Dynamics of dead wood in old-growth hemlock–hardwood forests of northern Wisconsin and northern Michigan
We studied the dynamics of coarse woody debris (logs and snags) in old-growth forests by estimating rates of tree mortality, snag change, and log decay in hemlock–hardwood stands located in northern Wisconsin and northern Michigan. To estimate mortality and snag changes, we recensused live trees and dead snags in permanent plots in 15 stands. We also recorded recent mortality along transects, and noted category of mortality (standing death, breakage, or uprooting) for gap-maker trees and logs in 25 stands. Decay rates were estimated from a simple exponential model of wood density from log sections against the age of the wood since tree death, and from ages of trees growing on decaying "nurse" logs. From data obtained in permanent plots, annual tree mortality averaged 0.9% original basal area, 0.9% original live tree density, and 4.8 trees/ha. Of the three categories of tree mortality, standing death accounted for 62% of all mortality, while breakage represented 25%, and uprooting, 13%. Based on origin of logs, species composition affected category of mortality. Eastern hemlock (Tsugacanadensis (L.) Carr.) was significantly more prone to uprooting while yellow birch (Betulaalleghaniensis Britton) and paper birch (Betulapapyrifera Marsh.) were less prone to uprooting than expected if independence of species and categories of mortality were assumed (χ2 = 216.5, df = 14, p < 0.001). Changes in snags (fragmentation to shorter snags or collapse at the base) occurred for <7% of the original snags annually. Fragmentation occurred for 2.1 snags/(ha × year), and collapse for 1.3 snags/(ha × year). We estimated that it takes nearly 200 years for hemlock logs to lose structural integrity and become partially incorporated into the soil. At >350 years, the two oldest hemlock–hardwood stands had accumulated volumes of logs >65 m3/ha distributed among all decay classes, and appeared to be at a dead wood equilibrium in which rates of log production from mortality balance rates of wood loss by decay.