Experimental design and data.

In Feb. 2007, bare-root pecan trees were planted at the Noble Research Institute’s McMillan Research and Demonstration Farm near the community of Madill, OK (lat. 34°2′10.4238″N, long. 96°56′27.0378″W; elevation 236 m). Trees were sourced from a nursery as a compound tree (3-year-old roots and a 1-year-old scion) with a single trunk, planted in augured 8 × 18-inch holes. The single-trunk trees (N = 21) had diameters (± sd) of 136 ± 8 inches. The trees were grafted ‘Kanza’ pecans on ‘Apache’ rootstock, were planted at a spacing of 35 × 35 ft on Bastrop fine sandy loam soil (taxonomic class: fine-loamy, mixed, active, thermic Udic Paleustalfs). The soil was well drained, with a depth to the water table of more than 80 inches. Tree roots on all trees were trimmed to 18 inches using cutting shears to fit into the augured hole without twisting roots (McCraw and Smith, 1998; Smith and Johnson, 1981). Trees were assigned to one of three pruning treatments in a completely randomized design: 0%, 50%, and 75% pruning of above-ground height. After pruning, mean tree height was 136, 68, and 34 inches for the 0%, 50%, and 75% treatments, respectively. The 50% and 75% treatments each made use of seven trees with each individual tree serving as a replication. The 0% control treatment only used six trees because one tree died shortly after planting. A 6-ft vegetation-free strip was maintained using glyphosate at a rate of 1.5 qt/acre per year along the row of trees throughout the study. Trees were fertilized with zinc sulfate foliar sprays at a rate of 2 lb/acre every 2 weeks from budbreak through August. Each tree received 0.5 lb/tree of 46N–0P–0K (urea) in May of year 1, and 2 lb/tree 46N–0P–0K the following years. No diseases were present during the study; however, foliar-feeding insects, including pecan catocala (Catocala sp.), serpentine leafminer (Stigmella junglandifoliella), unicorn caterpillar (Schizura unicornis), pecan budmoth (Gretchena boliana), glassy-winged sharpshooter (Homaladiaca vitripennis), grasshopper (Caelifera), and june beetles (Phyllophaga sp.) were controlled using 0.75 lb/acre chlorpyrifos (Lorsban 4E; Dow AgroSciences, Indianapolis, IN) as needed throughout the study. Trees were irrigated with a solid-set sprinkler system with a riser located next to each tree receiving 0.5 acre-inch irrigation water per week, except when natural rainfall occurred.

One year after planting, and each subsequent year of the trial, the trunk diameter (millimeters) for each tree was measured at 2 ft above the soil line using a digital caliper (Traceable Products, Webster, TX). In addition, the number of shoots were counted and recorded, and the length of shoots measured. Measurements were recorded each year after leaf drop during tree dormancy between December and February each year. The growth of every shoot on the tree was measured each year. Total shoot growth was calculated and recorded for each tree in each year as the summation across all shoots of growth in shoot length.

Statistical analysis.

To determine whether pruning alters trunk development, a mixed model with treatment fixed effect (capturing the effect of pruning on trunk diameter) and replicate random effect (capturing the intrinsic differences between experimental subjects) was used. The trunk diameters were compared across the three treatment groups (0%, 50%, and 75% pruned). The MIXED Procedure in SAS (SAS/STAT version 14.1 in SAS version 9.4; SAS Institute, Cary, NC) was used to estimate the parameters of the model (Littell et al., 1996).

The MIXED procedure was used to analyze both the number and average length of shoots. Both models were assumed to be mixed with tree random effect and the fixed effect of the treatment. For determining the pruning effects on total shoot growth (number of shoots times average length of shoots) a similar mixed model was used.

Trunk diameter.

The mean diameter, the annual incremental growth, and the percent change in diameter for each treatment group are presented in Table 1. For measurements taken 1 year after establishment, the descriptive statistics show that the average trunk diameter was higher for the untreated group (25 mm) followed by the 75%-pruned group (22 mm), and the 50%-pruned group had the lowest diameter (20 mm).

Table 1.

Average trunk diameter and annual growth of young ‘Kanza’ pecan trees with respect to 0%, 50%, and 75% pruning intensity.

View Table

During the second year of observation (2009), the annual increment analysis showed that the group with 50% intensity of pruning had increased on average by 8.6 mm (44% change in diameter), whereas the control group and the 75% group both increased 6.8 mm. However, in terms of percentage change in diameter, the control group had the lowest change (30%).

During the final year of observation (2010), the results indicate that the 50% pruning group had an increment increase in diameter of 18.8 mm compared with 2009. The control had increased by 13.6 mm, whereas the 75% pruning group had increased by 13.5 mm. Overall, trees in the control group had increased by 20.4 mm, trees that received 50% pruning had increased by 27.5 mm, and those with the 75% pruning intensity had increased by 20.2 mm during the 3-year period following their establishment.

Although there were some variations in trunk diameter across treatment groups, the analysis of variance did not find enough evidence of a statistically significant effect of pruning on trunk diameter. The probability value of the Duncan test was greater than the threshold level of 0.05.

Number and length of shoots.

The analysis looked at the number of shoots added on average per plant and the average shoot growth across the three treatment groups at 1, 2, and 3 years after establishment, as well as the average growth per shoot. In the first 2 years, the unpruned trees had more shoots (Table 2). However, by year 3 the pruned trees had added more shoots than the control trees, leading to no difference in mean number of shoots at the end of the observation period (2010).

Table 2.

Number of shoots and growth per shoot of ‘Kanza’ pecan trees in relation to year of planting and pruning intensity.

View Table

Average growth per shoot for the treated groups was significantly higher than for the control group. Trees that received 50% pruning at planting had an average limb growth 5.6 cm higher than the control trees. Those that were initially pruned at 75% had an average growth of 11 cm higher than the control group and 5 cm more than the 50% treatment group.

These findings were supported by previous work by Smith and Johnson (1981) who found that pecan trees that received either 25% or 50% top pruning maintain high shoot growth rates compared with trees that did not receive any pruning.

Shoup et al. (1980) found similar results for 5-year-old pecan trees. Essentially, their study suggested that 15% top pruning increases shoot growth over no pruning.

Total shoot growth.

Although the total shoot growth (Table 3) was higher for the pruned trees, the differences were not statistically significant (P > 0.05). Wood (1996) found that pruned trees eventually grew more than unpruned trees so that there was no difference in total shoot length growth after 3 years. Note that although statistical significance was not found, Bayesian analysis with a prior distribution that implies pruning leads to more growth would yield a posterior distribution quite similar to the prior distribution. Thus, the results are consistent with the current practice of pruning trees at the time of transplanting.

Table 3.

Total shoot growth of ‘Kanza’ pecan trees in response to 0%, 50%, and 75% pruning of scion at planting.

View Table