Comparison of spatial vegetation patterns following clearcuts and fires in Ontario's boreal forests

Abstract

The goal of this study was to compare spatial vegetation patterns, based on Landsat TM data, within post-clearcut and post-fire disturbances. Landscapes disturbed during the four decades prior to the collection date of the Landsat data were used for comparison. The disturbed landscapes were clustered according to their spatial edaphic factor patterns. A suite of indices representing patch geometry, contagion, and composition were used to describe spatial vegetation and edaphic factor patterns. A general linear model was used to compare the effects of disturbance type, time since disturbance, and edaphic factors (clusters) on seven indices of spatial vegetation patterns. Patch size and patch density differed following clearcuts and fires. It appears that clearcuts may result in greater spatial heterogeneity among landcover types compared to fires. I propose that fires were more severe than clearcuts; thus, creating larger and fewer patches. Time since disturbance had the greatest effect on spatial vegetation patterns. One decade old disturbances had larger patches, higher contagion and fewer landcover types than older disturbances. I suggest that spatial vegetation patterns reflected the destruction of overstory vegetation in one decade old disturbances, and revegetation in the form of small patches in older disturbances. It appears that the effects of disturbance on spatial vegetation patterns are temporary. Edaphic factor patch shapes may influence the shape of vegetation patches.