The effect of thinning on structural attributes of a low rainfall forest in eastern Australia

Forest thinning is a widely used silvicultural practice that has potential to provide benefits for biodiversity, but there is little direct evidence for this, particularly for regrowth in low rainfall forests and woodlands. This study examined the long-term effects of thinning on structural attributes of white cypress pine (Callitris glaucophylla) in the Pilliga forests of eastern Australia. A chronosequence approach was used which included six replicates of five forest management treatments; Unthinned (>80 years) with high density regrowth; Recent thinning (<8 years); Intermediate thinning (8-20 years); Old thinning (21-40 years); Long undisturbed (>80 years) with large Callitris and Eucalyptus trees present. Thinning reduced the dominance of species that form dense single-aged stands (Allocasuarina luehmannii and Callitris spp.). Following thinning there was a ∼4-fold reduction in small stems (<10 cm, diameter breast height (DBH)); 6030 stems ha−1 (Unthinned) compared to 1583 stems ha−1 (Recently thinned). The reduction in small stemmed vegetation was associated with lower cover in both the mid-storey (2-6 m) and sub-canopy (6-14 m) which persisted for 21-40 years while the long undisturbed treatment had mid-storey densities which were similar to the unthinned treatment. Density of medium-sized Eucalyptus spp. (10-30 cm DBH) was highest (>90 trees ha−1) where thinning had occurred (recent, intermediate and old thinning treatments) or where stem density of Allocasuarina and Callitris regrowth was low (long undisturbed), suggesting both these species competitively exclude recruitment of Eucalyptus spp.. The post-thinning reduction in woody vegetation was accompanied by an initial increase in the volume of downed coarse woody debris (DCWD), which was long lasting (21-40 years) and four to eight times greater than in long undisturbed sites, with greatest mean hollow diameter occurring in the old thinning treatment. Commercial thinning, in which some residue is removed from sites, still supported up to four times the DCWD volume as the long undisturbed treatment. There was a trend for a negative effect of thinning on the density of dead trees, but no effect on density of hollow-bearing trees and large trees (>50 cm DBH). Overall, our results indicate thinning had a mixed effect on key structural attributes that contribute to habitat structural complexity, indicating a need to record the direct responses of biodiversity.