LAND-COVER MAPPING, FIRE REGENERATION, AND SCALING STUDIES IN THE CANADIAN BOREAL FOREST WITH 1 KM AVHRR AND LANDSAT TM DATA

A multitemporal 1 km advanced very high resolution radiometer (AVHRR) land cover analysis approach was used as the basis for regional land c over mapping, fire disturbance-regeneration, and multiresolution land cover scaling studies in the boreal forest ecosystem of central Canada . The land cover classification was developed by using regional field observations from ground and low-level aircraft transits to analyze sp ectral-temporal clusters that were derived from an unsupervised cluste r analysis of monthly normalized difference vegetation index (NDVI) im age composites (April-September 1992). Quantitative areal proportions of the major boreal forest components were determined for a 821 km x 6 19 km region, ranging from the southern grasslands-boreal forest ecoto ne to the northern boreal transitional forest. The boreal wetlands (mo stly lowland black spruce, tamarack, mosses, fens, and bogs) occupied approximately 33% of the region, while lakes accounted for another 13% . Upland mixed coniferous-deciduous forests represented 23% of the eco system. A SW-NE productivity gradient across the region is manifested by three levels of tree stand density for both the boreal wetland coni fer and the mixed forest classes, which are generally aligned with iso pleths of regional growing degree days. Approximately 30% of the regio n was directly affected by fire disturbance within the preceding 30-35 years, especially in the Canadian Shield Zone where large fire-regene ration patterns contribute to the heterogeneous boreal landscape. Inte rcomparisons with land cover classifications derived from 30-m Landsat Thematic Mapper (TM) data provided important insights into the relati ve accuracy of the 1 km AVHRR land cover classification. Primarily due to the multitemporal NDVI image compositing process, the 1 km AVHRR l and cover classes have an effective spatial resolution in the 3-4 km r ange therefore fens, bogs, small water bodies, and small patches of dr y jack pine cannot be resolved within the wet conifer mosaic. Major di fferences in the l-km AVHRR and 30-m Landsat TM-derived land cover cla sses are most likely due to differences in the spatial resolution of t he data sets. In general, the 1 km AVHRR land cover classes are vegeta tion mosaics consisting of mixed combinations of the Landsat classes. Detailed mapping of the global boreal forest with this approach will b enefit from algorithms for cloud screening and to atmospherically corr ect reflectance data for both aerosol and water vapor effects. We beli eve that this 1 km AVHRR land cover analysis provides new and useful i nformation for regional water, energy, carbon, and trace gases studies in BOREAS, especially given the significant spatial variability in la nd cover type and associated biophysical land cover parameters (e.g., albedo, leaf area index, FPAR, and surface roughness). Multiresolution land cover comparisons (30 m, 1 km, and 100 km grid cells) also illus trated how heterogeneous landscape patterns are represented in land co ver maps with differing spatial scales and provided insights on the re quirements and challenges for parameterizing landscape heterogeneity a s part of land surface process research.