Response of understory trees to experimental gaps in old-growth Douglas-fir forests

Two 0.2-ha circular openings were created during fall 1990 in an old-growth Douglas-fir forest in the southern Washington Cascade Mountains. All trees >2 m tall were removed with care to minimize the disturbance to the understory and soil. A total of 1250 understory trees were monitored for growth and mortality over the next four years. Four main biological variables were measured: timing of spring budbreak, branch elongation, leader growth, and years of needle retention. Over the same area, measurements and models of direct light, diffuse light, and root competition were generated for correlations with the above biological variables. During 1992 and 1994, budbreak patterns were correlated with diffuse light conditions (0.42 to 0.55). In 1993, however, snow fall and cold temperatures in early spring caused poor correlations with these two predictors (0.06 to 0.24). Instead, direct light was correlated with the budbreak patterns (0.41 to 0.42), which were delayed by 2 wk relative to the other two years. Branch growth was radically affected by the creation of the gaps. During the first growing season following gap creation, branch elongation was suppressed in areas that received elevated levels of direct light, as indicated by a significant negative correlation of branch growth with direct light (-0.23 to -0.36). However, by the third growing season, surviving trees were growing rapidly (up to 30 cm/yr) in most portions of the gaps. Highest growth rates were south of center, where diffuse light levels were high but direct light levels were low. Needle retention patterns were significantly, although weakly, correlated negatively with direct light (-0.14 to -0.25).