Snag Characteristics and Dynamics in Douglas-Fir Forests, Western Oregon

We studied snags in 30 stands, 5-445 years old, of unmanaged and managed Douglas-fir (Pseudotsuga menziesii) in western Oregon to gain information about snag populations and status after logging. As snag production rates (snags/ha/year) declined from about 100 to 1, mean snag density decreased from 190 to 18/ha in age-classes 35 and 200+, respectively; remnant snags (formed in previous stands) represented 5-14% of current densities. Meanwhile, average snag dbh increased from 13 to 72 cm, and as dbh increased, snags stood longer. Douglas-fir was the dominant species among snags in all forest age-classes. Linear regression analysis showed a correlation (P < 0.001) between snag age and deterioration; populations consisted of fewer young (sound) and old (highly decayed) than middle-aged (partially decayed) snags. Cluster analysis revealed 5 stages of deterioration based upon snag size and decay condition. In unmanaged stands, most (62%) snag populations were distributed randomly, but patches of snags were found in all age-classes. Fewer snags (P < 0.001) remained after thinning and clear-cutting unmanaged forests, and natural snag production was disrupted. Large snags should be retained within forests managed over long (>200-year) rotations; in riparian forests; in extensively managed, slow-growing forests; and within intensively managed forests, safety permitting.