Estimation of stand mean crown diameter from high-spatial-resolution imagery based on a geostatistical method

Using the panchromatic band of QuickBird high-spatial-resolution satellite images as a data source, we estimated the mean crown diameters of different stands and of stands with different canopy densities by applying the semivariogram methodology used in geostatistics. The results indicated that relatively accurate estimates of the mean crown diameters of stands with a relatively high canopy density could be obtained. The estimated errors increased proportionally with decreasing canopy density. The whole-stand-image (matrix) method we used provided fairly accurate estimates of stand structural parameters, but would not readily permit the evaluation of patterns in stand structure. The transect method exhibited periodicity as a function of distance and, although it permits the detection of directional patterns, it does not depict stand structural parameters as accurately as the matrix method. Our results suggest that this approach could provide a new reference method for solving the problem of automated estimation of mean crown diameter at the stand level.