Effects of Pruning in the Management of Dogwood and Pine Branch Dieback in the Landscape

Materials and Methods

In 1996, 2 plots of Scots pines in Centre County, Pennsylvania, were located: Plot #1 consisted of a single row of 14-year-old pines, averaging 8.4 m (27.7 ft) tall and at least 1.6 m (5.3 ft) apart. These trees were situated along the edge of a golf course rough and their branches formed a closed lower crown. Most of the vegetation among the trees was tall grasses. Two blocks of trees with approximately the same incidence and severity of branch dieback, primarily Sphaeropsis tip dieback (Sphaeropsis sapinae (Fr.) Dyko & Sutton = Diplodia pinea (Desm.) Kickx.), were selected. Six control trees were not pruned, while 10 trees were pruned to remove as many dead twigs and branches as feasible. Plot #2 originally consisted 12 rows of 8 Scots pines planted on a 1.6-m spacing. The plot was modified by removing every other tree in a staggered pattern to increase spacing. All dead, damaged, and severely distorted trees were also removed. The trees were approximately 11 years old and averaged 3.7 m (12.2 ft) tall. The area under the trees was covered with fallen needles. Two blocks of trees were established. One block of 23 trees served as the unpruned control, while the second block of 17 trees was pruned. Initially, the blocks contained trees with similar ranges of branch dieback.

A plot of mature dogwoods affected primarily by anthracnose and Botryosphaeria canker (Botryosphaeria spp.) was located at Longwood Gardens, Kennett Square, Pennsylvania The trees, scattered on a hillside, were at least 10 m (33 ft) apart and were estimated to be 30 to 40 years old. They averaged 5.9 m (19.5 ft) tall, and the trunks averaged 86.3 cm (34.5 in.) in circumference measured at a point 30 cm (12 in.) above the soil. The trees were divided into 2 groups. Five trees were not pruned and served as the control, while 6 trees were pruned. Trees within each treatment had similar ranges of branch dieback.

In spring 1996, the radii of drip lines and the heights and circumferences of all the trees were measured, and the number of dead branches was determined. A dead branch index (dbi) was calculated (dbi = 3 × number of primary branches completely dead + 2 × number of primary branches only the ends of which were dead + percentage of secondary branches that were dead) to quantify the visual appearance of the trees. Trees were also rated for overall branch disease severity using the Horsfall-Barratt scale: 1 = no disease; 2 = 0 to 3% disease; 3 = 3% to 6%; 4 = 6% to 12%; 5 = 12% to 25%; 6 = 25% to 50%; 7 = 50% to 75%; 8 = 75% to 87%; 9 = 87% to 94%; 10 = 94% to 97%; 11 = 97% to 100%; 12 = 100% (Horsfall and Barratt 1945).

In spring 1997, comprehensive pruning was performed and pruned material was removed from the vicinity of the trees. In 1997 and 1998, all the above measurements and ratings were taken again and further pruning was done. Although Sphaeropsis tip dieback was present in all the pines and anthracnose and Botryosphaeria canker in all the dogwoods, other diseases were also present and contributed to branch death. Because the objective of the experiment was to determine whether pruning significantly improves the general health of trees in a landscape setting, regardless of the cause branch death, no effort was made to determine the cause of branch death in each case. Tukey’s Studentized Range test was applied to the data following analysis of variance (SAS 1985).

Results and Discussion

Branch disease severity (Table 1) was not significantly reduced by pruning the pines in plot #1 or the dogwoods. Pruning significantly reduced branch disease severity in pines in plot #2. Disease pressure was great at Scots pine plot #1 because a second row of large, severely affected pines not in the experiment had branches touching the experimental trees and the air circulation among all the trees was poor. It is likely that inoculum coming from neighboring trees offset the effect of removing inoculum sources from treated trees. Treatments should be continued in order to evaluate the effects of pruning over a longer term.

Pruning infected twigs and branches greatly enhanced the appearance of the pines and dogwoods, as indicated by the significant reduction in the dead branch index, a numerical value devised to quantify the visual effect of the presence of dead twigs and branches (Table 2). The labor involved in the first year’s pruning was intensive; several primary branches and the ends of primary branches were removed. Multiple cuts were necessary in order to avoid tearing bark and to ensure that the final cuts were in positions where callusing would occur. Initially, many secondary branches and twigs also had to be removed.

However, subsequent pruning was rapid, involving small twigs and branches. The improvement in aesthetic appearance justified the time and labor involved in comprehensive pruning.

Conclusion

Based on these experiments, we conclude that pruning significantly improves the appearance of Scots pines and dogwoods affected by branch dieback. However, it may require several years of pruning before overall disease severity is significantly reduced.