVIRONMENTS

The objectives of this study were to test the usefulness of various sp ectral channel combinations of AVHRR multitemporal composites for deri ving land cover information in northern environments, and to assess th e effect of AVHRR spatial resolution on the classification accuracy. A sequence of operations was carried out to remove radiometric distorti ons from AVHRR composites (1 km pixel sire) prepared for the landmass of Canada using multidate NOAA-11 data for the 1993 growing season: at mospheric corrections for AVHRR Channels 1, 2, and 4; identification a nd replacement of cloud-contaminated pixels; bidirectional reflectance corrections of Channels 1 and 2; and principal component (PC) calcula tions to retain significant Independent PC channels. Input principal c omponents were classified using an unsupervised clustering algorithm, and accuracies were assessed through a comparison to 30 m Landsat TM p ixels at five different sites in three biomes. We found that the norma lized difference vegetation index (NDVI) was the most effective single spectral dimension to derive land cover types, but other channels (es pecially 1 and 2) were needed to obtain highest accuracies. Overall, c lassification accuracies for the 30 m pixels were between 45% and 60%. Mixes of land cover classes within AVHRR pixels were the principal re ason for the low accuracies. When considering only AVHRR pixels with o ne dominant band cover type, the accuracy increased up to 80% or more in proportion to the mixed types retained. The accuracy also increased when a dispersed class (mixed forest) was combined with the more ubiq uitous coniferous forest class. The intrinsic AVHRR resolution and the compositing process are the major factors influencing the impact of m ixed cover types on the classification accuracy. The impact of these f actors is discussed and strategies for optimizing the use of multitemp oral AVHRR data in land cover classification are suggested.