Thinning Promotes the Restoration of Branch Structure in Second-Growth Redwoods at Redwood National Park

As the extent of old-growth coast redwood (Sequoia sempervirens) forests has diminished, forest restoration has become a dominant paradigm for the management of young second-growth forests throughout northern coastal California (Porter et al. 2007). Restoration efforts usually focus on promoting spatial heterogeneity, tree age diversity, multi-layered canopies, and large woody debris and snags, as they are the defining characteristics of Pacific Coast old-growth forest structure. In recent years this challenge has fostered the study of novel thinning techniques to advance the development of old-growth characteristics (O'Hara et al. 2010, Keyes et al. 2010).

With the current restoration focus on stand structure, aspects of individual tree crown architecture are often overlooked. Yet, the large branches and complex crowns (including reiterated trunks: tree crowns with multiple tops resulting from stem damage) that are associated with old-growth trees play an important role in providing habitat for species associated with late-successional forests. In particular, they provide vital nesting platforms for the marbled murrelet (Brachyramphus marmoratus), a threatened seabird that nests in coastal redwood forests (Baker et al. 2006). The widespread decline in nesting habitat as old-growth forests were converted to second-growth forests was the main reason that the murrelet was placed on the federal threatened species list (Baker et al. 2006). To more rapidly restore murrelet nesting habitat to the region's forests, ecologists have proposed that silvicultural treatments, such as thinning, might help stimulate branch growth and reiterated trunks in second-growth trees (Franklin et al. 2007).

Redwood National Park was originally established to protect rare old-growth redwood forests, but its boundaries were greatly expanded in 1978 with the annexation of 20,650 ha of second-growth forestlands that were even-aged and dense, with trees of low vigor and slow growth. The Holter Ridge Thinning Study (HRTS) was established in 1978 to evaluate the potential of thinning (at varying levels of thinning intensity, or quantity of trees removed) to promote development of old-growth structural properties. The results of stand responses to the thinnings were addressed in a recent analysis of 25-yr post-treatment response (Chittick and Keyes 2007). The objective of the present study was to evaluate branch diameters among a subset of the largest second-growth redwood trees within the HRTS plots, and infer the effect of thinning on branch development. The HRTS is an unevenly replicated case study, but it is a rarity in the region, and its treatments presented a population of trees of common origin but different thinning intensities that was well suited to this objective.

The HRTS was established in an 80-ha, 25-30 yr old naturally-regenerated, second-growth, even-aged stand of redwood, Douglas-fir (Pseudotsuga menziesii), tanoak (Lithocarpus densiflorus), western hemlock (Tsuga heterophylla), and Pacific madrone (Arbutus menziesii) trees. In 1978, 6 low thinning prescriptions resulted in post-treatment densities of 346-6,101 trees per ha (Chittick and Keyes 2007). Prior to thinning, stand densities averaged more than 400 trees per ha, with some plots exceeding 1,200 trees per ha (Veirs 1986). To minimize bias resulting from heterogeneous site factors (slope, aspect, slope position) and residual old-growth trees, we selected 10 of the HRTS plots (ranging from 370-2,915 trees per ha) for analysis.

Figure 1.

Among second growth redwoods in thinned stands at Redwood National Park, CA, the relationship of average maximum branch diameter ('BD4'; average of 4 largest branches per tree) to tree diameter at breast height (dbh) was positive. Overall, trees with stem damage had larger average maximum branch diameters (5.89 cm) than did undamaged trees (4.42 cm).

We selected up to 3 redwoods of greatest diameter at breast height (dbh) at each plot (N = 28 trees; only 2 redwoods present in 2 plots). We did not exclude trees with evidence of past crown damage, due to their ubiquitous presence throughout the study area. We measured dbh and the diameters of the 4 largest branches in the live crown of each tree...