Summary of main findings:

This analysis has shown that the natural white spruce regeneration present on cutblocks in aspen-dominated boreal mixedwood stands occurs in a spatially clustered arrangement. It was also shown that the location of seedlings is spatially correlated to the location of mature white spruce trees.

The observed trends were consistent across the southern extent of the boreal forest. Seedlings were found to be more clustered than mature trees, and the degree of clustering was found to be higher during mast years. The degree of association between seedlings and mature trees was found to be higher during mast years than in non-mast years.

On average, seedlings occurred in patches approximately 20 meters long.

Interpretations:

Many factors affect the distribution of white spruce regeneration, of these the location of seed sources is considered to be one of the most important (eg. Asselin et al 2001). This analysis confirms the importance of residual mature seed sources left within a cutblock during harvest.

White spruce seed dispersal is limited to 100-200 m with the vast majority of seed falling within tens of meters of the source tree (Greene et al 1996). Therefore, natural regeneration will tend to be concentrated near mature trees. This seems to be especially true when cutblocks are harvested during mast years, when seed production peaks and the area surrounding the source tree may be saturated with seeds. During non-mast years, long-distance seed dispersal may distribute seed more evenly across the cutblock.

The relatively large average patch size is a positive result for managers attempting to restore natural aspen-spruce mixedwoods. Smaller patches of regeneration would likely cause the majority of regeneration to become crowded over time, with the eventual result of few trees reaching maturity due to self-thinning within the small patch. The larger patch sizes indicated in this analysis should be capable of supporting several healthy, mature spruce.

Study limitations and suggestions for future research:

1. Low replication
   Only 15% of the transects originally sampled in the field were used in this analysis. This may have limited the possibility of finding statistically significant results. While the SADIE method is more suited to intensive, rather that extensive analyses, additional confidence in the observed trends may be gained by analyzing the remaining transects.
   
   
   
2. High variability
   Small scale factors such as microsite quality and the strength of local competition play a large role in determining where seeds are able to germinate, and where seedlings are able to establish and survive. These factors have not been accounted for in this analysis, this limits the ability of this analysis to fully account for the variation in seedling distribution. Further field work would be required to collect data about these other factors
   
   
   
3. Experimental design
   The SADIE method employed is best suited to study plots more square in proportion. With the nearly linear transect design used to collect the data for this study, the influence of seed sources along the unsampled edges may overwhelm the effects of seed sources encountered within the plots. Ideally, a few large areas (i.e. whole cutblocks) would be intensively sampled, rather than an extensive sample of small, linear plots.