Characterizing the canopy gap structure of a disturbed forest using the Fourier transform

Diseases and other small-scale disturbances alter spatial patterns of heterogeneity in forests by killing trees. Canopy gaps caused by tree death are a common feature of forests. Because gaps are caused by different disturbances acting at different times and places, operationally determining the locations of gap edges is often difficult. In this study, digital image analysis using aerial photos was used to formulate an operational definition of gaps and to compare spatial attributes of canopy gap structure in previously unharvested and harvested stands of ponderosa pine forests in the Black Hills of South Dakota. Fourier transforms of monochrome images of the study areas were edited to sequentially remove higher frequency spatial data implementing a type of moving window analysis where window sizes varied over a range of 1-1300 m(2). Boolean thresholds were chosen such that the mean gap size approximated the window size for each original image. Plots of mean gap size, mean fractal dimension, double log fractal dimension, and gap size density all versus window size showed distinct differences between treated and untreated plots. (C) 2000 Elsevier Science B.V. All rights reserved.