Abstract
Management regimes and geographical variations in potential forest productivity significantly influence the dynamics of carbon sources and sinks. Spatially explicit data on land cover, stand age class, and harvesting practices can be effectively acquired through satellite remote sensing. When combined with regional climate records, carbon-cycle process models can estimate potential production rates and associated decomposition processes. This integration of remote sensing and modeling enables the generation of spatially explicit information on carbon storage and flux. Using this approach, carbon flow between 1992 and 1997 was analyzed across two 165 km² regions in western Oregon: the West Cascades and the Coast Range. The West Cascades study area, predominantly composed of less-productive public lands, experienced minimal harvesting during the 1990s, with only 1% of its land base harvested between 1991 and 2000. In contrast, the Coast Range study area, largely managed for timber production on private lands, saw 17% of its land base harvested during the same period. Despite hosting a substantial proportion of young, highly productive stands that acted as carbon sinks, the Coast Range's mean annual harvest removals exceeded its mean annual net ecosystem production. Conversely, the West Cascades region functioned as a net carbon sink. The spatially and temporally explicit nature of this integrated approach allows for detailed identification of the mechanisms driving carbon flux across forested landscapes.