An algorithm for automated individual tree measurement was developed that is driven by a morphological analysis of a high-resolution LIDAR-based canopy surface model Binary and grayscale mathematical morphology were used to relate structure within a three-dimensional forest canopy model to the location of individual tree crown apexes This information was used to extract LIDAR measurements of individual tree position and height Algorithm measurements were compared to photogrammetric measurements from large (1: 3000) scale aerial photography Given a range of optimal input parameters, the algorithm was successful in locating and measuring individual tree crown heights The algorithm identified individual tree crown apexes in a mature forest with closed canopy within 2 meters of photogrammetrically-measured crown apexes with a User s accuracy of 89% and a Producer s accuracy of 83% The difference between algorithm and photogrammetric tree crown apex height measurements was approximately 1 meter in both study areas crown image models (Pollock, 1998; Larsen, 1998) Researchers in Scandinavia have attempted to model the relationship between the spatial distribution of individual trees and the position of spectral maxima in a digital image (Dralle and Rudemo, 1997; Lund and Rudemo, 2000) Another study has utilized two-dimensional mathematical morphology to analyze the spatial structure of individual trees composing the canopy in color aerial photography (Zheng et al, 1995) The use of airborne laser scanning for the acquisition of forest measurement data has also been an active area of research Research efforts investigating the use of small footprint (< 1 m) LIDAR for forest measurement have primarily concentrated on estimating forest stand-level parameters (Naesset, 1997; Nelson, 1988) A study conducted in Oregon demonstrated the use of LIDAR for predicting forest stand characteristics using plot-level LIDAR heights and canopy cover percentiles, and found very strong relationships between LIDAR-derived measurements and stand parameters (Means et al, 2000) Researchers in Canada have used a probabilistic model-based approach to estimate stand height from LIDAR data (Magnussen et al, 1999) Another study related the distribution of LIDAR canopy height measurements to the vertical distribution of foliage area (Magnussen and Boudewyn, 1998) Nelson found that the shape of the individual tree crowns composing the forest canopy can have an effect on the LIDAR-based prediction of stand-level parameters (biomass, basal area, volume), as LIDAR-based forest height estimates over canopies com-