Environmental risk assessment based on semi-quantitative analysis of forest management data

Abstract

The paper deals with environmental risk assessment in prevailingly unnatural spruce (Picea abies [L.] Karst.) forests in three regions with different patterns of forest damage in the Slovak part of the West Carpathians. Logistic regression was used to estimate the effect of 7 site-related, 5 stand-related and 2 anthropogenic factors on the probability that critical forest damage will occur. The results show that regression models can describe cause-effect relationships in regions with different regimes of forest decline. Stand age, proportion of spruce, and distance from the focus of biotic agent activity predicted decline in two regions with generally lower elevation in northern Slovakia (Kysuce and Orava). In a mountain region (Low Tatras), the importance of factors contributing to the static stability of trees and position towards dangerous winds increased significantly. The quality of the derived models and prospects for their usefulness in risk assessment are discussed.