Abstract

Wind damage to forests is determined by numerous factors that interact to produce complex, seemingly random damage patterns. However, the complexity may lie mostly among stands and be less within stands: in this study, I attempted to discern how predictable tree fall risk is within five southern boreal forest stands in northeastern Minnesota. I sampled five stands in the Boundary Waters Canoe Area Wilderness, following a July 1999 catastrophic windstorm. Levels of damage varied from 29.5% to 86.8% of basal area fallen and 23.3% to 63.4% of stems fallen. In all sites, the disturbance reduced mean trunk diameter of standing trees. In general, Abies balsamea (L.) Mill. was the most vulnerable species. I split the data set from each site into predictor and test portions and used the predictor data sets to derive logistic regression parameters for the relationship of tree size (trunk diameter) to probability of tree fall. Models based on these parameters allowed quite accurate predictions of the levels of damage in the test portion of each stand. For the five sites, the proportion of test trees predicted to fall differed from the proportion observed to fall by 5.7%, 3.9%, 8.3%, 1.4%, and 3.7% of the total test sample size. This suggests that while numerous factors indeed influence tree fall risk, the sizes and identities of trees may account for most of the within-stand variation in damage.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call