Coupe logging effects on tree health and mortality in terrace Rimu (Dacrydium cupressinum) forest, South Westland, New Zealand

Abstract

This study investigated the effect of coupe logging on residual forest condition 18 years after logging in terrace rimu (Dacrydium cupressinum) forest, south Westland, New Zealand. In 1981, four 10 ha blocks in Okarito Sustained Yield Forest were allocated to a logging treatment. Three blocks were assigned to different coupe felling sizes (coupe type); small (0.2ha), medium (0.5ha), large (2ha) coupes, and one block was left as a control. In line with the policy at the time, the level of harvest was set at 25% of the total merchantable volume in each block. Therefore, ten 0.2 ha coupes were cut in the small block, four 0.5 ha coupes were cut in the medium block and a single 2.0 ha coupe was cut in the large block. In the summer of 1999 the condition of the residual rimu trees (≥20cm dbh) was assessed along ten systematically located 10 x 250m transects in each block. Each tree found in the transects was categorized as dead, unhealthy (<50% foliage) or healthy (>50% foliage). In addition, for each tree, the initial diameter as recorded in the 1981 inventory, distances to the nearest and subsequent coupes and tracks, and the direction of the tree (aspect) from the coupes and tracks were recorded. The tree condition (dead, unhealthy and healthy) was modelled as a function of coupe type, diameter, and distance and direction to coupes and tracks using logistic regression. The results indicate that tree condition differed significantly in blocks with different logging treatments. Mortality was higher in the large block. The proportion of poor health trees was higher in the small block. Two factors significantly predicted tree condition. Dead trees were more likely to be located close to coupes or tracks and were more likely to be medium (40-50cm dbh) or large trees. Similarly, trees in poor health were more likely to be found near coupes or tracks, but small trees (20-30cm dbh) were more likely to be unhealthy. Having controlled for differences in diameter and distance and direction from coupes or tracks, coupe type did not explain any additional variation in tree mortality, suggesting that mortality rate was not influenced by other differences in coupe felling blocks. In contrast, coupe type explained additional variation in tree condition (healthy versus unhealthy) suggesting that something about the different coupe felling blocks not measured in this study was important. The natural mortality rate and incidence of poor health, as indicated by the control, were also high. This, along with the lack of replication makes it difficult to isolate the effects of the different coupe felling sizes on the residual forest condition. The implications are that management should aim at minimizing logging damage to residual forest stands by minimizing the creation of abrupt edges.