 The study found that automated tree crown isolation and classification using multi-spectral imagery was effective in isolating trees with fewer missions, but had issues with grouping trees within delineations, poor icealls at the edges of stands, and minor splitting. Spectral characteristics of 18 species showed considerable variability within species and overlap of signatures, which required creating spectral subclasses for certain species and dividing the site into broad and localized zones. Mapping accuracy and classification accuracy of manually delineated trees ranged from 40% to 85%, with manual tree class accuracy averaging 76%. The study suggests that production of individual tree species maps in complex forests will require judicious use of human judgment and intervention, regardless of the sophistication of techniques used. This article was authored by Donald G. Lecky, Francois Gougeon, Ryan McQueen, and others.